(Biofuels and Biorefineries 2) Zhen Fang, Chunbao (Charles) Xu (Eds.) - Near-Critical and Supercritical Water and Their Applications For Biorefineries-Springer Netherlands (2014)

Biofuels and Biorefineries 2
Zhen Fang
Chunbao (Charles) Xu Editors
Near-critical and
Supercritical
Water and Their
Applications for
Biorefineries
Biofuels and Biorefineries
Volume 2
Editors-in-Chief:
Professor Zhen Fang, Chinese Academy of Sciences, Kunming, China
Editorial Board Members:
Professor Liang-shih Fan, Ohio State University, USA;

Professor John R. Grace, University of British Columbia, Canada;
Professor Yonghao Ni, University of New Brunswick, Canada;
Professor Norman R. Scott, Cornell University, USA;
Professor Richard L. Smith, Jr., Tohoku University, Japan
For further volumes:

http://www.springer.com/series/11687
Aims and Scope of the series
Book Series in Biofuels and Biorefineries aims at being a powerful and integrative
source of information on biomass, bioenergy, biofuels, bioproducts and biorefinery.
It represents leading global research advances and opinions on converting biomass
to biofuels and chemicals; presents critical evidence to further explain the scientific
and engineering problems in biomass production and conversion; and presents
the technological advances and approaches for creating a new bio-economy and
building a clean and sustainable society to industrialists and policy-makers.
Book Series in Biofuels and Biorefineries provides the readers with clear and
concisely-written chapters on significant topics in biomass production, biofuels,
bioproducts, chemicals, catalysts, energy policy and processing technologies. The
text covers areas of plant science, green chemistry, economy, biotechnology,
microbiology, chemical engineering, mechanical engineering and energy studies.
Series description
Annual global biomass production is about 220 billion dry tons or 4,500 EJ,
equivalent to 8.5 times the world’s energy consumption in 2008 (532 EJ). On
the other hand, the world’s proven oil reserves at the end of 2011 amounted
to 1652.6 billion barrels, which can only meet 54.2 years of global production.
Therefore, alternative resources are needed to both supplement and replace fossil
oils as the raw material for transportation fuels, chemicals and materials in
petroleum-based industries. Renewable biomass is a likely candidate, because it
is prevalent throughout the world and can readily be converted to other products.
Compared with coal, the advantages of using biomass are: (i) it is carbon-neutral and
sustainable when properly managed; (ii) it is hydrolysable and can be converted
by biological conversion (e.g., biogas, ethanol); (iii) it can be used to produce
bio-oil with high yield (up to 75%) by fast pyrolysis because it contains highly
volatile compounds or oxygen; (iv) biofuel is clean because it contains little sulfur
and its residues are recyclable; (v) it is evenly distributed geographically and can
be grown close to where it is used, and (vi) it can create jobs in growing energy
crops and building conversion plants. Many researchers, governments, research
institutions and industries are developing projects to convert biomass (including
forest woody and herbaceous biomass) into chemicals, biofuels and materials
and the race is on to create new “biorefinery” processes. The development
of biorefineries will create remarkable opportunities for the forestry sector,
biotechnology, materials and the chemical processing industry, and it will stimulate
advances in agriculture. It will help to create a sustainable society and industry
based on renewable and carbon-neutral resources.
Zhen Fang • Chunbao (Charles) Xu
Editors
Near-critical and
Supercritical Water
and Their Applications
for Biorefineries
123
Editors
Zhen Fang Chunbao (Charles) Xu
Chinese Academy of Sciences Western University
Xishuangbanna Tropical Botanical Garden London, Canada
Kunming, China
ISSN 2214-1537 ISSN 2214-1545 (electronic)

ISBN 978-94-017-8922-6 ISBN 978-94-017-8923-3 (eBook)
DOI 10.1007/978-94-017-8923-3
Springer Dordrecht Heidelberg New York London
Library of Congress Control Number: 2014941958
© Springer Science+Business Media Dordrecht 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered
and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of
this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publisher’s location, in its current version, and permission for use must always be obtained from Springer.
Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations
are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

Preface
Supercritical water (SCW) is water above its critical point (CP; 374 ı C and
22.1 MPa). SCW or near critical water (NCW) is a unique green solvent with
reduced polarity and a lower dielectric constant (") value, which provides it
high solubility for many weakly-polar organics and light inorganic gases such as
hydrogen and oxygen. NCW or SCW exhibits very high reactivity for conversion of
biomass materials due to its greatly increased value of ion product ([HC ][OH ]),
which could efficiently hydrolyze biomass, and hence promote many biomass
conversion processes such as hydrolysis, liquefaction, carbonization, oxidation
and gasification. Thus, NCW and SCW have great promise for biorefineries and
conversion of biomass to produce biofuels and bio-based chemicals.
Although there are many books on the topic of either biomass conversion or
SCW, the unique feature of this book is that it links biomass conversion with
SCW/NCW and reactor design. This book is the second book of the series entitled
Biofuels and Biorefineries.
This book consists of 17 chapters contributed by leading world-experts on
biomass conversion with SCW/NCW. Each chapter was subjected to peer-review
and carefully revised by the authors and editors so that the quality of the material
could be improved. The chapters are arranged in three parts:
Part I: Fundamentals of Supercritical Water (Chaps. 1, 2, 3, and 4)
Part II: Reactor Design (Chaps. 5, 6, 7, and 8)
Part III: Near-critical and Supercritical Water Applications (Chaps. 9, 10, 11, 12,
13, 14, 15, 16, and 17)
Chapter 1 introduces the fundamentals of hydrogen bonding in SCW and its
relation to ion product and thermodynamic data of SCW for biorefining. Chapter
2 develops a local mapping concept to describe thermodynamically consistently the
saturation curve of water and biomass components, and gives results of calculations
of phase equilibria and critical curves for some main biomass components in
SCW. Chapter 3 mainly provides basic characteristics of co-solvents in sub- and
supercritical water and analyzes the effect of co-solvents on reactions to provide
v
vi Preface
readers with comprehensive information of co-solvents interactions. Chapter 4 uses

thermodynamic equilibrium modeling of SCW gasification (SCWG), and shows
how to build a process model based on Gibbs energy minimization. Chapter
5 introduces unique flow-through optical reactors designed to allow microscopic
observations of chemical processes of biomass in sub- and supercritical water.
Chapter 6 describes a fused silica capillary reactor, which allows in situ optical
observations of the sample in NCW using a microscope and spectroscopic tools
(e.g., Raman spectroscopy) for kinetic studies. Chapter 7 focuses on the use of
different types of reactors for SCW oxidation (SCWO) processes. Chapter 8 aims
to discuss the effects of the reactor wall properties, operating parameters on SCWG
of real wet biomasses and the associated operating challenges. Chapter 9 discusses
sub- and supercritical water treatment of biomass feedstocks for hydrochars, and
presents systematic characterization of fuel properties of the resultant chars. Chap-
ter 10 discusses the effects of operational parameters and operational problems
such as corrosion, salt deposition and carbonization in the treatment of organic
wastes with SCWO processes. Chapter 11 provides an overview on the yield
of syngas and hydrogen from SCWG of biomass, and discusses the influence
of different types of biomass, key process conditions as well as homogeneous
and heterogeneous catalysts on gasification. Chapter 12 introduces methane gas
production via SCWG of biomass, and focuses on the effects of operating conditions
and types of catalysts on the gasification efficiency and methane yield. Chapter
13 presents a comprehensive review on catalysis for SCWG, and summarizes the
development status and role of different homogeneous and heterogeneous catalysts.
Chapter 14 deals with hydrothermal bio-crude production, chemical reaction
pathways and upgrading pathways of bio-crude components while focusing on
hydrodeoxygenation reactions. Chapter 15 demonstrates the generation of energy
from bio-fuels using SCWO processes, and analyzes energy recovery studies of
SCWO processes. Chapter 16 reviews the production of chemicals in terms of feed-
stocks (biomass, plastics, inorganics and wastewaters) and reactions (gasification,
oxidation, depolymerization, precipitation and hydrothermal synthesis). Chapter
17 presents techno-economic analyses of SCWG processes for producing hydrogen
from glucose and sewage sludge.
This book covers a wide range of scientific and technical aspects of SCW/NCW
and the reactor design techniques necessary for efficient conversion of biomass
resources to bioenergy, bio-fuels and bio-based chemicals. The text is of interest to
students, researchers, academicians and industrialists in the areas of hydrothermal
processing of biomass, thermo-chemical conversion of biomass, and bioenergy and
bioproduct development.
Kunming, China Zhen Fang

London, Canada Chunbao (Charles) Xu
Acknowledgements
First and foremost, we would like to cordially thank all the contributing authors
for their great efforts in writing the chapters and insuring the reliability of the
information given in their chapters. Their contributions have really made this project
realizable.
Apart from the efforts of authors, we would also like to acknowledge the indi-
viduals listed below for carefully reading the book chapters and giving constructive
comments that significantly improved the quality of many aspects of the chapters:
Dr. Mesut Akgün, Yıldız Technical University, Turkey;
Dr. Ayten Ates, Cumhuriyet University, Turkey;
Prof. William Akers Bassett, Cornell University, USA;
Prof. Olivier Boutin, Aix Marseille University, France;
Dr. Satinder Kaur Brar, Institut national de la recherche scientifique, Canada;
Dr. Anand G Chakinala, University of Twente, the Netherlands;
Dr. Stefano Chiaberge, Eni Donegani Institute, Italy;
Prof. I-Ming Chou, Chinese Academy of Sciences, China;
Ms. Annamaria Croce, University of L’Aquila, Italy;
Dr. Miet Van Dael, Hasselt University, Belgium.
Prof. Marc Deshusses, Duke University, USA;
Prof. Tiziana Fornari, University Autónoma de Madrid, Spain;
Mr. Antonio CD Freitas, University Estadual de Campinas, Brazil;
Dr. Pierre Gallezot, CNRS/University de Lyon, France;
Prof. Jinlong Gong, Tianjin University, China;
Dr. Raed Hashaikeh, Masdar Institute, United Arab Emirates;
Dr. Attila R Imre, HAS Centre for Energy Research, Hungary;
Dr. Mª Belen Garcia Jarana, University de Cádiz, Spain;
Dr. Andrey G Kalinichev, Ecole des Mines de Nantes, France;
Prof. Birgit Kamm, Research Institute Biopos, Germany;
vii
viii Acknowledgements
Dr. Selhan Karagöz, Karabuk University, Turkey;

Dr. Shinji Kudo, Kyushu University, Japan;
Dr Witold Kwapinski, University of Limerick, Ireland;
Dr. Irene Leonardis, University of L’Aquila, Italy;
Dr. Youjun Lu, Xi’an Jiaotong University, China;
Dr. Tülay Güngören Madenoğlu, Ege University, Turkey;
Prof. Yizhak Marcus, the Hebrew University of Jerusalem, Israel;
Prof. Yukihiko Matsumura, Hiroshima University, Japan;
Dr. Violeta Vadillo Márquez, University de Cádiz, Spain;
Prof. Juan Ramon Portela Miguelez, University de Cádiz, Spain;
Prof. Lucía García Nieto, University de Zaragoza, Spain;
Dr. Jude Onwudili, University of Leeds, UK;
Prof. Francisco Javier Gutiérrez Ortiz, University de Sevilla, Spain;
Prof. Zhiyan Pan, Zhejiang University of Technology, China;
Dr. Francesco Picchioni, University of Groningen, the Netherlands;
Dr. Fernando Resende, University of Washington, USA;
Dr. Yohan Richardson, International Institute for Water and Environmental
Engineering, Burkina Faso;
Prof. María Dolores Bermejo Roda, University de Valladolid, Spain;
Prof. Frederik Ronsse, Ghent University, Belgium;
Dr. Takafumi Sato, Utsunomiya University, Japan;
Dr. Osamu Sawai, University of Tokyo;
Dr. Amr Sobhy, Heliopolis University for Sustainable Development, Egypt;
Dr. Alan K Soper, STFC Rutherford Appleton Laboratory, UK;
Dr. Zhongchao Tan, University of Waterloo, Canada;
Prof. Shuzhong Wang, Xi’an Jiaotong University, China;
Dr. Masaru Watanabe, Tohoku University, Japan;
Prof. Bert Weckhuysen, Utrecht University, the Netherlands;
Dr. Chunfei Wu, University of Leeds, UK;
Dr. Yulong Wu, Tsinghua University, China;
Dr. Zhirong Xu, Hohai University, China;
Mr. Onursal Yakaboylu, Delft University of Technology, the Netherlands;
Dr. Sudong Yin, IWR Technologies, Canada;
Dr. Linghong Zhang, Queen’s University, Canada;
Dr. Shicheng Zhang, Fudan University, China;
We are also grateful to Ms. Becky Zhao (senior editor) and Ms. Abbey Huang
(editorial assistant) for their encouragement, assistance and guidance during prepa-
ration of the book.
Acknowledgements ix
Finally, we would like to express our deepest gratitude towards our families for
their love, understanding and encouragement, which helped us in completion of this
project.
Zhen Fang, January 28, 2014 in Kunming, China (Zhen Fang)
Chunbao (Charles) Xu, January 28,

2014 in London, Canada (Chunbao (Charles) Xu)
Contents
Part I Fundamentals of Supercritical Water
1 Hydrogen Bonding in Supercritical Water . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 3

Yizhak Marcus
2 Phase Behavior of Biomass Components in Supercritical Water . . . . . 41
Sergey Artemenko, Victor Mazur, and Pieter Krijgsman
3 Role of Co-solvents in Biomass Conversion Reactions
Using Sub/Supercritical Water .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 69
Yulong Wu, Yu Chen, and Kejing Wu
4 Thermodynamic Analysis of the Supercritical Water
Gasification of Biomass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 99
Luca Fiori and Daniele Castello
Part II Reactor Design
5 Optical Reactors for Microscopic Visualization

of Chemical Processes in Sub- and Supercritical Water . . . . . . . . . . . . . . . 133
Shigeru Deguchi and Sada-atsu Mukai
6 Fused Silica Capillary Reactor and Its Applications . . . . . . . . . . . . . . . . . . . 157
I-Ming Chou and Zhiyan Pan
7 Reactors for Supercritical Water Oxidation Processes .. . . . . . . . . . . . . . . . 179
Pablo Cabeza, Joao Paulo Silva Queiroz,
M. Dolores Bermejo, Angel Martín, Fidel Mato,
and M. José Cocero
8 Effects of Reactor Wall Properties, Operating Conditions
and Challenges for SCWG of Real Wet Biomass. . . .. . . . . . . . . . . . . . . . . . . . 207
Mohammad S.H.K. Tushar, Animesh Dutta,
and Chunbao (Charles) Xu
xi
xii Contents
Part III Near-Critical and Supercritical Water Applications
9 Production of Renewable Solid Fuel Hydrochar

from Waste Biomass by Sub- and Supercritical Water Treatment .. . . 231
Zhengang Liu, Rajasekhar Balasubramanian,
and S. Kent Hoekman
10 Supercritical Water Oxidation (SCWO) for Wastewater
Treatment .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 261
Mesut Akgün and Onur Ömer Söğüt
11 Production of Hydrogen from Biomass via Supercritical
Water Gasification .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 299
Jude A. Onwudili and Paul T. Williams
12 Production of CH4 from Biomass via Supercritical Water
Gasification .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 323
Izad Behnia, Zhongshun Yuan, Paul Charpentier,
13 Catalysis in Supercritical Water Gasification of Biomass:
Status and Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 343
Youjun Lu, Sha Li, and Liejin Guo
14 Hydrothermal Conversion in Near-Critical Water – A
Sustainable Way of Producing Renewable Fuels . . . .. . . . . . . . . . . . . . . . . . . . 373
Jessica Hoffmann, Thomas H. Pedersen,
and Lasse A. Rosendahl
15 Supercritical Water Oxidation (SCWO) of Solid, Liquid
and Gaseous Fuels for Energy Generation . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 401
M. Dolores Bermejo, Ángel Martín,
Joao Paulo Silva Queiroz, Pablo Cabeza, Fidel Mato,
and M. José Cocero
16 Production of Chemicals in Supercritical Water .. . .. . . . . . . . . . . . . . . . . . . . 427
Yukihiko Matsumura and Tau Len-Kelly Yong
17 Techno-economic Analysis of Renewable Hydrogen
Production via SCWG of Biomass Using Glucose as
a Model Compound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 445
Dawood Al-Mosuli, Shahzad Barghi, Zhen Fang,
Editors’ Biography .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 473

Contributors
Mesut Akgün SCFT—Supercritical Fluid Technologies Research Group,

Chemical Engineering Department, Yıldız Technical University, Istanbul, Turkey
Sergey Artemenko Institute of Refrigeration, Cryotechnologies, and Eco-Power
Engineering (Former Academy of Refrigeration), Odessa National Academy of
Food Technologies, Odessa, Ukraine
Rajasekhar Balasubramanian Department of Civil and Environmental Engineer-
ing, National University of Singapore, Singapore, Singapore
Shahzad Barghi Department of Chemical and Biochemical Engineering, Western
University, London, ON, Canada
Izad Behnia Institute for Chemicals and Fuels from Alternative Resources,
Department of Chemical and Biochemical Engineering, Western University,
London, ON, Canada
M. Dolores Bermejo High Pressure Process Group, Department of Chemical
Engineering and Environmental Technology, University of Valladolid, Valladolid,
Spain
Pablo Cabeza High Pressure Process Group, Department of Chemical Engineering
and Environmental Technology, University of Valladolid, Valladolid, Spain
Daniele Castello Department of Civil, Environmental and Mechanical Engineer-
ing, University of Trento, Trento, Italy
Paul Charpentier Institute for Chemicals and Fuels from Alternative Resources,
London, ON, Canada
Yu Chen Institute of Nuclear and New Energy Technology, Tsinghua University,
Beijing, People’s Republic of China
xiii
xiv Contributors
I-Ming Chou Laboratory for Experimental Study Under Deep-Sea Extreme Con-
ditions, Sanya Institute of Deep-Sea Science and Engineering, Chinese Academy of
Sciences, Sanya, People’s Republic of China
M. José Cocero High Pressure Process Group, Department of Chemical
Spain
Shigeru Deguchi R&D Center for Marine Biosciences, Japan Agency for Marine-
Earth Science and Technology (JAMSTEC), Yokosuka, Japan
Animesh Dutta Mechanical Engineering Program, School of Engineering,
University of Guelph, Guelph, ON, Canada
Zhen Fang Biomass Group, Key Laboratory of Tropical Plant Resource and
Sustainable Use, Xishuangbanna Tropical Botanical Garden, Chinese Academy of
Sciences, Kunming, Yunnan, China
Luca Fiori Department of Civil, Environmental and Mechanical Engineering,
University of Trento, Trento, Italy
Liejin Guo State Key Laboratory of Multiphase Flow in Power Engineering
(SKLMFPE), Xi’an Jiaotong University, Xi’an, Shaanxi, China
S. Kent Hoekman Division of Atmospheric Sciences, Desert Research Institute,
Reno, NV, USA
Jessica Hoffmann Department of Energy Technology, Aalborg University,
Aalborg Ø, Denmark
Pieter Krijgsman Ceramic Oxides International, Wapenveld, The Netherlands
Sha Li State Key Laboratory of Multiphase Flow in Power Engineering
Zhengang Liu Department of Civil and Environmental Engineering, National
University of Singapore, Singapore, Singapore
Youjun Lu State Key Laboratory of Multiphase Flow in Power Engineering
Yizhak Marcus Institute of Chemistry, The Hebrew University of Jerusalem,
Jerusalem, Israel
Ángel Martín High Pressure Process Group, Department of Chemical Engineering
Fidel Mato High Pressure Process Group, Department of Chemical Engineering
Yukihiko Matsumura Division of Energy and Environmental Engineering,
Institute of Engineering, Hiroshima University, Higashi-Hiroshima, Japan
Contributors xv
Victor Mazur Institute of Refrigeration, Cryotechnologies, and Eco-Power

Engineering (Former Academy of Refrigeration), Odessa National Academy of
Food Technologies, Odessa, Ukraine
Dawood Al-Mosuli Department of Chemical and Biochemical Engineering,
Western University, London, ON, Canada
Sada-atsu Mukai R&D Center for Marine Biosciences, Japan Agency for Marine-
Earth Science and Technology (JAMSTEC), Yokosuka, Japan
Akiyoshi Bio-Nanotransporter Project, JST, ERATO, Akiyoshi Bio-
Nanotransporter Project, Katsura Int’tec Center, Nishikyou-ku, Kyoto, Japan
Department of Polymer Chemistry, Graduate School of Engineering, Kyoto
University, Nishikyo-ku, Kyoto, Japan
Jude A. Onwudili Energy Research Institute, Faculty of Engineering, University
of Leeds, Leeds, UK
Zhiyan Pan Department of Environmental Engineering, Zhejiang University of
Technology, Hangzhou, People’s Republic of China
Thomas H. Pedersen Department of Energy Technology, Aalborg University,
Aalborg Ø, Denmark
Joao Paulo Silva Queiroz High Pressure Process Group, Department of Chemical
Spain
Lasse A. Rosendahl Department of Energy Technology, Aalborg University,
Aalborg Ø, Denmark
Onur Ömer Söğüt SCFT—Supercritical Fluid Technologies Research Group,
Chemical Engineering Department, Yıldız Technical University, Istanbul, Turkey
Mohammad S.H.K. Tushar Mechanical Engineering Program, School of
Engineering, University of Guelph, Guelph, ON, Canada
Paul T. Williams Energy Research Institute, Faculty of Engineering, University of
Leeds, Leeds, UK
Kejing Wu Institute of Nuclear and New Energy Technology, Tsinghua University,
Yulong Wu Institute of Nuclear and New Energy Technology, Tsinghua University,
Chunbao (Charles) Xu Institute for Chemicals and Fuels from Alternative
Resources, Department of Chemical and Biochemical Engineering, Western
University, London, ON, Canada
xvi Contributors
Tau Len-Kelly Yong Department of Mechanical Science and Engineering,

Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, Japan
Zhongshun Yuan Institute for Chemicals and Fuels from Alternative Resources,
London, ON, Canada
Part I
Fundamentals of Supercritical Water
Chapter 1
Hydrogen Bonding in Supercritical Water
Yizhak Marcus
Abstract Hydrogen bonding (HB) in supercritical water (SCW) is much less

extensive than in water at ambient conditions. Still, it plays an important role in
the structural and dynamic properties of SCW and its capacity as a solvent. In
order to deal with the HB it is necessary to have definite criteria that specify
when a hydrogen bond exists or not, and these are provided. The extent of HB
in SCW is expressed by means of the mean number of hydrogen bonds per water
molecule, the fractions of water molecules with 0, 1, 2 : : : hydrogen bonds, and the
percolation limit determining regions of continuous HB. The HB in SCW deduced
from diffraction, spectroscopy, and computer simulations is described. The ionic
dissociation of SCW itself is more extensive at low temperatures and high pressures
than of ambient water. The solvent power of SCW regarding organic solutes is
described by its solubility parameters, but solvatochromic probes have found little
use in SCW. The solubility of salts in SCW and the hydration and pairing of their
ions are discussed. HB as well as its related ion product and thermodynamic data
of water plays an important role in biomass dissolution and reactions in SCW.
The appendix contains tables of physical properties of SCW at temperatures and
pressures relevant to biorefining.
Keywords Hydrogen bonding • Criteria for hydrogen bonds • Hydrogen bonds

per water molecule • Percolation limit • X-ray and neutron diffraction • IR and
NMR spectroscopy • Computer simulations • Ionic dissociation of SCW • Solu-
bility parameters • Salt solubilities
Y. Marcus ()
Institute of Chemistry, The Hebrew University of Jerusalem, Jerusalem 91904, Israel
e-mail: [email protected]
Z. Fang and C. Xu (eds.), Near-critical and Supercritical Water and Their Applications 3
for Biorefineries, Biofuels and Biorefineries 2, DOI 10.1007/978-94-017-8923-3__1,
© Springer ScienceCBusiness Media Dordrecht 2014
4 Y. Marcus
1.1 Introduction
The properties of supercritical water (SCW) depend on its structure and dynamics,
and these, in turn, depend on the hydrogen bonding between the water molecules.
The structure is governed also by the repulsion between neighboring molecules at
short distances and by dispersion and dipole interactions. The dynamics are mainly
expressed in terms of the orientation relaxation times of water molecules and the
life-times of the hydrogen bonds. Experimental methods, such as diffraction of
x-rays and neutrons for the structure and Nuclear Magnetic Resonance (NMR),
dielectric relaxation, and vibrational spectroscopies for the dynamics, yield the
required information. These are augmented by Monte Carlo (MC) and molecular
dynamics (MD) computer simulations and theoretical studies.
The structure of SCW can be described as the variety of preferential molecular
arrangements due to (thermal) fluctuations in the nearest environment of a given
water molecule [1]. There is abundant evidence that hydrogen bonds do exist in
SCW, but that the tetrahedral hydrogen bonded network present in ambient water is
no longer present in SCW. The extent of hydrogen bonding in SCW increases with
increasing densities but diminishes with increasing temperatures. The experimental
methods mentioned above and computer simulations provide information on the
extent of this hydrogen bonding in SCW, whether it consists of dimers only or
whether larger clusters also exist.
A water molecule may participate in up to four hydrogen bonds (HBs): two HBs
as an acceptor of HBs to its two oxygen lone pairs of electrons and two HBs as a
donor of its own hydrogen atoms. The term “the mean number of hydrogen bonds
per water molecule” must, however, be carefully defined. There are necessarily
two water molecules involved in each hydrogen bond, so that the mean number
of hydrogen bonds per water molecule in the system, symbolized by < nHB >, is
one half of the number of hydrogen bonds water molecules have on the average,
symbolized by :
< nHB >D =2: (1.1)
Ordinary ice has a regular tetrahedral hydrogen bond network with D 4, so

for ice < nHB > D 2. On melting of the ice this number diminishes and for liquid
water at ambient conditions (25 ı C and 0.1 MPa) < nHB > D 1.73, as adopted by
Hoffmann and Conradi [2] in their NMR study and by Kohl et al. [3] from their
Raman results. On heating of liquid water < nHB > decreases further and as SCW
is approached it may have quite small values that depend on the temperature and
pressure, hence on the density, .
The fraction of water molecules in the system with i D 0, 1, 2, 3, and four HBs
is denoted by fi . The value of f4 in SCW is negligible (0.02) at even the lowest
temperature and highest density. On the other hand, the fraction f0 of monomeric,
that is, non-hydrogen-bonded, water is a quantity of interest concerning the extent
of hydrogen bonding in SCW. For the purpose of the elucidation of such quantities
a precise definition of the criteria for an intact hydrogen bond to exist or for its
complete breakage is required.
1 Hydrogen Bonding in Supercritical Water 5
The percolation threshold is another quantity that describes the hydrogen

bonding in SCW. It is the thermodynamic state (T,) beyond which no con-
nected hydrogen-bonded chains, but only small clusters (down to dimers) of water
molecules, exist in SCW. The criterion for the existence of percolation D 1.55 has
been specified [4] and employed in some studies of SCW.
Physicochemical properties of SCW at temperatures and pressures that are
relevant to the subject of this book are shown in the appendix of this chapter [5].
1.2 Criteria for Hydrogen Bonds to Exist
MD computer simulations of water at ambient conditions and ab initio theoretical

computations lead to the criteria for the existence of a hydrogen bond between two
water molecules, see Fig. 1.1.
The criteria are geometric and energetic [6], as well as temporal [7], as follows:
(i) the distance between two neighboring oxygen atoms has to be d(O–O)
0.330 nm, and
(ii) the hydrogen bond distance has to be d(O H) 0.240 nm, and
(iii) the angle of the O–H O configuration has to be 130ı, and
(iv) the interaction energy eHB between the hydrogen bonded water molecules
should be more negative than 12.9 kJ mol–1 , and
(v) the occupancy of the ¢ * orbitals of the acceptor molecule has to be 0.0085,
and
(vi) the HB persists over at least 0.1 ps.
Criteria (i), (ii), and (iii) place geometrical constraints on the water molecules
between which a hydrogen bond may exist. The values of d(O–O) and d(O H)
are obtained from the partial pair correlation functions g(O–O, r) and g(O H, r)
defined in Sect. 1.4 below. Criterion (iii) permits considerable bending of the
hydrogen bond from linearity before it is regarded as broken. Criterion (iv)
is energetic, and some investigators allow a less negative binding energy, as
low as 10 kJmol–1 . The ¢ OH * anti-bonding molecular orbital of an acceptor
water molecule is empty in monomeric (non-hydrogen-bonded) water, hence its
Fig. 1.1 Schematic

representation of a hydrogen
bonded pair of water
molecules (Reprinted with
permission from [5],
Copyright © 2002, Wiley)
6 Y. Marcus
occupancy, once HBs are formed in liquid water, is the criterion (v) to be used [6]. It
is necessary to distinguish between spurious breaking and spurious making of HBs
and genuine breaking and making of such bonds, lasting for a time considerably
longer than a single oscillation [7]. Spurious breaking of HBs are due to stretching
vibrations leading to instantaneous, very short time (10–20 fs) excursions of the
HB length beyond its dO H cut-off criterion. Spurious making of HBs is caused by
instantaneous collisions of water molecules to less than the dO–O cut-off criterion.
This consideration leads to criterion (vi) for the existence of an HB between two
adjacent water molecules in SCW.
If the criteria for the existence of a hydrogen bond are somewhat relaxed and
more lenient requirements are used, naturally the extent of hydrogen bonding in
water in general and in SCW in particular is expected to increase. Some such
results dealt with below indeed use different criteria as noted, with the expected
consequences.
1.3 The Extent of Hydrogen Bonding in SCW
The extent of hydrogen bonding in SCW can be described in terms of several

quantities. These are the average numbers of hydrogen bonds per any water
molecule, (a maximum of 4) or the average numbers of hydrogen bonds per
water molecules in the system, <nHB > (a maximum of 2). A more detailed view
is provided by the fractions fi of water molecules with i D 0, 1, 2, 3, and four
hydrogen bonds. A further indication of the extent of hydrogen bonding in SCW
is the sizes of clusters of hydrogen bonded water molecules as is the position (T,)
among the thermodynamic states of the percolation threshold, beyond which no
connected hydrogen-bonded chains exist anymore.
Several scaling equations of state (EoSs) of SCW permit the deduction of the
fraction f0 of monomeric, non-hydrogen-bonded, water molecules in SCW. Gupta
et al. [8] derived am expression, based on the lattice fluid hydrogen bonding model
(LFHB), which relates f0 to the temperature, the pressure, and the density by means
of five specified parameters. The chemical potential expression resulting from the
EoS yields the fraction f0 of monomeric water (not hydrogen bonded at all) as:
h 1=2 i.
f0 D A2 C 8A A 4 (1.2)
Here

A D rVr00 exp E ı TSı C PV ı =RT (1.3)
r is the segment length, a dimensionless quantity, Vr 00 is the reduced

specific volume, 1/00r D */ (with * D 853 kgm–3) and Eı D 15.5 kJmol–1 ,
Sı D 16.6 JK–1 mol–1 , and V ı D 4.2 cm3 mol–1 are the hydrogen bonding
Table 1.1 The fraction of water molecules in SCW present as monomers (non-hydrogen-bonded),
f0 , obtained from various equations of state and PVT data
t/ ı C /kgm–3 f0 from [8] f0 from [9] f0 from [11] f0 from [12]
P D 50 MPa
374 637 0.17 0.03
400 582 0.60 0.19 0.08 0.33
450 404 0.67 0.28 0.28 0.62
500 247 0.75 0.42 0.51 0.73
550 186 0.82 0.52 0.63 0.80
600 151 0.85 0.58 0.72 0.84
P D 100 MPa
374 721 0.14 0.00
400 688 0.56 0.17 0.01 0.04
450 612 0.61 0.21 0.13 0.21
500 521 0.66 0.26 0.27 0.37
550 430 0.70 0.32 0.39 0.47
600 350 0.74 0.38 0.52 0.55
parameters. Values of f0 are shown in a figure in [8] for three temperatures, 400,
600 and 800 ı C and 0 P/MPa 100.
Smits et al. [9] preferred the three-HB-site associated perturbed anisotropic
chain theory (APACT). The theory requires four parameters: a characteristic
energy, T * D 183.8 K and a characteristic volume v* D 11.64 cm3 mol–1 , as well
as the enthalpy H ı D 20.10 kJmol–1 and entropy Sı D 10.97 JK–1 mol–1 of
hydrogen bonding. The latter yield the equilibrium constant for hydrogen bonding
D exp(H ı/RT C Sı /R). The fraction of monomeric water molecules is:
. n o1=2
f0 D 2 1 C 4 ./ C .1 C / 1 C 6 C ./
2 2
(1.4)
Smits et al. [9] compared values of f0 from their approach with values from the
earlier LFHB approach and the statistical associated fluid theory (SAFT) of Huang
and Radosz [10], finding considerable differences.
Vlachou et al. [11] developed the LHFB approach [8] and used different values
for the hydrogen bonding energy and entropy parameters and a temperature- and
pressure-dependent volume of hydrogen bonding. They obtained a still different set
of the fraction of monomeric water f0 .
Marcus [12] derived f0 values from the PVT data (see below) that tend to agree
with those of the LFHB model of Gupta et al. [8] at the lower pressure and
densities at temperatures 450 ı C and at the higher pressure more nearly with
those of Vlachou et al. [11]. Table 1.1 shows the values of f0 of SCW obtained
according to these models at two characteristic pressures, P D 50 and 100 MPa
for several temperatures. Although considerable differences in the estimated values
are apparent, all agree that f0 diminishes with lowered temperatures and increasing
8 Y. Marcus
Table 1.2 The average number of hydrogen bonds per water molecule in the system, <nHB >,
according to different methods and authors from the references in column 1
/ kgm–3 150 250 350 450 600 750 1,000
400 ı C
[2] 0.18 0.27 0.40 0.47
[13] 0.35 0.56 0.67 0.72 0.72 0.75
[14] 0.35 0.52 0.71 0.88 1.08
[15] 0.64 1.24
[16] 0.44 0.81 1.26
[17] 0.60 1.00
500 ı C
[2] 0.11 0.22
[14] 0.25 0.39 0.61 0.81 1.01
[15] 0.48 1.07
[17] 0.53 0.93
600 ı C
[2] 0.09
[14] 0.97
[15] 0.39 0.97
[17] 0.48 0.87
pressure (and density). The higher values of f0 obtained by the LFHB [8] theory
appear to lead to more reliable values than those from the APACT [9] theory or the
modified LFHB theory [11]. The latter two approached appear to overestimate the
extent of hydrogen bonding in SCW, leading to too low values of f0 .
The number of hydrogen bonds per water molecule and < nHB > in SCW has
been reported by several authors, based on a variety of methods. The < nHB > for
SCW is obtained from the NMR chemical shifts ı against benzene internal standard
reported by Hoffmann and Conradi [2], yielding the hydrogen bonding relative to
ambient water, . The ı value taken for dilute water vapor, where < nHB > D 0,
was 6.6 ppm relative to dilute benzene. The ı for ambient water, –2.5 ppm,
was taken from other authors and other internal standards, but ı D 2.8 for
ambient water relative to internal benzene standard should have been used instead.
Then < nHB > D 1.73 D 2.785 C 0.391ı would result, where 1.73 D <nHB > for
ambient water [2]. This procedure yields < nHB > values commensurate with most
others, see Table 1.2.
The NMR study of Matubayasi et al. [18] at 400 ı C yielded < nHB > values
ranging from 0.6 to 1.5 for densities from 190 to 660 kgm–3 (and a d(O H)
cutoff of 0.23 nm). Soper et al. [19] reported the value < nHB > D 1.50 at 400 ı C
and 658 kgm–3 (80 MPa), from neutron diffraction with isotope substitution. These
values are much larger than values obtained by others.
Franck and Roth [13] reported the integrated areas below the O–D stretch
peak of dilute HOD in supercritical H2 O at 400 ı C. These were divided by the
corresponding area at 30 ı C to obtain the relative amount of hydrogen bonding, ,
and were multiplied by 1.73 to yield < nHB > shown in Table 1.2.
The values of < nHB > obtained by computer simulations: MC by Kalinichev and
Bass [20] and MD by Mountain [14] and by Yoshii et al. [21] agree well with each
other. They agree also with values deduced from the Infrared (IR) absorption data of
Franck and Roth [13] and with those calculated from the EoS by Gupta et al. [15].
It should be noted that the temperature quoted for MD simulations is an output, but
must be corrected in the case where the corresponding critical temperature differs
from the true value. The MD data of Yoshii et al. [21] were reported for the nominal
temperature T D 600 K, but the critical temperature emerging from the model used
was Tc D 561 K instead of the true value 647 K. Therefore, the reduced temperature
was Tr D T/Tc D 1.07, that corresponds to 430 ı C.
As expected, the values of < nHB > increase with increasing densities and dimin-
ish with increasing temperatures, the values of < nHB >, interpolated from the
reported data, are shown in Table 1.2.
Marcus [12] used the semi-empirical model of Lamanna et al. [22], recognizing
the existence of water molecules with zero, one and two hydrogen bonds:
corresponding to monomers, dimers, and trimers. He calculated values of < nHB > in
SCW primarily from the PVT data at 400–600 ı C (35–100 MPa, 200–700 kgm–3).
Species with three or four hydrogen bonds are subsumed into those of species with
two hydrogen bonds since they are very rare. The probability of hydrogen bond
formation between two water molecules, p, is given by [22]:
p D 1 exp ŒE.P /=R .T T0 / (1.5)
recognizing the pressure dependence of the energy parameter. The fractions fj are
given by:
h i
f2 D 6p 2 f0 =.1p/2 I f1 D4p f0 = .1p/ I f0 D1= 1C4p= .1p/ C6p 2 =.1p/2
(1.6)
the value of f0 being obtained by difference, because f0 C f1 C f2 D 1.

The molar volumes of the species with 0, 1, and two HBs (neglecting for now
species with more than two HBs) are temperature- and pressure-dependent as:
h i
Vi D Vi 0 .P / 1 C ˛i .P /.T T0 /2 .i D 0; 1; 2/ (1.7)
To limit the number of fitting parameters required, the molar volume of the trimer
(with two hydrogen bonds) was taken as 3 /2 that of the dimer (with one hydrogen
bond) and their expansibilities ˛ i (P) were taken to be equal. The molar volume of
the SCW is then a weighted sum of the volumes of the species:
V .T; P / D f0 V0 C f1 V1 C f2 V2 (1.8)
The fractions fi (T,P) depend only on T0 and the expression for E(P),
requiring three parameters, via the probabilities p according to Eqs. 1.5
10 Y. Marcus
Fig. 1.2 The average number of hydrogen bonds per water molecule in SCW, <nHB >, the ordinate
numbers being expanded by a factor of 103 (Reprinted with permission from [12], Copyright ©
(2000), the Royal Society of Chemistry)
and 1.6: T0 D 637 K and E(P)/Jmol–1 D 18 C 0.028(P/ MPa)2. In order to

fit the experimental PVT data to Eqs. 1.5, 1.6, 1.7, and 1.8, the following
additional parameters were then required: V00 /cm3 mol–1 D 27.1–0.27(P/ MPa),
V10 /cm3 mol–1 D 29.6–0.072(P/MPa), ˛ 0 /K–2 D 1.7 10–5 C 0.44(P/MPa)–2, and
˛ 1 /K–2 D ˛ 2 /K–2 D 0.20 10–5 C 0.55(P/MPa)–2. The fi (T,P) resulting from
fitting the PVT data can be presented in a three-dimensional grid as in Fig. 1.2.
The same fractions fi should be applicable to measurements other than the
extensive PVT data, from which they were derived [12]. An expression analo-
gous to Eq. 1.8 was used with these fi (T,P) to fit the constant volume heat
capacities, Cv . For monomeric water Cv0 D 21.79 C 0.0158(t/ıC) JK–1 mol–1 . For
SCW the Cv values are appreciably larger than that, due to the vibrations of the
hydrogen bonds and the larger moments of inertia for the rotation of the dimers
and trimers. Fairly good fits were obtained with the above Cv0 , and the values
for the dimer: Cv1 D 65.73 C 0.0350(t/ ıC) JK–1 mol–1 (obtained independently
from spectroscopy) and of the single fitting parameter for the linear trimer:
Cv2 D 2.3 107 (P/MPa)3 [12].
Self-diffusion coefficients could also be fitted with an expression analogous to
Eq. 1.8 and the fi (T,P) derived from the PVT data. Three fitting parameters were
required: D0 D 157, D1 D 11, and D2 D 49, all in 10–9 m2 s–1 . That the value of D2
is larger than that of D1 is attributed to more efficient diffusion of the trimer by
rotation. Fair agreement is also achieved [12] with the values of 1 – f0 deduced from
the NMR chemical shifts [2].
Kirkwood dipole orientation parameters g:

g D .9"0 kB =NA / M T = 2 ."r "1 / .2"r C "1 / ="r ."1 C 2/2 (1.9)
have been calculated for SCW from relative permittivity data "r [23]. The model and
the calculated fi (T,P) [12] were applied to these values in analogy with Eq. 1.8:
g D 1Cf0 N .T; P / cos < 0 > Cf1 N .T; P / cos < 1 > Cf2 N .T; P / cos < 2 >
(1.10)
where N(T,P) is the number of nearest neighbors and < > is the average angle
between the dipoles of neighboring water molecules. The average random angle
for non-hydrogen-bonded molecules is 90ı , hence cos < 0 > D 0 and the term
in f0 vanishes. Either free rotation around the HB in singly hydrogen bonded
water molecules or minimal oxygen lone pair repulsion was assumed, yielding
cos < 1 > between 0.49 and 0.39. Calculations were limited to temperatures and
pressures where f2 was negligible, because the estimation of the average angle
for the trimers with two hydrogen bonds was difficult. General agreement was
found [12] between 1 C f1 N(T,P)cos < 1 > and values of g calculated from the
permittivities, but the latter themselves cover a range of values depending on
uncertainties in the infinite frequency permittivity "1 in Eq. 1.9 [23].
The values of < nHB > D f1 C 2f2 obtained from the model discussed here [12]
are shown in Fig. 1.2 and those of f0 , the fraction of non-hydrogen-bonded water
molecules in SCW, are compared with values obtained by other authors shown in
Table 1.1.
The extent of hydrogen bonding in SCW may also be described in terms of
the percolation threshold. SCW in thermodynamic states at which 1.55, i.e.,
<nHB > 0.77, permits percolation, since then their hydrogen bonded clusters
have the required degree of connectivity. Jedlovszky et al. [24–26] discussed the
percolation probability in SCW, based on MC computer simulations. Initially they
concluded that the required connectivity for percolation was lost at the critical
point [24]. Subsequently they located the percolation threshold line in SCW more
precisely [25, 26], the results devolving on whether a geometric criterion for
the hydrogen bond (d(O–O) 0.35 nm and d(O H) 0.25 nm) or an energetic
criterion (eHB < 11.5 kJmol–1 ) were used. The criteria for percolation were that
50 % of the configurations in their simulations were of “infinite” clusters in at least
one dimension. The geometric criterion permitted percolation at 430 and 480 ı C,
and at 430 ı C the threshold appeared to be a density 410 kgm–3 [26].
Bernabei et al. [27, 28] made recently neutron diffraction measurements with
isotope substitution (NDIS: H2 O, equimolar H2 O C D2 O, and D2 O) and analyzed
them by means of the empirical potential structure refinement (EPSR) method, see
Sect. 1.4. They applied these to three states of SCW (400 ı C and 580 and 750 kgm–3
and 480 ı C and 750 kgm–3) and used the geometric criterion of d(O H) 0.24 nm
for the existence of a hydrogen bond. They concluded that the three liquid-like states
were above the threshold, with D 2.2 for the higher density states and D 1.9 for
12 Y. Marcus
the lower density one. Variation of the d(O H) cutoff by ˙0.1 nm did not change
the results. The resulting clusters have up to 900 molecules altogether, larger than
the percolation threshold of 350 molecules, and their linear dimensions involved
13 water molecules.
1.4 Hydrogen Bonding in SCW Deduced

from Diffraction studies
The arrangements of water molecules in SCW are described by the radial

distribution function, N(r). Its differential, dN, is the probability of finding a water
molecule in a spherical shell of thickness dr at a distance r from the center of a
given water molecule. At large distances dN is proportional to the number density
of the water molecules, , there being no interactions between the molecules so that
dN(r ! 1,dr) D 4 r2 dr. At short distances water molecules are correlated by
attraction and repulsion forces between them. Hence the pair correlation function
g(r) is defined by the conditional probability of finding a water molecule at a
distance r from another one:
dN .r; dr/ D g.r/4 r 2 dr (1.11)
As no correlation exists between water molecules at large distances from each

other g(r ! 1) D 1. On the other hand, at distances shorter than the diameter of
the water molecules the large repulsion of the electronic shells of the atoms prevent
their overlapping and g(r ) D 0. The number of neighboring molecules around a
water molecule at the origin is the coordination number Nco up to a given distance
and is obtained by integration of Eq. 1.11. The coordination numbers reaches a
plateau at r corresponding to the first coordination shell.
X-ray and neutron diffraction measurements yield the pair correlation functions
g(r) in SCW as follows. The structure factors S(k) are
S.k/ ŒI.k/ I.0/ = ŒI .1/ I.0/ (1.12)
where the I(k) are the intensities of the beams for a fixed wavelength of the
radiation diffracted from the sample at the angles using the defined variable k:
k D 1 4 sin .=2/ (1.13)
The structure factors obtained by both x-ray and neutron diffraction require
corrections for effects, such as incoherent and non-elastic scattering, that may
introduce uncertainties if not correctly applied. Finally, the pair correlation function
g(r) is obtained from the experimental structure factors after application of a Fourier
transform:
Z 1
1
g.r/ D 2 2 r 0 .S.k/ 1/ k sin .kr/ dk (1.14)
The diffraction of x-rays from SCW occurs mostly from the electrons of the
oxygen atoms: g(r) g(O–O, r). Gorbaty and Dem’yanets [29–31] studied water
by x-ray diffraction up to 500 ı C and 100 MPa. They concluded that some
hydrogen bonding persists in the supercritical range of temperatures. Gorbaty and
Kalinichev [32] calculated the quantity f4 , the fraction of tetrahedral hydrogen
bonding persisting in high temperature water, from the area under the 0.28 nm peak
in the g(r) function divided by the coordination number Nco that corresponds to the
sum of the areas under the peaks at 0.28 and 0.45 nm resolved as Gaussians. They
found that f4 decreased linearly from ambient to supercritical conditions:
f4 D 0:851 8:68 104 .T =K/ (1.15)
up to 500 ı C and densities between 700 and 1,100 kgm–3 with an accuracy of ˙0.1
units. This corresponds to f4 0.29 at the critical point. This interpretation of the
x-ray diffraction data is consistent with infrared absorption spectral data, but the
resulting f4 values are much larger than noted for SCW by other measurements.
However, the molecular pair correlation function g(r) of SCW does not provide
full information on its structure because it is integrated over all the orientations
of the molecules. More detained information is provided by the partial pair
correlation functions pertaining to specific pairs of atoms: g(O–O, r), g(O–H, r), and
g(H–H, r).
Neutrons are diffracted from the nuclei of both oxygen and hydrogen atoms, the
hydrogen atoms being exchangeable, at least partly, for deuterium ones. The total
pair correlation function on partial substitution of D atoms for H atoms in the water
molecules is made up from contributions of three partial correlation functions:
h i .h
g.r/ D bO 2 gOO C 4bO bD gOH.D/ C 4bD 2 gHH.DD/ .bO C 2bD /2 (1.16)
where bO and bD are the scattering lengths of the O and D nuclei, the latter being
prorated for the presence of H atoms. The gOO etc. are notations for the partial
pair correlation functions: g(O–O, r), g(O–H, r), and g(H–H, r). Hence, when
three distinct experiments at the same thermodynamic states are made where the
fractional substitution of H by D atoms is different, thereby changing the values
of bD , neutron diffraction is capable of determining the partial pair correlation
functions.
The most recent neutron diffraction studies of SCW [27, 28] resolved earlier
controversies arising from insufficient accuracy and corrections for extraneous
effects. The authors applied the isotope substitution method, using pure D2 O, pure
H2 O, and equimolar mixtures to maximize the variation of the bD coefficient in
Eq. 1.16. They also applied the state of the art treatment to the raw diffraction
data and in addition applied the empirical potential structure refinement (EPSR)
14 Y. Marcus
computer program to the analysis of the data. They provided new measurements
on SCW in three thermodynamic states: one gas-like (400 ı C, at 116 kgm–3) and
two liquid-like (density 750 kgm–3 and at 400 and 480 ı C) and reconsidered also
the experimental results at 400 ı C and a density of 580 kgm–3 [33]. The resulting
partial pair correlation functions of the liquid-like states confirm the presence of the
hydrogen bond signature peak at 0.19 nm in g(O–H, r) and an absence of a feature
at 0.45 nm in g(O–O, r). The EPSR data were interpreted in terms of the presence
of percolation at the liquid-like states, i.e., some hydrogen bonding network, and
its absence in the gas-like state [25]. The average number of hydrogen bonds of a
water molecule in the gas-like state is only 0.8, below the percolation threshold of
D 1.55 [4]. The threshold is exceeded in the liquid-like states, where D 2 on the
average. Some 5–10 % of the molecules in the denser SCW do form four hydrogen
bonds, so that it can be considered to be a percolating system.
Incoherent neutron scattering was used by Tassaing et al. [34] to obtain the
diffusion coefficients in SCW at 380 ı C and six densities (200–900 kgm–3). The
hydrogen bond life-time derived from the data remained nearly constant near 0.19 ps
up to a density of 600 kgm–3 and decreased then to 0.13 ps at 900 kgm–3 ,
reflecting the higher incidence of collisions at the higher densities.
1.5 Hydrogen Bonding in SCW Deduced from Spectroscopy
Luck [35, 36] applied infrared absorption spectroscopy to just-supercritical water,

the measurements extending up to 390 ı C. The 8,749 cm–1 2v3 C 2 combination
band is characteristic of vibrations of free O–H groups, that is, those in monomeric
molecules that do not participate in any hydrogen bonding at all. This 2v3 C 2
combination band is somewhat more intense beyond 350 ı C than the 8,696 cm–1
band of the free O–H vibration of a molecule in which the other OH group does
participate in a hydrogen bond. Luck and Ditter subsequently [37] extended these
studies to dilute (5.2 mol%) HOD in D2 O and presented the IR spectrum at 386 ı C,
assigning the overtone 2 3 at 7,163 cm–1 to the free O–H vibration. They calculated
the fraction of free O–H groups from various relevant IR bands just beyond the
critical point, arriving at f0 D 1, contrary to findings by others and other methods,
see Table 1.1. Franck and Roth [13] used 8.5 mol% HOD in H2 O (rather than in
D2 O) and measured the IR spectra at 400 ı C at various densities of SCW from
150 to 900 kgm–3. The authors concluded that almost all the O–D groups are
hydrogen bonded to some extent at densities 100 kgm–3 , contrary to the findings
of Luck [36].
More recently, Kandratsenka et al. [38] measured IR spectra of 2 mol% HOD in
D2 O at one SCW state: 397 ı C and 252 kgm–3, confirming the findings of Franck
and Roth [13]. They related the MD-calculated 3,600 cm–1 band frequencies to
the local d(O–O) distances and in turn to the mean number of hydrogen bonds
per water molecule in the system, <nHB >. Tassaing et al. [39] measured the IR
spectrum of water at 380 ı C, i. e., just above the critical point. The resulting spectra
were de-convoluted into bands corresponding to small clusters: monomer ( free

O–H), dimer, and linear trimer. As the density of the SCW increased from zero to
430 kgm–3 , the fractions of the species changed and reached f0 63 %, f1 21 %,
and f2 16 % at the largest density.
Kohl et al. [3] applied Raman spectroscopy to SCW containing 9.7 mol%
HOD in H2 O at 400 ı C and a variety of densities between 40 and 800 kgm–3 .
They de-convoluted the 2,730 cm–1 O–D 1 stretching vibration band into two
components. These corresponded to hydrogen-bonded and non-hydrogen-bonded
water molecules, indicating that only 33 % of the water molecules are hydrogen-
bonded. Frantz et al. [40] studied the Raman spectra of SCW at three temperatures:
401, 451, and 505 ı C and various pressures, from 23 to 202 MPa. They interpreted
the spectra obtained as the temperature was increased and/or the density was
decreased in terms of increasing fractions of non- or weakly-hydrogen-bonded water
molecules.
Walrafen and Chu [41] studied one SCW state, at 400 ı C and an unspecified
density above 800 kgm–3 , and discussed the relation between the intensity of the
correlated Raman scattering at the 3,250 cm–1 shoulder, corresponding to collective
motions of hydrogen bonded molecules, and the 3,400 cm–1 peak corresponding to
un-correlated 1 stretching vibrations of the O–H bonds. Over a wide temperature
range, including the SCW state, the intensity ratio Rc/u of the correlated to
uncorrelated bands was linear with the structural correlation lengths (SCL) obtained
from diffraction measurements: Rc/u D 2.55(SCL/nm) – 0.304. The persistence of
the linearity to SCW suggested that hydrogen-bonded clusters are present there that
have some degree of collective motions of their molecules. According to Walrafen
et al. [42], a blue shift 1 was seen as the temperature was raised, indicating
the breaking of hydrogen bonds, but 1 did not quite reach the monomer value of
3,657 cm–1 , so that some hydrogen bonding persisted. A cooperative network of
hydrogen bonds in SCW at densities >400 kgm–3 was absent and only dimers were
the predominant species for the hydrogen bonding.
According to Yasaka et al. [43] the Raman frequencies of SCW at 400 ı C were
red-shifted linearly as the density increased from 200 to 600 kgm–1 , as a result of
hydrogen bonding. The same was noted by Yui et al. [44], in whose study of water
at 400 ı C the pressure was increased from 19.5 to 36.7 MPa. The frequency of the
O–H stretching vibration was 1 /cm–1 D 3645.5–0.03831(/ kgm–3). A very short
and intense laser pulse [45] at the 1 band frequency has the same effect as heating
water to supercritical conditions: hydrogen bonds of the excited water molecule are
disrupted, to be re-formed with an appropriate relaxation time.
Wernet et al. [46] applied x-ray Raman scattering at the oxygen K-edge to
SCW at 380 ı C (just supercritical) and a density of 540 kgm–3. The resulting
spectrum showed a near-edge peak at 534 eV, a shoulder at 536 eV, and a major
contribution between 537 and 542 eV (post-edge). The intensity of the post-edge
region, an energy region absent in water vapor, points to a large degree of hydrogen-
bonding, whereas the former two features correspond to those found in dilute water
vapor, i.e., monomeric water molecules, the fraction f0 , of which was estimated as
35 ˙ 20 %. The remaining 65 ˙ 20 % of the water molecules appear in clusters
16 Y. Marcus
of 5–10 water molecules presumed to have four (!) hydrogen bonds each, though
elongated and bent, with no allowance for molecules with fewer than four hydrogen
bonds.
Jonas et al. [47] applied 1 H NMR to SCW at 400–600 ı C and fairly low densities,
50–350 kgm–3. They reported the spin-lattice relaxation times T1 that increased
with increasing densities but diminished with increasing temperatures. They were
unable to obtain accurate estimates of the angular momentum correlation times J ,
which were some five times shorter than the reported Enskkog relaxation times,
ranging from 0.2 to 2.8 ps. The shortness of the J was ascribed, in part, to
strongly anisotropic forces between water molecules (i.e., hydrogen bonding but
not specified as such). A more detailed study by Lamb and Jonas [48] followed, in
which the spin lattice relaxation times T1 were measured for SCW from 400 ı C up to
700 ı C. The estimated angular momentum correlation times J were now reported,
increasing with the increasing temperatures at a constant density (at 350 kgm–3
from 0.0641 ps at 400 ı C to 0.103 ps at 700 ı C) but decreasing with the density
at a constant temperature. (from 0.777 ps at 500 ı C and 50 kgm–3 to 0.0468 ps at
600 ı C and 350 kgm–3 ).
Hoffmann and Conradi [2] shifted the attention from relaxation times dealt with
above to the chemical shifts ı measurable by proton NMR in SCW at 400–600 ı C,
using dilute benzene as an internal reference, as described in Sect. 1.3. Applied to
SCW at 400 ı C and 520 kgm–3, for instance, the chemical shift of ı D 5.4 ppm
corresponds to 29 % of hydrogen bonding relative to ambient water.
Matubayasi et al. [18, 49] measure the proton chemical shifts in SCW at
380, 390, and 400 ı C at the densities, 290, 410, 490 and 600 kgm–3 . Magnetic
susceptibility corrections were applied in order to refer the chemical shifts to an
isolated water molecule. Over this narrow temperature range ı is only weakly
temperature dependent, but increases with the density. The resulting ı values were
in good agreement with those of Hoffmann and Conradi [2], and both sets of authors
stressed that no cooperativity of hydrogen bonding needs to be taken into account
for SCW. More recently Yoshida et al. [50, 51] used proton NMR with a high
resolution (500 MHz) instrument to study SCW at 400 ı C. The relaxation time
of the translational velocity, D , is related to the self-diffusion coefficient D. The
solvation relaxation time, S, is the relaxation time in the solvation shell of a given
water molecule. The latter declines with the number n of molecules in this shell,
from 0.04 ps at n D 4 to 0.01 ps at n D 22 at the supercritical density of 600 kgm–3 .
These values correspond to nHB in the solvation shell or cluster (not per molecule)
of from 0 to 8.
Matubayasi et al. [52] concluded from several ab initio molecular orbital theories
for the proton chemical shift at the critical temperature and density that water
is made up from 80 % monomer and 20 % dimer. Sebastiani and Parinello [53]
calculated for just critical water (374 ı C, 320 kgm–3 ) and SCW slightly above the
critical point (380 ı C, 730 kgm–3) that there are 14 and 37 % hydrogen bonded
water molecules relative to ambient water under these conditions.
Tsukahara et al. [54] employed 17 O NMR in order to study SCW, up to
425 ı C, reporting both chemical shift, •, and spin lattice relaxation times, T1 .
Magnetic susceptibility corrections were applied and ambient water was used as
an external reference. The chemical shift of dilute water vapor was ı D 36.1 ppm
against the reference. The hydrogen bonding relative to ambient water (cf. Hoff-
mann and Conradi [2] above) was D 1 C 0.0274ı. At Tc and 480 kgm–3 D 0.40,
falling to 0.17 at 430 ı C and 240 kgm–3 (read from a figure).
Okada et al. [55] presented detailed Debye relaxation time, D , data of the
permittivity " of a fluid in an alternating field of frequency up to ! D 40 GHz for
both supercritical H2 O and D2 O up to 600 ı C and densities 700 kgm–3 . The
values for D2 O are some 30 % larger than those for H2 O at a given temperature
and density. Yao and Hiejima [56], following Okada et al. [55], introduced the
hydrogen bonding of a fraction fb of the water molecules in SCW into the relaxation
expression:
" .!/ D " .1/ C .".0/ " .1// Œ.1 fb / = .1 C i ! D f / C fb = .1 C i ! D b /

(1.17)
Here D f and D b are the relaxation times of the ‘free’ and ‘bound’ water
molecules and the measured relaxation time is D D (1–fb ) D f C fb D b . In the dilute
limit fb 0, so that D f is the binary collision time of water molecules in dilute
vapors. The relaxation time of the hydrogen bonded water molecules D b is related
to the librational motions of the water molecules, assumed to cause the braking
of the relaxation. Therefore, D b D D f C < lib > exp(hb H/RT), where < lib > is
the inverse of the mean librational frequency and hb H is the enthalpy for the
formation of a hydrogen bond. Values of the fraction of hydrogen bonded water
molecules, fb , were obtained on setting < lib > as 0.067 ps and the enthalpy as
hb H D 10.6 ˙ 0.4 kJmol–1 . The resulting values fall between the NMR values
of Matubayasi et al. [18] or Hoffmann and Conradi [2] and those obtained from
neutron diffraction by Jedlovszky et al. [24].
1.6 Hydrogen Bonding in SCW Deduced

from Computer Simulations
In the computer simulations, a specified number of water molecules are placed in

a cubic box with periodic boundary conditions that simulate to some extent infinite
systems. This device avoids problems with the surface of an ensemble of a limited
number of molecules, the number growing over the years, from only 108 used in
1980 [57] to as many as 1,000 in 2005 [25]. The potential functions for water-
water interactions involve Coulombic interactions that take place between the partial
charges on the atoms, so that the long range electrostatic effects have to be taken
into account according to various schemes. The simulations are classified as MC
ones, in which the equilibrium structure of SCW is established, and MD ones, in
which some dynamic features of SCW are additionally obtained. In MD simulations
the computationally derived critical point of the system, Tc , is determined by the
18 Y. Marcus
potential function employed. This Tc , generally deviates from that of the real fluid,
374 ı C: it may be considerably lower, down to 301 ı C [52], or larger, up to 437 ı C
[58] according to the potential functions employed. It is, therefore, expedient to
report the reduced temperature Tr D T/Tc model , rather than the nominal temperatures
T reported by the authors.
The results of the MC simulation are the thermodynamic functions of the system
as well as the pair correlation functions, g(r), as averages over the ensemble of
configurations, assuming the equivalence of the time and ensemble averages (the
ergodic principle). Kalinichev et al. published a series of papers on the application
of MC simulations to SCW [20, 59–61] and previous papers referenced there.
Kalinichev used the empirical TIP4P potential on SCW at 400 ı C and densities
of 187, 390, 600, and 1,050 kgm–3 and at 500 ı C at densities of 257, 529, 1,014,
and 1,262 kgm–3 [59]. Partial pair correlation functions exhibited the hydrogen
bond signature peak at 0.19 nm in g(O–H, r), see Sect. 1.4, the absence of the
tetrahedral structural feature at 0.45 nm in g(O–O, r), and hardly any rotational
orientation correlation in g(H–H, r). Kalinichev and Bass [20] confirmed the
previous findings and augmented them with criteria for hydrogen bond formation:
d(O H) 0.24 nm for the hydrogen bond length and eHB 10 kJmol–1 for
its energy. The average O–H O angle was 150 ı for the hydrogen bonds at all
the thermodynamic states examined. A combination of these criteria showed the
average number of hydrogen bonds per water molecule < nHB > in the system at
500 ı C, increasing gradually from 0.1 at 10 MPa (30.5 kgm–3) to 1.25 at 10 GPa
(1,666 kgm–3). Further simulations were made [60] at 400 ı C and densities ranging
from 578 to 1,670 kgm–3 as well as at 500 and 1,000 ı C at densities of 440
and 170 kgm–3 respectively. The hydrogen bond signature peak at 0.19 nm in
the g(O–H, r) corresponded to D 2.1 HBs per water molecule at 400 ı C, but at
1,000 ı C it degenerated to a barely discernable shoulder with D 0.2. Over the
entire range from ambient to 1,000 ı C the resulting fraction of monomeric water
molecules, f0 , correlated with the average number of hydrogen bonds of a water
molecule D f(T,P,):
f0 D 0:948 0:886 C 0:2862 0:0303 (1.18)
The percolation threshold was exceeded only for temperatures below 600 ı C
at any of the available densities. Kalinichev and Churakov [61] dealt mainly with
the structures of the water clusters in SCW that existed at conditions below the
percolation threshold. Chain-like trimers, tetramers and pentamers predominated
over branched and ring structures, their maximal size being seven water molecules.
Matubayasi et al. [62] employed both the TIP4P and the simple point charge
(SPC) potentials for MC simulations with 648 SCW molecules at 400 and 500 ı C at
densities of 200, 400, and 600 kgm–3. They computed the expected NMR chemical
shifts (Sect. 1.5) and the partial pair correlation functions. The latter depend strongly
on the dipole moments used as parameters in the range 1.85 / D 2.35, and
2.15 D was required to conform to results generally accepted for the pair
correlation functions in SCW. The presence of a third water molecule near a
hydrogen bonded pair enhanced the probability of hydrogen bonding over a wide
range of O–O–O angles (>72ı ) rather than the tetrahedral angle (109ı ) as in ambient
water.
Mountain [14] applied MD simulations to supercritical water using the TIP4P
potential on 108 water molecules and derived the partial pair correlation functions
g(O–O, r), g(O–H, r), and g(H–H, r) for several SCW states. He reported the
average number of hydrogen bonds per water molecule, <nHB >, based on the
geometrical criterion for the existence of a hydrogen bond: d(O H) 0.24 nm.
This < nHB > decreased mildly with increasing temperatures, up to Tr D 2.34,
but strongly with diminishing densities: <nHB > is 1.36 at 999 kgm–3, 0.83 at
600 kgm–3, but only 0.40 at 250 kgm–3 all at Tr D 1.36.
Krishtal et al. [63] summarized the earlier finding by Kalinichev et al. referenced
there from the use of the flexible Bopp-Janscó-Heinzinger BJH potential function
for MD simulations of SCW at 1.00 Tr 1.21 and densities between 200 and
600 kgm–3. They presented values for < nHB > (read from a small scale figure)
diminishing from 1.3 to 0.6 as the temperature increased, based on the geometric
criterion. The fractions of water molecules with 0, 1, 2, 3, and four hydrogen bonds
were presented in a small scale figure too. Practically no molecules with four bonds
existed but a small fraction (from 10 % down to 2 % as Tr increased) with three
bonds did.
Mizan et al. [64] discussed critically the choice of the water model potential
function for MD simulations of SCW. Rigid models, such as SPC or TIP4P,
yielded critical points far from the experimental one, whereas flexible models
succeed much better in reproducing the thermodynamic properties. The Teleman-
Jönsson-Engström (TJE) model appeared to yield the best potential for simulating
SCW (but Tc was not specified). The authors applied this potential to 256 water
molecules at nominally 500 ı C at four densities: 115.3 (gas-like), 257.0, 405.8,
and 659.3 kgm–3. The 0.19 nm HB signature peak was clearly shown by the
g(O–H, r) curves of all the SCW states, but its intensity diminished with increasing
density. A rigid four-site polarizable model applied by Dang [65] to SCW at
Tr D 1.09 and 660 kgm–3 showed agreement with the neutron diffraction-with-
isotope-replacement experimental results (Sect. 1.4), confirming the presence of a
significant hydrogen bonding signature peak in g(O–H, r) at 0.2 nm.
Liew et al. [66] used a flexible four-site water model for MD simulations on 500
water molecules. This model was capable of near reproduction of the experimental
critical temperature and density of water, contrary to rigid models. Bifurcated
hydrogen bonds between adjacent water molecules were found beside the linear
ones, less stable than the latter by only 5.8 kJmol–1 .
Yoshii et al. [21] applied a polarizable SPC model to SCW at a reduced
temperature Tr D 1.07, i.e., at 400 ı C and six densities from 27 to 1,000 kgm–3 .
They performed MD calculations on 256 water molecules and used d(O H)
0.25 nm as a definition of an intact hydrogen bond. The coordination numbers
and < nHB > values increased steadily with the density, up to 12 and 1.9 respec-
tively at the highest density, the hydrogen bonds being bent at 21ı from linearity on
average. The was larger than the percolation threshold of 1.55 at densities above
20 Y. Marcus
800 kgm–3. Petrenko et al. [67] used the TIP4P potential modified for the explicit
formation of hydrogen bonds for MD simulations with 216 molecules. Along the
50 MPa isobar the average number of hydrogen bonds of a water molecule, ,
decreased from 2.0 at Tr D 1.0 to 0.45 at Tr D 1.27.
Dyer and Cummings [17] applied MD simulation at two SCW densities: 1,000
and 600 kgm–3 over the temperature range 423–723 ı C, using the Gaussian
charge polarizable model and the Carr-Parinello ab-initio one to 256 and 32 water
molecules respectively. They estimated the number of hydrogen bonds per water
molecule, , and the values of < nHB > D / 2 interpolated from a figure are shown
in Table 1.2.
Skarmoutsos and Guardia [68] used the SPC/E model for MD simulations of 500
water molecules at Tr D 1.03 over a wide density range: 61–644 kgm3 to calculate
the life times of hydrogen bonds, defined geometrically: d(O–O) 0.36 nm and
d(O H) 0.24 nm. The intermittent life times, HBi , decrease from 0.62 ps at
61 kgm3, to a minimum of 0.44 ps near the critical density and increases to 0.47 ps
at 644 kgm3 . On the other hand, the continuous life time, HBc , remained near
0.075 ps over the density range. As expected, the re-orientation correlation times
increased with both the number of hydrogen bonds a water molecule is engaged in
and with the density.
Kandratsenka et al. [38] stressed the local densities as being more relevant to the
hydrogen bond connectivity than the bulk density. The g(O–O, r) curve from MD
simulations with the SPC/E model potential on 108 water molecules at Tr D 1.09 at
428 kgm–3 was resolved into contributions from 1st, 2nd, 3rd, and 4th neighbors.
The d(O–O) distances were then correlated with the maxima of the O–H stretching
vibrations of dilute HOD in D2 O. Swiata-Wojcik and Szala-Bilnik [16] also used
MD simulations with the BJH flexible model to study the inhomogeneity in SCW
(at Tr D 1.10 and 1.04) with results shown in Table 1.2 for < nHB >, being larger than
other estimates.
1.7 Ionic Dissociation of SCW
The self-dissociation of water in the supercritical state has to be written as a

bimolecular reaction, because no free protons have been detected in any of the states
examined:
2H2 O H3 OC C OH (1.19)
Each of these three nominal species is hydrogen-bonded to further water

molecules. The ion product KW is, therefore, related to the extent of the hydrogen
bonding in SCW.
Quist and Marshall [69] used the conductance of a solution of NH4 Br, a
hydrolysable salt, in SCW in comparison with those of KBr and HBr to obtain
Table 1.3 The p(KW /mol2 kg–2 ) of SCW according to data from [70] and Eq. 1.21 and thermody-
namic functions H ı /kJmol–1 and Sı / JK–1 mol–1 from [71], at relatively low pressures
P/ MPa
25 30 40 50 60 70
t/ ı C pKW –H ı –Sı pKW
400 13.16 12.97 12.61 12.24 197 520 11.88 11.51
500 16.97 16.64 15.97 15.30 417 849 14.63 13.96
600 19.88 19.44 18.56 17.68 191 569 16.80 15.93
700 21.88 21.38 20.39 19.40 18.41 17.42
800 22.97 22.47 21.46 20.45 122 495 19.45 18.45
the ion product, KW , values. The equilibria involved are: NH4 Br NH3 C HBr,
NH3 C H2 O NH4 C C OH– , and HBr C H2 O H3 OC C Br– , where, again, all
the species are hydrated and hydrogen bonded to water molecules. Quist equated
the ionic conductivities of NH4 C and KC and included the mean ionic activity
coefficients, y˙ D exp(AI1/2/(1 C I1/2 )). Here A is the Debye-Hückel coefficient
at the prevailing temperature and relative permittivity and I is the ionic strength.
The latter resulted from the concentrations of all the ionic species involved, so
that an iterative calculation was necessary. It yielded the desired ion product
constant of SCW, KW D [HC ][OH– ]y˙ 2 , where [ ] denote molar concentrations. For
temperatures ranging up to 800 ı C, pressures up to 400 MPa, and the densities
between 450 and 950 kgm–3 the expression on the molar scale is [69]:
h 2 i
log KW = mol dm3 D 33:05 3050= .T =K/ C 16:8 log cW (1.20)
involving the density-dependent molar concentration of water, cW. The estimated

uncertainty of the values from Eq. 1.20 was about 0.3 to 0.5 units.
Marshall and Franck [70] provided a six parameter expression for the ion product
of water (on the molal scale) at states corresponding to SCW, based on the results
of Quist and Marshall [69] in terms of the temperature and pressure:
h 2 i
log KW = mol kg1 D .15:68C0:181 .P=MPa//CŒ9:730:0736 .P=MPa/
2
102 t=ı C Œ5:73 0:048 .P=MPa/ 105 t=ı C (1.21)
with an estimated uncertainty of ˙0.03 to ˙0.10.

Compared to the well known value of log[KW /(molkg–1)2 ] D 14.00 for liquid
water at the ambient conditions of 0.1 MPa and 25 ı C, the values for SCW can
be considerably less negative at 400 ı C, see Table 1.3, and at larger pressures than
shown there increasing to 11.19 at 450 ı C and 100 MPa and to 8.85 at 600 ı C
and 500 MPa. Thus, there can be appreciably more ionic dissociation in SCW than
in ambient water.
22 Y. Marcus
Mesmer et al. [71] obtained the standard thermodynamic functions H ı , Sı ,

Cp ı , and V ı for the ionization of water from the equilibrium constants, the
ionization being entropy-controlled (see Table 1.3). They used the volume change
V ı during ionic dissociation of water in SCW to estimate that about 14 water
molecules are associated through hydrogen bonding with the ions formed at a
density of 800 kgm–3, a value that increased with decreasing densities.
Tanger and Pitzer [72] applied to SCW a different approach, based on a
thermodynamic cycle involving estimates of the Gibbs energies of hydration of
H3 OC and OH– , thus not depending on the conductance of electrolytes as used
by Quist and Marshall [69]. The results, applicable up to 2,000 ı C and 500 MPa,
agreed with those of Marshall and Franck [70] at 450 kgm–3 but were said to
produce more correct values for SCW at lower densities than in [70].
2H2 O .g; T / ! H3 OC .g; T / C OH .g; T /

#" #" #" (1.22)
C
2H2 O .SCW; P; T / ! H3 O .SCW; P; T / C OH .SCW; P; T /
The Gibbs energies of hydration were estimated according to a semi-continuum

model, involving the enthalpy and entropy changes of successive hydration of
the ions.
Other semi-theoretical and computer simulation approaches, such as those of
Tawa and Pratt [73], Bandura and Lvov [74], Takahashi et al. [75], and Halstead
and Masters [76], did not provide insight to the hydrogen bonding involved with the
self-ionization of SCW.
1.8 The Solvent Power of SCW for Organic Solutes
The diminished extent of hydrogen bonding in SCW relative to water under ambient
conditions is an important factor in the enhanced solubility of non-polar organic sub-
stances in SCW. This enhanced solubility is often ascribed to the lower permittivity
of the SCW, but this is a secondary effect. As for other solvents, one measure for
the estimation of the solubility of a given substance in a solvent is their Hildebrand
solubility parameters, ı H , relative to each other. The smaller their difference, the
larger is the solubility. In the case of SCW, as for other supercritical solvents,
it is not possible to use the usual expression employed in obtaining solubility
parameters: ı H D [(V H – RT)/V]1/2 , because their enthalpy of vaporization V H
is meaningless. The appropriate expression is ı H D [U/V]1/2 , where –U is the
configurational potential energy of the supercritical solvent (expressed for ordinary
solvents by V H – RT).
The configurational potential energy is related to the attractive potential
parameter of the equation of state (EoS): –U D a’/V. For conformal supercritical
fluids, those interacting by dispersion forces only, the Redlich-Kwong EoS is
applicable [77]:
P D RT= .V b/ NA 2 ˛a=V .V C b/ (1.23)
The parameters describing the attractive forces and excluded volume are a and b
respectively, and ’ D T –1/2 . However, SCW is a non-conformal fluid, hence mod-
ifications of the Redlich-Kwong EoS, e.g., those of Soave-Redlich-Kwong (SRK)
[78] or of Peng-Robinson (PR) [79] are required. These two approaches specify the
attractive parameter as a D cR2 T 2c /Pc and the factor ˛ D [1 C m(1 T 1/2 2
r )] instead
of ’ D T –1/2
. Here Tr D T/Tc is the reduced temperature and m is a quadratic in the
Pitzer acentric factor (! D 0.3443 for water). However, the differences between the
SRK- and PR-derived values of ı H /MPa1/2 for SCW differ by only 0.1–0.2 units, so
that their averages may be used. According to these considerations, the expression
for the Hildebrand solubility parameter of SCW [80], slightly revised in [81], is:

ıH .SCW/ =MPa1=2 D 14:11 1 C 1:145 1 Tr1=2 Tr1=4 r (1.24)
At all practically available thermodynamic states of T and the Hildebrand

solubility parameter ı H of SCE is much below that for liquid water at ambient
conditions, 48 MPa1/2 , i. e., much nearer the values of ı H of non-polar organic
substances, in agreement with their enhanced solubility in SCW.
However, the total Hildebrand solubility parameter of SCW, ı H as presented
by Eq. 1.24 is not the best indicator for the solubility of specific solutes. The
three Hansen solubility parameters: ı d for dispersion interactions, ı p for polar
interactions, and ı H for hydrogen bonding should be preferably employed:
ıH2 D ıd2 C ıp2 C ıh2 (1.25)
For water at 25 ı C the respective values are ı d D 15.6, ı p D 16.0, and ı H D 42.3
(all in MPa1/2 ) to be used as reference values (subscript ref ). Following Williams
et al. [82] the expressions for the calculation of the Hansen solubility parameters ı d
and ı p for SCW are according to Marcus [81]:
ıd D ıd ref .V =Vref /1:25 (1.26)
ıp D ıp ref .V =Vref /0:5 (1.27)
The expression for ı H in SCW was not given explicitly in [81] but the tabulated
values for various thermodynamic states lead to:
ıh D ıh ref exp f0:00132 .T Tref / C 0:5 ln .V =Vref /g (1.28)
with Vref D 18.07 cm3 mol–1 at Tref D 298.15 K for all three parameters. The molar
volume V/ cm3 mol–1 of SCW at the desired temperature T and pressure are obtained
from the Steam Tables or other relevant references, see the appendix. Calculated
values of these Hansen solubility parameters of SCW are reported in [81], and
24 Y. Marcus
Table 1.4 The Hansen P/ MPa ıd ıp ıH ıH

solubility parameters, ı d , ı p,
and ı hb in MPa1/2 , and the 400 ı C
total Hildebrand solubility 25 1.7 6:5 5:2 8:5
parameter ı H calculated from 40 6.9 11:6 12:7 18:5
Eq. 1.25 in MPa1/2 60 8.4 12:5 16:3 22:2
80 9.3 13:0 18:5 24:5
100 9.8 13:3 20:0 25:9
500 ı C
25 0.8 4:8 2:0 5:2
40 1.8 6:8 4:0 8:1
60 4.0 9:3 7:5 12:6
80 5.9 10:9 10:8 16:4
100 7.0 11:7 13:4 19:1
600 ı C
25 0.6 4:3 1:4 4:5
40 1.1 5:6 2:6 6:3
60 2.2 7:3 4:7 8:9
80 3.4 8:7 7:0 11:7
100 4.5 9:8 9:3 14:2
700 ı C
25 0.5 3:9 0:9 4:1
40 0.9 5:1 1:8 5:5
60 1.6 6:4 3:4 7:4
80 2.3 7:5 5:1 9:4
100 2.9 8:2 6:2 10:7
a figure showing ı H for 380 t/ı C 480 and 15 P/ MPa 40 was shown by
Morimoto et al. [83]. Table 1.4 shows the Hansen and the total Hildebrand solu-
bility parameters of SCW at a few pressures 25 P/MPa 100 and temperatures
400 t/ı C 700. Over these temperature and pressure ranges they increase with
pressure at isotherms and diminish with rising temperatures at isobars.
Solvent properties at ambient conditions, such as solvation ability, polarity,
and hydrogen bonding ability, have been extensively estimated by means of
solvatochromic probes. However, little use has been made of otherwise suitable
probes in SCW, because they tend to be unstable at the high temperatures involved.
If air is not carefully excluded they are oxidized and their oxidation products have
similar but not the same solvatochromic properties.
Still, Bennet and Johnston [84] applied benzophenone and acetone as UV-
visible spectroscopic Kamlet-Taft * (polarity/polarizability) probes in SCW up
to 440 ı C. Acetone is miscible with water as a supercritical fluid mixture [85],
so that a wide range of temperatures is available for its spectroscopic study. The
n– * band of acetone is red-shifted from 262 nm in ambient water to 280 nm at
380 ı C. Above 28 MPa the band position is affected by both solvatochromism and
thermochromism, but is independent of the pressure. The spectral shift of the band
maximum in SCW is:

.T; / D .T; / 42:8 1:71 t=ı C (1.29)
where is in wave numbers and the second term is the thermochromic correction,
obtained from spectral data of acetone vapor in argon gas. An appreciable red-shift
of the band maximum occurred when the reduced density r of SCW diminished
from 3.2 to 2.0 independently of the temperature, an effect ascribed to the decreased
hydrogen bonding donicity of the water. When the density was further decreased
to r D 0.5 about one half of (T,) was due to the changed hydrogen bonding,
the other half to physical effects. A plateau in the (r ) curve was observed for
0.5 r 1.5, the conclusion being that the local density of water molecules near
the acetone ones was much larger that the bulk density and changed little when the
latter diminished.
For benzophenone in SCW, according to Bennet and Johnston [84], an increase
in temperature caused a decrease in the solvent density relative to ambient water
and reduced its polarizability per unit volume, causing a blue shift of the absorp-
tion spectrum of the benzophenone, as is known also for conventional solvents.
Conversely, an increase in pressure (and density) in SCW caused a red shift, by
increasing the polarizability density. The probe itself, however, affected the density
of the nearby SCW, augmenting it by interactions of the dipoles of the water
molecules with the highly polarizable probe and by hydrogen bond donation from
the water to the carbonyl group. The spectral shift corresponded to an increase in the
local density, up to r local D 1.3 at r D 1.0, when the reduced density of the SCW
at 380 ı C was diminished down from r D 1.5. MC computer simulation with a
polarized benzophenone model [86] reproduced the spectral shift in SCW at 400 ı C
and 34.5 MPa.
In spite of the instability of nitroaromatics in SCW, Oka and Kajimoto [87]
did study the spectral properties of 4-nitroaniline, N,N-dimethyl-4-nitroaniline,
and 4-nitroanisole in SCW at 380, 390, and 410 ı C by employing a flow system
that restricted the residence time of the indicators in SCW. The absorption peak
of 4-nitroaniline at 380 nm in ambient water is blue-shifted in SCW at 380 ı C,
the more the lower the density of the SCW, down to 300 nm at a density of
123 kgm–3. Similar blue shifts were obtained with the other indicators and at
the other temperatures, but at densities above 300 kgm–3 the effects were much
smaller. A comparison of results from the three indicators led to the conclusion that
the spectral shifts were due to the extent of hydrogen bonding in the SCW itself
and that no specific hydrogen bonding between the water molecules of SCW with
either the nitro-group or the amino group of 4-nitroaniline took place. The Raman
spectroscopic study of Fujisawa et al. [88] did find hydrogen bonds to be formed in
SCW between the water molecules and the amino group of p-nitroaniline, contrary
to the conclusion of Oka and Kajimoto [87] that such bonds are absent.
Minami et al. [89] measured the UV spectrum of 4-nitroanisole in SCW, again in
a flow system in order to minimize the decomposition, and calculated the Kamlet-
Taft solvatochromic parameter * . This was defined as

D max =cm1 33; 985 = .31; 612 33; 985/ (1.30)
26 Y. Marcus
where max (cyclohexane, 0) D 33,985 cm–1 and max (dimethylsulfoxide,

* 1) D 31,612 cm–1 at ambient conditions were used for the normalization
of the scale. A linear dependence of * on the density was found for SCW at 400
and 420 ı C:
D .0:71 ˙ 0:06/ C .1:77 ˙ 0:09/ (1.31)
but at 380 and 390 ı C deviations from linearity (curves concave downward)
occurred. These deviations were explained by local density augmentation (cluster-
ing), paralleling the maximum of the isothermal compressibility of SCW at the same
bulk density, 300 kgm–3 , an effect no longer present at 400 ı C. Dipole-dipole
interactions take place between the indicator molecules and the water ones rather
than hydrogen bonding, and these are weaker than the hydrogen bonding of the
water molecules among themselves in SCW. The local density augmentation was
the same for N,N-dimethylaminobenzonitrile as that of 4-nitroanisole, but that near
N,N-dimethyl-4-nitroaniline at 380 ı C was found to be somewhat larger.
The proton exchange between an organic anion A– and hydroxide anions:
HA C OH– A– C H2 O takes place with no net change in the charges in the
system. Therefore, as the density of the SCW is varied the equilibrium constant
KBHA is not sensitive to the changes in the permittivity of the system, but only to
the interactions with the solvent. Xiang and Johnston [90] studied the acid-base
reaction of dilute solutions of ˇ-naphthol in SCW at 400 ı C by UV-spectroscopy.
The peak absorption of the acidic form is at 326.7 nm whereas that of the
basic form (naphtholate anion) was shifted to 370 nm. The value of logKBHA at
400 ı C was 3.1 at 200 kgm–3, diminishing to 1.2 as the density was increased
to 1,000 kgm–3 and the reaction was endothermic. The negative charge resided
preferentially on the larger naphtholate anion than on the smaller hydroxide one.
The equilibrium constant for the ionization of ˇ-naphthol in SCW, HA A– C HC ,
is Ka D KBHA KW , where KW is the ion product of water, dealt with in Sect. 1.7, and
the ionization reaction is exothermic.
The local density of SCW near organic molecules may differ from the bulk
density if enhancement or diminution of its self-hydrogen-bonding takes place,
irrespective of whether water molecules are hydrogen bonded to the solute. UV-
visible and Raman spectroscopic methods have been used to study indicator probe
molecules in SCW, and in particular the local density enhancement or diminution
near the critical temperature. Osada et al. [91] measured the UV-visible spectrum of
quinoline in SCW at 380–430 ı C and up to 40 MPa. Negative solvation takes place,
i. e., the local density of water around the quinoline is lower than in bulk water
and the hydrogen bonding between water and quinoline is strongly diminished,
at conditions where the compressibility of water is large, 0.5 < r < 1.5. On the
contrary, Aizawa et al. [92, 93] found local density augmentation in SCW around the
exciplex between acetophenone and tetramethylbenzidine at 380–410 ı C, the more,
the lower the density. At a bulk density of 150 kgm–3 the local density of the water
near the exciplex is raised to 270 kgm–3. Similarly, Minami et al. [94] found that the
local water density is augmented in solutions of pyridazine in SCW from UV-visible

spectroscopy at 380–420 ı C. At 380 ı C and a bulk density of 200 kgm–3 there are
30 % as many hydrogen bonds (<nHB > D 0.52) as in the absence of the solute.
Osawa et al. [95] applied Raman spectroscopy to study p-aminobenzonitrile in SCW
and supercritical methanol, finding a red shift of the C N stretching vibration with
increasing solvent density up to r 2, beyond which there occurred a blue shift.
Takebayashi et al. [96] measured the 13 C and 1 H NMR chemical shifts in acetone
dissolved in SCW up to 400 ı C for the investigation of the hydration and hydrogen
bonding in this system, supplemented by MC computer simulation. The NMR
chemical shifts, both ı(13 C D O) and ı(1 H2 O) increase somewhat as the temperature
is reduced at a given density in SCW. These shifts increase more appreciably as
the density is raised isothermally from gas-like densities (100 kgm–3) to liquid-
like ones (600 kgm–3). The non-linear density increase of ı(13 C D O), a concave
downward curve, from 196.7 ppm for an isolated acetone molecule to 208.2 at a
density of 600 kgm–3 parallels closely the blue shift of the UV absorption peak
reported by Bennett and Johnston [84]. The results are interpreted in terms of
competition between hydrogen bonding of water molecules to acetone (up to 0.7
water molecules at 600 kgm–3) and among themselves, in a manner confirmed by
the computer simulations.
Summarizing these studies, it must be concluded that so far the use of solva-
tochromic probes in SCW has yielded little information on the properties of this
solvent regarding other solutes than the probes themselves. The only exception to
this appears to be the study by Minami et al. [89] of p-nitroanisole, yielding Kamlet-
Taft * values that are descriptive of the polarity/polarizability of the SCW. No
values of the HB donation and acceptance abilities of SCW resulted from the studies
of solvatochromic probes in it so far, and this is a challenge to future work.
1.9 The Solvent Power of SCW for Salts
Contrary to the solubilities of non-polar gases and organic solutes in SCW that are
strongly enhanced relative to the solubilities in ambient water, those of ordinary salts
are greatly diminished in SCW at gas-like densities (200 kgm–3). The enhanced
solubility of the former kinds of solutes is ascribed to the breakdown of the hydrogen
bonded network in SCW, a network which demands the expenditure of considerable
energy to create a cavity in it to accommodate a weakly interacting solute. In
the case of crystalline salts their large lattice energies must be compensated
by large hydration energies of their ions in order for solubility to take place.
The low permittivity of SCW does now play a role, enhancing ion association as
the permittivity is decreased, forming less well hydrated ion pairs compared to the
hydration of the ions. The lower density of SCW compared to liquid water also
plays a role, as fewer water molecules are available for the hydration of the ions to
overcome the lattice energies.
28 Y. Marcus
As the density of the SCW is increased towards liquid-like ones ( 600 kgm–3)
the solubility of many salts increases too, with hydration and ion pairing competing,
but both work energetically against the lattice energy that must be invested to effect
solubility. In some cases even complete miscibility is attained, whereas in others a
two-fluid equilibrium occurs above the melting point of the salt. No phase diagrams
of mineral acids, such as HCl, HNO3 , H2 SO4 , and H3 PO4 , in SCW are known,
indicating their miscibility at all proportions.
Valyashko and Urusova [97] presented a small scale figure with the mole fraction
solubilities xsalt (t) of Na2 CO3 , Li2 SO4 , Na2 SO4 , K2 SO4 , and BaCl2 in SCW, ranging
up to ca. 15 mol%, and of NaCl, NaBr, KCl, and Sr(NO3 )2 , ranging up to at
least 30 mol%. The solubilities increase with increasing temperatures. Pressures as
large as 40 MPa are involved in the saturated solutions, but they are lower than
those in pure SCW at corresponding temperatures. The phase diagrams of such
systems can be complicated by liquid immiscibility and other features. Valyashko
[98] reviewed the phase equilibria of water-salt systems at high temperatures and
pressures, including SCW conditions.
Crystalline sodium hydroxide (“-form) melts at 319.1 ı C, and above this
temperature it is completely miscible with water [99, 100]. At higher pressures the
single fluid phase extends to SCW. The conductivity of NaOH in SCW [100] and its
molar volume [101] were measured over the entire composition range at 100 MPa
and 400 ı C.
The solubility of sodium chloride in SCW has been studied extensively over the
years. Anderko and Pitzer [102] presented an EoS for the solubility of solid NaCl
in SCW up to 900 ı C and 500 MPa. Up to the melting point of NaCl its mole
fraction solubility in SCW at liquid-like densities is described by means of a cubic
expression in the temperature:
2 3
xNaCl satd D 0:090 C 1:1183 107 t=ı C C 1:6643 109 t=ı C (1.32)
According to Bischoff [103] the pressure of the saturated solution at 380 ı C,

xNaCl satd D 0.197, is 12 MPa and at 600 ı C, xNaCl satd D 0.490, it is 40 MPa. On
the other hand, SCW is gas-like and has a low density, of the order of 100 kgm–3, at
relatively low pressures. Tester and coworkers [104, 105] reported the solubility of
sodium chloride at 450–550 ı C and 10–25 MPa. The solubility of sodium chloride
under such conditions is measured in ppm:

log CNaCl;ppm D 7:772 C 3:866 log W =g cn3 1233:4= .T =K/ (1.33)
The ppm solubility corresponds to mg(NaCl)(kg SCW)–1 . Its division by the

molar mass of NaCl, 58,450 mgmol–1, converts it to the molality and further
division of the latter by 55.51 (mol H2 O)kg–1 converts it to the mole fraction of the
salt in the saturated solution. The large difference between the solubility of NaCl
in low density SCW, of the order of hundreds of ppm, Eq. 1.33, and the significant
solubility at high density SCW, of mole fractions of some tenths, Eq. 1.32 should be
noted. The coefficient 3.866 of logW represents the hydration number of the NaCl
(see below).
Table 1.5 The parameters Salt Ref. a b n

for the molal solubilities of
salts in SCW, Eq. 1.34 LiCl [108] 271 3:79 2.48
LiNO3 [108] 815 4:28 4.33
NaCl [107] 575 4:94 4.57
NaCl [108] 981 4:53 4.88
NaNO3 [107] 341 4:11 3.06
NaNO3 [108] 269 4:86 3.93
NaH2 PO4 [110] 1;859 7:96 3.47
Na2 CO3 [107] 453 6:90 3.52
Na2 SO4 [107] 1;637 8:78 7.13
Na2 HPO4 [110] 7; 439 18:35 5.37
KOH [108] 698 4:23 3.24
KCl [108] 685 5:00 4.65
KNO3 [108] 407 5:85 3.72
KH2 PO4 [110] 4;418 14:17 4.33
MgCl2 [109] 0 5:61 3.44
MgSO4 [110] 431 8:06 3.31
CaCl2 [109] 441 4:32 2.52
CuO [107] 1;241 5:42 1.34
PbO [107] 1;457 3:03 1.97
Dell’Orco et al. [106] measured the solubility of NaCl in low density SCW at
500 ı C and 27.5 MPa by precipitation from supersaturated solutions, obtaining
227 ppm and confirming well the result of 229 ppm according to [104]. They
proceeded to use their experimental method to study the solubilities of LiNO3 ,
NaNO3 , and KNO3 at 450–525 ı C and 24.8–30.2 MPa. At these temperatures the
pure alkali metal nitrates are molten rather than solid, but in SCW their hydrolysis
leads to formation of HNO3 that in turn decomposes, so that after some time at SCW
conditions nitrates are converted partly to nitrites.
Leusbrock et al. [107] applied a semi-empirical approach to the dissolution of
the solid salt Ca Ac (s) in SCW to form the hydrated fully associated salt Ca Ac nH2 O.
The solubility constant Ks uses W (T,P) to represent the activity of the solvent and
on the assumption of temperature independent enthalpy sln H and entropy sln S of
solution the molality of the saturated solution becomes:
log m .Ca Ac nH2 O/ D a=T C b C n log W (1.34)
The coefficient a, b, and n represent sln H, sln S, and the hydration number
of the salt, respectively. Leusbrock et al., in a series of papers [107–110] studied
the solubility of various salts in SCW, but again at relatively low pressures, hence
densities 200 kgm–3. The solubilities in SCW were described well by Eq. 1.34,
with the parameters shown in Table 1.5. It is not clear why some of the a values
are positive, signifying a negative enthalpy of solution, not discussed in the relevant
publications.
Once a salt is dissolved in SCW there are two competing interactions that have
to be considered: ion hydration and ion association. Contrary to the behavior of
30 Y. Marcus
Table 1.6 The densities / kgm–3 of SCW at some thermodynamic states

P/ MPa
t/ ı C 25 30 40 50 60 70
380 446:40 533:70 594:60 629.37 654.5 674.8
400 166:28 353:28 523:81 578.34 612.7 636.1
450 109:04 148:70 272:12 401.27 479.1 528.0
500 89:86 115:19 178:06 257.60 338.8 405.3
550 78:61 98:37 143:23 195.58 266.3 303.7
600 70:79 87:44 123:64 163.64 207.2 251.7
650 64:87 79:48 110:46 143.68 179.1 215.4
700 60:13 73:28 100:70 129.53 159.8 190.7
750 56:21 68:24 93:04 118.76 145.3 172.4
800 52:89 64:02 86:80 110.18 134.0 158.3
Table 1.7 The molar heat capacity at constant pressure, Cp/ /JK–1 mol–1 , of SCW at some
thermodynamic states
P/ MPa
t/ ı C 25 30 40 50 60 70
380 421:1 177:4 125:3 109:2 100:7 95:28
400 238:3 451:8 157:0 123:9 108:3 100:4
450 92:04 123:2 201:9 169:4 135:4 117:1
500 67:32 77:68 104:5 130:4 135:7 126:0
550 58:11 63:74 76:86 61:07 102:5 107:7
600 53:54 56:93 64:80 73:18 81:07 87:30
650 50:71 53:25 58:66 64:26 69:07 74:49
700 49:16 51:09 55:07 59:14 63:11 66:75
750 48:24 49:76 52:84 55:95 59:00 61:85
800 47:72 48:93 51:49 53:90 56:31 58:60
electrolytes in ambient water, where they are more or less completely dissociated
into free ions, in SCW ionic dissociation is the exception, occurring generally only
to a small extent. At gas-like densities of SCW (<200 kgm–3) ionic dissociation
can be neglected altogether, but may also be unimportant at liquid-like densities
(Table 1.6).
The ion association of electrolytes in SCW is best studied by means of conductiv-
ity measurements and for many of the electrolytes studied the limiting conductivity
1 in SCW at 400 ı C and 600 kgm–3 is nearly independent of the temperature
but decreases with the density. The association constant can often be fitted with four
parameters as a function of the temperature and the density of the water:
log Kassoc D a C b=T C .c C d=T / log W (1.35)
Equation 1.35 is a fitting expression and its parameters have no direct

physical significance, except the coefficient of logW. Marshall [111] expressed
the association constant of binary electrolytes in terms of the number nW
Table 1.8 The relative P/ MPa

permittivity, "r , of SCW at
some thermodynamic states t/ ı C 30 40 50 60 70
380 10:49 13:09 14:27 15:10 16:01
400 6:05 9:62 11:77 13:10 13:80
450 2:18 4:41 6:49 8:25 9:37
500 1:78 2:41 3:46 4:93 6:28
550 1:60 1:96 2:52 3:19 4:24
600 1:50 1:79 2:12 2:46 3:00
650 1:42 1:63 1:90 2:15 2:34
700 1:36 1:50 1:74 1:98 2:05
750 1:32 1:44 1:64 1:85 1:97
Table 1.9 The dynamic P/ MPa

viscosity, / Pas, of SCW at
some thermodynamic states t/ ı C 25 30 40 50 60 70
380 52.5 61.3
400 29.0 43.8 61.3 67.9 72.4 76.0
450 28.9 30.8 39.1 50.7 58.2 63.8
500 30.6 31.7 35.2 40.7 47.4 53.4
550 32.5 33.4 35.7 38.9 43.0 47.4
600 34.5 35.2 37.0 39.3 42.1 45.3
650 36.5 37.1 38.6 40.4 42.6 45.0
700 38.4 38.9 40.2 41.8 43.6 45.5
750 40.3 40.8 41.9 43.3 44.8 46.4
800 42.1 42.6 43.7 44.9 46.2 47.6
Table 1.10 The thermal P/ MPa

conductivity,
th / mWK–1 m–1 , of SCW at t/ ı C 25 30 40 50 60 70
some thermodynamic states 375 411.4 438.0 473.2 498.5 519.4 537.7
400 169.3 330.1 414.0 451.6 477.7 498.7
450 108.8 136.0 227.6 315.6 371.2 408.5
500 100.3 113.7 151.6 202.7 255.6 301.5
550 100.6 109.8 133.5 163.7 198.0 232.4
600 104.1 111.7 130.0 152.1 177.0 202.7
650 109.1 115.8 131.5 149.8 170.1 191.0
700 114.6 120.7 134.8 150.9 168.5 186.7
750 120.3 126.0 138.8 153.4 169.1 195.1
800 126.0 131.3 143.1 156.4 170.6 185.2
of water molecules released when the hydrated ions form the ion pair. On the
molar scale:
log Kassoc.c/ D log K ı C nW log .0:05551W/ (1.36)
where W is the density of water in kgm–3 . In the case of NaCl at 500–800 ı C at

pressures from 100 to 400 MPa nW D 9.7 [111]. The expression (c C d/T) of Eq. 1.35
may thus be equated with nW .
32 Y. Marcus
Ryzhenko and Bryzgalin [112] modeled on an electrostatic basis the association

of acids in SCW: HF, HF C – – C
2 , HNO3 , H HS , HS , HSO4 , H HCO3 , HCO3

among some others. The key variable determining the association constant is the
permittivity that depends on the thermodynamic conditions. It should be noted that
electrolyte solutions at appreciable concentrations have lower permittivities than
pure SCW or dilute solutions, hence the association should be enhanced relative to
low solute concentrations.
In SCW, as in common solvents and in water at ambient conditions, coordinate
bonds between cations and the electron pair donor oxygen atoms of the water
molecules and between anions and the hydrogen bond donor atoms are formed.
The number of available water molecules in SCW, depending on its density, the
self-association of the water molecules, and the competition between the hydration
of the ions and the association of oppositely charged ions are the factors that should
be considered. Sodium chloride has obtained the major attention in this respect as
for other properties of salts in SCW.
Cummings and coworkers [113, 114] obtained the pair correlation functions
g(ion–water, r), where the ion is NaC or Cl– from computer simulations in SCW
with the SPC water model. At the reduced temperature Tr D 1.05 and density
r D 1.00 a hydration number of four around each of the NaC and Cl– ions resulted
from the simulations. The excess of water molecules, N(R), in a correlation region
near a NaC or Cl– ion extending up to a distance R from its center, is:
Z R
N.R/ D W0 4 r 2 .g .ion water; r/ 1/ dr (1.37)
In the local regions extending to R corresponding to four water molecule

diameters there were 26 and 20 water molecules, respectively, having a density
manifold larger than the average bulk water density. The excess water density is
however somewhat smaller at a denser state, Tr D 1.00 and r D 1.50. The average
number of hydrogen bonds per water molecule, , was found to be independent of
the central particle, whether a water molecule or an ion, being near 1.6 at the lower
density and near 2.0 at the higher one.
Cui and Harris [115], applying the SPC model for water in their MD simulations
over the temperature interval from 427 to 727 ı C and at a density of 300 kgm–3
of SCW, found the excess water molecules to a cutoff distance of R D 1 nm
(approximately 3.6 water molecule diameters) diminishing near NaC ions from 7.2
to 4.4 and near Cl– ions from 10.1 to 4.9. The hydration numbers of the ions on
the basis of these numbers of excess water molecules were 4.3 ˙ 0.4 for NaC and
7.1 ˙ 1.7 for Cl– .
Gupta and Johnston [15] calculated the mean hydration numbers of the sodium
and chloride ions in SCW from a molecular thermodynamic model that included
physical interactions (repulsion and dispersion forces) and hydrogen bonding. At
a constant pressure of 30 MPa the hydration numbers diminish with increasing
temperatures from 400 to 850 ı C as the density of the SCW decreases from
350 to 60 kgm–3: for NaC from 5.7 to 3.2 and for Cl– from 5.4 to 1.9.
Bondarenko et al. [116] studied the structure of NaCl in SCW up to 500 ı C at

the constant pressure of 100 MPa with the SPC/E water model at a constant density.
The strong hydrogen bonding from the water to the Cl– anions were partly replaced
by ion pairing, and same-charge association (e.g., NaC -NaC ) was not excluded by
the results.
Flanagan et al. [117] used MD simulations with the SPC/E water model at
Tr D 1.05 and densities of 500, 290, and 87 kgm–3 to study the hydration structure
of several ions. The hydration numbers hi at the lower density were: LiC 4.1, NaC
4.8, KC 4.8, Be2C 4.0, Mg2C 6.0, Ca2C 8.0, F– 5.6, and Cl– 7.5, similar to the values
in ambient water. Luo and Tucker [118] studied the compression of the SCW as a
result of the electric field around a model ion (singly charged, in a cavity of 0.2 nm
radius) at three thermodynamic states of 1.01 Tr 1.30 and 0.875 r 0.50.
They concluded that the compression-induced hydration effects are confined to the
first two hydration shells of the ion.
Mesmer et al. [71] studied the hydration of the ions and undissociated molecules
of hydrogen chloride experimentally. At a density of 800 kgm–3 the hydration
number change for the dissociation of HCl ! HC C Cl– in SCW obtained from
Eq. 1.36 diminished slightly (from 15 to 14) as the temperature was raised from 400
to 800 ı C, but the numbers are larger at lower densities and much more strongly
temperature dependent.
Yamaguchi et al. [119] applied neutron diffraction with isotope substitution
to LiCl in SCHW (supercritical D2 O, Tc D 370.7 ı C) at Tr D 1.03, a method
that permits the individual ion-water distances and coordination numbers to be
determined. In 5 molal LiCl hydrogen bonding to the chloride anions by 2.8
deuterium atoms was found, diminishing to 2.5 in 3 molal LiCl.
1.10 Conclusions
Over the years both the experimental and the computational methods employed for
the study of the hydrogen bonding and other properties of SCW have been improved
significantly. Therefore, on the whole, the more recent results could be the more
reliable ones. The average number of hydrogen bonds per water molecule in the
system, <nHB >, increases with the pressure, hence with the density of the SCW at
a given temperature and diminishes with increasing temperatures at given pressures
and densities, as shown in Fig. 1.2. The fraction of monomeric, non-hydrogen-
bonded water molecules, f0 , increases in the opposite manner, i.e., increases with the
temperature and diminishes with the pressure, as shown in Table 1.1, most reliably
with the entries from [8]. The percolation threshold, <nHB > 0.77, is not exceeded
at temperatures > 600 ı C, and at lower temperatures requires a minimal density for
percolation to exist, e.g., 410 kg/m3 at 430 ı C.
The hydrogen bonding in SCW pertains to the ionic dissociation of the water,
because the products, H3 OC and OH– , are hydrated by hydrogen-bonded water
molecules. The ion product increases (pKW becomes smaller) as the pressure
34 Y. Marcus
increases but fewer ions are produced as the temperature of the SCW increases
(pKW becomes larger), Table 1.3. The solvent power of SCW depends on the
hydrogen bonding too, as shown by the Hansen solubility parameter ı hb , Table 1.4,
which is an important part of the total (Hildebrand) solubility parameter, ı H , that
governs the solubilities of non-ionic solutes. The solubility of salts depends on the
hydrogen bonding ability of water molecules to the anions and is described by a
three parameter expression, Eq. 1.34, which involves the hydration number of the
salt ions that associate to ion pairs as the permittivity of the water diminishes with
increasing temperatures, due to the diminished hydrogen bonding.
Appendix
Some physicochemical properties of SCW at the temperatures and pres-

sures applicable to bio-refining are required in order to benefit fully from
the book. The values recorded here are adapted from Marcus [5], where
references to the original publications are available. The critical tempera-
ture of water is Tc D 647.096 K D 373.946 ı C and its critical pressure is
Pc D 22.064 MPa D 217.8 atm. The critical density of water is c D 322 kgm–3
and its critical molar volume is Vc D 56.0 10–6 m3 mol–1 . Molar volumes, V,
corresponding to the densities, , are obtained by division of the molar mass of
water, 0.01802 kgmol–1 by the densities (and multiplication by 106 cm3 m–3 in
order to obtain the volumes in cm3 mol–1 ).
References
1. Gorbaty Yu E, Gupta RB. The structural features of liquid and supercritical water. Ind Eng
Chem Res. 1998;37:3026–35.
2. Hoffmann MM, Conradi MS. Are there hydrogen bonds in supercritical water? J Am Chem
Soc. 1997;119:3811–7.
3. Kohl W, Lindner HA, Franck EU. Raman spectra of water to 400 ı C and 3000 bar. Ber
Bunsenges Phys Chem. 1991;95:1586–93.
4. Blumberg RL, Stanley HE, Geiger A, Mausbach P. Connectivity of hydrogen bonds in liquid
water. J Chem Phys. 1984;80:5230–40.
5. Marcus Y. Supercritical water. A green solvent. New York: Wiley; 2012.
6. Kumar R, Schmidt JR, Skinner JL. Hydrogen bond definitions and dynamics in liquid water.
J Chem Phys. 2007;126:204107-1-12.
7. Voloshin VP, Naberukhin Y. Hydrogen bond lifetime distributions in computer-simulated
water. J Struct Chem. 2009;50:78–89.
8. Gupta RB, Panayiotou CG, Sanchez IC, Johnston KP. Theory of hydrogen bonding in
supercritical fluids. AICHE J. 1992;38:1243–53.
9. Smits PJ, Economou IG, Peters CJ, de Swaan Arons J. Equation of state description of thermo-
dynamic properties of near-critical and supercritical water. J Phys Chem. 1994;98:12080–5.
10. Huang SH, Radosz M. Equation of state for small, large, polydisperse, and associating
molecules. Ind Eng Chem Res. 1990;29:2284–94.
11. Vlachou T, Prinos I, Vera JH, Panayiotou CG. Nonrandom distribution of free volume in
fluids and their mixtures: hydrogen-bonded systems. Ind Eng Chem Res. 2002;41:1057–63.
12. Marcus Y. Supercritical water: relationships of certain measured properties to the extent
of hydrogen bonding obtained from a semi-empirical model. Phys Chem Chem Phys.
2000;2:1465–72.
13. Franck EU, Roth K. Infra-red absorption of HDO in water at high pressures and temperatures.
Discuss Faraday Soc. 1967;43:108–14.
14. Mountain RD. Molecular dynamics investigation of expanded water at elevated temperatures.
J Chem Phys. 1989;90:1866–70.
15. Gupta RB, Johnston KP. Lattice fluid hydrogen bonding model with a local segment density.
Fluid Phase Equilib. 1994;99:135–51.
16. Swiata-Wojcik D, Szala-Bilnik J. Transition from patchlike to clusterlike inhomogeneity
arising from hydrogen bonding in water. J Chem Phys. 2011;134:054121-1-9.
17. Dyer PJ, Cummings PT. Hydrogen bonding and induced dipole moments in water. Predictions
from the Gaussian charge polarizable model and Car-Parinello molecular dynamics. J Chem
Phys. 2006;125:144519-1-6.
18. Matubayasi N, Wakai C, Nakahara M. Structural study of supercritical water. I. Nuclear
magnetic resonance spectroscopy. J Chem Phys. 1997;107:9133–40.
19. Soper AK, Bruni F, Ricci MA. Site-site correlation functions of water from 25 to 400 ı C:
revised analysis of new and old diffraction data. J Chem Phys. 1997;106:247–54.
20. Kalinichev AG, Bass JD. Hydrogen bonding in supercritical water: a Monte Carlo simulation.
Chem Phys Lett. 1994;231:301–7.
21. Yoshii N, Yoshie H, Miura S, Okazaki S. A molecular dynamics study of sub- and
supercritical water using a polarizable potential model. J Chem Phys. 1998;109:4873–84.
22. Lamanna R, Delmelle M, Cannistrato S. Role of hydrogen-bond cooperativity and free-
volume fluctuations in the non-Arrhenius behavior of water self-diffusion: a continuity of
states model. Phys Rev E. 1994;49:2841–50.
23. Marcus Y. The structuredness of supercritical water up to 600 ı C and 100 MPa as obtained
from relative permittivity data. J Mol Liq. 1999;81:101–13.
24. Jedlovszky P, Brodhold JP, Bruni F, Ricci MA, Soper AK, Vallauri R. Analysis of the
hydrogen-bonded structure of water from ambient to supercritical conditions. J Chem Phys.
1998;108:8528–40.
25. Pártay L, Jedlovszky P. Line of percolation in supercritical water. J Chem Phys.
2005;123:024502-1-5.
26. Pártay L, Jedlovszky P, Brovchenko I, Oleinikova A. Percolation transitions in supercritical
water: a Monte Carlo simulation. J Phys Chem B. 2007;111:7603–9.
27. Bernabei M, Botti A, Bruni F, Ricci MA, Soper AK. Percolation and three-dimensional
structure of supercritical water. Phys Rev E. 2008;78:021505-1-9.
28. Bernabi M, Ricci MA. Percolation and clustering in supercritical aqueous fluids. J Phys Cond
Matt. 2008;20:494208-14-7.
29. Gorbaty Yu E, Dem’yanets Yu N. X-ray diffraction study of liquid and supercritical water
at high temperatures and pressures. I. Molecular functions of structurally sensitive scattering
components at 100 bar and 298–773 K. Zhur Strukt Khim. 1982;23:73–85.
30. Gorbaty Yu E, Dem’yanets Yu N. X-ray diffraction study of liquid and supercritical water at
high temperatures and pressures. II. Functions of the radial distribution of molecular density
and pair correlation functions. Zhur Strukt Khim. 1983;24:66–74.
31. Gorbaty Yu E, Dem’yanets Yu N. The pair correlation function of water at a pressure of
1000 bar in the temperature range 25–500 ı C. Chem Phys Lett. 1983;100:450–4.
32. Gorbaty YE, Kalinichev AG. Hydrogen bonding in supercritical water. 1. Experimental
results. J Phys Chem. 1995;99:5336–40.
33. Botti A, Bruni F, Ricci MA, Soper AK. Neutron diffraction study of high density supercritical
water. J Chem Phys. 1998;109:3180–4.
36 Y. Marcus
34. Tassaing T, Danten Y, Besnard M. Supercritical water: local order and molecular dynamics.
Pure Appl Chem. 2004;76:133–9.
35. Luck WAP. Zur Assoziation des Wassers. III Die Temperaturabhängigkeit der Wasserbanden
bis zum kritischen Punkt. Ber Bunsenges Phys Chem. 1965;60:626 C 63.5.
36. Luck WAP. Spectroscopic studies concerning the structure and the thermodynamic behaviour
of H2 O, CH3 OH, and C2 H5 OH. Discuss Faraday Soc. 1967;43:115–27.
37. Luck WAP, Ditter W. Die Temperatur-abhängigkeit der D2 O und HOD Spektren im nahen IR
bis in überkritische Bereiche. Z Naturforsch. 1969;24b:482–94.
38. Kandratsenka A, Schwarzer D, Vöhringer P. Relating linear vibrational spectroscopy to
condensed-phase hydrogen-bonded structures: liquid-to-supercritical water. J Chem Phys.
2008;128:244510-1-6.
39. Tassaing T, Garrain PA, Bégué D, Baraille I. On the cluster composition of supercritical
water combining molecular modeling and vibrational spectroscopic data. J Chem Phys.
2010;133:0324103-1-9.
40. Frantz JD, Dubessy J, Mysen B. An optical cell for Raman spectroscopic studies of
supercritical fluids and its application to the study of water to 500 ı C and 2000 bar. Chem
Geol. 1993;106:9–26.
41. Walrafen GE, Chu TC. Linearity between structural correlation length and correlated-proton
Raman intensity from amorphous ice and supercooled water up to dense supercritical steam.
J Phys Chem. 1995;99:1122511229.
42. Walrafen GE, Yang W-H, Chu YC. Raman spectra from saturated water vapor to the
supercritical fluid. J Phys Chem B. 1999;103:1332–8.
43. Yasaka Y, Kubo M, Matubayasi N, Nakahara M. High sensitivity Raman spectroscopy
of supercritical water and methanol over a wide range of density. Bull Chem Soc Jpn.
2007;80:1764–9.
44. Yui K, Uchida H, Itatani K, Koda S. Raman OH stretching frequency shifts in supercritical
water and in O2 – and acetone-aqueous solutions near the water critical point. Chem Phys Lett.
2009;477:85–9.
45. Wang Z, Pang Y, Dlott DD. Hydrogen-bond disruption by vibrational excitations in water.
J Phys Chem A. 2007;111:3196–208.
46. Wernet P, Testemale D, Hazemann J-L, et al. Spectroscopic characterization of microscopic
hydrogen-bonding disparities in supercritical water. J Chem Phys. 2005;123:154503-1-7.
47. Jonas J, DeFries T, Lamb WJ. NMR proton relaxation in compressed supercritical water.
J Chem Phys. 1978;68:2988–89.
48. Lamb WJ, Jonas J. NMR study of compressed supercritical water. J Chem Phys.
1981;74:913–21.
49. Matubayasi N, Wakai C, Nakahara M. NMR study of water in super- and subcritical
conditions. Phys Rev Lett. 1997;78:2573–6.
50. Yoshida K, Matubayasi N, Nakahara M. Solvation shell dynamics studied by molecular
dynamics simulation in relation to the translational and rotational dynamics of supercritical
water and benzene. J Chem Phys. 2007;127:174509-1-13.
51. Yoshida K, Matubayasi N, Uosaki Y, Nakahara M. Self-diffusion in supercritical water
and benzene in high-temperature high-pressure conditions studied by NMR and dynamic
solvation-shell model. J Phys Conf Ser. 2010;215:012093-1-6.
52. Matubayasi N, Nakao N, Nakahara N. Structural study of supercritical water. III. Rotational
dynamics. J Chem Phys. 2001;114:4107–15.
53. Sebastiani D, Parrinello M. Ab-initio study of NMR chemical shifts of water under normal
and supercritical conditions. ChemPhysChem. 2002;3:675–9.
54. Tsukahara T, Harada M, Tomiyasu H, Ikeda Y. 17 O Chemical shift and spin-lattice relaxation
measurements of water in liquid and supercritical states y using high-resolution multinuclear
NMR. J Supercrit Fluids. 2003;26:73–82.
55. Okada K, Yao M, Hiejima Y, Kohno H, Kajihara Y. Dielectric relaxation of water and heavy
water in the whole fluid phase. J Chem Phys. 1999;110:3026–36.
56. Yao M, Hiejima Y. Dielectric relaxation of supercritical water and methanol. J Mol Liq.
2002;96–97:207–20.
57. O’Shea SF, Tremaine PR. Thermodynamics of liquid and supercritical water to 900 ı C by a
Monte-Carlo method. J Phys Chem. 1980;84:3304–6.
58. Bastea S, Fried LE. Exp6-polar thermodynamics of dense supercritical water. J Chem Phys.
2008;128:174502–4.
59. Kalinichev AG. Monte Carlo simulations of water under supercritical conditions. I. Thermo-
dynamic and structural properties. Z Naturforsch. 1991;46a:433–44.
60. Kalinichev AG, Bass JD. Hydrogen bonding in supercritical water. 2. Computer simulations.
J Phys Chem A. 1997;101:9720–7.
61. Kalinichev AG, Churakov SV. Thermodynamics and structure of molecular clusters in
supercritical water. Fluid Phase Equilib. 2001;183–184:271–8.
62. Matubayasi N, Wakai C, Nakahara M. Structural study of supercritical water. II. Computer
simulations. J Chem Phys. 1999;110:8000–11.
63. Krishtal S, Kiselev M, Puhovski Y, et al. Study of the hydrogen bond network in sub- and
supercritical water by molecular dynamics simulation. Z Naturforsch. 2001;56a:579–84.
64. Mizan TI, Savage PE, Ziff RM. Molecular dynamics of supercritical water using a flexible
SPC model. J Phys Chem. 1994;98:13067–76.
65. Dang LX. Importance of polarization effects in modeling the hydrogen bond in water using
classical molecular dynamics techniques. J Phys Chem B. 1998;102:620–4.
66. Liew CC, Inomata H, Arai K, Saito S. Three-dimensional structure and hydrogen bonding
of water in sub- and supercritical regions: a molecular dynamics study. J Supercrit Fluids.
1998;13:83–91.
67. Petrenko VE, Antipova ML, Ved’ OV, Borovkov AV. Molecular dynamic simulation of sub-
and supercritical water with new interaction potential. Struct Chem. 2007;18:505–9.
68. Skarmoutsos I, Guardia E. Effect of the local hydrogen bonding network on the reorientational
and translational dynamics in supercritical water. J Chem Phys. 2010;132:074502-1-10.
69. Quist AS, Marshall WL. Electrical conductances of aqueous potassium nitrate and tetram-
ethylammonium bromide solutions to 800 deg. and 4000 bars. J Chem Eng Data.
1970;15:375–6.
70. Marshall WL, Franck EU. Ion product of water substance, 0–1000 ı C, 1–10,000 bars New
international formulation and its background. J Phys Chem Ref Data. 1981;10:295–304.
71. Mesmer RE, Marshall WL, Palmer DA, Simonson JM, Holmes HF. Thermodynamics of
aqueous association and ionization reactions at high temperatures and pressures. J Solut
Chem. 1988;17:699–718.
72. Tanger IV JC, Pitzer KS. Calculation of the ionization constant of H2 O to 2,273 K and
500 MPa. AICHE J. 1989;35:1631–8.
73. Tawa GJ, Pratt LR. Theoretical calculation of the water ion product KW . J Am Chem Soc.
1995;117:1625–8.
74. Bandura AV, Lvov SN. The ionization constant of water over wide ranges of temperature and
density. J Phys Chem Ref Data. 2006;35:15–30.
75. Takahashi H, Sato W, Hori T, Nitta T. An application of the novel quantum mechani-
cal/molecular mechanical method combined with the theory of energy representation: an ionic
dissociation of a water molecule in the supercritical water. J Chem Phys. 2005;122:044504-
1-9.
76. Halstead SJ, Masters AJ. A classical molecular dynamics study of anomalous ionic product
in near critical and supercritical water. Mol Phys. 2010;108:193–203.
77. Goldman S, Gray CG, Li W, et al. Predicting solubilities in supercritical fluids. J Phys Chem.
1996;100:7246–9.
78. Soave G. Equilibrium constants from a modified Redlich-Kwong equation of state. Chem Eng
Sci. 1972;27:1197–203.
79. Peng DY, Robinson DBA. A new two-constant equation of state. Ind Eng Chem Fundam.
1976;15:59–64.
38 Y. Marcus
80. Marcus Y. Are solubility parameters relevant to supercritical fluids? J Supercrit Fluids.
2006;38:7–16.
81. Marcus Y. Hansen solubility parameters for supercritical water. J Supercrit Fluids.
2012;62:60–4.
82. Williams LL, Rubin JB, Edwards HW. Calculation of Hansen solubility parameter values for
a range of pressure and temperature conditions, including the supercritical fluid region. Ind
Eng Chem Res. 2004;43:4967–72.
83. Morimoto M, Sato S, Takanohashi T. Conditions of supercritical water for good miscibility
with heavy oils. J Jpn Petrol Inst. 2010;53:61–2.
84. Bennett GE, Johnston KP. UV-visible spectroscopy of organic probes in supercritical water. J
Phys Chem. 1994;98:441–7.
85. Marshall WL, Jones EV. Liquid-vapor critical temperatures of several aqueous-organic and
organic-organic solution systems. J Inorg Nucl Chem. 1974;36:2319–23.
86. Fonseca TL, George HC, Coutinho K, Canuto S. Polarization and spectral shift of benzophe-
none in supercritical water. J Phys Chem A. 2009;113:5112–8.
87. Oka H, Kajimoto O. UV absorption solvatochromic shift of 4-nitroaniline in su7percritical
water. Phys Chem Chem Phys. 2003;5:2535–40.
88. Fujisawa T, Terazima M, Kimura Y. Solvent effects on the local structure of p-nitroaniline in
supercritical water and supercritical alcohols. J Phys Chem A. 2008;112:5515–26.
89. Minami K, Ohashi T, Suzuki M, et al. Estimation of local density augmentation and hydrogen
bonding between pyridazine and water under sub- and supercritical conditions using UV-vis
spectroscopy. Anal Sci. 2006;22:1417–23.
90. Xiang T, Johnston KP. Acid-base behavior of organic compounds in supercritical water. J Phys
Chem. 1994;98:7915–22.
91. Osada M, Toyoshima K, Mizutani T, et al. Estimation of the degree of hydrogen bonding
between quinoline and water by ultraviolet-visible absorbance spectroscopy in sub- and
supercritical water. J Chem Phys. 2003;118:4573–7.
92. Aizawa T, Kanakubo M, Ikushima Y, et al. Local density augmentation around acetophe-
none N, N, N0 , N0 -tetramethylbenzidine exciplex in supercritical water. Chem Phys Lett.
2004;393:31–5.
93. Aizawa T, Kanakubo M, Hiejima Y, et al. Temperature dependence of local density augmen-
tation for acetophenone N, N, N0 , N0 -tetramethylbenzidine exciplex in supercritical water.
J Phys Chem A. 2005;109:7375–8.
94. Minami K, Mizuta M, Suzuki M, et al. Determination of Kamlet-Taft solvent parameter *
of high pressure and supercritical water by UV-vis absorption spectral shift of 4-nitroaniline.
Phys Chem Chem Phys. 2006;8:2257–64.
95. Osawa K, Hamamoto T, Fujisawa T, et al. Raman spectroscopic study on the solvation of
p-aminobenzonitrile in supercritical water and methanol. J Phys Chem A. 2009;113:3143–54.
96. Takebayashi Y, Yoda Y, Sugeta T, et al. Acetone hydration in supercritical water: 13 C NMR
spectroscopy and Monte Carlo simulation. J Chem Phys. 2004;120:6100–10.
97. Valyashko V, Urusova M. Solubility behavior in ternary water-salt systems under sub- and
supercritical conditions. Monatsh Chem. 2003;134:679–92.
98. Valyashko VM. Phase equilibria of water-salt systems at high temperatures and pressures. In:
Palmer DA, Fernandez-Prini R, Havey AH, editors. Aqueous systems at elevated temperatures
and pressures. Amsterdam: Elsevier; 2004. p. 597–641.
99. Cohen-Adad R, Tranquard A, Perrone R, et al. The system water-sodium hydroxide. Compt
Rend C. 1960;251:2035–7.
100. Kerschbaum S, Franck EU. High pressure supercritical PVT-data of water-sodium hydroxide
mixtures and of liquid sodium hydroxide to 673 K and 400 MPa. Ber Bunsenges Phys Chem.
1995;99:624–32.
101. Eberz A, Franck EU. High pressure electrolyte conductivity of the homogeneous, fluid
water-sodium hydroxide system to 400 ı C and 3000 bar. Ber Bunsenges Phys Chem.
1995;99:1091–103.
102. Anderko A, Pitzer KS. Equation-of-state representation of phase equilibria and volumetric
properties of the system NaCl-H2 O above 573 K. Geochim Cosmochim Acta.
1993;57:1657–80.
103. Bischoff JL. Densities of liquids and vapors in boiling sodium chloride-water solutions: a
PVTX summary from 300ı to 500 ı C. Am J Sci. 1991;291:309–38.
104. Armellini FJ, Tester JW. Solubility of sodium chloride and sulfate in sub- and supercritical
water vapor from 450-550 ı C and 100–250 bar. Fluid Phase Equilib. 1993;84:123–42.
105. DiPippo MM, Sako K, Tester JW. Ternary phase equilibria for the sodium chloride-sodium
sulfate-water system at 200 and 250 bar up to 400 ı C. Fluid Phase Equilib. 1999;157:229–55.
106. Dell’Orco P, Eaton H, Reynolds T, Buelow S. The solubility of 1:1 nitrate electrolytes in
supercritical water. J Supercrit Fluids. 1995;8:217–27.
107. Leusbrock I, Metz S, Rexwinkel G, Versteeg GF. Quantitative approaches for the description
of solubilities of inorganic compounds in near-critical and supercritical water. J Supercrit
Fluids. 2008;47:117–27.
108. Leusbrock I, Metz S, Rexwinkel G, Versteeg GF. Solubility of 1:1 alkali nitrates and chlorides
in near-critical and supercritical water. J Chem Eng Data. 2009;54:3215–23.
109. Leusbrock I, Metz S, Rexwinkel G, Versteeg GF. The solubility of magnesium chloride and
calcium chloride in near-critical and supercritical water. J Supercrit Fluids. 2010;53:17–24.
110. Leusbrock I, Metz S, Rexwinkel G, Versteeg GF. The solubility of phosphate and sulfate salts
in supercritical water. J Supercrit Fluids. 2010;54:1–8.
111. Marshall WL. Complete equilibrium constants, electrolyte equilibria, and reaction rates.
J Phys Chem. 1970;74:346–55.
112. Ryzhenko BN, Bryzgalin OV. Electrolytic dissociation of acids during hydrothermal pro-
cesses. Geokhimiya. 1987;1:137–42.
113. Cummings PT, Cochran HD, Simonson JM, Mesmer RE, Karaboni S. Simulation of
supercritical water and of supercritical aqueous solutions. J Chem Phys. 1991;94:5606–21.
114. Cochran HD, Cummings PT, Karaboni S. Solvation in supercritical water. Fluid Phase
Equilib. 1992;71:1–16.
115. Cui ST, Harris JG. Ion association and liquid structure in supercritical water solutions of
sodium chloride: a microscopic view from molecular dynamics simulations. Chem Eng Sci.
1994;49:2749–63.
116. Bondarenko GV, Gorbaty Yu E, Okhulkov AV, Kalinichev GA. Structure and hydrogen
bonding in liquid and supercritical aqueous NaCl solutions at a pressure of 1000 bar and
temperatures up to 500 ı C: a comprehensive experimental and computational study. J Phys
Chem A. 2006;110:4042–52.
117. Flanagan LW, Balbuena PB, Johnston KP, Rossky PJ. Ion solvation in supercritical water
based on adsorption analogy. J Phys Chem B. 1997;101:7998–8005.
118. Luo H, Tucker SC. A case against large cluster-induced nonequilibrium solvent effects in
supercritical water. Theor Chem Acc. 1997;96:84–91.
119. Yamaguchi T, Ohzono H, Yamagami M, et al. Ion hydration in aqueous solutions of lithium
chloride, nickel chloride, and cesium chloride in ambient to supercritical water. J Mol Liq.
2010;153:2–8.
Chapter 2
Phase Behavior of Biomass Components
in Supercritical Water
Sergey Artemenko, Victor Mazur, and Pieter Krijgsman
Success depends upon previous preparation, and without such

preparation there is sure to be failure.
Confucius
Abstract The importance of thermodynamic and phase behavior is fundamental

to supercritical water (SCW) technologies of biomass treatment. Considering the
extremely large number of biomass components, it is obvious that there is need
for developing theoretically sound methods of the prompt estimation of their phase
behavior in aquatic media at supercritical conditions. A local mapping concept is
introduced to describe thermodynamically consistently the saturation curve of water
and biomass components. The global phase diagram studies of binary mixtures
provide some basic ideas of how the required methods can be developed to visualize
the phase behavior of biomass decomposition products in supercritical aqueous
media. The mapping of the global equilibrium surface in the parameter space of the
equation of state (EoS) model provides the most comprehensive system of criteria
for predicting binary mixture phase behavior. Analytical expressions for selection
the azeotropic states in binary mixtures are given. Results of calculations of phase
equilibria and critical curves for main biomass components in supercritical water
are described.
S. Artemenko • V. Mazur ()

Institute of Refrigeration, Cryotechnologies, and Eco-Power Engineering (former Academy of
Refrigeration), Odessa National Academy of Food Technologies, 1/3 Dvoryanskaya Str., 65082
Odessa, Ukraine
P. Krijgsman
Ceramic Oxides International, Wapenveld, The Netherlands
42 S. Artemenko et al.
Keywords Biomass components • Supercritical water • Widom line • Phase

equilibria • Global phase diagram • Parameter identification
2.1 Introduction
Biomass is the prevailing form of renewable energy which can provide the con-
version to heat and power, biofuels and biobased chemicals as the replacement of
conventional fossil. Three key macromolecules: lignin, cellulose, and triacylglyc-
erides can generate the wide opportunities to transform biomass into goal products –
energy and desired materials [1]. Proteins are not primary components of biomass
and account for a lower proportion than do the previous three macromolecules.
Protein properties depend on the kinds and ratios of constituent amino acids,
and the degree of polymerization. The amounts of the other organic components
vary widely depending on specie, but there are also organic components. Biomass
comprises organic macromolecular compounds, but it also contains inorganic
substances (ash) in trace amounts. The primary metal elements include Ca, K, P,
Mg, Si, Al, Fe, and Na.
There are many classical technologies that ensure the fulfillment of biomass
conversion problem but modern challenges of sustainable development require the
emergence of environmentally friendly green technologies. Water in supercritical
state i.e. at temperature and pressure high enough to vanish differences between
liquid and vapor phases, above 647 K and 22 MPa, is recognized as green solvent.
Thermodynamic and phase behavior of SCW are differed qualitatively from the
“normal” state (for example it can act as a quasi-organic solvent). Not only
the chemistry, but also the physics can change drastically. For example, density
variation with a factor of 7 for 25 MPa around a few Kelvin range in the supercritical
region across the Widom line representing the maximum of response functions can
change heat capacity and compressibility with several order of magnitudes, making
almost impossible to use traditional flow- and thermal transport calculations.
Development of new SCW technologies is impeded by the lack of fundamental
understanding of many aspects of the supercritical fluid state itself, and particularly
of their thermodynamic and phase behavior. Properties and phase equilibria of fluid
mixtures define the scientific platform for developing green chemical processes and
predetermine the success of emergent technologies [2].
The main aim of this Chapter is to review existing experimental data and
theoretical models for analyzing, correlating, and predicting the phase behavior of
biomass components in supercritical water.
The Chapter is organized as follows. In the first section, main biomass com-
ponents are selected. Critical (or pseudocritical) parameters are considered as
fundamental characteristics of a fluid by themselves and also often as the input from
which the other physical properties can be generated. The quantitative structure-
property relationships (QSPRs) to estimate the critical properties of biomass
components are discussed. The boundaries of the Widom line for supercritical water
2 Phase Behavior of Biomass Components in Supercritical Water 43
where reaction mechanisms accelerate the biomass conversion maximally are given.
In the second part, a theoretical analysis of the topology of phase diagrams as useful
tool for understanding the phenomena of phase equilibrium that are observed in
biomass components – supercritical water mixtures in vicinity of the Widom line
is given. We review the global phase behavior of binary mixtures and derive an
analytical expression for determining the boundaries of azeotropy in terms of the
critical parameters of mixture components and the binary interaction parameter
k12 for the one fluid models of the equation of state. The knowledge of binary
interaction gives possibility to predict all types of phase behavior in supercritical
water – biomass component mixtures. The results of simulation of phase equilibria
for binary mixtures are illustrated using examples of various classes of biomass
components.
2.2 Main Biomass Components
The estimation of the properties of main biomass components is an important

prerequisite for the design of processes and equipment, environmental impact
assessment and other major chemical engineering activities related to phase equi-
libria. The main components of biomass resources are typically 40–45 wt%
cellulose, 25–35 wt%, hemicellulose, 15–30 wt% lignin and up to 10 wt% for other
compounds [3]. Elemental composition of conventional biomass includes following
components: C, H, O, N, and S. Chemical formula Ca Hb Oc Nd Se simulates the
various types of biomass, e.g., sewage sludge C2.83 H4.86 O1.25 N0.34 S0.04 , microalga
Spirulina C3.66 H6.81 O2.16 N0.47 S0.01 , the zoomass CH2.06 O0.52 N0.12 S0.01 , and others
[4]. Possible reaction products depend on goals of biomass conversion and thermo-
dynamic conditions of SCW treatment processes. The gasification of biomass and
organic wastes generates the gas components such as H2 O, CO2 , CO, N2 , N2 O, NO,
NO2 , SO2 , SO3 , H2 , CH4 , and hydrocarbons (HC). Biomass products produced from
different materials and by different external conditions may differ greatly from one
another. As a result, the compositions of different reaction products usually vary in
wide ranges.
To establish the thermodynamic properties of great variety of compounds under
SCW treatment the simple models of the EoS P D P (T,V,X), e.g., the Peng-Robinson
(PR) [5] or the Redlich-Kwong-Soave (SRK) [6, 7] models are a better choice to link
key characteristics of biomass components and target properties. Critical parameters
of biomass components can be considered as their information characteristics which
could generate a set of target properties for designed SCW system. This class of
EoS has simple relationships between their model parameters and critical constants
derived from critical conditions. During the many years of super critical research,
a variety of fluids have been investigated, and accordingly, both their critical
temperature and pressure have been determined. An exposure of these physical
properties is condensed in Table 2.1.
Table 2.1 Critical properties of selected chemical compounds (sorted by critical temperatures)
Critical Critical pressure Critical density Acentric
Product components temperature, (K) (MPa) (kg/m3 ) factor
Hydrogen 33:145 1:296 31:26 0:219
Nitrogen 126:19 3:396 313 0:0372
Carbon monoxide 132:86 3:444 303:91 0:0497
Oxygen 154:60 5:013 436:14 0:0222
Methane 190:56 4:599 162:66 0:0114
Ethylene 282:16 5:042 214:25 0:0866
Carbon dioxide 304:13 7:377 467:6 0:2239
Ethane 305:46 4:872 206:18 0:0995
Propane 309:52 4:251 228:48 0:1521
Nitrous oxide 369:90 7:245 452 0:1613
Ammonia 405:40 11:333 225 0:2560
Sulfur dioxide 430:64 7:884 525 0:2557
Sulfuric oxide 491:00 8:200 630 0:4810
Ethanol 513:90 6:148 276 0:6440
Water 647:096 22:064 322 0:3443
To define more exactly equation of state parameters for pure substances,

especially for water, concept of local mapping is developed. Standard approach
to determination of the EoS parameters X D X (a, b, ˛) is derived from critical
conditions. But for phase equilibria prediction this parameter set is not sufficient to
satisfy the limiting conditions for pure components (x ! 1 and x ! 0). To provide a
reliable thermodynamic consistency between exact and model equations of state the
equalities of pressures, isothermal compressibility’s and internal energies should be
applied. As an example, a refinement of the SRK EoS parameters is considered. To
satisfy thermodynamic consistency conditions at given point of P – V – T surface
the set of equations (2.1) should be solved:
Pexact .T; V / Pmod .X; T; V / D 0;

@Pexact .T; V / @Pmod .X; T; V /
D 0;
@T @T
@Pexact .T; V / @Pmod .X; T; V /
D0 (2.1)
@V @V
where PÈØÃÔt – fundamental EoS for H2 O [8], Pmod – the SRK model [7], ¸ – vector
of the SRK model parameters, ˛(T/Tc,!) as function of reduced temperature and
acentric factor (!) is the attractive term in the original Redlich–Kwong equation.
Results of calculations, which exactly describe data along saturation curve, are
presented in Table 2.2.
More sound consideration of problem should be guided on quantitative “struc-
ture – property” relationships (QSPR). The basic idea of QSPR is to find a
Table 2.2 Parameters of the SRK EoS for water

T, K P, MPa Vl , cm3 /mole aH2 O ; Jm3 /mole2 bH2 O , cm3 /mole ˛
400 0:24577 19.217 0.49846 14.542 0.92771
450 0:93220 20.234 0.44438 14.612 1.0563
500 2:6392 21.671 0.42379 14.628 1.0922
550 6:1172 23.836 0.40995 14.515 1.0812
600 12:345 27.741 0.39366 14.141 1.0466
relationship between the structure of compound expressed in terms of constitutional,

geometrical, topological, and other quantum-chemistry descriptors and target prop-
erty of interest.
The QSPR employs two databases – the critical property database and structure
database. The correlations between databases are established in the form of the
property model – M(P), the parameters of which are determined by minimizing the
“distance” between the experimental property P j and its model Mj. For the most of
compounds experimental data are available via on-line databanks e.g., NIST, KDB,
CHEMSAFE, BEILSTEIN, GMELIN et al. If direct measurement results are not in
option, thermodynamic models from process simulators like ASPEN PLUS can be
partly used to estimate the missing properties.
There are many group contribution methods estimating the critical properties of
possible biomass compounds from molecular structure and success of any model
depends on the amount of data used in determining the contribution of independent
variables (molecular descriptors). The start point of group contribution technique is
a decomposition of the molecular structure into particular groups and the counting of
atoms in those groups. Increments are assigned to the groups by regression of known
experimental data for the chosen property. The molecular structure can be retrieved
by summation of the contributions of all groups. The least sophisticated atom count
technique was proposed by Joback [9]. More general cases for molecular descriptors
determination are described elsewhere, e.g. [10, 11]. At present time reliable
correlations are established by heuristic, Multi-Linear Regression techniques, e.g.
[12], by nonlinear techniques like Genetic Function Approximation [13] or by
Artificial Neural Networks, e.g. [14]. The choice of appropriate technique to predict
critical properties from molecular structure depends on the statement problem and
should correspond to the final aim of molecular design. Distribution of critical points
for hydrocarbons and key biomass components versus saturation curve of water are
given in Fig. 2.1.
Critical temperature and pressure data for the hydrocarbons were taken from
correlation Wakeham et al. [15]. Data for selected chemical compounds were
taken from Table 2.1. Critical (or pseudocritical) parameters are considered as
fundamental characteristics of a fluid by themselves and also often as the input
from which the other physical properties can be generated. The chain starting from
descriptors of molecular structure to target property of biomass components should
be constructed through critical parameters of technological fluid and equation of
state model, accordingly.
25
C
20
Pressure, MPa
15
10
0
200 400 600 800
Temperature, K
Fig. 2.1 Distribution of critical points of hydrocarbons ( ) and selected chemical compounds
() versus location of H2 O saturation curve
2.3 SCW Thermodynamic Behavior
The critical point defines the vertex of the vaporization boundary; hence, geometric
considerations ensure that isothermal compressibility, isobaric expansion and other
specific partial derivatives will diverge to infinity at critical conditions. This
inherent divergence of derivatives effectively controls not only the thermodynamic,
electrostatic, and transport properties of water, but they also influence on transport
and chemical processes in SCW systems. The behavior of water as solvent, with
some simplification, one can describe by following way while the subcritical water
is good solvent for inorganic and bad solvent for organic materials, it turns into the
opposite in the SC region; it could dissolve several organic materials but could not
dissolve some inorganic ones. By using pure SCW, after days or months various
corrosion products can still be dissolved into it.
Both compressibility and thermal expansion behaviour are very important factors
in order to understand the shifts of the equilibrium constant in supercritical solvents
as a system response on both thermal and mechanical perturbations. The coefficient
of isothermal compressibility is defined as:

1 ı
ˇD ; (2.2)
ıp T
and diverges sharply (critical exponent 1.241) from ideal gas values to infinity at
the critical point. It is an interesting fact that the ˇ divergence depends on pathways
of approaching to the critical state along the critical isochore (sharpest slope) or
0.4
0.35
Isothermal compressibility, 1/MPa
0.3
0.25
0.2
0.15
0.1
0.05
0
800
750 20
30
700 40
50
60
650 70
80
90
600 100
Pressure, MPa
Temperature, K
Fig. 2.2 ˇ – p – T – surface of supercritical water
along any other pathway in the p-T plane. There are three major regions in which it
is possible to mark the p-T plane (Fig. 2.2):
• the low-density region along an isobar at T > 723 K where the isothermal
compressibility decreases with both pressure and temperature;
• a region to the low-density side along an isotherm where the isothermal
compressibility increases with pressure and decreases with temperature;
• a region to the high-density side along an isobar where the isothermal compress-
ibility decreases with pressure and increases with temperature.
In supercritical fluid, the isothermal compressibility in the low-density region is
typically several times larger than those in the high-density region. The coefficient
of isobaric expansion is defined as:

1 ı
˛T D : (2.3)
ıT p
The qualitative behaviour of the ˛ T (p,T) and ˇ(p,T) surfaces is similar, but
relative maxima of ˛ T along the isotherms and isobars are shifted to larger densities
relative to ˇ. Isochoric heat capacity behaviour, like isothermal compressibility and

isobaric expansion, changes from ideal gas values to infinity at the critical point.
In contrast to the strong divergence of the isothermal compressibility and isobaric
expansion, the CV has a near logarithmic divergence. Supercritical maxima of
isochoric heat capacities in the pressure – temperature diagram tend to go to the
side of decreasing density. At pressures less than 30 MPa, and temperatures less
than 673.15 K, the surface CV (p,T) looks like to the isothermal compressibility ˇ.
Isobaric heat capacity from thermodynamic relationships is
˛T2
CP D CV C T : (2.4)
ˇ
The near-critical region has the similar asymptotic behaviour for all fluids and
theoretically is described by the universal critical exponents and scaling functions,
and is based on the scaling concept. The power laws, that represent asymptotic
behaviour along the iso-lines on thermodynamic surfaces, are described in numerous
literature sources. The main conclusion is that – H2 O behaviour is not unusual in
critical region and has the same anomalies as all other fluids.
2.4 Local Extrema of Thermodynamic Properties.

The Widom Line
Thermal effects and volume changes of reactions are linearly dependent on com-
pressibility behavior of fluids. Local extrema in ˛ T , ˇ and have a significant
influence on the nature of both heat and mass transport processes. As an example,
for the design of high-pressure units for SCW biomass treatment the local rate of
convection heat transfer is proportional to both fluid flux and CP . Usually fluid flux
is also proportional to ˛ T and . Therefore, the maximum flow rates occur at state
conditions intermediate between ˛ T and maxima. Because CP is a function of ˛ T ,
its extreme projections are nearly coincident. As a result, convection heat transfer
rates in supercritical conditions attain local maxima in the vicinity of ˛ T extremes.
Near the critical point, the extrema lines for response functions are merged. An
asymptotic line as continuation of vapour pressure curve to the one phase region is
often named – the Widom line often referred as pseudo-critical or pseudo-spinodal
line. The extrema lines (isobaric heat capacity – CP , compressibility coefficients –
˛ T and ˇ, speed of sound – a, inflection points – IP, the Joule – Thomson
coefficient – œ) were calculated on the basis of the Wagner and Pruss equation of
state [8]. A quantitative presentation of extreme behavior in the properties of H2 O
in the P – T diagram is illustrated in Fig. 2.3.
The pressure range, where all extrema lines are merged, is located within 22
: : : 25 MPa. Similar picture giving the same lines for water with fittings based on
IAPWS EoS is considered in [16]. The main conclusion from these calculations is
as follows. No universal curve as continuation of vapor pressure curve exists.
Fig. 2.3 Extrema of response functions in the P – T diagram
In practical design of SCW units, we need to consider non-purity of water. It

leads to a shift of extrema lines in the pressure – temperature plane. The typical
impurity in supercritical water is a small concentration of NaCl. To study the
changes in behaviour along the Widom line, we have calculated the shift of the
max CP line at different concentrations (100, 200, and 300 ppm) [17]. We suggested
that at small concentrations of NaCl in water the correspondence state principle is
working. The NaCl additives displace the position of critical point along critical line
of aqueous solution (Fig. 2.4). Detailed analysis of data was given by Shibue [18].
Critical temperatures and critical pressures of aqueous NaCl solutions also were
taken from [18].
To those versed in the art, it is known that enhanced solubility is able to accelerate
reaction rates, particularly supercritical water can be a medium for either ionic or
single-radical chemistry. A variety of compounds are hardly soluble, if at all, in a
medium at room temperature, however they could become soluble in super critical
media. Contrary to this observation, those compounds that are soluble in media at
room temperature could become less soluble at supercritical conditions.
Solubility growth in water – salt systems at supercritical conditions plays a
principal role for the SCW biomass treatment and other industrial applications.
Experimental data on solubility of solids in supercritical water have testified that
pressure increases the solubility of most of the solids above the H2 O critical
temperature (systems H2 O–SiO2 , H2 O–Na2 SO4 , H2 O–Li2 SO4 , etc. are examples
confirming this observation). At the present time, there are many computer programs
for modelling salt/water systems, taking into account the irreversible dissolution
Fig. 2.4 Shift of critical point and max CP lines in aqueous solution of NaCl [17]
of the reactants and the reversible precipitation of secondary products. From

a mathematical point of view, the most successful approach to calculate the
equilibrium assemblage is based upon a solution of a set of non-linear mass-action
equations, involving equilibrium constants for all relevant equilibrium and auxiliary
sets of linear mass-balance and charge balance equations. The primary aim of
computer programs is to calculate the speciation of aqueous solutions and their
equilibrium to salts (e.g., http://h2o.usgs.gov/software). Most comprehensive review
of experimental data on aqueous phase equilibria and solution properties at elevated
temperatures and pressures is given by Valyashko [19].
2.5 Phase Equilibria in Binary Mixtures.

Global Phase Diagram
Theoretical analysis of the topology of phase diagrams is a very useful tool for
understanding the phenomena of phase equilibrium that are observed in biomass
components – supercritical water mixtures in vicinity of the Widom line. The
pioneering work of van Konynenburg and Scott [20] demonstrated that the van
der Waals one-fluid model has opened opportunities of qualitative reproducing the
main types of phase diagrams of binary fluids. The proposed classification was
successful, and is now used as a basis for describing the different types of phase
behavior in binary mixtures. At present, the topological analysis of equilibrium
surfaces of binary fluid systems contains 26 singularities and 56 scenarios of
evolution of the p–T diagrams [21]. The introduction of liquid–solid equilibria into
classification schemes makes it possible to outline a strategy of continuous topo-
logical transformation for constructing complete phase diagrams. The mapping of
the global surface of a thermodynamic equilibrium onto the space of parameters of
an equation of state gives the possibility to obtain the most comprehensive system
of criteria for predicting of the binary mixture phase behaviour. The influence
of critical parameters of compounds on phase topology is visualized via global
phase diagrams. Such diagrams are presented not in the pressure – temperature
variables but in the space of the equation of state parameters, e.g. the van der Waals
geometric – b and energetic – a parameters or critical parameters.
For understanding and classifying a great number of phase diagrams, the
determination of critical points is an important practical and theoretical problem.
A mixture of a given composition can have one, more than one, or none critical
point. The types of phase behavior that are of interest for the SCW technologies are
characterized as follows (Fig. 2.5).
(I) The simplest type that has a continuous critical curve between the two
critical points C1 and C2 . It can shape of the critical curve and the position
of the azeotropic line.
(II) This type is characterized by the presence of an immiscibility zone at
temperatures below the critical temperature of a more volatile component,
by a critical curve that connects two critical points of pure components, and
by a critical line that starts from the critical end point where the liquid–liquid
equilibrium line ends.
(III) This type comprises two different critical curves. One of them starts from
the critical point of a pure component with a higher critical temperature
and goes to the range of high pressures. The other critical curve starts from
the critical point of the other component and leads to the critical end point
at the end of the three-phase line. The type is divided into five subtypes.
The main subtypes differ in the arrangement of the three-phase line at
pressures above the saturation pressure of components and the azeotropic
line that is bounded by the azeotropic end point from below and by the
critical azeotropic point from above (subtype III–A). A distinctive feature
for subtype III–H is the occurrence of azeotropy.
(IV) Type IV is characterized by two curves of the liquid–liquid–gas equilibrium.
The high temperature three-phase line is bounded by two critical end points
(lower (LCEP) and upper (UCEP)). In the vicinity of the UCEP, the solution
becomes immiscible with decreasing temperature. In the region of the
LCEP, an immiscibility zone appears with increasing temperature.
(V) This type resembles type IV which has no liquid–liquid critical line and
three-phase line at low temperatures. For this type, the occurrence of
azeotropic states and multiextremal critical lines is possible.
(VI) This type of phase behavior is characterized by the liquid–vapor critical line
that connects two critical points of pure components and by the liquid–liquid
critical line with a pressure peak that connects the UCEP and the LCEP in
the three-phase line.
Fig. 2.5 Types of phase diagrams: C1 are critical points of pure components; CE1 mean the critical
end points; the continuous line denotes the equilibrium curves of pure components; the dashed lines
correspond the critical lines; and the dash-dotted lines denote the three-phase equilibrium curves
(VII) This type, unlike I–VI, has not been confirmed experimentally and differs
from type VI by the behavior of the liquid–vapor critical curve, which is
divided into two lines that start at the critical points of pure components and
end at the LCEP and UCEP in the second three-phase line.
(VIII) This type is characterized by three critical lines. One critical line starts at
the critical point of one of the pure components and goes towards the range
of high pressures as in type III. The other critical curves start at the LCEP
and UCEP in the three-phase line and end in the region of infinitely high
pressures and the critical point of the other pure component, respectively.
The systems of interests, for example, H2 O C n-alkanes, H2 O C CO2 belong to
the type III phase diagrams according to van Konynenburg and Scott [20], with
three phase equilibrium and interrupted critical line, of which the lower branch
connects the critical point of n-alkane and the upper critical end point and the
upper branch runs from the critical point of pure water to high pressures (valid
to water – n – hexacosane). Other systems include H2 , CH4 , CO2 and etc. Figs. 2.6
and 2.7 illustrate collection of experimentally determined critical curves of aqueous
systems.
The system SCW – O2 . The critical curve, an envelope of the isopleths, begins
at the critical point of water (647 K), has a temperature minimum (639 K) at about
75 MPa and proceeds to 250 MPa at 659 K. Excess volume, VE , values have been
calculated for 673 K from 30 to 250 MPa. All VE values are positive. The maximum
is 57 cm3 mol1 at 70 mol % of H2 O at 30 MPa and about 2 cm3 mol1 at 40 mol %
H2 O and 250 MPa [22].
The water-carbon dioxide system exhibits type III phase behavior in the classifi-
cation of van Konynenburg and Scott [20] with a discontinuous vapor–liquid critical
curve, a wide region of liquid–liquid coexistence below the critical temperature
of CO2 , and very limited mutual solubility in the regions of two- and three-phase
equilibria. This system was studied by Takenouchi and Kennedy [23] to pressures
of 160 MPa and at temperatures of 383–623 K. The critical curve of the system
trends toward higher pressures at lower temperatures and departs strongly from the
critical point of pure water. At low pressures the CO2 rich phase is the light phase,
but at higher pressures this phase is the denser fluid phase. In a natural system of
H2 O–CO2 complete miscibility will not exist below 538 K; at higher temperatures a
completely mixed supercritical fluid may exist, but at lower temperatures this fluid
will segregate into two fluid phases. Fluid-fluid equilibria and critical curves in the
system H2 O–CO2 have been studied, among others by Tödheide and Franck [24],
Heidemann and Khalil [25] and Blencoe et al. [26].
The water – nitrogen system has been investigated by Tsiklis and Maslennikova
[27], Prokhorova and Tsiklis [28], Japas and Franck [29]. Despite some differences
in p – T – x conditions, the phase diagram topology of the three systems is not
changed. The available data indicate that immiscibility is generally restricted to
temperature below about 673 K. The critical curves consistently exhibit temperature
minima. These minima occur at 60–70 MPa and 638 K in the H2 O–N2 system
[29], and in vicinity 155–190 MPa and 538 K in the H2 O–CO2 system [25].
Fig. 2.6 Experimental P – T data along critical curve for the SCW – key biomass component
binary mixtures
The system SCW – H2 has been studied isochorically from 0.5 to 90 mol-%
H2 and up to 713 K and 250 MPa pressure using an autoclave containing two
sapphire windows through which phase transitions could be observed at elevated
temperatures and pressures. The system was found to exhibit so-called “gas-gas”
immiscibility with a critical curve proceeding to higher temperatures and pressures
from the critical point of pure water. Within the range of these experiments, the
Fig. 2.7 Collection of experimentally determined critical curves of aqueous systems
critical temperature of H2 –H2 O mixtures does not change any noticeable from that
of pure water (e.g. Tc D 654.5 K at pc D 25.2 MPa for 38 mol % H2 ) [30].
The system water – ethanol at temperatures up to 673.15 K, including the
saturation curve and the critical and supercritical regions, and at pressures up to
50 MPa for ethanol concentrations of 0.2, 0.5, and 0.8 mole fractions was studied
particularly by Abdurashidova et al. [31]. The data of p, , T, x – measurements are
used to determine the critical parameters of mixtures. The thermal decomposition
of ethanol molecules is observed at a temperature above 623.15 K [31–34]. Both
systems water – ammonia [53–55] and water – ethanol [31–34] are a classic example
of simplest type I of phase behavior. We should emphasize that, although presented
in the context of key components, the topics tackled could apply to thermodynamic
models of other biomass components.
Theoretical comprehension of the topology of fluid phase equilibrium has proven
to be very useful for the description of complex fluid phase relations, which are
observed in multiple-component systems. Indeed, whereas the phase diagram of
a one-component system is very simple, at least eight different types of phase
diagrams have already been observed for binary systems. Many studies have shown
that equations of state are able to generate the different kinds of fluid phase
equilibrium (liquid-vapour, liquid-liquid, gas-gas and liquid-liquid-gas). The first
pioneering work of this type was the study of van Konynenburg and Scott [20].
They have shown that the simplest van der Waals model reproduces qualitatively the
major types of phase diagrams of binary fluids. A classification was proposed and is
currently used as a basis, for the discrimination of different kinds of phase diagrams,
in binary systems. Following Varchenko’s approach [35], generic phenomena

encountered in binary mixtures when the pressure p and the temperature T change,
correspond to singularities of the convex envelope (with respect to the x variable)
of the “front” (a multifunction of the variable x) representing the Gibbs potential
G(p,T,x). Pressure p and temperature T play the role of external model parameters
like k12 . Although there are no theoretical limits in the extension of this approach for
three- or multiple-component mixtures, classification studies – even for ternaries –
are just at their beginning because of the amount of work that has to be carried out.
Our aim is to recognise a wide variety of phase diagrams from analysis of
variations in geometry and energy characteristics (e.g., in critical volume and critical
temperature) of mixture components. The influence of these two parameters on the
phase diagram topology could be conveniently visualised on the master diagram,
called a “global phase diagram”. Such a diagram shows the different areas of
occurrence of the possible phase diagrams as a function of the geometry and energy
factors of the compounds used. Ever since the work of van Konynenburg and Scott
[20], numerous studies have been carried out on other, more realistic EoS [36, 37].
The global phase diagrams of such different models, as the one-fluid EoS of binary
Lennard-Jones fluids [36] and the Redlich-Kwong model [37], are almost identical
in their main features including such a sensitive phenomenon as the closed-loops of
liquid-liquid immiscibility. Therefore, most of the considerations and conclusions
made on the basis of one of the realistic models can be transferred vice versa.
The boundaries, between the various types in the global phase diagram, can
be calculated directly using the thermodynamic description of the boundary states
(tri-critical line, double critical end-points, etc.). The dimensionless co-ordinates
that are used for the representation of the boundary states, depend on the equation
of state model, but normally they are designed similar to those proposed by van
Konynenburg and Scott for the van der Waals model [20]. In this case, the global
phase diagrams of all realistic models have a very similar structure, in particular for
the case of equal sized molecules. We consider here the classical Redlich-Kwong
model as an example.
The Redlich-Kwong EoS [6] is used in its classical, non-modified, form:
RT a
pD ; (2.5)
.V b/ T 0:5 V .V b/
where R is the universal gas constant, and parameters a and b of the mixture
EoS depend on the mole fractions xi and xj of the components i and j and on the
corresponding parameters aij and bij for different pairs of interacting molecules:
X
2 X
2
aD xi xj aij
i D1 j D1
(2.6)
X 2 X
bD xi xj bij :
i D1
The set of dimensionless parameters for the Redlich-Kwong model is as follows

[37]:
d22 d11
Z1 D ;
d22 C d11
d22 2d12 C d11
Z2 D ;
d22 C d11
b22 b11
Z3 D ;
b22 C b11
b22 2b12 C b11
Z4 D (2.7)
b22 C b11
where
2
. " !#1 1.
Tij bij ˝b aij 3 1. 2 3 1
dij D ; Tij D ; ˝a D 9 2 3 1 ; ˝b D :
bi i bjj R˝a bij 3
(2.8)
It should be brought to the attention that the dimensionless parameter Z1

represents the difference of the critical pressures of the components, and that the
dimensionless parameter Z3 represents the difference of the critical volumes. So,
there is a direct correlation of the global phase behaviour between mixtures and
critical properties, i.e. geometry and energy parameters of real binary fluids.
Recently, Cismondi and Michelsen [38] introduced a procedure to generate
different type of phase diagrams classified by van Konynenburg and Scott. Their
strategy does not take into account an existence of solid phase. At present time
Patel and Sunol [39] developed an automated and reliable procedure for systematic
generation of global phase diagrams for binary systems. The approach utilizes
equation of state, incorporates solid phase and is successful in generation of type
VI phase diagram. The procedure enables automatic generation of GPD which
incorporates calculations of all important landmarks such as critical endpoints
(CEP), quadruple point (QP, if any), critical azeotropic points (CAP), azeotropic
endpoints (AEP), pure azeotropic points (PAP), critical line, liquid–liquid–vapor
line (L1L2V, if any), solid–liquid–liquid line (SL1L2, if any), solid–liquid–vapor
line (SLV) and azeotropic line. The proposed strategy is completely general in that
it does not require any knowledge about the type of phase diagram and can be
applied to any pressure explicit equation of state model. Figure 2.8 shows the global
phase diagram for binary mixtures of equal-sized molecules, plotted in the two-
dimensional (Z1 – Z2 ) space [40].
The simplest boundary is a normal critical point when two fluid phases are
becoming identical. Critical conditions are expressed in terms of the molar Gibbs
energy derivatives in the following way:
Fig. 2.8 Global phase

diagram of the
Redlich-Kwong model

@2 G @3 G
D D 0: (2.9)
@x 2 p;T @x 3 p;T
Corresponding critical conditions for the composition – temperature – volume

variables are:
Axx WAxV D 0I
Axxx 3WAxxV C 3W 2 AxV V 3W 3 AV V V D 0I (2.10)
where A is the molar Helmholtz

nCm energy,

W D AAVxxV , AmV nx D @x @ A
n @V m is a contracted notation for differentiation
T
operation which can be solved for VC and TC at given concentration x.
Without exception, one of the most important boundaries is visualised by the
tri-critical points (TCP). This boundary divides the classes I and V, II and IV, or
III and IV. The tri-critical state is a state, where the regions of the liquid-liquid-gas
immiscibility shrink to one point, which is named the TCP. Three phases become
identical at a TCP. Another important boundary in the global phase diagrams is
the locus of double critical end-points (DCEP) that divides types III and IV, or II
and IV. Type IV is characterised by two liquid-liquid-gas curves. One is at high
temperatures and is restricted by two critical end points [one lower critical end point
(LCEP) and an upper critical end point (UCEP)]. At the upper critical end point, the
solution (compounds) become(s) immiscible as the temperature is lowered. At a
lower critical end point, the solution separates into two phases as the temperature
is increased. A DCEP occurs in a type IV when LCEP high-temperature three-
phase region joins the UCEP of low-temperature three-phase region. A DCEP is
produced in a type III system when the critical curve cuts tangentially the three-
phase line in a pressure – temperature diagram. The types I and II, or IV and V,
differ in the existence of a three-phase line, which goes from high pressures to an
UCEP. For this case, the boundary situation is defined by the zero-temperature end
point (ZTEP). Thermodynamic expressions and mathematical tools are given in the
literature [37, 38].
Azeotropy in binary fluids can be easily predicted in the framework of global
phase diagrams. The corresponding boundary situation is called the degenerated
critical azeotropic point (CAP) and represents the limit of the critical azeotropy at
xi ! 0 or at xi ! 1. This results in solving the system of thermodynamic equations
for a degenerated critical azeotrope.
One may obtain the relationships for azeotropy boundaries from the global phase
diagram [shaded A (Azeotropy) and H (Hetero-azeotropy)] regions in Fig. 2.8. The
above azeotropic borders are straight lines in the (Z1 , Z2 )-plane that cross at a single
point in the vicinity of the centre for equal sized molecules. It opens the opportunity
for obtaining the series of inequalities to separate azeotropic and non-azeotropic
regions of the global phase diagram. For the Redlich – Kwong EoS a corresponding
relationship was obtained in an analytical form [41]:

1 Z4
Z2 D
Z1 .1 ˙ Z1 / 1 0:6731: (2.11)
1 ˙ Z3
Global phase diagrams of binary fluids represent the boundaries between differ-
ent types of phase behaviour in a dimensionless parameter space. In a real P – T –
x space, two relatively similar components usually have an uninterrupted critical
curve between the two critical points of the pure components. An upper branch
extends from the critical point of the higher boiling component to higher pressures,
sometimes passing through a temperature minimum as in the C6 H6 –H2 O system.
A general simplification of the theoretical description of ternary mixtures has
been achieved by the consideration of the water C salt system as a single component
mixture. This component has interaction parameters (or EoS parameters) dependent
on the concentration of the salt. The details of the water-salt interactions have to
be taken as averaged by an effective, spherically symmetric, interaction potential.
This approach is justified by the phase behaviour of the system water C sodium
chloride that belongs to the simplest type II of phase behaviour and has a very
simple critical curve in the vicinity of the critical point of water. Three main types
of phase topologies have been known in binary water-electrolyte systems. These are
type-III (H2 O–HCl) system where hydrogen chloride is more volatile than water
and a negative azeotropic line for water-rich compositions, at pressures below the
saturation pressure of the water and the liquid-liquid-gas equilibrium, occur at
pressures near saturation pressures of HCl, type-II (H2 O–NaCl, H2 O–NaOH), and
type-IV phase diagrams (H2 O–Na2 WO4 , H2 O–K2 SO4 , H2 O–UO2 SO4 ).
The topological predictions on the basis of a global phase diagram will become
a convenient method in the analysis of supercritical water mixtures of scientific and
industrial interest. Topologically, there is no difference in isoproperty behaviour for
any pure fluid. This fact allows us to find the parameters of the equation of state
model, which can reproduce thermodynamic properties of an arbitrary substance in
a local region of a phase diagram.
Fig. 2.9 Allocation of phase behavior types for hydrocarbon – water and key biomass compo-
nents – water mixtures in reduced variables. – selected aqueous solutions of hydrocarbons
performing III type of phase behavior: (1) cyclohexane; (2) benzene; (3) n-decane; and (4) n-
hexane [43]; ♦ – selected aqueous solutions of key biomass components
To simulate the behavior of different biomass components in the SCW media,

models based on equations of state are more than sufficient. Considering a huge
array of reaction products, which wait in queue to be destroyed or recycled, it
is obvious that there is need to develop theoretically sound methods for reliable
assessment of their thermodynamic and phase behaviour, especially, at supercritical
water conditions.
If combination rules are known, then it is possible to determine the global phase
diagram via the ratio of critical parameters of pure components only [42]. The
boundaries between different types of phase behavior taken from their paper and
distribution of reduced critical temperatures and volumes for aqueous solutions of
hydrocarbons are given in Fig. 2.9.
Critical parameters of hydrocarbons were taken from correlation [15]. The
simple estimation of transition from III type of phase behaviour to II type
(Tr D TCH2O /TC 0.5) in terms of pure components for the one fluid model of
the Redlich-Kwong EoS [44] and Carnahan-Starling – van der Waals EoS [42] with
the Lorentz – Berthelot combination rules (k12 D 0) shows the serious discrepancies
with experimental evidence of III type phase behaviour for systems water – alkanes
[45, 46, 43]. Experimental values of k12 for water–alkanes systems show significant
deviations from zero. Therefore, for a correct classification of phase behavior types,
we should shift the lines DCEP to DCEP* (Fig. 2.9), which is achieved due to
variations in the interaction parameters k12 and l12 . The DCEP* lines in Fig. 2.9
classify experimentally observed types of phase behavior more correctly for the
SRK model with parameters corresponding the values k12 D 0.1 and l12 D 0.01. The
Lorentz – Berthelot model is not working also for ammonia water and ethanol –
water systems.
The traditional classification of fluid phase behavior can easily be discussed with
the aid of the p-T projections of fluid phase diagrams. There are two kinds of phase
diagrams. Phase diagrams of types I, V, and VI have the vapour pressure curves
that are started and ended in nonvariant points with equilibria where the no solid
phase exist. In the case of types II, III and IV, some critical curves, starting in high-
temperature nonvariant points, are not ended by the nonvariant points from the lower
temperature side, where the solid phase should exist. A solid phase is absent in
calculations of fluid phase diagrams using previously discussed equations of state
and the nonvariant equilibria with solid could not be obtained even at 0 K. Therefore
the monovariant curves remain incomplete on the theoretical p – T projections. As a
result these diagrams can be considered as the ‘derivative’ versions. It demonstrates
not only the main types of fluid phase behavior but also the fluid phase diagrams
which appear when the heterogeneous fluid equilibria are bounded not only by
another fluid equilibrium but also by the equilibrium with solid phase that is usually
observed in the most real systems [47].
To give an exhaustive description of phase behavior in binary systems, the
systematic classification of the main types of complete phase diagrams that
considers any equilibria between liquid, gas and/or solid phases in a wide range
of temperatures and pressures was proposed by V. Valyashko [47].
In the new nomenclature, suggested by Valyashko [48] for systematic classifi-
cation of binary complete phase diagrams, four main types of fluid phase behavior
are designated in order of their continuous topological transformation as a, b, c and
d types. In order to simplify the construction of phase diagrams he accepted the
following limitations for the main types of binary complete phase diagrams:
1. The melting temperature of the pure nonvolatile component is higher than the
critical temperature of the volatile component.
2. There are no solid-phase transformations such as polymorphism, formation of
solid solutions and compounds, and azeotropy in liquid-gas equilibria in the
systems under consideration.
3. Liquid immiscibility is terminated by the critical region at high pressures and
cannot be represented by more than two separated immiscibility regions of
different types.
Complete phase diagram presentation is very sophisticated and we refer the
readers to the cited original Valyashko’s papers.
The design and development of SCW technology depend on the ability to model
and predict accurately the solid-supercritical fluid equilibrium. It has been known
for more than 100 years that a supercritical fluid can dissolve a substance of low
volatility and that the solubility is dependent on the pressure. The ability to control
solubility by means of pressure as well as temperature has brought about the use
of supercritical fluids in different applications. For SCW technologies, knowledge
of the solubility of substances in supercritical fluids is very important and a large
number of experimental data has become available in recent years.
Hence, a more correct estimation of binary interaction parameters is needed to
predict phase behaviour of biomass components in supercritical water. The best
Table 2.3 Recommended Solute AKr , MPa [50] AKr , MPa [49]
values of the Krichevskii
parameters for biomass H2 169.9 ˙ 7.8 170 ˙ 8
component solutes N2 177.5 ˙ 7.1 178 ˙ 7
O2 171.3 ˙ 7.3 177 ˙ 10
CO 173.3 ˙ 5 174 ˙ 10
CO2 124.3 ˙ 5 127 ˙ 10
NH3 42 ˙ 1 46 ˙ 5
CH4 164.6 ˙ 5.7 164 ˙ 6
Methanol 32 ˙ 6 29 ˙ 5
Ethanol 43 ˙ 2 44 ˙ 5
source of information for model parameter determination to restore critical curves

in the SCW – biomass component mixtures is experimental data in vicinity of
solution critical line. In dilute near-critical solutions, the partial molar properties of
solutes, the coordinates of the critical lines of binary mixtures, and the temperature
variations of the vapor–liquid distribution andHenry’s c constants,
are controlled by
c
the critical value of the derivative AKr D @V@A@x T;xD0 D @P @x T;V;xD0 which is
called the Krichevskii parameter.
Here A is the Helmholtz energy of the binary system, x is the mole fraction
of a solute, P is the pressure, V is the volume of the system, and the superscript
c indicates evaluation at the solvent critical point. The Krichevskii parameter is
finite and has the dimension of pressure. The physical meaning of the Krichevskii
parameter is that it gives the change in pressure at the critical point of the solvent
when an infinitesimal amount of solvent molecules is replaced by solute at constant
volume and temperature: for most neutral solutes the pressure will rise, and for
solutes with strong attractions to the solvent (for example, ions in water) the pressure
of the system will decrease [49]. The review of Fernández-Prini et al. [50] presents
data for 14 gases (He, Ne, Ar, Kr, Xe, H2 , N2 , O2 , CO, CO2 , H2 S, CH4 , C2 H6 , SF6 )
in water with the accuracy of 2–6 % at high temperatures; similar uncertainties are
expected for derived AKr results. Most comprehensive review of the significance
of the Krichevskii parameter for the analysis of the phase equilibria, including the
solubility, in binary systems are presented by Plyasunov [49]. Recommended values
of the Krichevskii parameters for selected biomass components from [49] and [50]
are given in Table 2.3.
2.6 Uncertainties and Conflicts in the Parameter Estimation

for Thermodynamic Models
The reliable estimation of model parameters from experimental data in phase

equilibria simulation is an important requirement in many applications. Such models
offer the useful tools of aggregating large amounts of data, allow both data for
interpolation of data and extensions beyond regions in which measurements have
been made, and provide insight into physical and chemical phenomena. Transition
from real phenomenon to its model entails the appearance of uncertainty caused
by the statistical pattern of experimental information, inadequacy and ambiguity of
used models, etc. Experimental data, which are generated by different experimental
units, have as a rule different dimensions, different physical meaning, and different
statistical distribution. It results to conflict situation when the set of parameters
restored according to the one category of data does not correspond to parameters
from other data sources. Therefore, the conflict appears in model parameter
estimation and it is desirable to reduce an arising uncertainty by the simultaneous
consideration of all data fitness criteria for each property.
Advanced methods of probability theory consider uncertainty as some specific
value, characterizing an emergence of predetermined chance outcomes. Statistical
methods interpret all variety of uncertainty types in the framework of the ran-
domness concept. Nevertheless, there are ill-structured situations, which have not
any strictly defined boundaries and cannot be accurately formulated. Examples
of such situation are standard statements in the problem of model parameter
estimation: “mean-square deviation of calculated values from measured data must
be in neighborhood of feasible deviations •” or “to provide the best fit to measured
data”. The expressions “in neighborhood” or “best fit” have not exact boundaries
which separate one class of objects from others. The vague verbal models such
as “equation of state is valid in critical point neighborhood”, “high temperature
approximation”, “adequate model”, and etc. are fuzzy formulated targets, which
depend on biased assessment of boundaries for used approximations.
The problem of optimum parameter estimation in models of thermodynamic and
phase behavior under the uncertainty is a multicriteria problem of mathematical
programming. The first step in solution of multicriteria problem is a search of
the Pareto set, i.e. such compromise domain in the permissible parameter space X
where the value of any criterion cannot be improved without the value of the others
criteria being worsened. The diverse computational methods of the Pareto-optimum
parameter estimation and different (crisp and fuzzy) convolution schemes to reduce
a vector criterion into the scalar one should be used. As examples of case study, we
demonstrate the Pareto-optimum estimation of parameters in the Soave–Redlich–
Kwong equation of state (EoS), using different conflicting data sets (simultaneous
description of the phase equilibria and critical line data in binary mixtures).
The last two decades have been characterized by a growing comprehension of the
fact that the ideas of uncertainty should not be neglected in the fluid phase equilibria
modeling. Since one of the meanings of uncertainty is randomness, a conventional
approach is utilized via probability and random process theories. However, as
recognized recently, the probabilistic methods are accompanied by serious problems
during its implementation in practice of phase equilibria modeling and lead to the
unreliable estimation of parameters.
Development of reliable models for thermodynamic and phase behavior descrip-
tion of binary mixtures in most cases is associated with the estimation of the
binary interaction parameters from the restricted set of VLE data. Although there
are rigorous relationships to obtain from EoS model the different derivatives of
thermodynamic values, for instance, the prediction of critical lines from the EoS
with parameters restored from VLE data is questionable due to diverse sources of
uncertainty both in the used models and experimental data.
To illustrate conflict between phase equilibria and critical line description we
consider the water – carbon dioxide system in the vicinity of critical point of water.
Experimental data on p – V – T – x properties have been taken from [51, 44] data
on phase equilibria and critical curve have been extracted from [52, 24]. The critical
lines have been calculated using algorithm developed in [25]. We performed phase
equilibria and critical line calculations for the SRK EoS with parameters binary
interaction parameters k12 and l12 from standard mixing rules,

b11 Cb22
b D b11 x 2 C l12 x .1 x/ C b22 .1 x/ 2
2
p (2.12)
a D a11 x 2 C k12 x .1 x/ a11 a22 C a22 .1 x/ 2
restored from different crisp convolution schemes.

For simplicity the results of calculation for phase equilibria and critical line
are presented in Figs. 2.10 and 2.11 only for additive compromise scheme. The
parameters restored from P-V-T-x data [51, 44] are not suitable for phase equilibria
description and completely unusable for critical curve description. These parameters
are skipped here. There are many speculations about problem of EoS singular
behavior near critical point. Here we do not discuss the possible approaches to
consistent description of regular and singular behavior of thermodynamic functions,
but point out only the fact that classical EoS models, which are widely used in
phase equilibrium calculations, cannot describe simultaneously phase equilibria and
critical curves. The problem of conflict between parameters restored from different
sections of thermodynamic surface is remained valid for more sophisticated EoS
with singular behavior also. As a whole, it is observed the general conflict situation:
if the cross-interaction parameters are restored from the one category of properties
the prognosis of other properties is dubious in spite of indubitable validity of
thermodynamic relationships. The final solution has subjective nature and depends
on the statement of problem and decision maker experience.
2.7 Conclusion
This study is one of the first attempts to establish and demonstrate multiple links
existing among the phase equilibria phenomena in supercritical aqueous mixtures
with biomass components and their models. From the very beginning of these
efforts, the global phase diagrams have been a very useful tool for scientists
and engineers working in the field of emerging SCW technologies. There is no
doubt that extension of our knowledge about global phase behaviour of two- and
multicomponent fluids will lead to the creation of reliable engineering recipes for
solving the actual problems of SCW technology applications in biomass conversion.
Fig. 2.10 Critical lines of °2 ±–´±2 system. – experimental data [24]. – best fit of VLE
data. ıııı – compromise solution. — — – best fit of critical curve
Fig. 2.11 Phase equilibria in °2 ±–´±2 system. – experiment T D 548.15 K. – best fit of
VLE data. ıııı – Ôompromise solution. — — – best fit of critical curve
Unfortunately, there are no rigorous mathematical methods and sound physical

concepts to construct the adequate model without enlisting of subjective point of
view due to inaccuracy of our knowledge. To minimize this source of unavoidable
uncertainty the different compromise schemes of criteria convolution are considered
and their selection is strongly depended on decision maker experience. Here, by
way of illustration, it has been applied to estimate the Redlich-Kwong-Soave EoS
parameters for simultaneous description of the phase equilibria and critical line data
in binary mixtures.
References
1. Peterson A, Vogel F, Lachance R, Froling M, Antal M, Tester J. Thermochemical biofuel

production in hydrothermal media: a review of sub- and supercritical water technologies.
Energy Environ Sci. 2008;1:32–65.
2. Smith Jr RL, Fang Z. Properties and phase equilibria of fluid mixtures as the basis for
developing green chemical processes. Fluid Phase Equilibria. 2011;302:65–73.
3. Dorrestijn E, Laarhoven L, Arends I, Mulder P. The occurrence and reactivity of phenoxyl
linkages in lignin and low rank coal. J Anal Appl Pyrolysis. 2000;54:153–92.
4. ECN. Phyllis, database for biomass and waste. 2012; Available from: http://www.ecn.nl/
phyllis2
5. Peng D-Y, Robinson DB. A new two-constant equation of state. Ind Eng Chem Fundam.
1976;15:59–64.
6. Redlich O, Kwong JNS. On the thermodynamics of solutions V. An equation of state.
Fugacities of gaseous solutions. Chem Rev. 1949;44:233–44.
7. Soave G. Equilibrium constants from a modified Redlich–Kwong equation of state. Chem Eng
Sci. 1972;27:1197–2003.
8. Wagner W, Pruß A. The IAPWS formulation 1995 for the thermodynamic properties
of ordinary water substance for general and scientific use. J Phys Chem Ref Data.
2002;31(2):387–535.
9. Joback K. Knowledge bases for computerized physical property estimation. Fluid Phase
Equilibria. 2001;185:45–52.
10. Somayajulu GR. Estimation procedures for critical constants. J Chem Eng Data.
1989;34(1):106–20.
11. Labanowski JK, Motoc I, Damkoehler RA. The physical meaning of tological indexes. Comput
Chem. 1991;15:47–53.
12. Katritzky AR, Dobchev D, Karelson M. Physical, chemical, and technological property
correlation with chemical structure: the potential of QSPR. Z Naturforsch. 2006;61b:373–84.
13. Rogers D, Hopfinger AJ. Applications of genetic function approximation (GFA) to quantita-
tive structure-activity relationships (QSAR) and quantitative structure property relationships
(QSPR). J Chem Inf Comp Sci. 1994;34:854–66.
14. Mosier P, Jurs P. QSAR/QSPR studies using probabilistic neural networks and generalized
regression neural networks. J Chem Inf Comput Sci. 2002;42(6):1460–70.
15. Wakeham W, Cholakov G, Roumiana S. Liquid density and critical properties of hydrocarbons
estimated from molecular structure. J Chem Eng Data. 2002;47:559–70.
16. Imre A, Deiters U, Kraska T, Tiselj I. The pseudocritical regions for supercritical water. Nucl
Eng Des. 2012;252:179–83.
17. Imre A, Hazi A, Horvath MC, Mazur V, Artemenko S. The effect of low-concentration
inorganic materials on the behaviour of supercritical water. Nucl Eng Des. 2011;241:296–300.
18. Shibue Y. Vapor pressures of aqueous NaCl and CaCl2 solutions at elevated temperatures. Fluid
Phase Equilibria. 2003;213(1–2):39–51.
19. Valyashko V. Experimental data on aqueous phase equilibria and solution properties at elevated
temperatures and pressures. New York: Wiley; 2009.
20. van Konynenburg PH, Scott RL. Critical lines and phase equilibria in binary van der Waals
mixtures. Philos Trans R Soc Lond. 1980;298:495–540.
21. Aicardi F, Valentin P, Ferrand E. On the classification of generic phenomena in one–parameter
families of thermodynamic binary mixtures. Phys Chem Chem Phys. 2002;4:884–95.
22. Japas M, Franck E. High pressure phase equilibria and PVT-data of the water-oxygen system
including water-Air to 673 K and 250 MPa. Ber Bunsenges Phys Chem. 1985;89:1268–75.
23. Takenouchi S, Kennedy GC. The binary system H2 O–CO2 at high temperatures and pressures.
Am J Sci. 1964;262:1055–74.
24. Todheide K, Frank E. Das zweiphasengebiet und die kritische kurve im system kohlendioxid-
wasser bis zu drucken von 3500 bar. Z Phys Chem. 1963;37:387–401.
25. Heidemann R, Khalil A. The calculation of critical points. AICHE J. 1980;26:769–78.
26. Blencoe JG. The CO2 –H2 O system: IV. Empirical, isothermal equations for representing
vapor–liquid equilibria at 110–350 ı C, P < D150 MPa. Am Mineral. 2004;89(10):1447–55.
27. Tsiklis DS, Maslennikova VY. Limited mutual solubility of the gases in the H2 O–N2 system.
Dokl Akad Nauk SSSR. 1965;161:645–7.
28. Prokhorova VM, Tsiklis DS. Gas-gas equilibrium in nitrogen-water system. Russ J Phys Chem.
1970;44:1173.
29. Japas ML, Franck EU. High pressure phase equilibria and PVT data of the water – nitrogen
system to 673 K and 250 MPa. Ber Bunsenges Phys Chem. 1985;89:793–800.
30. Seward TM, Franck EU. The system hydrogen – water up to 440 ı C and 2500 bar pressure.
Ber Bunsenges Phys Chem. 1981;85:2–8.
31. Abdurashidova A, Bazaev A, Bazaev E, Abdulagatov I. The thermal properties of water-
ethanol system in the near-critical and supercritical states. High Temp. 2007;45(2):178–86.
32. Barr-David F, Dodge BF. Vapor-liquid equilibrium at high pressures. The systems ethanol-
water and 2-propanol-water. J Chem Eng Data. 1959;4:107–21.
33. Bazaev AR, Abdulagatov IM, Magee JW, Bazaev EA, Ramazanova AE, Abdurashidova AA.
PVTx measurements for a H2 O C methanol mixture in the subcritical and supercritical regions.
Int J Thermophys. 2004;25:805–38.
34. Bazaev EA, Bazaev AR, Abdurashidova AA. An experimental investigation of the critical state
of aqueous solutions of aliphatic alcohols. High Temp. 2009;47:195–200.
35. Varchenko AN. Evolution of convex hulls and phase transition in thermodynamics. J Sov Math.
1990;52(4):3305–25.
36. Mazur VA, Boshkov LZ, Murakhovsky VG. Global phase behaviour of binary mixtures of
Lennard–Jones molecules. Phys Lett. 1984;104A:415–8.
37. Deiters UK, Pegg JL. Systematic investigation of the phase behavior of binary fluid mixtures. I.
Calculations based on the Redlich–Kwong equation of state. J Chem Phys. 1989;90:6632–41.
38. Cismondi M, Michelsen M. Global phase equilibrium calculations: critical lines, critical
end points and liquid-liquid-vapour equilibrium in binary mixtures. J Supercrit Fluids.
2007;39(3):287–95.
39. Patel K, Sunol A. Automatic generation of global phase equilibrium diagrams for binary
systems from equations of state. Comput Chem Eng. 2009;33:1793–800.
40. Mazur V, Boshkov L, Artemenko S. Global phase behaviour of natural refrigerant mixtures.
In: Proceedings of the IIR – Gustav Lorentzen conference: natural working fluids. Oslo; 1998.
p. 495–504.
41. Artemenko S, Mazur V. Azeotropy in the natural and synthetic refrigerant mixtures. Int J
Refrig. 2007;30:831–9.
42. Sadus R, Wang J. Phase behaviour of binary mixtures: a global phase diagram solely in terms
of pure component properties. Fluid Phase Equilibria. 2003;214:67–78.
43. Brunner E, Thies M, Schneider G. Fluid mixtures at high pressures: phase behavior and critical
phenomena for binary mixtures of water with aromatic hydrocarbons. J Supercrit Fluids.
2006;39:160–73.
44. Shmulovich KI, Mazur VA, Kalinichev AG, Khodorevskaya LI. P-V-T and component
activity-concentration relations for systems of H2 O-nonpolar gas type. Geochem Int.
1980;17(6):18–31.
45. Tsonopoulos C, Wilson GM. Phase behavior of binary mixtures: a global phase diagram solely
in terms of pure component properties. AICHE J. 1983;29(6):990–3.
46. Tsonopulos C. Thermodynamic analysis of the mutual solubilities of normal alkanes and water.
Fluid Phase Equilibria. 1999;156:21–33.
47. Valyashko VM. Derivation of complete phase diagrams for ternary systems with immiscibility
phenomena and solid–fluid equilibria. Pure Appl Chem. 2002;74(10):1871–84.
48. Valyashko V. Hydrothermal properties of materials. Experimental data on aqueous phase
equilibria and solution properties at elevated temperatures and pressures. New York: Wiley;
2008. p. 1–134.
49. Plyasunov A. Values of the Krichevskii parameter, AKr, of aqueous nonelectrolytes evaluated
from relevant experimental data. J Phys Chem Ref Data. 2012;41(3):1–27.
50. Fernández-Prini R, Alvarez JL, Harvey AH. Henry’s constant and vapor – liquid distribution
constants for gaseous solutes in H2 O and D2 O at high temperatures. J Phys Chem Ref Data.
2003;32:903–16.
51. Sterner M, Bodnar R. Synthetic fluid inclusions. X: experimental determination of P – V – T –
x properties in the CO2 –H2 O system to 6 KB and 700 C. Am J Sci. 1991;64:1–54.
52. Mather A, Franck E. Phase equilibria in the system carbon dioxide-water at elevated pressures.
J Phys Chem. 1992;6:6–8.
53. Hicks CP, Young CL. The gas-liquid critical properties of binary mixtures. Chem Rev.
1975;75(2):119–75.
54. Rizvi SSH, Heidemann RA. Vapor- liquid equilibrium in the ammonia – water system. J Chem
Eng Data. 1987;32:183–91.
55. Sassen CL, van Kwartel RAC, van der Kooi HJ, de Swaan AJ. Vapor – liquid equilibrium for
the system ammonia C water up to the critical region. J Chem Eng Data. 1990;35:140–4.
Chapter 3
Role of Co-solvents in Biomass Conversion
Reactions Using Sub/Supercritical Water
Yulong Wu, Yu Chen, and Kejing Wu
Abstract Water above its critical point (Tc D 647 K, Pc D 22.1 MPa), which is
regarded as supercritical water (SCW), is being given increasing attention as a
medium for organic chemistry. This interest in SCW is mainly driven by the search
for more “green” or environmentally benign chemical processes. The use of SCW
instead of organic solvents in chemical processes offers environmental advantages
and may lead to pollution prevention. SCW has been applied in synthetic fuels
production, biomass processing, waste treatment, materials synthesis, and geo-
chemistry. However, higher critical parameters of water indicate that the operation
process is performed at harsh conditions, thereby increasing cost. Other drawbacks
of the use of water as the medium for biomass liquefaction reaction include lower
biofuel yield and higher oxygen content. Organic solvents, such as methanol,
ethanol, and 2-propanol, have been utilized as co-solvents of SCW to enhance the
biofuel yield with lower oxygen content and higher heating value. This paper mainly
expounds the basic characteristics of the co-solvent in sub/supercritical water and
analyzes the function of the co-solvent in reactions to provide readers with a more
comprehensive knowledge of the co-solvent. Based on literature and related studies
conducted by our group, systematic analysis about selection and application of
co-solvent was conducted.
Keywords Supercritical water (SCW) • Co-solvent • Properties • Application
Y. Wu () • Y. Chen • K. Wu
Institute of Nuclear and New Energy Technology, Tsinghua University,
Beijing 100084, People’s Republic of China
70 Y. Wu et al.
3.1 Introduction
This chapter describes the application and current status of the properties of co-
solvents in supercritical water (SCW) for the production of liquid fuels (bioethanol
and biocrude), gaseous fuels (methane, hydrogen, and synthesis gas), and solid
fuels (biochar and other functional carbonaceous materials) from biomass. Energy
security, sustainability, and climate change concerns have motivated the search for
renewable and alternative energy resources. Large-scale substitution of petroleum-
based fuels and products with renewable sources are needed to minimize environ-
mental issues. The use of renewable biomass resources (such as terrestrial biomass
and aquatic biomass) is becoming increasing important. Converting biomass into
biofuels compatible with existing petroleum refinery infrastructure requires the
removal of oxygen from biomacromolecules, which include carbohydrate, lipid,
protein, cellulose, hemicellulose, and lignin molecules. The unique physicochemical
properties of supercritical fluids provide an attractive medium for chemical reactions
and other processes. Fluids near their critical points have solvent power comparable
with those of liquids, and they are significantly more compressible than dilute gases.
The transport properties of such fluids lie intermediate between gas- and liquid-like.
Supercritical fluids are being given significant attention in various fields in science
and technology.
In the last two decades, the use of water as a reaction medium has been given
remarkable importance because water is a nontoxic, nonflammable, and inexpensive
solvent. Water is an ecologically safe and abundantly available solvent in nature.
Water has a relatively high critical point (Tc D 647 K, Pc D 22.1 MPa) [1] because
of the strong interaction between the molecules caused by strong hydrogen bonds.
Liquid water below the critical point is referred to as subcritical water, whereas
water above the critical point is called SCW. The density and dielectric constant
of the water medium are important in solubilizing different compounds. Water at
ambient conditions is a good solvent for electrolytes because of its high dielectric
constant; however, most of the organic compounds are poorly soluble at these
conditions [2].
When water is heated, H-bonds start to weaken, thereby allowing the dissociation
of water into acidic hydronium ions (H3 OC ) and basic hydroxide ions (OH ). The
water structure changes significantly near the critical point because of the breakage
of the infinite network of hydrogen bonds, and water exists as separate clusters with
a chain structure. In addition, the dielectric constant of water decreases considerably
near the critical point, thereby changing the dynamic viscosity and increasing the
self-diffusion coefficient of water.
SCW has liquid-like density and gas-like transport properties, with behavior that
is significantly differently from that of water at room temperature. For example,
SCW is highly nonpolar, which permits the complete solubilization of most organic
compounds and oxygen. The resulting single-phase mixture does not have many of
the conventional transport limitations that are encountered in multiphase reactors.
However, the polar species present, such as inorganic salts, are no longer soluble,
3 Role of Co-solvents in Biomass Conversion Reactions Using. . . 71
Fig. 3.1 Physical properties of water with temperature at 24 MPa (Reprinted with permission from
James [2], Copyright © 2013, Springer)
and they start to precipitate. The physicochemical properties of water, such as

viscosity, ion product, density, and heat capacity, also change dramatically in the
supercritical region with only a small change in temperature or pressure, thereby
resulting in a substantial increase in the rates of chemical reactions. The dielectric
behavior of 473 K water is similar to that of ambient methanol, 573 K water is
similar to ambient acetone, 643 K water is similar to methylene chloride, and 323 K
water is similar to ambient hexane (Fig. 3.1) [2].
In addition to the unusual dielectric behavior, the transport properties of SCW
are significantly different from those of ambient water (Table 3.1). The use
of SCW media, which also known as hydrothermal media and can be broadly
defined as water-rich phase above 473 K, offers several advantages over other
biofuels production methods. The substantial changes in the physical and chemical
properties of water in the vicinity of its critical point can be used to convert
biomass to desired biofuels. Moreover, the use of SCW also provides a novel
approach to conduct tunable reactions for the synthesis of specialty chemicals from
biomass.
In the subcritical region, the ionization constant (Kw ) of water increases with
temperature and is approximately three times higher than that of ambient water
(Table 3.1), and the dielectric constant (") of water decreases from 80 to 20. A low
72 Y. Wu et al.
Table 3.1 Comparison of ambient and SCW

Ambient water SCW
Dielectric constant (") 78 <5
Solubility of organic compounds Very low Fully miscible
Solubility of oxygen 6 ppm Fully miscible
Solubility of inorganic compounds Very high 0
Diffusivity (cm2 s1 ) 105 103
Viscosity (g cm1 s1 ) 102 104
Density (g cm3 ) 1 0.2–0.9
Reprinted with permission from James [2], Copyright © 2013, Springer
dielectric constant allows subcritical water to dissolve organic compounds, whereas

a high ionization constant allows subcritical water to provide an acidic medium
for the hydrolysis reactions. These ionic reactions can be dominant because of the
liquid-like properties of SCW. Moreover, the physical properties of water, such
as viscosity, density, dielectric constant, and ionic product, can be tuned by small
changes in pressure and/or temperature in the subcritical region. In the supercritical
region, the density of water decreases, which means that the ionic product of water is
significantly lower and that ionic reactions are inhibited because of the low relative
dielectric constant of water. A lower density favors free-radical reactions, which
may be favorable for biomass conversion [2].
However, Li et al. [3] proposed that in the liquefaction of biomass, the biofuel
fractions obtained in SCW have lower carbon content and higher oxygen content,
with a lower heating value. To enhance the yield of liquid products with lower
oxygen content [hence, higher heating value (HHV)], the use of organic solvents
has been adopted [3–5]. On one hand, the solvent has a remarkable effect on the
liquefaction reaction [6]. On the other hand, the ethanol–water solvent has been
shown to produce a synergistic effect in the thermochemical liquefaction of biomass
[7–9]. As mentioned above, a higher critical temperature and critical pressure of
water meant challenging operation conditions [10], thereby increasing cost. The
critical value for a mixture could be reduced when water is mixed with an organic
solvent with a critical value lower than that of water. Organic solvents, such as
methanol, ethanol, 2-propanol, acetone, and tetralin, have been utilized as reaction
media instead of water to enhance the yield of biofuel with a lower oxygen content
[6, 11–13]. Given the low critical temperatures and pressures of these organic
solvents (e.g., methanol: Tc D 512.5 K, Pc D 8.1 MPa; ethanol: Tc D 513.8 K,
Pc D 6.1 MPa; isopropanol: Tc D 508.2 K, Pc D 4.8 MPa), the critical point and the
dielectric constant of the water C organic solvent mixture are lower than those of
pure water, thereby resulting in milder conditions for the reaction and an increase in
the solubility of relatively high molecular weight products from the biomass [14].
Therefore, the co-solvent selected as the reaction medium of SCW should have
broad application prospects in biomass conversion.
3.2 Main Properties of Co-solvents in Supercritical

Condition
3.2.1 Preface
Although water is a cheap and common medium in the liquefaction process,

the critical value for water means that the SCW liquefaction process requires
challenging operative conditions. Brennecke and Eckert [15] found that the sol-
ubility of supercritical fluids could be significantly improved by adding a small
amount of a second solvent, which is commonly called co-solvent. Co-solvents can
enhance the properties of SCW for target applications, such as bio-oil conversion
of biomass, chemical recycling of plastic via depolymerization, and biodiesel
production [16]. Ethanol, 2-propanol, and methanol are often used as co-solvent
for biomass conversion in supercritical state [8, 12, 16], and significantly higher
biofuel yields can be achieved in the co-solvent–water mixture than in single SCW.
Table 3.2 shows the critical values of particular co-solvent-water mixtures. The
critical values are significantly dependent on the content. Thus, studies about the
critical characteristics of co-solvent–water systems are important.
Water is a hydrogen-bonding fluid that has a 3D tetrahedral network structure,
which strongly influences its characteristic properties as a solvent. By contrast,
methanol is a hydrogen-bonding fluid that not only has hydrogen-bonding sites
but also a hydrophobic alkyl group. Methanol–aqueous solutions involve attractive
hydrogen bond interactions and hydrophilic interactions between the hydrophobic
alkyl group and water molecules, thereby changing the 3D network structure [16].
Imagining the hydrogen-bonded structure of water shows many large cavities. By
adding small amounts of alcohol to water, alcohol molecules can be accommodated
in those cavities, and the whole structure is stabilized. At higher temperatures,
the thermal motion of the molecules weakens the structure of water and the
penetration of alcohol molecules into cavities becomes rare. Filling up the cavities
Table 3.2 The critical values Solvent: water Ethanol-water 2-Propanol-water

(Tc , Pc ) of mixtures of water
and solvents that can be used (v/v) Tc /K Pc /MPa Tc /K Pc /MPa
as co-solvents 0:10 647.30 22:10 647.30 22:10
1:9 636.07 19:89 635.53 19:61
3:7 612.53 15:56 610.89 14:70
5:5 587.12 11:67 584.35 10:23
6:4 572.64 10:04 570.31 8:35
8:2 544.94 7:82 540.50 6:05
10:0 516.40 6:30 508.00 4:80
Reprinted with permission from Yuan et al. [12],
Copyright © 2011, Elsevier
74 Y. Wu et al.
is also reduced with increasing the pressure, when the structure of water becomes
compacter and this leads to diminishing the size and number of cavities. The
non-ideality decreases with increasing temperature and pressure. Irrespective of
temperature and pressure with increasing concentration of ethanol, the stabilization
of water structure soon reaches a maximum when no more alcohol molecules
can be accommodated in the cavities. With further addition of ethanol volumetric
properties of alcohol becomes dominant [17]. Dixit et al. [18] demonstrated the
microscopically in homogeneous evidence of methanol-water mixtures at room
temperatures via MD simulation and attributed this behavior to small but non-
negligible hydrophobic interactions in aqueous solution. Considering their results,
solvent properties at high temperature will be greatly affected by hydrophobic
groups because the hydrogen bond is considerably weakened at high temperatures as
verified by Ikushima et al. [19] via Raman measurements. There is a long-standing
controversy regarding the effect of methanol on the three-dimensional hydrogen
bonded network of water, because there is a delicate balance between “structure-
making” and “structure-breaking” effects, i.e., both hydrophilic and hydrophobic
effects play equally important roles in the mixture [20].
Investigation of hydrophobic hydration in alcohol-water mixture under supercrit-
ical conditions is of a great practical importance for chemical engineering [21, 22].
Computer simulation was successfully employed to study hydrophobic phenomena
in SCW by Gao [23] some years ago.
3.2.2 Phase Diagrams
In order to understand the properties of the co-solvent with water as reaction

medium, the phase diagrams of many aqueous mixtures, including two components,
three components or even more, have been studied firstly. But the interaction
between molecules seems to be much more complex with more components, and
it’s not convenient for study and application. So, most studied are focused on two-
component mixtures.
Vankonynenburg and Scott [24] classified phase diagrams of binary Van der
waals mixtures into six types. In the simplest cases, liquids are mutually soluble
completely or near critical region. The end of liquid-gas curve of a pure component
is a critical point (CP) and the L-G critical curve of the mixtures connects the two
CPs. When the liquids are miscible in all regions, the phase diagram is called type I
shown in Fig. 3.2. This type occurs when both molecules have similar polarity and
size, such as ethanol-water and acetone-water mixtures.
When the dissimilarities of the liquids become larger, the L-L equilibrium
happens. If the liquids are immiscible at low temperature while miscible at high
temperature, the L-L curve ends before reaching the critical states and the critical
curve remains the same, which is classified as type II in Fig. 3.2 [25]. This
behavior occurs for aqueous solutions of organics, for example phenol-water and
diacetyl-water. In fact, the critical curve is not always one simple line. When the
L-L curve develops and interrupts the critical curve, that is, L-L equilibrium still
Fig. 3.2 P-T phase diagram

of type I, II, and III phase
behavior of the
Scott-Konynenburg
classification. The projection
shows the L–G critical curve
( ), vapor-pressure curves
of the pure components (- - -),
and L-L-G three-phase line
( : : : ). CP indicates the
critical points of the pure
fluids. CE is the critical end
point (Reprinted with
permission from Weingartner
and Franck [25], Copyright
2005 ©, Wiley Online
Library)
Fig. 3.3 (a) Phase diagram for benzene-water mixtures. The thick solid lines: one-phase critical
curve and liquid-liquid-gas three phase curve; CP is the critical points and CE is the critical end
point (Reprinted with permission from Carlos et al. [26], Copyright © 2004, EBSCOhost). (b)
Phase diagram for n-C5 H12 C H2 O mixtures. 1 the vapor-pressure curve of pure water (IAPWS);
2 the vapor-pressure curve of pure n-C5 H12 . CP1 critical point of pure water; CP2 critical point
of pure n-C5 H12 ; Brunner; de Loos et al.; Brunner; Jou and Mather.; Gillespie and
Wilson; ı Connolly; : : : three-phase (L-L-G) curve; UCEP upper critical end point (Reprinted
with permission from Rasulov and Abdulagatov [27], Copyright © 2010, American Chemical
Society)
exists in critical state, the critical curve is divided into two parts by the L-L curve
at the critical end point (CE), which is called type III shown in Fig. 3.2. The
phase behaviors of aqueous alkane mixtures, such as n-pentane-water and ethane-
water, are exemplified for this type. The differences of the other three types of
Konynenburg and Scott classification are mainly the L-L equilibrium behaviors.
The phase diagrams of n-pentane-water and toluene-water are shown in Fig. 3.3
[26]. It’s clear that all of these follow type III phase behaviors, but the CEs
76 Y. Wu et al.
Table 3.3 Upper and lower critical curve of the H2 O C n-C5 H12 mixture
Upper critical curve Lower critical curve
x/mole fraction of n-C5 H12 TC /K PC /MPa x/mole fraction of n-C5 H12 TC /K PC /MPa
0.1185 646.30 75.50 1.0000 469.90 3.36
0.0967 629.95 49.70 0.8898 467.93 3.70
0.0750 625.50 33.90 0.8570 466.68 3.90
0.0580 627.05 27.10 0.8260 465.45 4.17
0.0270 632.35 21.78 0.7970 464.15 4.50
0.0000 647.10 22.15 0.7910 463.85 4.57
Reprinted with permission from Rasulov and Abdulagatov [27], Copyright © 2010, American
Chemical Society
are not located between the CPs as shown in Fig. 3.2. The critical values for
n-pentane-water mixture is shown in Table 3.3 [27]. In different critical curves, the
Tc and Pc vary differently with n-pentane content. On the upper critical curve, the
Tc first decreases and then increases and the Pc almost always decrease with the
decreasing of n-pentane content. But on the lower critical curve, the Tc decreases
while Pc increases with the decreasing of n-pentane content. Not only the critical
values, but also we get to know the mole volume, density and other important
characteristics through PVTx data. These will be further discussed later.
3.2.3 Thermodynamics and Molecule Interactions
3.2.3.1 Cubic Equations of State
Introduction of state equations is needed for better understanding of

thermodynamics. SRK equation is widely used for PVT simulation and it is
expressed below:
RT a
P D (3.1)
V b V .V C b/
Another important state equation is PR equation as shown below:
RT a
P D (3.2)
V b V .V C b/ C b .V b/
Other equations can be easily found in many books and literatures.

For aqueous mixtures simulation, mixing rules are very necessary. For example,
van der Waals rule is specially used with SRK and PR equations. The parameters a
and b are given by:
X
n X
n X
n X
n
aD xi xj aij bD xi xj bij (3.3)
i D1 j D1 i D1 j D1
Where n is the component number. The parameters aij and bij can be calculated from
ai and bi of pure fluids.
p bi C bj
aij D ai aj 1 kij bij D 1 lij (3.4)
2
Where kij and lij are fitted though experiment data. Other rules are available in many
books and literatures.
3.2.3.2 Micro Structures and Interactions of Co-solvent-Water Mixture
Equations of state are more of macro properties and empirical formulas, but the
micro interaction of the molecules is missed. And it is difficult to describe polar
mixtures and electrolyte solutions using cubic equations, even more difficult at
supercritical state. So molecule dynamics calculations (MD) are used for the
purpose to study the micro structure and interaction in aqueous mixtures. As
alcohols, especially methanol, are commonly used as co-solvents for SCW, we
take methanol for detailed discussions. Kiselev et al. [28] introduced molecule
dynamics calculations to investigate the hydrophobic hydration and heat capacity of
methanol-water mixture. The hydrophobic hydration is that water around methanol
reorganizes through water-methanol and water-water H-bonds to form cage like
structures. It’s well known that methanol has both hydrophilic and hydropho-
bic groups, and there exists a delicate balance between “structure-making” and
“structure-breaking” effects.
At supercritical region, the strength of H-bonds is much weaker than that at
ambient conditions and the structure of hydrophobic hydration becomes drastically
loose. However, the isochoric heat capacity (Cv ) reaches maximum, shown in
Fig. 3.4a, when the mole fraction of methanol (Xm ) is 0.12, which confirms the
existence of hydrophobic hydration.
H-bonds are the most important for hydrophobic hydration formation, and the
water O-H radial distribution functions is observed to describe the strength of the
H-bonds. The study shows that the peak values of radial distribution functions
are higher for Xm D 0.12 than other methanol mole fractions, especially when the
experimental density is 0.98 g cm3 . It indicates that the H-bonds between water
molecules is stronger when Xm D 0.12 and the results are in great agreement with the
isochoric heat capacity behavior. This phenomenon can be explained as below: the
H-bonds are gradually broken with the increasing of temperature, and more energy
is needed for breaking stronger H-bonds; then, the strong H-bonds for Xm D 0.12
result in the largest isochoric heat capacity.
3.2.3.3 Co-solvent Influence and Micro Explanations
Takumi et al. [16] further studied the critical properties of methanol-water mixture.
As one of the most important characters, the density has great dependence on
methanol mole fraction, temperature and pressure. At 623 K, The densities of
78 Y. Wu et al.
Fig. 3.4 (a) Isochoric heat capacity

as a function of Xm at 660 K and experimental density of
0.98 g/cm3 (•) and 0.5 g/cm3 ( ). (b-1b-5) Experimental densities (¡), excess molar volume
(VE m ), excess internal energy (UE ), self-diffusion coefficient of water (Dw) and self-diffusion
coefficient of methanol (Dm) as a function of Xm at 623 K and () 20 MPa, (ı) 25 MPa, ()
30 MPa, (♦) 35 MPa (Reprinted with permission from Takumi et al. [16], Copyright © 2011,
Elsevier)
mixtures studied decreased with increasing the methanol mole fraction (Xm ) at all
pressures. For a given Xm , density increased with pressure. In the larger Xm region,
the densities of methanol-water mixture did not change much with Xm . We can also
find that the densities decreased much faster around Xm D 0.2.
The excess molar volumes (VE ) of methanol aqueous mixture were evaluated to
observe the deviation of the PVT behavior of real mixture from that of ideal one. VE
is given by:

V E D V Xm Vm0 C .1 Xm / VW0 (3.5)
Where Xm is a mole fraction of methanol (m) in water (w), V is the molar volume of
the mixtures, and V0 is molar volume of pure components. Lower VE usually results
from stronger attractive interaction between co-solvent and water than that in ideal
mixture.
The experiment VE values are given in Fig. 3.4b-2. There are negative and
positive regions of excess molar volumes for all pressures, and the transition of
the two regions is Xm D 0.2. It indicates that the attractive force is slightly larger at
lower Xm region and the repulsive force is gradually increases with methanol content
at the critical region. The excess internal energy, which is related to the stability of
the molecules in the given potential field, is given similarly with VE :

U E D U Xm Um0 C .1 Xm / UW0 (3.6)
Where U is the molar internal energy of the mixtures, and U0 is molar internal
energy of pure components. It seems like that the internal energy of the methanol-
water mixture decreases if structures are more stable.
The UE values of methanol-water mixtures obtained through molecule dynamics
calculations are shown in Fig. 3.4b-3. When Xm is less than 0.2, the UE is quit
complex. But the UE largely increases with Xm around Xm D 0.2. Besides, UE
decreases with pressure at a given Xm when Xm is larger than 0.2. The methanol
mole fraction Xm dependence of self-diffusion coefficients of water and methanol is
shown in Fig. 3.4b-4, b-5. The self-diffusion coefficients of water first decrease with
Xm , and then, increase with Xm . And the transition point is also around Xm D 0.2.
The self-diffusion coefficients of methanol increase with methanol mole fraction at
all pressures and vary slowly in larger Xm region. The relationships between self-
diffusion coefficients and pressures are similar with that between excess internal
energy and pressures.
Taking VE , UE and self-diffusion coefficients into consideration, we can conclude
that at lower methanol content (Xm < 0.2), water molecules reorganize to compose
weak H-bonds network structure and the hydrophobic hydration is formed though
H-bonds between methanol and water. Although H-bonds is largely weakened at
high temperature, the cage like structure still exists as indicated by MD calculations,
which results in special behaviors of methanol-water mixtures at supercritical
region. MD calculations also indicate that the number of crossed H-bonds between
the hydrogen atom of water molecules and the electron ring of the benzene molecule
is considerable [26].
3.3 Related Chemical Reactions in Supercritical Co-solvents
3.3.1 Alcohol (Methanol, Ethanol, Isopropanol,

and 1-Butanol)
Water and alcohol have been proposed as green solvents for biomass conversion
[29, 30], chemical recycle of monomer via depolymerization [31, 32], particle
80 Y. Wu et al.
formation [33], and biodiesel production [34]. Several of these proposals use SCW,
whereas others use supercritical alcohol for which solvent selection is generally
based on the affinity of the solvent for the target reactants/materials. Thus, alcohol–
water mixtures can be used advantageously as a solvent medium because solvent
properties can be continuously varied between those of methanol and water without
phase separation at appropriate T and P conditions. Methanol and water have mutual
miscibility at ambient conditions, which can be attributed to the hydrogen bonding
interaction between –O– and –H– in both molecules.
The introduction of alcohol as the co-solvent has several major advantages [8]:
First, the critical temperature and critical pressure of alcohol are far below than
those of water, so the alcohol-water solvent, as the medium, can significantly relieve
reaction conditions. Second, alcohol has the ability to act as a hydrogen donor,
which can provide active hydrogen in the liquefaction process. Besides, alcohol can
react with the acidic components of biofuel products by esterification reaction to
obtain fatty acid ethyl esters similar to biodiesel, which would reduce the acidity of
the biofuel and improve its quality. Third, due to relatively lower dielectric constant
of alcohol compared with that of water, alcohol can readily dissolve relatively
high-molecular-weight products such as in the case of microalgae biomass. Finally,
among all the supercritical organic solvents, many alcohols (e.g. ethanol, methanol,
and butanol) can be derived as a renewable resource.
3.3.1.1 Liquefaction of Lignocellulose Biomass to Biofuel
In the liquefaction process of lignocelluloses biomass, macromolecule compounds

are decomposed into fragments of lighter molecules, and these reactive fragments
are then repolymerized into oily compounds (biofuel) [35]. However, the biofuel
obtained from the liquefaction process with pure water is a viscous tarry lump,
which is difficult to handle. When an organic solvent (e.g., 2-propanol [36], butanol
[37], ethyl acetate [38], acetone [39], and glycerine [40]) is added to the reaction
system, biofuel with low viscosity is obtained. In addition Yao [41] studied the
liquefaction of lignocellulose with a mixture (e.g., methanol mixed with acetone)
to investigate the effect of the mixture on the liquefaction process. The liquefaction
of lignocellulose with the methanol–acetone mixture might be more significantly
enhanced than that with a single solvent.
Cheng et al. [9] reported that alcohol (methanol or ethanol) and water showed
synergistic effects on biomass liquefaction, and the 50 % co-solvent of either
methanol-water or ethanol-water was found to be the most effective solvent for the
liquefaction of eastern white pine sawdust. The 50 % aqueous alcohol at 573 K for
15 min produced a biofuel yield at approximately 65 wt.% and a biomass conversion
of >95 wt.%. At a temperature higher than 573 K, conversion of biofuel to char
was significant by repolymerization. Analyses of the obtained biofuels confirmed
the presence of primarily phenolic compounds and their derivatives, followed by
aldehyde, long-chain (and cyclic) ketone and alcohol, ester, organic acid, and ether
compounds [9]. Inspired by this research work, liquefaction of woody biomass in the
Fig. 3.5 Yields of biofuel,

SR, gas, and AP from the
liquefaction at 573 K in
ethanol-water solvent
mixtures (Reprinted with
permission from Cheng et al.
[9], Copyright © 2010,
American Chemical Society).
SR solid residue, AP aqueous
soluble products
alcohol-water mixture at a higher temperature will be of interest for the production

of biofuel and phenolic feedstock for the synthesis of bio-based phenolic resins.
Using water alone, the liquefaction produced 70 wt.% biomass conversion and
40 wt.% biofuel yield compared to about 42–43 wt.% biomass conversion and 23–
26 wt.% biofuel yield from the operations using pure ethanol. Ethanol could reduce
the surface tension of the liquefied products and, hence, improve the diffusion of
solvent to the lignin matrix. Ethanol is also expected to readily dissolve relatively
high-molecular-weight liquid products/intermediates derived from cellulose, hemi-
cellulose, and lignin because of their lower dielectric constants when compared
to that of water. The high activity of hot-compressed water (HCW) might also
be related to many special properties of water in subcritical state; e.g., it has a
lower dielectric constant, fewer and weaker hydrogen bonds, a higher isothermal
compressibility, and an enhanced solubility for organic compounds than ambient
liquid water. Moreover, HCW has been found very effective for promoting ionic,
polar non-ionic, and free-radical reactions [42], which make it a promising reaction
medium for biomass liquefaction.
The synergistic effects of water and alcohol has also been provided previously
in organosolv delignification of woody biomass at 463 K, where 50 % methanol-
water solution or 50 % ethanol-water solution was found to be very effective
for wood delignification [43, 44]. Although a methanol-water solvent is as active
as an ethanol-water solvent for biofuel production and biomass conversion, the
ethanol-water mixture has attracted more interest, simply because ethanol is a
renewable resource, which can be obtained readily by fermentation of sugars.
Figure 3.5 revealed that mono-solvent of either pure ethanol or pure water was less
82 Y. Wu et al.
Table 3.4 Effect of volume ratio on the yield of products (biofuel, residue, and total product:
wt.%) and the HHV of biofuel (MJ kg1 ) using mixed solvent at 573 K
Ethanol-water 2-Propanol-water
Total Total
Ratio (v/v)a Biofuel Residue product HHV Biofuel Residue product HHV
0:1 29.05 20.51 49.56 27.33 29.05 20.51 49.56 27.33
1:9 35.83 20.48 56.31 28.96 31.53 16.75 48.28 28.09
3:7 36.78 18.52 55.30 27.86 38.20 16.43 54.63 28.40
5:5 38.35 19.21 57.36 29.66 39.70 19.66 59.36 30.75
6:4 36.00 21.82 57.82 31.13 35.75 22.18 57.93 31.56
8:2 21.21 27.27 48.48 33.11 22.20 21.83 44.03 33.93
10:0 13.02 27.91 40.93 33.46 15.11 21.92 37.03 35.78
Reprinted with permission from Yuan et al. [12], Copyright © 2007, Elsevier
a
Ratio as the ratio of alcohol/water
effective for biomass conversion or biofuel yields compared to the ethanol-water

mixture consisting of <50 % ethanol. The biofuel yield and biomass conversion
peaked with the 50 % ethanol solution, but interestingly, both the biofuel yield and
conversion were greatly reduced as the ethanol content in the ethanol-water co-
solvent increased to 75 wt.%.
For the biomass liquefaction process at an elevated temperature in a closed
system, ionic and radical reactions, including nucleophilic, electrophilic, hydrolysis,
and thermal cracking/pyrolysis reactions, would occur [45, 46]. In addition to the
special properties [42] of the HCW and subcritical water as described before,
alcohols are slightly weaker acids than water. As a result, the lower activity of
the pure ethanol for biomass liquefaction than that of pure water may be predicted
because of the limited hydrolysis reactions in the system with pure ethanol. The
addition of ethanol as a co-solvent into water would thus enhance the solvolytic
liquefaction of biomass, which was previously observed by Pasquini et al. [44].
Yuan et al. [12] reported the liquefaction of rice straw to biofuel with
sub/supercritical mixtures (ethanol-water and 2-propanol-water mixture) as medium
in a 1,000 mL autoclave at 533–623 K, 6–18 MPa, respectively. The results showed
that the maximum yield of biofuel was 39.7 wt.% for the 2-propanol/water volume
ratio of 5/5 at 573 K, while the higher heating value (HHV) of biofuel increased with
the reaction temperature and solvent volume ratio. Using a mixture could inhibit the
formation of residue and then promote the conversion of rice straw with the ratio
of 1/9-5/5 (See Table 3.4). It can be concluded that the formation mechanism of
low-boiling-point materials from subcritical liquefaction of rice straw with diluted
hydrogen-donor solvent.
Published work of Liu et al. [47] showed, during the liquefaction of rice husk
for biofuel production, the mixed solvent (ethanol-water) showed synergistic effect,
which combined the advantages of pure water and ethanol. The compositions of
biofuel from rice husk consisted mainly of phenolic compounds, long-chain alkanes,
ketones and esters. Compared with pure water or ethanol, the relative concentration
of phenolic compounds in mixed solvent run was higher, which suggested that
mixed solvent could promote the formation of phenolic compounds. When using
mixed solvent as reaction solvent, the water could act as a nucleophile and reacted
with some active centers in the proto-lignin [3]. Ye et al. [7] also proposed that
ethanol-water mixed solvent showed synergistic effects on lignin hydrothermal
degradation.
The critical temperature of the ethanol-water co-solvent [when the ethanol-water
ratios are 5/5] is about 587 K [11]. The hydrogen donor capability of ethanol was
promoted in the sub/supercritical condition [12]. On the one hand, water could
lead to high biofuel yield, but that decreased at the higher water amount for the
formation of solid residue (SR) via self-condensation reactions [48]. On the other
hand, ethanol, as extraction solvent, has much higher dissolving power for oils [49].
The adding ethanol could act as both reaction substrate and hydrogen-donor in
the liquefaction process [4]. The combination of these advantages may prevent the
formation of residue and improve the biofuel yield.
A mechanism was proposed in earlier literature [12] that ethanol had a hydrogen
donor capability to stabilize the free radicals generated and thus reduce the
repolymerization reaction, finally, low-boiling-point materials were produced which
might be vaporized and vented in subsequent separation procedure. With proper
amount of water, ethanol was diluted by it, and the formation of high-boiling-point
materials (labeled as biofuel) was promoted. Former works [12, 14] have reported
that the critical point and the dielectric constant of the alcohol/water mixture would
be lower than that of pure water, which led to milder conditions for the reaction
and the increase of the solubility of relatively high molecular weight products from
cellulose, hemicelluloses, and lignin.
All of these aspects demonstrate that ethanol–water mixed solvent is a better
choice than a single solvent at relatively low temperatures. Thus, the highest biofuel
yield is obtained with the use of ethanol–water mixture.
3.3.1.2 Conversion of Aquatic Biomass (Algae)
D. tertiolecta was liquefied in sub/supercritical ethanol-water at different ethanol

contents, varying from 0 to 100 % at 593 K for 30 min to determine the optimal
ethanol content by Chen et al. [8]. The biofuel, SR, and others yields as well
as the conversion, were plotted against the ethanol content as shown in Fig. 3.6.
Figure 3.6 reveals that using either ethanol or water as the medium is less effective
for conversion or biofuel yield compared with the ethanol-water mixed solvent.
Ethanol and water showed synergistic effects on direct liquefaction of D. tertiolecta.
The conversion and biofuel yield peaked with the 40 % ethanol solution, but inter-
estingly, both the conversion and biofuel yield were greatly reduced as the ethanol
content in the co-solvent increased to 80 %. This phenomenon can be explained as
follows, when the ethanol content is 40 %, the critical temperature of the co-solvent
is about 599 K [12], and the actual liquefaction temperature (593 K) is very close to
the critical point. With increasing of ethanol content, the critical pressure and critical
84 Y. Wu et al.
Fig. 3.6 Effect of ethanol content on the biofuel, others and SR yields for direct liquefaction of
D. tertiolecta at 593 K in ethanol-water co-solvent. The value of D. tertiolecta conversion (wt.%) is
shown in parentheses (Reprinted with permission from Chen et al. [8], Copyright © 2012, Elsevier)
temperature of the mixed solvent decrease, therefore, the liquefaction process was
carried out under sub/supercritical conditions in this study. Sub/supercritical water
can provide ionic, polar non-ionic and free radical in biomass liquefaction for
biofuel [9]. Since ethanol is slightly weaker acids than water, the increase ethanol
content resulted in the decrease of free-radical derived from ionization water, which
is not conducive to D. tertiolecta liquefaction, and brought on the biofuel yield
reduce. The best ethanol content was chosen as 40 %, where the optimal conversion
and biofuel yield were 98.24 wt.% and 64.68 wt.%, respectively.
The results indicated that the added ethanol not only reacted with the amides
and/or acids to form ethyl esters but was also used as a hydrogen donor in the
liquefaction process. Previous studies have mentioned the function of ethanol as
hydrogen donors in the algae liquefaction processes for biofuel [9, 12]. However,
the liquefaction of algae with sub/supercritical ethanol–water as the medium is a
complex process, and numerous reactions could occur. The function of ethanol as
a hydrogen donor cannot be regarded as a simple dehydrogenation of ethanol, and
direct evidence of the hydrogen donor effect could be obtained via the hydrogen
isotopic tracer method.
Based on the research results by Chen et al. [8], a plausible reaction mechanism
as described in Fig. 3.7, there is a competitive reaction with carboxylic acid
between ammonia and ethanol to form amide and ester during the liquefaction
Ester, Ketone, Hydrocarbon, etc.
+H+ +H+ +H+

Protein * HN CH CO * H N CH-COOH
+H2O 2
R
R n
NH3
R
+H+
O
R1COO CH2 C2H5OH
R O
+H+
D. tertiolecta Fat
2
R COO CH HO C2H5OH +H+
+H2O O
3 R
R COO CH2 Carboxyl acid
CH2 OH +H+
Triacylglycerides H2N
NH3
CH OH O
HOH CH2 OH
+ HO
+H +
Carbohydrate HO +H
HO OH
H OH Ester, Ketone, Hydrocarbon, etc.
H H
Glucose
Ester, Ketone, Hydrocarbon, etc.
C 2H5OH
H2O
H OH
HO
HO
HO O
H OH
H H
Glucose-anhydride
Fig. 3.7 Liquefaction mechanisms of the main chemical components in D. tertiolecta (Reprinted
with permission from Chen et al. [8], Copyright © 2012, Elsevier)
process. Thus, the amide amount increases compared with the reduced amount of
the carboxylic acid. When the ethanol content is above 40 %, the amides begin
to react with ethanol. Accordingly, the amide and acid contents are reduced. This
finding suggests that the hexadecanoic acid ethyl esters are not only obtained
from the esterification reaction of hexadecanoic acid and ethanol, but also from
alcoholysis of hexadecanoic amides. Sub/supercritical ethanol-water are weak
acid, the D. Tertiolecta liquefaction is considering as an acid-catalyzed process.
Under acidic conditions, proteins first form a long peptide chain, which is then
hydrolyzed to form amino acids. The amino acids undergo cracking, condensation,
decarboxylation, and deamination, etc. to form a liquefied product. Cellulose and
other carbohydrates undergo dehydration to form monosaccharides, a part of which
may then react with ethanol to form the ether that exists in the solid residues. Most of
the monosaccharides may further react to generate carboxylic acid or other organic
compounds that undergo the acid-catalyzed process [1]. The carboxylic acid can
then react with the solvent (ethanol) via alcoholysis to form carboxylic acid esters
and to undergo ammonolysis with ammonia molecules from the decomposition
of protein to obtain amides. The amides can also undergo alcoholysis to generate
carboxylic acid esters when the concentration of ethanol is high. The esterification
reaction can occur for fat in the supercritical ethanol-water. In addition, during
the acid-catalyzed decomposition of proteins, aside from carboxylic acid and its
derivatives, ammonia molecules may also be produced, which may be used as an
ammonolysis reagent in the direct liquefaction process.
86 Y. Wu et al.
Overall, the reduction of organic acids and amides, as well as the increase in
fatty acid esters, could significantly improve the biofuel quality. Generally, the low
organic acids reduce acid value and corrosivity of the biofuel. Similar to biodiesel,
the biofuel obtained in liquefaction contains numerous organic acid esters, which
improve the biofuel quality. In brief, the addition of ethanol not only makes reaction
condition mild and elevates the biofuel yield, but also improves the quality of
the biofuel. The study presents the novel approach to produce biofuel by direct
liquefaction of D. tertiolecta.
3.3.1.3 Co-liquefaction of Biomass and Coal
Biomass, an environmentally friendly and renewable energy, is one of the largest

energies in the world because of its many advantages, including reproducibility, low
pollution, and availability. The co-liquefaction of coal and biomass has received
increasing attention because of the advantages of hydrogen of biomass, thereby
resulting in a decrease in hydrogen consumption. This process also involves mild
operation compared with direct coal liquefaction.
Yang et al. [50] presented the co-liquefaction of coal and D. tertiolecta to produce
liquid fuel with sub/supercritical water-ethanol as the reaction medium in a batch
autoclave at high temperature and pressure. The optimal liquid fuel yield and
conversion were 40.29 wt.% and 70.62 wt.%, respectively, in the sub/supercritical
water-ethanol with ethanol content as 60 %, at 633 K, feedstock/reaction medium
ratio of 1:10 (g mL1 ), and reaction time of 30 min. The experimental results
showed that co-liquefaction of coal and D. tertiolecta could not be estimated as a
weighted average sum of the pure feedstock. Under the optimal reaction conditions,
the synergetic effect values of liquid fuel yield and conversion were 12.53 wt.% and
15.74 wt.%, respectively. On the basis of elemental analysis and FT-IR spectroscopy,
the generation of esters from the esterification reaction of ethanol and acid enabled
the increase of the quality of liquid fuel. The HHV of co-liquefaction liquid fuel
(Co-LF) was 36.70 MJ kg1 .
With a increase of the ethanol content from 80 to 100 %, both conversion and
liquid fuel were decreased slightly, because the “water-gas shift reaction” could not
occur without water, where CO can derive from the oxidation reaction of coal [51].
The variation tendency of preasphaltene and asphaltene (PAA) is contrary to liquid
fuel. The generally acceptable essence of co-liquefaction of coal and biomass is
that the weak chemical bonds in the coal or biomass macromolecular, such as C-O-
C or C-C, are pyrolyzed to form some large free radical fragments, which can be
terminated by hydrogen or other smaller fragments to form PAA, and then PAA is
converted to liquid fuel and gas [52]. Thereby, the sub/supercritical water-ethanol
co-solvent is selected as the promising reaction medium with broad application
prospects.
Scheme 3.1 Possible lignin hydrothermal degradation mechanism in water-ethanol mixture under
hydrogen (Reprinted with permission from Chandler et al. [55], Copyright © 2002, American
Chemical Society)
3.3.1.4 Depolymerization of Biomacromolecule
Lignin is a natural phenolic polymer displaying high molecular weights which

accounts for 20–35 wt.% in woody biomass (40–50 wt.% in bark) and 10–20 wt.%
in agricultural stems. As well known, one practical industrial application of lignin
is to substitute petroleum-derived phenol for the synthesis of phenol-formaldehyde
(PF) resin. Direct use of lignin as a replacement for phenol in PF resins can hardly
be achieved due to fewer reactive sites and steric hindrance effects caused by its
complex chemical structure [53]. In this regard, an effective way to improve the
reactivity of lignin by depolymerization of lignin into phenolic compounds with
lower molecular weights is crucial for the synthesis of bio-based phenolic resins.
Hydrothermal degradation of an alkali lignin (AL) was achieved in sub/
supercritical ethanol-water or pure ethanol with and without a catalyst [54]. Solvent
and reaction temperature played a major role in the product yields and properties.
Effective degradation of AL was achieved in 50/50 water-ethanol at 473–723 K
under 5 MPa H2 . Hydrothermal treating AL at 573 K for 2 h without a catalyst
led to an 89 wt.% yield of degraded lignin (DL). Compared to the 50/50 water-
ethanol treatment, the pure ethanol treatment at 573 K led to a lower yield of DL
(<15 wt.%). The lower DL yield in pure ethanol than that in water might be due
to less or no hydrolysis reaction. The addition of ethanol as a co-solvent into water
would thus enhance the solvolytic/hydrolytic degradation of the lignin as markedly
shown.
The cleavage of carbon-carbon bonds needs more severe conditions. The possible
hydrothermal degradation mechanisms for lignin in sub/supercritical water-ethanol
may be proposed and illustrated in Scheme 3.1. Under sub/supercritical water-
ethanol, hydrogenolysis and hydrolysis occurred at ether bonds to form aryl
hydroxyl groups and alcohol hydroxyl groups and alkyl groups. A fraction of
degraded products/intermediates will undergo further hydrocracking to form low-
molecular-weight products, such as monophenols, benzene derivatives and alcohols.
88 Y. Wu et al.
Addition and condensation reactions among the degraded products/intermediates

may take place simultaneously to form oligomers, char and coke. The addition and
condensation reactions of the highly reactive lignin degradation intermediates would
explain the yield of SR increasing at a higher temperature than 598 K in 50/50 water-
ethanol.
Beyond that, hydrolysis of cellulose by hot-compressed ethanol-water mixture is
a promising way to obtain reducing sugar (RS) for biofuel production. McCormick
et al. [56] conducted the hydrolysis of microcrystalline cellulose in hot-compressed
alcohol-water mixtures. Among three kinds of alcohol-water mixtures, the hot-
compressed ethanol-water mixture was the most appropriate system to hydrolyze
cellulose into RS. The RS yield of cellulose hydrolysis was reached as high
as 98.22 wt.%, under the optimal conditions of ethanol mole fraction of 0.22,
temperature of 533 K, and pressure of 5.75 MPa. The equation expressed the
relationship between the RS yield and the density as well as the ethanol mole
fraction in ethanol-water mixture system was proposed. The greater the density of
the ethanol-water mixture, the higher the yield of the RS. Hydrolysis of cellulose
by ethanol-water mixture cannot only obtain a high yield of RS but also reduce
energy consumption, without pollution. This work proposed a green and efficient
RS production method.
3.3.1.5 Transesterification for Production of Biodiesel
Over the past years, biodiesel has been recognized as a feasible alternative fuel
to mineral diesel either as an additive or replacement, and its advantages include
being non-toxic, biodegradable, domestically produced, and renewable. Moreover,
biodiesel has a cetane number that is higher than diesel from petroleum and a better
combustion emissions profile, such as reduced levels of particulate matter, carbon
monoxide, and nitrogen oxides [57].
The most common way to produce biodiesel is widely known to be via
transesterification reaction and such a reaction is conventionally performed using
alkaline, acid or enzyme catalysts [57]. Chemical transesterification through alkali-
catalyzed processes affords high conversion in short reaction times, but it suffers
from several drawbacks. On the other hand, transesterification using acid catalysts
is much slower than that catalyzed by alkali catalysis, may lead to the formation of
undesirable by-products [58]. It has been suggested the use of enzyme-catalyzed
transesterification methods can overcome the drawbacks of chemical-catalyzed
processes, but at present, the high cost of enzyme production still remains the major
obstacle to commercialization [59].
The supercritical method, a catalyst-free technique for the transesterification of
vegetable oils by using an alcohol at supercritical conditions, has been successfully
proposed [60]. Several advantages, such as improved phase solubility, reduction of
mass-transfer limitations, higher reaction rates, easier separation and purification
steps, and higher quality of the glycerol produced, are attributed to the use of this
method. This method is more tolerant to the presence of water and free fatty acids
than the conventional alkali-catalyzed technique, thus, the method more versatile in
terms of the use of various types of vegetable oils as feedstock [60].
Many previous studies were directed to investigate the effect of water content
on the efficiency of soybean oil transesterification with supercritical ethanol (SC-
EtOH) [60] and the effect of reaction conditions on the composition profile of fatty
acid ethyl esters at temperatures ranging from 523 to 623 K and 20 MPa with an oil:
ethanol ratio of 1:40 in a continuous process [61]. Results obtained in those works
showed that although high conversions of oil to biodiesel were achieved, even with
a 10 wt.% of water in the water-ethanol mixture, the ester content in the product
was not higher than 77.5 wt.%. This relatively low ester content was due to the
degradation of the fatty acids, to a certain extent, to by-products.
It has been also reported by other researchers that supercritical methanolysis
of vegetable oils (SC-MeOH) has resulted in full ester yields for substantially
high water contents (up to 50 wt.% on the alcohol basis) in the reaction medium,
using a batch reaction system [60], without reporting the occurrence of fatty acid
degradation. These results suggests that ethanolysis could be more susceptible than
methanolysis to water presence on the reaction system or that, in spite of the
difference between both alcohols, the water effect is highly dependent on the type
the process: bath or continuous mode.
Ignacio et al. [63] affirmed the presence of 10 wt.% water in the reaction medium
presented higher ester content than those performed without the addition of water,
it can be inferred that water played a favorable effect on the conversion of oil to
biodiesel. This favorable effect has been previously evidenced in studies performed
on the supercritical synthesis of ethyl esters from soybean oil or rapeseed oil [64].
In this case, water can be regarded as co-solvent of alcohol, so it is special case
about application of supercritical water-alcohol mixture on biomass conversion.
These results are in agreement with the hypothesis that the presence of water in
the reaction medium involves the occurrence of the reaction following a faster
mechanism parallel to the anhydrous transesterification, involving the hydrolysis
of the triacylglycerols followed by the fast esterification of the free fatty acids with
the alcohol [64]. The addition of water to a level of 10 wt.% to the reaction medium
permitted to increase both methyl and ethyl ester contents by the combination of
two favorable effects: increasing the reaction rate and reducing the degradation of
fatty acids.
3.3.2 1, 4-Dioxane as the Co-solvent
Li et al. [3] studied the critical liquefaction of rice straw in sub/supercritical 1,

4-dioxane-water mixture with a 500 mL autoclave at temperature of 533–613 K,
resistance time of 0–20 min, and 1, 4-dioxane volume ratios of 0–100 %. The
optimal yields of biofuel and PAA were in the range of 29.64–57.30 wt.% and
6.42–22.68 wt.%. The synergistic effect of 1, 4-dioxane-water mixture could allow
the great decomposition of the tubular structure of lignocelluloses.
90 Y. Wu et al.
Fig. 3.8 Effect of volume

ratio on the products yield
with 1, 4-dioxane-water
mixture at 57 (Reprinted with
permission from Li et al. [3],
Copyright © 2009, Elsevier)
In Fig. 3.8, the yields of biofuel obtained at the ratios of 50, 80 and 100 %
were higher than that obtained in pure water run. It can be probably due to the
relation between the value of oxygen concentration and polarity of water soluble
fraction (biofuel) in biomass liquefaction products [65]. High oxygen composition
increases the polarity of the liquid components and makes them more water-
soluble. So, “oxygen-transfer” capability of 1, 4-dioxane may give a contribution
to the increment of the polarity of biofuel fractions. Among the three components
of biomass, lignin was the most difficult one to decompose, and consequently,
the amount of solid residue increased in proportion to the lignin content in the
liquefaction process.
Polycarbonate is an important thermoplastic material among other polymers
and has significant properties such as good stability, excellent flexibility and good
transparency [66]. Increasing of waste polycarbonate specially to form of plastics
and CDs, it is necessary to decompose it with suitable methods. One of the important
techniques for degradation of polymers is chemical recycling. Chemical recycling
process for polymers and plastics are mainly divided into glycolysis, methanolysis,
aminolysis and hydrolysis [67]. Process of conventional chemical recycling was
done usually at high pressure and temperature during long time processing [68].
Fu-sheng Liu et al. [69] investigate the hydrolysis degradation of polycarbonate
(PC) under moderate condition to achieve pure the solid product. The hydrolysis
degradation of PC was done by using 1, 4-dioxane, tetrahydrofuran and N-methyl-
2-pyrrolidone as co-solvent and also water as a main solvent. Using of different
types of co-solvent has significant effect on yield of BPA. The yield of BPA
was 70 and 68 wt.% respectively when using tetrahydrofuran (THF) and 1, 4-
dioxane as co-solvent. Depolymerization of PC was done by using SCW and
used methanol as solvent and BPA was achieved as a main monomer. BPA yields
has reached to 90–93 wt.% [70]. The methanolysis of PC was carried out by
CH3 OH without any solvent, the reaction time was 2 h and the yields of BPA
and DMC were not achieved. However, by using solvent such as 1, 4-dioxine, 1,
2-dichloroethane, N-methyl-2-pyrrolidone or tetrahydrofuran, the yield of monomer

is increased rapidly. According to their study yield of bisphenol A (BPA) reached
to 78 wt.% in the present of THF as a solvent however this value increased up to
79 wt.% at the present of 1, 4-dioxane as the solvent [71]. Nayeleh and Abdul [72]
confirmed the hydrolysis degradation of PC by using 1, 4-dioxane as the co-solvent
under microwave irradiation. The highest yield of the solid product (85 wt.%) was
achieved by using 20 g 1, 4-dioxane as co-solvent, 3.5 g H2 O as main solvent and
0.5 g NaOH as catalyst during 12.5 min. The obtained results show that using 1,
4-dioxane as a co-solvent lead to decrease the yield of BPA.
3.3.3 Phenol as the Co-solvent
Phenol is a kind of important organic compounds. The use of phenol and water-
phenol mixtures is an effective liquefaction technique for lignin [73]. Wayman
and Lora [74] reported that the addition of phenolic compounds suppressed the
condensation among reactive intermediates from the lignin decomposition products
during auto autohydrolysis at 448 K. Lin et al. [75] examined the reaction of
guaiacygrycerol-“-guaiacyl ether as lignin model compounds in a water-phenol
mixture at 523 K and concluded that the condensation reaction could be suppressed
in the presence of phenol. In subcritical and SCW, the reaction of phenol with
alcohols and aldehydes yields alkylphenols through alkylation without a catalyst
[55]. These phenomena suggest that phenol acts as a capping agent and prevents
char formation in SCW; therefore, the water-phenol mixture can also be an effective
solvent for the decomposition of lignin.
Motofumi et al. [73] examined the decomposition of lignin in SCW with and
without phenol at 673 K. In the absence of phenol, the yield of tetrahydrofuran
(THF)-insoluble (TIS) products decreased and the molecular weight distribution
of THF-soluble (TS) products shifted toward lower molecular weights as the
water density increased, so the increase in water density could enhanced the
lignin conversion. When adding phenol, the yield of TIS products was lower and
the molecular weight distribution of TS products shifted toward lower molecular
weights than those in the absence of phenol. Some alkylphenols were obtained only
in the presence of phenols, due to the reaction of phenol with the decomposition
products. These results show that the reaction of phenol with reactive sites occurred
in SCW and suppressed cross-linking reactions among reactive sites of large
fragments. This promoted the decomposition of lignin to lower-molecular-weight
compounds.
The hydrolysis is an important reaction to disassemble lignin in SCW. However,
the formation of char due to repolymerization occurs at the same time. This seems
a main reason why the yield of chemicals is so low in the lignin conversion. Saisu
et al. [76] demonstrated the conversion of organosolv lignin into chemicals with a
water-phenol (2.5 g: 0.75 g) mixture at 673 K. They suggested that phenol acted
as a capping agent to prevent char formation in SCW. Nevertheless, complete
92 Y. Wu et al.
Fig. 3.9 Reaction scheme of lignin conversion in water and water-phenol mixture (Reprinted with
permission from Okuda et al. [77], Copyright © 2004, Elsevier)
suppression of char formation could not been achieved by their method. Longer
reaction time increased the formation of char. A mixture of water and phenol gave
no char formation in the decomposition of lignin in supercritical water-phenol at
673 K. Although a detailed mechanism has not been elucidated, according with
the results reported by Kazuhide et al., the reaction scheme of lignin conversion
as shown in Fig. 3.9 [77]. The lignin structure was decomposed by hydrolysis
and dealkylation to form compounds with the average molecular weight of around
1,000 and low molecular weight compounds from 100 to 300 without formation
of the residual solids. This reaction mainly occurs at the sites in lignin (ether,
hydroxyl group, etc.) that are readily cleaved. At this stage, cross-linking reactions
are depressed due to entrapment of active fragments and capping of active sites
by excess phenol they are formed in rapid decomposition of lignin by hydrolysis
and dealkylation. The compounds with the average molecular weight of around
1,000 are captured by phenol and are probably relatively stable. During the longer
reaction time, the compounds around 1,000 of average molecular weight formed
slowly decompose to form lower molecular weight compound s from 100 to 300
Scheme 3.2 Proposed

conversion pathway of
organosolv lignin into BMP
in a water-p-cresol mixture at
673 K (Reprinted with
permission from Chandler
et al. [55], Copyright © 2002,
American Chemical Society)
without formation of the residual solids, the low molecular weight compounds are
syringols, guaiacols and catechols derived from lignin phenol unit. Alkylphenols,
xanthene, dihydroxy-phenylmethanes are also produced from the original lignin
or the reaction between phenol and formaldehyde [77] (Fig. 3.9). On the other
hand, reaction in water recombines large molecular weight fragments due to active
fragments (form aldehyde, etc.) and sites in heavier components, to elucidate the
detailed pathway. The mechanism and kinetics study is in progress.
3.3.4 p-Cresol as the Co-solvent
In order to convert lignin into single chemical species at high efficiency, Kazuhide
Okuda et al. [77] examined mixtures of water and p-cresol for selectively producing
chemicals from lignin without the formation of char. In both water and p-cresol
solvents, repolymerization occurred to some extent and gave the THF insoluble
components. The yield of the THF soluble components decreased as reaction
proceeded and the yields were about 82 wt.% after 60 min and 97 wt.% in pure
water and pure p-cresol, respectively. On the other hand, the products in the mixture
were completely soluble in THF throughout the reaction.
Kazuhide Okuda et al. [78] has realized chemical recovery of a phenolic
compound 2-(hydroxy-benzyl)-4-methyl-phenol (BMP), from organosolv lignin
without forming char in a mixture of SCW and p-cresol. On the basic of this, Seiichi
Takami et al. [79], from the same group, evaluated the reaction rate constants of
the depolymerization process of organosolv lignin using Monte Carlo simulation.
The reaction scheme is shown in Scheme 3.2. The obtained reaction rate constants
gave insight into the mechanism why the mixture of SCW and p-cresol suppressed
the formation of char. They also suggested the reaction path for the formation of
BMP. The results indicated that the faster reaction between p-cresol and fragmented
intermediate species enabled the chemical conversion of organosolv lignin without
forming char. The activation energy for the formation of BMP suggested that the
rate-limiting process in BMP formation from intermediate species was the reaction
94 Y. Wu et al.
of the ˛-carbon of glycerol in GGGE with p-cresol. In addition, these reaction rate
constants enable the design and optimization of the chemical recovery processes to
realize the highest yield.
Conversion of organosolv lignin into BMP involved hydrolysis (depolymeriza-
tion) by SCW to form reactive fragmented intermediate species. In the presence
of p-cresol, the intermediate species was efficiently stabilized by the reaction
with p-cresol to form BMP (Scheme 3.2). The proposed reaction path involves
dehydroxylation at the ˛-position, followed by the reaction with p-cresol.
Seiichi Takami et al. [79] suggested that lignin decomposed via hydration
and dealkylation to form low-MW fragments. These fragments having reactive
functional groups produced higher MW products through a cross-linking reaction.
However, in the presence of p-cresol, the capping of the carbonium ion by p-
cresol might prevent unfavorable polymerization. The rate constant for the capping
reaction is four to five times faster than the depolymerization rate of organosolv
lignin. This comparison supports the above analysis. The fragmented intermediate
species, which was produced by depolymerization of organosolv lignin, was quickly
stabilized by p-cresol. This quick capping possibly suppressed the formation of char
by prohibiting polymerization of reactive intermediate species.
3.4 Conclusions and Outlook
Although excellent experimental results have been achieved in the laboratory, and
the co-solvents of SCW are important in the process of biomass thermochemical
conversion, thereby providing many potential benefits over conventional methods of
processing biomass to biofuels or chemicals, specific issues need to be addressed.
First, given that the supercritical system involves harsh reaction conditions with
high temperature and pressure, solving the issue of industrial-scale production and
building an industrial production unit are important concerns.
Second, a large number of inorganic salts exist in the biomass, which results in
the plugging of reactors caused by the precipitation of inorganic salts above the
supercritical temperature and at low-density conditions. At normal temperatures
and pressures, water is an excellent solvent for most inorganic salts. However,
the solubility of most salts is very low in the supercritical system, particularly
with the addition of an organic solvent. Nevertheless, the problem may become an
opportunity because the by-product of the process can be used to produce valuable
fertilizer, if managed properly.
Third, inorganic matter such as halogens, sulfur, and phosphorous present in the
biomass are converted to their respective acids, thereby causing severe corrosion
on the reactor wall at harsh reaction conditions. The corrosion problem can be
reduced or avoided by selecting appropriate construction materials or by developing
a slightly modified reactor concept.
Therefore, the application of SCW and other supercritical solvents on biomass

conversion is still confined to basic laboratory research. Moreover, many problems
about industrial-scale production need to be addressed.
References
1. Andrew AP, Frédéric V, Russell PL, Morgan F, Michael Jr JA, Jefferson WT. Thermochemical
biofuel production in hydrothermal media: a review of sub- and supercritical water technolo-
gies. Energy Environ Sci. 2008;1:32–65.
2. James WL. Advanced biofuels and bioproducts. New York: Springer; 2013. ISBN 978-1-4614-
3348-4.
3. Li H, Yuan XZ, Zeng GM, Tong JY, Yan Y, Cao HT, Wang LH, Cheng MY, Zhang JC, Yang D.
Liquefaction of rice straw in sub- and supercritical 1, 4-dioxane-water mixture. Fuel Process
Technol. 2009;90:657–63.
4. Huang HJ, Yuan XZ, Zeng GM, Wang JY, Li H, Zhou CF, Pei XK, You Q, Chen L.
Thermochemical liquefaction characteristics of microalgae in sub- and supercritical ethanol.
Fuel Process Technol. 2011;92:147–53.
5. Karagoz S, Bhaskar T, Muto A, Sakata Y. Comparative studies of oil compositions produced
from sawdust, rice husk, lignin and cellulose by hydrothermal treatment. Fuel. 2005;84:
875–84.
6. Liu ZG, Zhang FS. Effects of various solvents on the liquefaction of biomass to produce fuels
and chemical feedstocks. Energy Convers Manag. 2008;49:3498–504.
7. Ye YY, Fan J, Chang J. Effect of reaction conditions on hydrothermal degradation of cornstalk
lignin. J Anal Appl Pyrolysis. 2012;94:190–5.
8. Chen Y, Wu YL, Zhang PL, Hua DR, Yang MD, Li C, Chen Z, Liu J. Direct liquefaction
of Dunaliella tertiolecta for bio-oil in sub/supercritical ethanol-water. Bioresour Technol.
2012;124:190–8.
9. Cheng SN, D’cruz I, Wang MC, Leitch M, Xu CB. Highly efficient liquefaction of woody
biomass in hot-compressed alcohol-water co-solvents. Energy Fuel. 2010;24:4659–67.
10. George WH, Sara I, Avelino C. Synthesis of transportation fuels from biomass: chemistry,
catalysts and engineering. Chem Rev. 2006;106:4044–98.
11. Xu CB, Etcheverry T. Hydro-liquefaction of woody biomass in sub- and super-critical ethanol
with iron-based catalysts. Fuel. 2008;87:335–45.
12. Yuan XZ, Li H, Zeng GM, Tong JY, Xie W. Sub-and supercritical liquefaction of rice straw in
the presence of ethanol-water and 2-propanol-water mixture. Energy. 2007;32:2081–8.
13. Wang G, Li W, Chen H, Li B. The direct liquefaction of sawdust in tetralin. Energy Source Part
A. 2007;29:1221–31.
14. Wang CW, Zhou FL, Yang Z, Wang WG, Yu FQ, Wu YX, Chi R. Hydrolysis of cellulose
into reducing sugar via hot-compressed ethanol/water mixture. Biomass Bioenergy. 2012;42:
143–50.
15. Brennecke JF, Eckert CA. Phase equilibria for supercritical fluid process design. AICHE J.
1989;35:1409–27.
16. Takumi O, Shunsuke K, Takaaki H, Yoshiyuki S, Hiroshi I. Volumetric behavior and
solution microstructure of methanol-water mixture in sub- and supercritical state via density
measurement and MD simulation. Fluid Phase Equilib. 2011;302:55–9.
17. Darja P, Valter D. Volumetric properties of ethanol-water mixtures under high temperatures
and pressures. Fluid Phase Equilib. 2005;230:36–44.
18. Dixit S, Crain J, Poon WC, Finney JL, Soper AK. Molecular segregation observed in a
concentrated alcohol-water solution. Nature. 2002;416:829–32.
96 Y. Wu et al.
19. Ikushima Y, Hatakeda K, Saito N, Arai M. An in situ Raman spectroscopy studies of subcritical
and supercritical water the peculiarity of hydrogen bonding near the critical point. J Chem
Phys. 1998;108:5855–60.
20. Lamanna S, Cannistraro R. Effect of ethanol addition upon the structure and the cooperativity
of the water H bond network. Chem Phys. 1996;213:95–110.
21. Lee JH, Foster NR. Oxidation of methanol in supercritical water. J Ind Eng Chem. 1999;5:
116–22.
22. Anitescu G, Zhang ZH, Tavlarides LL. A kinetic study of methanol oxidation in supercritical
water. Ind Eng Chem Res. 1999;1999(38):2231–7.
23. Gao J. Supercritical hydration of organic compounds. The potential of mean force for benzene
dimer in supercritical water. J Am Chem Soc. 1993;115:6893–5.
24. Van Konynenburg PH, Scott RL. Critical lines and phase equilibria in binary Van Der Waals
mixtures. Phil Trans R Soc Lond A. 1980;298:495–540.
25. Weingartner H, Franck EU. Supercritical water as a solvent. Angew Chem Int Ed.
2005;44:2672–92.
26. Carlos ND, Josep BA, Oliver C, Philippe U, Jacqueline R. Dynamical and structural properties
of benzene in supercritical water. J Chem Phys. 2004;121:10566–76.
27. Rasulov SM, Abdulagatov IM. PVTx measurements of water-n-pentane mixtures in critical
and supercritical regions. J Chem Eng Data. 2010;55:3247–61.
28. Kiselev M, Puhovskia Y, Kerdcharoen T, Hannongbua S. The study of hydrophobic hydration
in supercritical water-methanol mixtures. J Mol Graphics Model. 2001;19:412–6.
29. Antal MJ, Allen SG, Schulman D, Xu XD, Divilio RJ. Biomass gasification in supercritical
water. Ind Eng Chem Res. 2009;39:4040–53.
30. Aida TM, Sato Y, Watanabe M, Tajima K, Nonaka T, Hattori H, Arai K. Dehydration of
d-glucose in high temperature water at pressures up to 80 MPa. J Supercrit Fluids.
2007;40:381–8.
31. Watanabe M, Matsuo Y, Matsushita T, Inomata H, Miyake T, Hironaka K. Chemical recycling
of polycarbonate in high pressure high temperature steam at 573 K. Polym Degrad Stab.
2009;94:2157–62.
32. Genta M, Iwaya T, Sasaki M, Goto M. Supercritical methanol for polyethylene terephthalate
depolymerization: observation using simulator. Waste Manag. 2007;27:1167–77.
33. Takesue M, Suino A, Hakuta Y, Hayashi H, Smith RL. Formation mechanism and lumines-
cence appearance of Mn-doped zinc silicate particles synthesized in supercritical water. J Solid
State Chem. 2008;181:1307–13.
34. Saka S, Kusdiana D. Biodiesel fuel from rapeseed oil as prepared in supercritical methanol.
Fuel. 2001;80:225–31.
35. Demirbas A. Mechanisms of liquefaction and pyrolysis reactions of biomass. Energy Convers
Manag. 2000;41:633–46.
36. Ogi T, Minowa T, Dote Y, Yokoyama S. Characterization of oil produced by the direct
liquefaction of Japanese oak in an aqueous 2-propanol solvent system. Biomass Bioenergy.
1994;7:193–9.
37. Ogi T, Yokoyama S. Liquid fuel production from woody biomass by direct liquefaction. Sekiyu
Gakkaishi. 1993;36:73–84.
38. Demirbas A. Effect of lignin content on aqueous liquefaction products of biomass. Energy
Convers Manag. 2000;41:1601–7.
39. Heitz M, Brown A, Chornet E. Solvent effects on liquefaction: solubilization profiles of a
Canadian prototype wood, populus deltoids, in the presence of different solvents. Can J Chem
Eng. 1994;72:1021–7.
40. Demirbas A. Conversion of wood to liquid products using alkaline glycerol. Fuel Sci Technol
Int. 1992;10:173–84.
41. Yao YG. Soluble properties of liquefied biomass prepared in organic solvents. Mokuzai
Gakkaishi. 1994;40:176–84.
42. Akiya N, Savage PE. Roles of water for chemical reactions in high-temperature water. Chem
Rev. 2002;102:2725–50.
43. Li L, Kiran E. Interaction of supercritical fluids with lignocellulosic materials. Ind Eng Chem
Res. 1988;27:1301–12.
44. Pasquini D, Pimenta MTB, Ferreira LH, Curvelo AAS. Extraction of lignin from sugar cane
bagasse and Pinus taeda wood chips using ethanol–water mixtures and carbon dioxide at high
pressures. J Supercrit Fluids. 2005;36:31–9.
45. Kabyemela BM, Adschiri T, Malalua R, Arai MK, Ohzeki H. Rapid and selective conversion
of glucose to erythrose in supercritical water. Ind Eng Chem Res. 1997;36:5063–7.
46. Antal MJ, Leesomboon T, Mok WS, Richards GN. Mechanism of formation of 2-furaldehyde
from D-xylose. Carbohydr Res. 1991;217:71–85.
47. Liu Y, Yuan XZ, Huang HJ, Wang XL, Wang H, Zeng GM. Thermochemical liquefaction
of rice husk for bio-oil production in mixed solvent (ethanol-water). Fuel Process Technol.
2013;112:93–9.
48. Liu HM, Xie XA, Li MF, Sun RC. Hydrothermal liquefaction of cypress: effects of reaction
conditions on 5-lump distribution and composition. J Anal Appl Pyrolysis. 2012;94:177–83.
49. Muppaneni T, Reddy HK, Patil PD, Dailey P, Aday C, Deng S. Ethanolysis of camelina oil
under supercritical condition with hexane as a co-solvent. Appl Energy. 2012;94:84–8.
50. Yang RL, Chen Y, Wu YL, Hua DR, Yang MD, Li C, Chen Z, Liu J. Production of liquid fuel
via co-liquefaction of coal and Dunaliella tertiolecta in sub/supercritical water-ethanol system.
Energy Fuel. 2013;27:2619–27.
51. Ross DS, Green TK, Mansani R, Hum GP. Coal conversion in CO/water. 1. Conversion
mechanism. Energy Fuel. 1987;1:287–91.
52. Guo ZX, Bai ZQ, Bai J, Wang ZQ, Li W. Co-liquefaction of lignite and sawdust under syngas.
Fuel Process Technol. 2011;92:119–25.
53. Alonso MV, Oliet M, Perez JM, Rodriguez F, Echeverria J. Determination of curing kinetic
parameters of lignin-phenol-formaldehyde resol resins by several dynamic differential scan-
ning calorimetry methods. Thermochim Acta. 2004;419:161–7.
54. Cheng SN, Wilks C, Yuan ZS, Leitch M, Xu CB. Hydrothermal degradation of alkali lignin
to bio-phenolic compounds in sub/supercritical ethanol and water–ethanol co-solvent. Polym
Degrad Stab. 2012;97:839–48.
55. Chandler K, Deng F, Dillow AK, Liotta CL, Eckert CA. Alkylation reactions in near-critical
water in the absence of acid catalysts. Ind Eng Chem Res. 1997;36:5175–9.
56. McCormick RL, Graboski MS, Alleman TL, Herring AM. Impact of biodiesel source material
and chemical structure on emissions of criteria pollutants from a heavy-duty engine. Environ
Sci Technol. 2001;35:1742–7.
57. Lotero E, Liu Y, Lopez DE, Suwannakarn K, Bruce DA, Goodwin Jr JG. Synthesis of biodiesel
via acid catalysis. Ind Eng Chem Res. 2005;44:5353–63.
58. Iso M, Chen B, Eguchi M, Kudo T, Shrestha S. Production of biodiesel fuel from triglycerides
and alcohol using immobilized lipase. J Mol Catal B: Enzym. 2001;16:53–8.
59. Madras G, Kolluru C, Kumar R. Synthesis of biodiesel in supercritical fluids. Fuel.
2004;83:2029–33.
60. Kusdiana D, Saka S. Effects of water on biodiesel fuel production by supercritical methanol
treatment. Bioresour Technol. 2004;91:289–95.
61. van Kasteren JMN, Nisworo AP. A process model to estimate the cost of industrial scale
biodiesel production from waste cooking oil by supercritical transesterification. Resour
Conserv Recycl. 2007;50:442–58.
62. Silva C, Weschenfelder TA, Rovani S, Corazza FC, Corazza ML, Dariva C, Vladimir OJ.
Continuous production of fatty acid ethyl esters from soybean oil in compressed ethanol. Ind
Eng Chem Res. 2007;46:5304–9.
63. Ignacio V, da Camila S, Isabella A, Gustavo RB, Fernanda CC, Vladimir OJ, Maria AG, Iván J.
Continuous catalyst-free methanolysis and ethanolysis of soybean oil under supercritical
alcohol/water mixtures. Renew Energy. 2010;35:1976–81.
64. Kusdiana D, Saka S. Methyl esterification of free fatty acids of rapeseed oil as treated in
supercritical methanol. J Chem Eng Jpn. 2001;34:383–7.
98 Y. Wu et al.
65. Zhang T, Zhou YJ, Liu DH, Petrus L. Qualitative analysis of products formed during the acid
catalyzed liquefaction of bagasse in ethylene glycol. Bioresour Technol. 2007;98:1454–9.
66. Zhi YP, Zhen B, Ying XC. Depolymerization of poly (bisphenol A carbonate) in subcritical
and supercritical toluene. Chin Chem Lett. 2006;17:545–8.
67. Vijaykumar S, Mayank RP, Jigar VP. Pet waste management by chemical recycling: a review.
J Polym Environ. 2010;18:8–25.
68. Mohammad NS, Dimitris S, Halim HR, Dimitris NB, Konstantios AGK, George PK.
Hydrolytic depolymerization of PET in a microwave reactor. Macromol Mater Eng.
2010;295:575–84.
69. Liu FS, Li Z, Yu ST, Cui X, Xie CX, Ge XP. Methanolysis and hydrolysis of polycarbonate
under moderate conditions. J Polym Environ. 2009;17:208–11.
70. Chen L, Wu Y, Ni Y, Huang K, Zhu Z. Depolymerization of polycarbonate in critical region of
methanol. Acta Sci Cir. 2004;24:604.
71. Raul P, Juan G, Maria JC. Chemical recycling of polycarbonate in a semi-continuous lab-plant.
A green route with methanol and methanol-water mixtures. Green Chem. 2005;7:380–7.
72. Nayeleh D, Abdul RR. Hydrolysis degradation of polycarbonate using different co–solvent
under microwave irradiation. APCBEE Procedia. 2012;3:172–6.
73. Motofumi S, Takafumi S, Masaru W, Tadafumi A, Kunio A. Conversion of lignin with
supercritical water-phenol mixtures. Energy Fuel. 2003;17:922–8.
74. Wayman M, Lora JH. Aspen autohydrolysis: the effects of 2-naphthol and other aromatic
compounds. Tappi Tech Assoc Pulp Pap Ind. 1978;61:55–7.
75. Lin L, Yao Y, Yoshioka M, Shiraishi N. Liquefaction mechanism of lignin in the presence of
phenol at elevated temperature without catalysts. Studies on “-O-4 lignin model compound. I.
Structural characterization of the reaction products. Holzforschung. 1997;51:316–24.
76. Saisu M, Sato T, Adschiri MWT, Arai K. Conversion of lignin with supercritical water-phenol
mixtures. Energy Fuel. 2003;17:922–8.
77. Okuda K, Umetsu M, Takami S, Adschiri T. Disassembly of lignin and chemical recovery-rapid
depolymerization of lignin without char formation in water-phenol mixtures. Fuel Process
Technol. 2004;85:803–13.
78. Okuda K, Man X, Umetsu M, Takami S, Adschiri T. Efficient conversion of lignin into single
chemical species by solvothermal reaction in water-p-cresol solvent. J Phys Condens Matter.
2004;16:S1325.
79. Takami S, Okuda K, Man X, Umetsu M, Ohara S, Adschiri T. Kinetic study on the selective
production of 2-(Hydroxybenzyl)-4-methylphenol from organosolv lignin in a mixture of
supercritical water and p-cresol. Ind Eng Chem Res. 2012;51:4804–8.
Chapter 4
Thermodynamic Analysis of the Supercritical
Water Gasification of Biomass
Luca Fiori and Daniele Castello
Abstract Thermodynamic equilibrium modeling is important in order to under-

stand which are the expected yields and the energy needs of the supercritical
water gasification (SCWG) process. A thermodynamic equilibrium model allows
calculating which is the system composition at equilibrium. Equilibrium compo-
sition can allow calculating, with good approximation, the overall process yields,
especially when high temperatures and/or catalysts are used. In this Chapter,
the current state of the art in the field of thermodynamic equilibrium modeling
of SCWG is reviewed. The most common modeling approaches are presented,
highlighting their advantages and disadvantages. Then, the practical implementation
of a non-stoichiometric thermodynamic equilibrium model, based on Gibbs energy
minimization, is performed. Finally, it is shown how thermodynamic modeling can
be used to build a process model for SCWG, which is a fundamental tool to evaluate
the process energy sustainability and its practical feasibility.
Keywords Supercritical water gasification • Modeling • Thermodynamics •

Thermodynamic equilibrium • Gibbs energy
Nomenclature
A dimensionless form for a in Peng-Robinson EoS

a Peng-Robinson attraction parameter (N m4 mol2 )
B dimensionless form for b in Peng-Robinson EoS
b Peng-Robinson repulsion parameter (m3 mol1 )
b* term in Peng-Robinson EoS (m3 mol1 )
L. Fiori () • D. Castello

Department of Civil, Environmental and Mechanical Engineering, University of Trento,
Trento, Italy
100 L. Fiori and D. Castello
cp isobaric heat capacity (kJ mol1 K1 )

E number of elements
ESSC energy self-sufficiency concentration (wt.%)
f fugacity (Pa)
G Gibbs’ energy – extensive property (kJ)
h activity
H molar enthalpy (kJ mol1 )
HHV high heating value (MJ kg1 )
Keq equilibrium constant
kil binary mixture parameter for intermolecular interactions
M number of reactions
N number of components
n number of moles (mol)
P pressure (Pa)
q1 , : : : ,q5 coefficients for cp calculation
R universal gas constant (J mol1 K1 )
S molar entropy (kJ mol1 K1 )
T temperature (K)
VQ molar volume (m3 mol1 )
x molar fraction
Z compressibility factor
Greek Symbols
variation
' fugacity coefficient
chemical potential (kJ mol1 )
˜il binary mixture parameter for packing of unlike components
¨ acentric factor
Subscripts and Superscripts
0 standard
c critical
i i-th component
j j-th reaction
l l-th component
m mixture
r reduced
sat saturation
x stoichiometric index for carbon
y stoichiometric index for hydrogen
z stoichiometric index for oxygen
4 Thermodynamic Analysis of the Supercritical Water Gasification of Biomass 101
4.1 Introduction
Mathematical modeling is a powerful tool for process engineering. Through mod-

eling, it is possible to understand how a process behaves under a great variety
of conditions. This allows reducing the number of expensive and time-consuming
experimental tests, whose execution could be therefore better oriented and planned.
Processes involving biomass may be extremely complex. Biomass, indeed, is a
highly heterogeneous material, without a precise and definite chemical composition.
During its processing, a very high number of chemical species come out, most of
which cannot even be easily identified. Thermodynamic modeling of the supercrit-
ical water gasification (SCWG) of biomass allows obtaining precious information,
even without deep knowledge of the single reactions and compounds (chemical
intermediates) involved in the process. Such information can be of fundamental
importance: first of all, in order to understand how the process works. Then,
thermodynamic modeling can be the starting point for feasibility and optimization
studies, with specific practical applications in the engineering field.
4.1.1 Why Thermodynamics?
When approaching a process for design several questions should be answered:

1. Which is the process yield?
2. Which is the conversion of reactants (if any)?
3. Which are the energy duties (thermal power and mechanical work required/
released)?
4. Which are the best operating conditions (temperature, pressure, etc.)?
5. At which rate the process occurs?
6. Which equipment are necessary to achieve the process target?
7. Which is their size?
The points above represent a preliminary list of the issues the process engineer
is asked to focus on, trying to find a substantial answer when dealing with process
development.
Thermodynamics allows answering some of the questions above (specifically:
1–4). Actually, in the path from idea to actual process implementation, thermo-
dynamics allows providing the first basic answers and stating if the process is
(thermodynamically) feasible. If this is the case, then it is worthy to deepen
the analysis towards process implementation. If this is not, the idea should be
disregarded without further analysis.
It is thus clear how thermodynamics represents the preliminary, fundamental
step for process development. Thermodynamics is useful for thermal cycles –
thermodynamic cycles aimed at developing power from heat – and for multiphasic
and/or reacting systems.
The focus here is on biomass-water reacting system. The reaction process

(gasification) is aimed at producing a syngas rich in combustible species, such as
H2 , CH4 and CO. According to the syngas final use (burner or fuel cells), it would
be preferable to obtain a syngas rich in CH4 or, conversely, H2 . Thermodynamics
allows foreseeing, for instance, which are the process yields in H2 and CH4 at
equilibrium; which are the ranges of temperature, pressure, and water to biomass
ratio useful for maximizing the yield in H2 or CH4 ; which is the gasification thermal
duty, i.e. how much heat is released by the reactions or, conversely, how much heat
should be supplied to sustain the reactions.
The results obtained by thermodynamics refer to systems at thermo-dynamic
equilibrium and, in the case of reacting systems, at chemical-reaction equilibrium.
If there are no kinetics constraints and the thermodynamic model used is reliable,
thermodynamics predicts the actual composition of the reacting system and the
actual energy balance involved in the reactions.
Systems evolve towards equilibrium: thermodynamics allows predicting the
limit, threshold values achievable by the considered system. In practical and
engineering terms, situations will exist where equilibrium is achieved to a good
extent, while in other situations the system will be very far from equilibrium.
Thermodynamics is useful in the former case, while it loses utility in the latter one.
4.1.2 Thermodynamics in Supercritical Conditions
The supercritical state is known for being a highly reactive state, where mass
and heat transfer resistances are limited. For water, the critical point occurs at a
temperature and pressure equal to 374 ı C and 22.1 MPa, respectively. For higher
values of pressure and temperature, the water becomes supercritical. High pressure
favors the contacting among molecules, while the effect of temperature depends
on the actual value of the temperature itself. In the general case, according to
the consolidated Arrhenius approach, high temperatures increase the reaction rates
and favor the achievement of chemical-reaction equilibria. Low temperatures, con-
versely, go along with significant kinetics constraints which limit the establishment
of thermodynamic equilibrium. To overcome reaction kinetics constraints, and to
drive the process towards the desired products, catalysts are usually adopted. In
order to assign an actual meaning to “low” and “high” temperatures in the case
of hydrothermal gasification, low-temperature catalytic gasification occurs in a
temperature range of 350–600 ı C, high-temperature SCWG occurs in a temperature
range of 500–750 ı C [1].
Whatever the temperature, considering that catalysts can be adopted, it is possible
to affirm that thermodynamics forecasts are in good agreement with a significant
portion of SCWG experimental data [2, 3]. This testifies that SCWG process can be
run close to equilibrium.
Therefore, with the limits and specifications above underlined, thermodynamics

represents a fundamental tool for a preliminary evaluation of the SCWG process per-
formances. It gives important information to understand which parameters should
be changed, and in which measure, in order to obtain the desired SCWG outputs
and yields. It allows evaluating the thermal tonality (endothermic or exothermic) of
the SCWG reactions and thus to establish the conditions which make the SCWG
self-sustainable from a thermal point of view or, conversely, the heat which has to
be provided to the reactor for sustaining the SCWG.
4.1.2.1 The Equations of State
When applied to SCWG, thermodynamics has to cope with high pressure fluids.
Specific thermodynamic tools must be used for the purpose.
An Equation of State (EoS) is a mathematic function defining the relationship
between molar volume, pressure and temperature of a fluid. The simplest EoS is
represented by the ideal gas EoS, which is expressed by:
RT
P D (4.1)
VQ
Such EoS is very effective when the ideal gas approximation is satisfied, that is
at low pressure and high temperature. At supercritical conditions, the ideal gas EoS
is not able to give a reliable estimate anymore. For example, for water at 400 ı C and
25 MPa, a molar volume of 108.2 cm3 /mol is experimentally measured [4], while
the ideal gas EoS would foresee a value of 223.9 cm3 /mol: an error higher than
100 %!
The need for more precise estimates of the properties of fluids in supercritical
conditions has led to develop more sophisticated EoS, which are able to account for
the non-idealities of the systems.
Peng-Robinson EoS is one of the most adopted EoS in the SCWG
thermodynamics literature. It can be expressed by the following formula:
RT a
P D (4.2)
VQ b VQ C 2b VQ b 2
2
Where a and b are two parameters that depend on the specific chemical
compound analyzed. Indeed, while the ideal gas EoS is valid for any chemical
compound, non-ideal EoS, like Peng-Robinson EoS, take into account the specific
nature of the species involved. The actual application of Peng-Robinson EoS will
be specifically addressed in Sect. 4.3.
Nevertheless, other EoS can be also chosen, with very good results. In the
literature there are also works dealing, for example, with Duan, Soave-Redlich-
Kwong, virial EoS. These works will be reviewed in Sect. 4.2.
4.2 The Approaches to Thermodynamic Equilibrium

Modeling
A thermodynamic equilibrium mathematical model is a calculation tool able to

foresee the chemical composition of a reacting system at equilibrium, given its
initial composition and the operating conditions (temperature and pressure).
Practically, every model includes:
1. A set of chemical species (components) involved.
2. A set of operating conditions (temperature/pressure).
3. The initial composition of the system.
Given these three elements, there are two ways to develop a thermodynamic
equilibrium mathematical model: the stoichiometric approach and the non-
stoichiometric approach. In the stoichiometric approach, a system of chemical
reactions leading from reactants to products is explicitly defined. Vice versa,
in the non-stoichiometric approach only a list of components is defined and the
minimization of Gibbs energy is used to predict equilibrium.
In the present paragraph, these two families of methods are presented and the
current state of the art is reviewed.
4.2.1 The Stoichiometric Approach
The stoichiometric approach is based on reaction equilibria. It involves a set of

chemical reactions through which the initial compounds are converted into products.
Let us consider a thermodynamic model involving M chemical reactions and N
components. Each chemical reaction can be endowed with an equilibrium constant
Keq . For example, for the generic j-th reaction:
˛A C ˇB ! C C ıD (4.3)
it can be defined an equilibrium constant Keq,j as the product of the activities of the
products, each elevated to its stoichiometric coefficient, divided by the product of
the activities of the reactants, each elevated to its stoichiometric coefficient:

hC hıD
Keq;j D ˇ
(4.4)
h˛A hB
The activity of the species i is defined as:
P
hi D 'i xi (4.5)
P0
Where ' i and xi are, respectively, the fugacity coefficient and the molar fraction
of component i, and P0 is the reference pressure of 0.1 MPa. ' i represents the ratio
between fugacity fi and the partial pressure of component i, Pxi :
fi
'i D (4.6)
P xi
Fugacity has the units of a pressure and can be calculated through the EoS by
means of the following equation:

fi VQi dP
d ln D (4.7)
P0 RT
It can be easily proved that, when the ideal gas EoS is considered, fugacity is
equivalent to the actual partial pressure: therefore, ' i is equal to 1. When an EoS for
a non-ideal gas is adopted, this does not happen. Therefore, the fugacity coefficient
' i can be seen as a measure of the non-ideality of the system.
The equilibrium constant Keq,j can be related to the other thermodynamic
quantities by means of the following expression:

Gj0 D RT ln Keq;j (4.8)
Where G0 j is the standard Gibbs energy for the j-th reaction, calculated at
atmospheric pressure but at the reaction temperature. G0 j can be also defined
as the difference between the standard chemical potential 0 of the products and
the reactants, each multiplied by its stoichiometric coefficient (number of moles of
each species involved in the j-th reaction). Referring to the reaction represented by
Eq. 4.3:

Gj0 D 0C C ı 0D ˛ 0A C ˇ 0B (4.9)
The standard chemical potential 0 i , which does not depend on the actual
reaction being considered, can be calculated by means of standard enthalpy and
entropy:
0i .T / D Hi0 .T / T Si0 .T / (4.10)
For each component, standard enthalpy and entropy are functions of the temper-
ature through the isobaric heat capacity cp . For each species i, it can be written:
Z T
Hi0 .T / D Hi0 .T0 / C cp;i d T (4.11)
T0
Z T
cp;i
Si0 .T / D Si0 .T0 / C dT (4.12)
T0 T
where T0 is the reference temperature (usually 25 ı C). Therefore, combining

Eqs. 4.8 and 4.9, it is easy to calculate the equilibrium constant Keq,j for each
reaction composing the model.
Here, some clarification about the number of chemical reactions must be given.
In order to ensure an univocal solution, the model must consider all the independent
reactions that can be established among the considered chemical species. A set of
reactions are independent if none of them can be written by the combination of other
reactions in the same set.
There is a general rule to determine the maximum number of independent
chemical equilibria that can be written for a given set of components. If a model
involves N chemical species and the total number of elements forming these species
is E, then the maximum number of independent reactions M is given by1 :
M DN E (4.13)
For example, if a model involving N D 12 species is considered and if such

chemical species are all compounds of carbon, oxygen and hydrogen (E D 3), the
maximum number of independent reactions is 9. Therefore, it will be possible to
write nine equations in the form of Eq. 4.8. Further three equations can be obtained
by imposing the conservation of the total moles of each element (C, O, H).
The model will thus result in an algebraic non-linear system of N equations with
N unknowns, which can be solved using analytical or (more commonly) numerical
techniques.
4.2.1.1 State of the Art
In the literature, the stoichiometric approach for SCWG modeling has been followed
by a limited number of works.
In this field, a relevant work is represented by the study of Letellier et al. [5].
Here, the authors performed the modeling of a system composed by a SCWG reactor
and a flash separator, where permanent gases are separated from solid and liquid
products. The authors considered seven independent chemical reactions, taking
place during the gasification operations, and ten chemical species. They performed
their simulations and compared the obtained data with some literature studies, with
good accordance to them.
The same model was utilized in a subsequent study by the same group [6], with
the aim of performing an energy analysis of the process. In such study, the authors
applied the model to the gasification of vinasse, a by-product of the alcohol industry.
1
Rigorously, the maximum number of independent reactions within a given set of chemical
compounds is given by M D N-rank(W), where W is the composition matrix. Its generic element
wij represents the number of moles of the element j (stoichiometric subscript) in the component i.
In practical applications, the number of components is much higher than the number of elements.
Therefore, rank(W) is actually most of the times coincident with the number of elements E.
In their results, they concluded that SCWG can be energetically self-sufficient,

especially if a small amount of oxidizing agent is added. They also pointed out
that a temperature of 600 ı C can ensure the best gasification efficiency in their
experimental conditions.
4.2.2 The Non-stoichiometric Approach
The other family of thermodynamic models is defined non-stoichiometric. Unlike

the previous class of models, in this case no reactions must be defined. This can be
a considerable advantage: indeed, when dealing with biomass, the reactions are not
known in advance. Thousands of reactions and chemical species could be involved
in the process, which makes their analytical description quite puzzling.
The non-stoichiometric approach has known a significant development during
the last decades; such development was made possible thanks to the progress of
computers. The non-stoichiometric approach has been widely used in those fields
of chemical engineering where high temperatures are involved, such as metallurgy
and high-temperature material science. There, such modeling technique allowed for
the calculation of solid phase equilibria, enabling to obtain reliable phase diagrams
[7]. Another application is related to hydrocarbons engineering, where refining and
upgrading processes can be modeled [8]. Non-stoichiometric models have been
also utilized to analyze the feasibility of combustion-based energy processes [9],
as well as for the calculation of the amounts of pollutants produced by them [10].
Another interesting field of application concerns concentrated aqueous solutions
[11]. Nowadays, non-stoichiometric modeling is widely used for biomass energy
processes (mainly combustion and gasification).
The non-stoichiometric approach does not need the knowledge of the involved
reactions. It is based on a thermodynamic property of the chemical systems:
when chemical equilibrium is achieved, the functional of Gibbs energy reaches a
minimum. It is thus necessary to write down the functional of Gibbs energy for the
whole system:
X
N
GD ni i (4.14)
i D1
It must be pointed out that Eq. 4.14 is obtained for an isothermal and isobaric
system, where no work other than that related to volume change (pressure-volume
work) is involved. The chemical potential of each component is a function of
temperature and pressure, according to:
fi
i .T; P / D 0i .T / C RT ln (4.15)
P0
The chemical potential is composed of two contributions. The first term after the
equal sign is the standard chemical potential: it is calculated at atmospheric pressure
and thus it only depends on the temperature of the system. Its calculation can be
easily performed by means of Eq. 4.10.
The second term after the equal sign in Eq. 4.15 represents the dependence of
the chemical potential on the pressure and accounts also for the non-ideality of the
system through fugacity (see Eq. 4.6).
The solution of the problem consists in finding the values of ni that allow
reaching the minimum value of G. This class of problems is commonly known
as optimization problems. Indeed, the function G can be regarded as a cost
function which must be minimized to obtain the “optimal” composition values: the
equilibrium ones.
The composition calculated in this way must be consistent with some constraints.
First of all, the composition must be obviously consistent with the mass balance
for each element. In other words, the total amount of each element (carbon,
hydrogen, etc.) initially present in the system must be conserved also in the products.
Moreover, it must be imposed that the model never calculates negative molar
fractions, which would obviously be physically meaningless. This is not a trivial
issue, since actually the model could still come to a minimized value of G, satisfying
the mass balances but with meaningless negative molar fractions. To solve such
problem, it is possible to implement the method of Lagrange multipliers, or using a
commercial software.
4.2.2.1 State of the Art
Unlike stoichiometric models, non-stoichiometric ones are largely present in the

literature. Tang and Kitagawa [3] followed such approach to predict the behavior of
a supercritical mixture. They adopted Peng-Robinson EoS and utilized LINGO®
to perform minimization calculations. Glucose, cellulose and real biomass were
analyzed, obtaining good agreement with experimental results. They also managed
to improve the fitting with experimental results by taking into account the rate
controlling steps of the process.
A similar study was conducted by Yan et al. [12], who treated the SCWG of
glucose. In their work, the authors adopted the Duan EoS, which was originally
developed for the geochemical field and was said to be very effective in modeling
water at supercritical state. Moreover, they introduced the “carbon conversion
efficiency” in order to account for the actual production of non-equilibrium char.
As it has been stated before, a non-stoichiometric model implies a quite high
computational effort, due to the solution of a highly non-linear optimization
problem. The possibility to reduce the computational effort is the goal of the work
by Voll et al. [13]. There, the authors utilized an approximated model, where the
fugacity coefficients ' i were all let equal to 1. In other words, their approach was
equivalent to the adoption of the ideal gas EoS. For calculations, they utilized the
software GAMS® 21.6 (General Algebraic Modeling System), with the CONOPT2
solver. They found out that the results (equilibrium compositions) were comparable
with the ones obtained utilizing Peng-Robinson EoS and with experimental data, but
the mathematical solution was easier and more reliable, without undesirable local
minimum points.
In all these works, only one phase is considered: the selected reaction products
are always in the homogeneous (supercritical) phase. On the other hand, SCWG is
able to produce, under certain equilibrium conditions, solid products, for example
char. Switching to a two-phase model was the aim of the work by Castello and Fiori,
who implemented a non-stoichiometric model by means of the software MatLab [2].
Such model will be presented in detail in Sect. 4.3. The possibility of solid carbon
formation at equilibrium was also included in the model by Gutiérrez-Ortiz et al.
[14], who adopted the predictive Soave-Redlich-Kwong (PSRK) EoS and studied
the SCWG of pure and pretreated crude glycerol from bio-diesel production.
Another example of a thermodynamic model for SCWG is represented by the
work of Freitas and Guirardello, who analyzed the gasification of two types of
microalgae (Clorella vulgaris and Spirulina) in the range 500–900 ı C [15]. They
adopted the virial EoS and implemented the model with the software GAMS® .
Besides the effect of temperature, pressure and biomass concentration, the authors
analyzed the addition of a co-reactant (CO2 and CH4 ), finding out that this improved
the yields and the quality of the produced syngas, especially for Fisher-Tropsch
applications. In their work, the authors also considered the possibility of solid char
formation at equilibrium, but they actually did not obtain any char at the chosen
experimental conditions (maximum biomass concentration of 20 wt.%).
A multi-phase model was also implemented by Yakaboylu et al., who performed
an extensive analysis of manure SCWG [16]. There, the authors investigated over
the partitioning of elements typically present in real biomass, like P, S, Si, Ca, Cl,
Mg and K. In this way, they were able to predict the composition of solid, liquid and
gaseous phases arising from gasification operations.
4.3 A Solid-Gas Thermodynamic Equilibrium Model
In the present section, the practical implementation of a non-stoichiometric thermo-

dynamic equilibrium model is discussed. Such model is able to predict the system
composition, dealing with the formation of two different phases: a supercritical
phase in equilibrium with a solid phase, represented by graphitic carbon. It is worthy
to state that, in reality, other solid products besides graphitic carbon can be found
at equilibrium. This is due to the mineral content of biomass. Different compounds
of Ca, Na, K, Cl, etc. can thus arise: their actual form (carbonates, nitrates, sulfates,
etc.) depends on the reaction conditions. To this purpose, an interesting work was
carried out by Yakaboylu et al. [16]. In the present work, however, solids are
schematized as mere graphitic carbon. Indeed, since the mineral content of biomass
was neglected, it seemed reasonable to foresee only such compound as the solid
product by SCWG. The reader can find more information about the thermodynamics
of solids in supercritical fluids referring to [17].
The two-phase model here presented can thus effectively foresee the possible
occurrence of char at equilibrium. Moreover, the composition of the produced
syngas can be predicted for whatever biomass composition at different pressures,
temperatures and biomass concentrations in the feed to the reactor.
4.3.1 Description of the Model
The model here presented is based on the minimization of Gibbs energy, thus
belonging to the family of non-stoichiometric models. In Sect. 4.2.2, the general
elements of such typology of models were already presented. The mathematical
formulation here presented is a simplified version of the work by Castello and
Fiori: here nitrogen compounds were omitted for the sake of simplicity. For a more
comprehensive treatise, the reader can refer to the original paper [2].
4.3.1.1 Thermodynamic Quantities
In order to calculate the thermodynamic quantities, it is necessary to calculate the

enthalpy H and entropy S, by means of Eqs. 4.11 and 4.12. Since cp,i depends on
the temperature T, it is necessary to know how this dependency is expressed. In the
present model, the so-called NASA polynomial formula was adopted. According to
this formulation, cp is expressed as a polynomial in the form:
cp;i
D q1 C q2 T C q3 T 2 C q4 T 3 C q5 T 4 (4.16)
R
The coefficients q1 , : : : ,q5 can be found in a report [18], which includes a large
number of compounds. Thanks to this formulation, the solution of Eqs. 4.11 and
4.12 is immediate. For the sake of completeness, it must be mentioned that, besides
NASA polynomials, other thermodynamic formulations are also available. Among
these, we can cite NIST-JANAF tables [19].
4.3.1.2 Equation of State
Non-idealities are taken into account by using Peng-Robinson EoS (Eq. 4.2) [20].
In such equation, the parameters a and b for each species involved in the model are
calculated as follows:
" s !#2
0:45724 R2 TC2 T
aD 1 C 0:37464 C 1:54226! 0:26992! 2 1
PC Tc
(4.17)
0:07780 RTC
bD (4.18)
PC
In Eq. 4.17, Pitzer acentric factor ! was utilized. For each component, such
parameter is defined as:
ˇ
Psat ˇˇ
! D 1 log (4.19)
PC ˇT r D 0:7
However, the EoS must predict the behavior of a mixture rather than of a
single pure compound. To this purpose, the parameters a and b, relative to a pure
compound, must be substituted by am and bm , which refer to the whole mixture.
Such new parameters may be calculated by means of van der Waal’s mixing rules,
resulting in:
X
N X
N
am D xi xl ai l (4.20)
i D1 lD1
X
N X
N
bm D xi xl bi l (4.21)
i D1 lD1
Where:
p
ai l D .1 ki l / ai al (4.22)
bi C bl
bi l D .1 i l / (4.23)
2
Parameters kil and il are called binary interaction parameters and express the
interaction between species i and l. Such parameters are experimentally determined
for each couple of compounds. As an alternative, they can be estimated using
appropriate correlations. More information can be found in [2].
The EoS so obtained can be solved for the molar volume, fixed the values of
temperature and pressure. Since Peng-Robinson is a cubic EoS, three solutions can
be found. Nevertheless, in supercritical conditions, only one of them is physically
meaningful, the remaining two being negative or complex.
4.3.1.3 Fugacity Coefficients
Once the EoS has been fully defined, the fugacity coefficient ' i of the i-th
component of the mixture can be obtained by the formula:
bi
ln .'i / D .Z 1/ ln .Z B/
bm
0 N 1
X
B2 xl ai;l C
A B B lD1 bi C
C Z C 2:414B
B C ln (4.24)
2:828B B am bm C Z 0:414B
@ A
Where:
am
AD P (4.25)
R2 T 2
bm
BD P (4.26)
RT
X
N
bi D2 xl bi l bm (4.27)
lD1
P VQ
ZD (4.28)
RT
4.3.1.4 Solid Phase
Since the model includes a solid phase (graphitic carbon), it is necessary to adopt
some slight modifications in order to correctly calculate its chemical potential.
Indeed, Eq. 4.15 is not applicable anymore, since it has been derived for a gas.
From the definition of Gibbs energy:
ZP
.T; P / D 0 .T / C VQ dP (4.29)
P0
For a solid, the molar volume VQ can be reasonably considered to be constant, and
thus taken outside the sign of integration. Furthermore, since the molar volume is
constant, the critical volume VQc can be adopted. The integration is, then, immediate,
resulting in:
.T; P / D 0 .T / C VQc .P P0 / (4.30)
4.3.2 Model Implementation
The implementation of the selected model was performed by means of the software
MatLab® (The Mathworks, Inc.). Here some technical details about the strategy
adopted are presented.
4.3.2.1 Initial Composition of the System
To run the model, it is necessary to declare the initial composition of the system
whose equilibrium composition has to be calculated. Considering SCWG, the
compounds fed to the system are water and biomass, in different relative amounts.
In the present model, biomass is modeled as a pseudo-compound with the generic
formula: Cx Hy Oz . The subscripts x, y and z can be calculated from the elemental
composition of each biomass.
It is important to state that the physical properties of the specific biomass are
useless for this model. Indeed, the definition of the initial composition of the system
is only functional to the mass balance. In other words, it merely represents a
constraint that must be respected by the minimization algorithm adopted to solve the
problem. As it can be observed in Sect. 4.3.1, the only thermo-physical parameters
required are those of the expected equilibrium products. Since it is assumed that no
unreacted biomass is found at equilibrium, the physical properties of the feedstock
are not required to run the model.
4.3.2.2 List of Compounds
Another basic aspect to be defined in a non-stoichiometric model is the list of

chemical compounds which can be reasonably found in the output stream. The
model here described foresees the utilization of 17 species. Besides the standard
products like H2 O, H2 , CO, CO2 and CH4 , C2-3 hydrocarbons and PAH were
considered. The complete list of output compounds, along with the physical
parameters required by the EoS, is reported in Table 4.1.
4.3.2.3 Calculation Procedure
It has been already stated (see Sect. 4.2.2) that the resulting problem is a constrained
optimization problem. In order to solve it, the software MatLab® (The Mathworks,
Inc.) was utilized. The function FMINCON was adopted to the purpose. This routine
implements the methods of Lagrange multipliers and is able to effectively solve this
class of problems.
Besides the function to be minimized and the constraints to impose (elemental
balance and non-negativity of the number of moles), it is necessary to provide an
Table 4.1 List of the compounds involved in the model, along with their critical parameters
(NIST-TRC databank)
Tc [K] Pc [MPa] ¨ []
Oxygen O2 154.6 5.042 0.0213
Hydrogen H2 33.2 1.240 0.2320
Water H2 O 647.3 22.140 0.3439
Solid carbon C 6,810 223.000 0.3268
Methane CH4 190.6 4.608 0.0106
Carbon monoxide CO 134.5 3.774 0.0371
Carbon dioxide CO2 304.2 7.378 0.2249
Propane C3 H8 369.9 4.256 0.1529
Ethane C2 H6 305.4 4.886 0.1002
Ethylene C2 H4 282.3 5.042 0.0864
Methanol CH3 OH 512.7 8.013 0.5597
Acethylene C2 H2 308.3 6.240 0.1857
Methyl-acethylene C3 H4 401.6 5.626 0.2091
Propylene C3 H6 364.9 4.594 0.1422
Benzene C6 H6 562.0 4.897 0.2105
Naphthalene C10 H8 748.2 4.082 0.3079
Formaldehyde CH2 O 418.0 6.854 0.2018
initial estimate for the equilibrium composition. This initial estimate could be any:
however, the more it is near to the actual equilibrium composition, the less the solver
will require in terms of computational time. Providing a good initial estimate is also
crucial to avoid false solutions, such as, for example, local minimum points. When
performing a series of simulations where a single parameter is varied, a tip could
be that of using the results of the previous simulation as first guess for the next
one. Indeed, the new solution would be reasonably close to the previous one, thus
considerably speeding up the calculations.
4.3.3 Model Validation
Before performing any further analysis, it is necessary to check if the model results
are in agreement with experimental data. This step is called “model validation”.
Validating a model with experimental data is a fundamental step, because it helps
understanding if the model results are reliable and can be utilized to describe
reality.
To this purpose, the model results were compared with the experimental data by
Byrd et al. [21], who carried out a study on the SCWG of glycerol (C3 H8 O3 ) with
a Ru/Al2 O3 catalyst. The results of model testing are shown in Fig. 4.1, where the
syngas composition is reported as a function of the feed concentration expressed
on weight basis. For instance, a feed concentration of 20 wt.% corresponds to a
feed consisting of water (80 %) and dry glycerol (20 %). The experimental data
Fig. 4.1 Model validation with the data of Byrd et al. [21] (Reprinted and adapted with permission
from Castello and Fiori [2]. Copyright © 2011, Elsevier)
were obtained operating at a temperature of 800 ı C, a pressure of 24.1 MPa and a

residence time of 1 s. The adopted tubular reactor had a volume of approximately
3.65 mL and was packed with 2.0 g Ru/Al2 O3 catalyst.
It is possible to appreciate how the predicted data (lines) are in very good
agreement with the experimental points. The fitting is extremely good for H2 and
CH4 , less good but still satisfying for CO2 and CO.
This allows concluding that the model forecasts are physically meaningful and
thus they can be used to predict the behavior of the SCWG process.
4.3.4 Results
The model was first run to state the dependence of the composition of the
reaction products (syngas and possibly solid carbon) as a function of the biomass
concentration in the feed.
4.3.4.1 Syngas Composition
Figure 4.2a shows the composition of the syngas produced through the SCWG of
glycerol at 800 ı C and 25 MPa. Molar fractions are given on a water-free basis.
Fig. 4.2 Equilibrium composition of the products from glycerol SCWG as calculated by the
model, expressed as molar fraction on a water-free basis: (a) Effect of the feed concentration at
T D 800 ı C, P D 25 MPa; (b) Same at T D 500 ı C, P D 25 MPa; (c) Effect of the temperature
at P D 25 MPa, feed concentration D 80 wt.%; (d) Effect of the pressure at T D 800 ı C, feed
concentration D 80 wt.%
The syngas composition significantly varies with the feed concentration (expressed
in weight basis as for Fig. 4.1). At low feed concentrations, H2 is the preferred
product, together with CO2 and a very small amount of CO. At these conditions, H2
can represent up to 70 mol.% of the permanent gases produced. However, when the
concentration of organics in the feed increases, H2 fraction tends to decrease, while
CH4 and CO are more and more present. CO2 exhibits a slight increase; however,
its trend can be said to be substantially constant.
It can be observed that the plots have a regular trend up to a biomass concentra-
tion of 72 wt.%. Starting from that point, a sudden change in all the trends is shown,
which is associated with the occurrence of solid carbon. CH4 , CO and CO2 trends
start decreasing, since carbon preferably forms the solid phase.
A change in the temperature determines completely different results. In Fig. 4.2b
it is shown a similar analysis conducted at 500 ı C. Now, H2 is the preferred product
only at very low biomass concentrations. Already at a concentration of 10 wt.%,
CH4 becomes the preferred product, together with CO2 . Unlike the previous case,
CO is not found among the products. The occurrence of solid carbon, which was
observed at a concentration of 72 wt.% in the previous case, now is shown at a lower
biomass concentration: around 60 wt.%. This allows concluding that the formation
of solid carbon is favored at lower operating temperatures. It must be pointed out
that the calculated concentrations for solid carbon formation at equilibrium are, in
both cases, well above the biomass concentrations usually adopted for SCWG (not
higher than 20–30 wt.%). As a consequence, it can be affirmed that, in real SCWG
applications, carbon formation at equilibrium is hardly possible. Nevertheless, even
though solid carbon is not found at equilibrium, the fact that it is more favored at
lower temperatures is documented by a number of experimental studies [22, 23].
In Fig. 4.2c the independent variable is changed. Temperature is now varied
between 400 and 1,200 ı C, while biomass concentration in the feed is held constant
at 80 wt.%. This graph demonstrates what has been just stated: solid carbon
formation is promoted at lower temperatures. Operating SCWG at high temperatures
would thus avoid the formation of solids. It can be also observed that temperature
favors the formation of H2 and CO, while an opposite behavior is shown for CH4
and CO2 .
In Fig. 4.2d the influence of pressure is shown. Operating at high pressure tends
to favor CH4 and CO2 formation, as well as solid carbon. On the other hand, H2 and
CO are produced in lower amounts at high pressures. However, it must be observed
that in the pressure range usually adopted for SCWG, which lays between 25 and
30 MPa, pressure plays only a very limited role and its influence is much lower than
that of the other operating variables.
In Fig. 4.2c, d it was decided to run the model at a high feedstock concentration
(unrealistic if dealing with solid biomass, still possible in practice in the case
of liquid substrates like glycerol) to show, together with the trend of the syngas
composition, the trend of char occurrence. When simulating the SCWG equilibrium
at more realistic biomass concentrations (10–25 wt.%), the trend of the syngas
constituents, H2 , CH4 , CO and CO2 , remains the one shown in Fig. 4.2c, d (not
reported data).
4.3.4.2 Solid Char Formation
The elemental composition of biomass is an important parameter affecting the

occurrence of char at equilibrium. Through thermodynamic modeling, a very high
number of different biomasses can be tested and such influence can be better
clarified.
In order to account for the different biomasses, a representation based on a
ternary diagram was adopted. Indeed, in this model biomass was schematized as
a pseudo-compound consisting of C, H and O (i.e. neglecting the presence of
ash and minor constituents). Therefore, each SCWG reactor feed composition (i.e.
water C biomass) could be represented by a point in a ternary diagram, with the
elemental fraction of C, H and O on each axis. Such representation can be seen in
Fig. 4.3.
For the simulations, each axis was divided into 50 intervals. Therefore, around
2,500 simulations were performed, corresponding to all the possible formulas
“Cx Hy Oz C H2 O” indicative of the various biomass types and feed compositions
(relative elemental amount of C, H and O in the feed; 0).
118
Fig. 4.3 Ternary diagrams for a generic biomass schematized by the pseudo-molecule Cx Hy Oz . Values on the axes refer to the molar ratios of C, H and O in the
biomass/water system. The red line is representative of the glycerol/water mixture at different glycerol concentration. (a) Char formation at varying temperature
(P D 25 MPa); (b) Char formation at varying pressure (T D 800 ı C) (Reprinted and adapted with permission from Castello and Fiori [2]. Copyright © 2011,
Elsevier)
L. Fiori and D. Castello
In Fig. 4.3a, three analyses were performed at increasing temperature values:

400, 800 and 1,200 ı C. All the points laying down the curves represent the
feed compositions that can lead to the formation of solid carbon at equilibrium.
As it could be expected, when the temperatures is raised the “char formation”
region becomes smaller. Indeed, high temperature reduces the possibility of carbon
formation.
A similar analysis was carried out with the pressure, and it can be observed in
Fig. 4.3b. As it was underlined during the discussion of Fig. 4.2d, a higher pressure
actually favors the production of carbon. Figure 4.3b confirms such conclusion.
Nevertheless, it can be also noticed that the influence of pressure is quite little. The
curves corresponding to 10 and 50 MPa are very close to each other, meaning that
changing the operating pressure only plays a very limited effect on the final result.
In both Fig. 4.3a, b, the example of glycerol SCWG was reported. All the possible
glycerol/water mixtures are represented by a line in the triangular plot; the line
goes from the point representative of pure water (lying on the right side of the
triangle) to the point representative of pure glycerol. As a consequence, each point
within the line represents a different glycerol concentration. It can be observed
that, for glycerol, carbon formation actually occurs only for quite high feedstock
concentrations. The portion of the line falling within the “char formation area” is
relevant to quite high glycerol concentrations. It can be therefore concluded that
glycerol does not yield solid carbon at equilibrium, when typical SCWG conditions
are adopted (feed concentrations not higher than 30 wt.%).
Through the proposed triangular diagram, the concentrations at which a specific
organic feedstock would produce char at equilibrium can be determined, which is a
precious piece of information for design and process engineering.
4.4 The Thermodynamic Approach for Process Design
In Sect. 4.3 it was shown how thermodynamics is able to give valuable and inter-
esting pieces of information to foresee the influence of process conditions on the
process outputs. The results obtained allowed understanding how the combination
of temperature, pressure and water-to-biomass ratio determines the composition of
the syngas produced.
Nevertheless, this approach is only focused on the SCWG reaction itself. In other
words, the analysis that was carried out considers a system which is only composed
by a mixture of water and biomass, that is reacted at high temperature and pressure
to form the equilibrium products. This is a too reductive model to describe what
takes place in real-life industrial applications. Indeed, if we want to turn SCWG into
an actual industrial process, we need to account for a pump to achieve the desired
pressure, a heat exchanger to heat the reactants by recovering part of the thermal
energy of the products, a cooler in order to separate water from the other permanent
gases, and so on.
This is crucial especially to calculate the energy requirements of the process. If

we only consider the mere SCWG reaction, we could be able to determine if it is
exothermic or endothermic and to calculate the energy produced or required by it.
However, this is not enough to understand which are the actual energy needs of
the complete process. It must be determined if part of the heat of the hot reaction
products can be recovered and, if yes, to what extent. Furthermore, the energy
requirements for pumping or cooling the reaction products must be taken into
account, as well as the energy that could be effectively produced by the process
itself.
In this framework, thermodynamics still plays a fundamental role. The thermo-
dynamic formulation we just considered is the core of the process model, since it
calculates the composition of the reaction products. Moreover, thermodynamics is
involved in the calculation of the energy needs of all the other unit operations of
the process. For example, it is useful to calculate the temperatures of the hot and
cold streams exiting a heat exchanger, or the temperature of a gas after its isentropic
expansion in a lamination valve. It is also necessary in order to calculate the phase
distribution achieved in the liquid-gas separators downstream the reactor [24].
All these reasons lead to the adoption of a process model. In this section, a simple
process model for biomass SCWG will be implemented and run in order to carry out
an energy feasibility analysis of the process. The software Aspen Plus® , a leading
product in this field, will be used for the purpose.
4.4.1 Using Aspen Plus® for Process Simulations
Aspen Plus® (AspenTech, Inc.) is one of the most worldwide utilized software for
chemical process simulation. Its utilization is common in many fields of science
and technology: petro-chemical industry, food processing industry, environmental
treatment, etc. Aspen Plus® can model even very complex systems, composed
of several unit operations, and simulate them under very different operating and
reaction conditions.
Although it can be also utilized for dynamic (transient-state) simulations, Aspen
Plus® is commonly used for steady state analyses.
In order to deal with a process simulation with Aspen Plus® , it is first necessary
to define a process layout, arranging the different unit operations that constitute the
process to simulate. Choosing among the different unit operations available in the
built-in library, it is possible to draw a block diagram of the process.
Aspen Plus® includes a library with a huge number of chemical species, along
with their chemic-physical and thermodynamic properties. It also allows defining
the EoS to be used, choosing among a very large number of available formulations.
Aspen Plus® has been utilized to simulate SCWG processes in a few works in
the literature. Gutiérrez Ortiz et al. performed a thermodynamic analysis [25] and
a subsequent process concept development [26] for the SCWG of glycerol. Their
comprehensive works include the implementation of a heat exchangers network and
the adoption of ancillary units, such as a reformer and a separator based on pressure-
swing adsorption (PSA). Withag et al. [27] used Aspen Plus® to simulate a simple
SCWG system, testing different biomass model compounds.
Another work in this field is the one by Fiori et al. [28], dealing with the SCWG
of several types of biomass. In their study, the authors proposed a conceptual
design for the SCWG process and performed a careful energy analysis, adopting
different process solutions. Their aim was to develop and simulate a quite simple
scheme, which could be implemented for small-medium scale applications. In the
next paragraphs, such model will be described in detail.
4.4.2 A Simple Process Layout
The first move with Aspen Plus® is defining a process layout for the SCWG process.
A process layout consists of a number of blocks, each representing a unit operation,
which are connected by arrows, representing the material and energy streams among
the units.
The process scheme was conceived in order to be as simple as possible, trying
to develop a concept that could be likely implemented in a real-life application. For
all the simulations, Peng-Robinson EoS was adopted. In Fig. 4.4, the implemented
process scheme is shown.
The selected process was designed for the production of pure H2 in order to
feed a proton exchange membrane (PEM) fuel cell for power generation. First,
fresh reactants, that is biomass and water, are mixed and pumped up to the process
pressure (for instance, 30 MPa). After that, the stream passes through a pre-heater
(HEATX-1), where it is heated up thanks to the hot reaction products from the
gasification reactor, which are consequently cooled down. The stream SYNGAS2 is
imposed to be completely in vapor phase (vapor fraction equal to 1).
4.4.2.1 The Reactor
The resulting stream PREHEAT1 enters the reactor, which is schematized with
two blocks: a heat exchanger (HEATX-2) and the outright reactor (REACTOR1).
Indeed, in reality the reactor would be represented by a vessel heated by an external
furnace. Therefore, it will act as a reactor and as a heat exchanger, at the same time.
The implemented reactor belongs to the typology “RGibbs”. It operates in the
very same way as the thermodynamic model presented in Sect. 4.3, that is by
calculating the equilibrium composition through performing the minimization of
Gibbs energy. The reactor operates at isothermal conditions; the temperature is set
for each simulation. The thermal energy required by the reactor is provided by
an external heater (BURNER), powered with the CH4 and CO produced by the
gasification reactions (COMB-2) and, in case of necessity, with an external supply
of methane (CH4).
122
Fig. 4.4 Schematic flow-sheet for the considered SCWG process (Reprinted and adapted with permission from Fiori et al. [28]. Copyright © 2012, Elsevier)
L. Fiori and D. Castello
In Fig. 4.4 it is possible to observe a stream (Q-REACT) exiting the reactor.

This is a heat stream, indicating the amount of thermal energy that is produced or
required by the reactor. In the former case, the value of Q-REACT will be positive;
in the latter case, it will be negative, meaning that an external supply of energy is
needed to ensure the desired reaction temperature. Q-REACT is calculated by Aspen
Plus® as the enthalpy change between the products (SYNGAS) and the reactants
(PREHEAT2).
4.4.2.2 The Water Separator
Downstream the reactor, the hot products are first cooled down by the fresh reactants
(HEATX-1) and then further quenched to 60 ı C in a flash separator (SEPAR). Here,
water reverts to the liquid state and is separated from the permanent gases. It is worth
noticing that such operation, which theoretically would imply that some energy is
produced by the process, actually results in an amount of energy required by the
process. It is indeed necessary to provide some electrical energy to get rid of the
produced heat, since its low temperature (60 ı C) does not allow any effective heat
recovery inside the process. The heat stream Q-SEPAR quantifies the amount of
thermal energy generated by such unit operation.
There are many possible ways to achieve an effective cooling in the considered
unit. One way could be using a cooling cycle based on a refrigerating fluid. In
this case, a coefficient of performance (COP) must be considered. Another way
is represented by the utilization of a cooling tower. In this case, the operating costs
are limited to the power required by a pump needed to circulate water and a fan to
create an air flux inside the tower itself. Further details on this topic can be found
in [28].
4.4.2.3 Gas Treatment
The gaseous stream then passes through a heat exchanger where its temperature
is raised in order to prevent its freezing when expanded in the lamination valve
(LAMINA1). At this point, the stream is fed to a special separator (SEPAR2), where
H2 is separated from the other gases produced by gasification (mainly CH4 , CO and
CO2 ). The palladium filter Hysep® [29] was considered for this purpose. According
to the manufacturer’s specifications, such device must be operated at 300 ı C and
6 MPa. Therefore, it was necessary to include the heater HEAT-AIR, using the hot
flue gases from the burner (BURNER) to increase the stream temperature to the
desired value. It must be noticed that the inclusion of such heater is also necessary
to counterbalance the effect of the lamination valve LAMINA1. Indeed, the well-
known Joule-Thompson effect will cause the gas to cool down when expanded.
Therefore, the heat exchanger HEAT-AIR must ensure a temperature higher than
the desired one of 300 ı C.
4.4.2.4 Gas Utilization
The separated H2 (H2STREAM) is thus fed to a PEM fuel cell in order to produce
electrical power with high efficiency: an electrical yield of 45 % was conserva-
tively assumed, according to the existing technical literature [30]. The remaining
gases are first expanded to atmospheric pressure through another lamination valve
(LAMINA2) and then fed to a burner (BURNER), which supplies thermal energy
to the SCWG reactor. In case the energy produced by the combustion of such gases
would not be sufficient to obtain the desired temperature in the reactor, an auxiliary
methane source (CH4) is included. The burner uses air as comburent, which is
preheated by recovering the sensible heat of combustion flue gases in the pre-heater
HEATAIR2.
4.4.3 Results
The simulations were executed utilizing different biomasses and model compounds:
glycerol, phenol, microalga Spirulina, sewage sludge and grape marc. Their elemen-
tal composition was retrieved from the database Phyllis [31].
The software Aspen Plus® allowed simulating the process under very different
operating conditions. Here, some of the most important results are presented.
In Fig. 4.5a, the results of the process simulations conducted by varying the
biomass concentration in the feed are shown. Such data were obtained when
operating the SCWG reactor at 700 ı C and 30 MPa and when no additional methane
was supplied through the stream CH4 (see Fig. 4.4). Indeed, the aim was that
of quantifying the actual energy uptake of the reactor, by means of the stream
Q-REACT.
It can be observed that the value of Q-REACT increases when the biomass
concentration in the feed increases. When biomass is very diluted, Q-REACT is
negative, meaning that an external energy supply is needed to ensure the desired
reaction temperature. In such conditions, the process is not self-sustainable. Vice
versa, when biomass concentration increases, Q-REACT also increases, until it
becomes positive: the burner produces more heat than the amount required by the
reactor.
Interestingly, for each compound analyzed there is a concentration at which
Q-REACT is equal to zero. Such concentration can be referred as “energy self-
sufficiency concentration” and it represents the minimum biomass concentration
to be supplied in order to make the SCWG energetically self-sustainable. As it
can be observed in Fig. 4.5a, each biomass shows its own energy self-sufficiency
concentration, ranging from 12 wt.% for phenol to 24 wt.% for sewage sludge.
In any case, the biomass concentration required to ensure energy self-sufficient
operations is much higher than the one adopted in several literature experimental
studies [32–34], where concentrations even lower than 5 wt.% were adopted
[24, 25].
4 Thermodynamic Analysis of the Supercritical Water Gasification of Biomass
Fig. 4.5 Energy analysis. The indicators represent the simulation outputs, the curves connecting the indicators are intended to help the reader in the
comprehension of the figure. (a) Thermal flux Q-REACT as a function of biomass concentration in the feed (from simulations where no CH4 is supplied
to the burner; T D 700 ı C, P D 30 MPa); (b) relation between the biomass HHV and the energy self-sufficiency concentration (T D 700 ı C, P D 30 MPa); (c)
thermal flux Q-REACT as a function of microalga Spirulina concentration in the feed at various temperatures and pressures (from simulations where no CH4
is supplied to the burner); (d) thermal flux Q-SEPAR as a function of biomass concentration in the feed (from simulations where the reaction step occurs at
T D 700 ı C and P D 30 MPa) (Reprinted and adapted with permission from Fiori et al. [28]. Copyright © 2012, Elsevier)
125
In Fig. 4.5b it can be observed the relationship between the energy self-
sufficiency concentration and the higher heating value (HHV) of the considered
biomass. Biomass with the lower HHV shows the higher energy self-sufficiency
concentrations and vice versa. Interestingly, the points dispose themselves along a
straight line. It can be also derived a correlation between the HHV (expressed in
MJ/kg) and the energy self-sufficiency concentration (wt.%), in the form:
ESSC D 0:6368 HHV C 31:6 (4.31)
This formula may be used as a rule-of-thumb to determine the energy self-

sufficiency concentration for a given biomass, at the considered process conditions
(700 ı C, 30 MPa).
Process conditions also influence the value of Q-REACT. In Fig. 4.5c, several
analyses were performed with microalga Spirulina at different temperatures and
pressures. Here, it can be observed that lower reaction temperatures result in
lower energy self-sufficiency concentrations. Indeed, at 700 ı C and 30 MPa, the
energy self-sufficiency concentration for such biomass is around 18 wt.%. When
the process is operated at 500 ı C, it drops to 13 wt.%. Conversely, if a temperature
of 900 ı C is adopted, an energy self-sufficiency concentration higher than 25 wt.%
is foreseen. In practical terms, this means that a self-sustainable process can hardly
be obtained, since it becomes difficult to pump a very concentrated water/biomass
mixture up to the desired pressure. The effect of pressure on the energy self-
sufficiency concentration is far less evident. Higher pressures result in lower energy
self-sufficiency concentrations, but the effect is extremely modest.
Based on such results, it could be affirmed that it is technologically simpler to
operate at low temperatures. Indeed, since the energy self-sufficiency concentrations
are lower, pumping issues are greatly reduced. Furthermore, plants operating at
lower temperatures also require reduced investment costs due to the possibility
to resort to cheaper construction materials. On the other hand, the goal of the
process must be carefully considered. Indeed, as it was shown in Sect. 4.3.4, low
temperatures are more favorable for the formation of CH4 , while they are not
suitable for H2 . For the process here considered, this would result in lower power
productions, since there will be less H2 in the syngas. In addition, it has been proved
in the literature that higher temperatures result in higher energy efficiencies [25]. It
is thus important to carefully evaluate the optimal tradeoff between efficiency and
technological feasibility.
In Fig. 4.5d, the focus is on the thermal energy generated by the separation of
water, which is represented by the stream Q-SEPAR (Fig. 4.4). It has already been
explained (see Sect. 4.4.2) that this energy stream actually represents a process cost,
since the heat must be dissipated somehow.
An increase in the concentration of the biomass fed to the reactor results in
lower amounts of energy to be dissipated. Indeed, higher biomass concentration
means lower water content, thus implying less thermal energy to be dissipated for
its condensation. Operating at higher biomass concentrations is therefore useful to

reduce the energy required for cooling. As far as biomass typology is concerned,
differences are consequently more evident at higher concentrations. Phenol shows
the lowest energy demand for cooling down, while glycerol exhibits the highest one.
4.5 Conclusions
Thermodynamic modeling is a powerful tool to analyze a complex process like

supercritical water gasification of biomass. The main advantage of this kind of
methodology is that it can provide a good way to model a complex system
without a specific and punctual knowledge about it. Modeling through the non-
stoichiometric approach can address this necessity and give very good results that
fit the experimental data in a very satisfactory way. Such family of models can
be implemented by means of common numerical calculus software, like MatLab® ,
thus they are a concrete practical tool that can be self-constructed by the scientists
involved.
Thermodynamic modeling is able to produce fundamental pieces of information.
We showed in this chapter how thermodynamics is able to foresee how the different
reaction conditions affect the composition of the syngas produced. This approach
enables to understand which are the process parameters that play a major role on
the final result, thus helping to concentrate the attention on the most significant
ones. We also showed the contribution that thermodynamic modeling can give to
the understanding of the problem of solid carbon formation, which is one of the
technical limitations of the SCWG process. Therefore, thermodynamic modeling
represents a concrete and effective usefulness for the technical field.
Thermodynamics also constitutes the “heart” of process simulations, which can
be carried out with commercial software like Aspen Plus® . In this way, results
directly connected to industrial practice can be obtained.
Several challenges are still open. First of all, thermodynamics is very effective
in modeling reality when high temperatures or effective catalysts are adopted,
otherwise equilibrium is not actually achieved in reality. More effective models
could be built, taking into account efficiency factors to reconcile numerical outputs
with non-equilibrium experimental results. Even better, reaction kinetics should be
accounted for in these cases [35]. This will also lead to a more accurate estimation
of the energy feasibility of the process.
Finally, as knowledge about the reaction mechanisms involved in SCWG
increases, new stoichiometric thermodynamic models could be implemented. In
this way, the contributions of the single reactions could be obtained, highlighting
the most significant ones and the conditions most favorable for their equilibrium.
Such piece of information will give the impulse for a more effective design of
SCWG reactors and processes.
References
1. Matsumura Y, Minowa T, Potic B, Kersten S, Prins W, van Swaaij W, van de Beld

B, Elliott D, Neuenschwander G, Kruse A, Antal Jr MJ. Biomass gasification in near-
and super-critical water: status and prospects. Biomass Bioenergy. 2005;29(4):269–92.
doi:10.1016/j.biombioe.2005.04.006.
2. Castello D, Fiori L. Supercritical water gasification of biomass: thermodynamic constraints.
Bioresour Technol. 2011;102(16):7574–82. doi:10.1016/j.biortech.2011.05.017.
3. Tang H, Kitagawa K. Supercritical water gasification of biomass: thermodynamic anal-
ysis with direct Gibbs free energy minimization. Chem Eng J. 2005;106(3):261–7.
doi:10.1016/j.cej.2004.12.021.
4. Wagner W, Pruß A. The IAPWS formulation 1995 for the thermodynamic properties
of ordinary water substance for general and scientific use. J Phys Chem Ref Data.
2002;31(2):387–535. doi:10.1063/1.1461829.
5. Letellier S, Marias F, Cezac P, Serin JP. Gasification of aqueous biomass in supercrit-
ical water: a thermodynamic equilibrium analysis. J Supercrit Fluid. 2010;51(3):353–61.
doi:10.1016/j.supflu.2009.10.014.
6. Marias F, Letellier S, Cezac P, Serin JP. Energetic analysis of gasification of
aqueous biomass in supercritical water. Biomass Bioenergy. 2011;35(1):59–73.
doi:10.1016/j.biombioe.2010.08.030.
7. Kaufman L, Bernstein H. Computer calculation of phase diagrams. New York: Academic;
1970.
8. Zhu J, Zhang D, King KD. Reforming of CH4 by partial oxidation: thermodynamic and kinetic
analyses. Fuel. 2001;80:899–905. doi:10.1016/S0016-2361(00)00165-4.
9. Gordon S, McBride BJ. Computer program for calculation of complex chemical equilibrium
compositions and applications. Cleveland: National Aeronautics and Space Administration
(NASA); 1994.
10. Zheng L, Furimsky E. Assessment of coal combustion in O2 C CO2 by equilibrium calcula-
tions. Fuel Process Technol. 2003;81(1):23–34. doi:10.1016/s0378-3820(02)00250-3.
11. Königsberger E, Eriksson G. Simulation of industrial processes involving concentrated aque-
ous solutions. J Solut Chem. 1999;28(6):721–30. doi:10.1023/A:1021768011887.
12. Yan Q, Guo L, Lu Y. Thermodynamic analysis of hydrogen production from biomass
gasification in supercritical water. Energy Convers Manag. 2006;47(11–12):1515–28.
doi:10.1016/j.enconman.2005.08.004.
13. Voll FAP, Rossi CCRS, Silva C, Guirardello R, Souza ROMA, Cabral VF, Cardozo-
Filho L. Thermodynamic analysis of supercritical water gasification of methanol,
ethanol, glycerol, glucose and cellulose. Int J Hydrog Energy. 2009;34(24):9737–44.
doi:10.1016/j.ijhydene.2009.10.017.
14. Gutiérrez Ortiz FJ, Ollero P, Serrera A, Sanz A. Thermodynamic study of the super-
critical water reforming of glycerol. Int J Hydrog Energy. 2011;36(15):8994–9013.
doi:10.1016/j.ijhydene.2011.04.095.
15. Freitas ACD, Guirardello R. Thermodynamic analysis of supercritical water gasification of
microalgae biomass for hydrogen and syngas production. Chem Eng Trans. 2013;32:553–8.
doi:10.3303/CET1332093.
16. Yakaboylu O, Harinck J, Gerton Smit KG, de Jong W. Supercritical water gasifica-
tion of manure: a thermodynamic equilibrium modeling approach. Biomass Bioenergy.
2013;59:253–63. doi:10.1016/j.biombioe.2013.07.011.
17. Chrastil J. Solubility of solids and liquids in supercritical gases. J Phys Chem.
1982;86(15):3016–21. doi:10.1021/j100212a041.
18. McBride BJ, Gordon S, Reno MA. Coefficients for calculating thermodynamic and transport
properties of individual species. Cleveland: National Aeronautics and Space Administration
(NASA); 1993.
19. NIST NIST-JANAF thermochemical tables. http://kinetics.nist.gov/janaf/. Accessed 10 Jan

2014.
20. Peng D-Y, Robinson DB. A new two-constant equation of state. Ind Eng Chem Fundam.
1976;15(1):59–64. doi:10.1021/i160057a011.
21. Byrd A, Pant K, Gupta R. Hydrogen production from glycerol by reforming in supercritical
water over Ru/Al2 O3 catalyst. Fuel. 2008;87(13–14):2956–60. doi:10.1016/j.fuel.2008.04.024.
22. Castello D, Kruse A, Fiori L. Biomass gasification in supercritical and subcritical water: the
effect of the reactor material. Chem Eng J. 2013;228:535–44. doi:10.1016/j.cej.2013.04.119.
23. Müller JB, Vogel F. Tar and coke formation during hydrothermal processing of glycerol and
glucose. Influence of temperature, residence time and feed concentration. J Supercrit Fluid.
2012;70:126–36. doi:10.1016/j.supflu.2012.06.016.
24. Feng W. Phase equilibria for biomass conversion processes in subcritical and supercritical
water. Chem Eng J. 2004;98(1–2):105–13. doi:10.1016/s1385-8947(03)00209-2.
25. Gutiérrez Ortiz FJ, Ollero P, Serrera A, Galera S. An energy and exergy analysis of
the supercritical water reforming of glycerol for power production. Int J Hydrog Energy.
2012;37(1):209–26. doi:10.1016/j.ijhydene.2011.09.058.
26. Gutiérrez Ortiz FJ, Ollero P, Serrera A, Galera S. Optimization of power and hydrogen
production from glycerol by supercritical water reforming. Chem Eng J. 2013;218:309–18.
doi:10.1016/j.cej.2012.12.035.
27. Withag JAM, Smeets JR, Bramer EA, Brem G. System model for gasification of biomass
model compounds in supercritical water – a thermodynamic analysis. J Supercrit Fluid.
2012;61:157–66. doi:10.1016/j.supflu.2011.10.012.
28. Fiori L, Valbusa M, Castello D. Supercritical water gasification of biomass for H2 production:
process design. Bioresour Technol. 2012;121:139–47. doi:10.1016/j.biortech.2012.06.116.
29. ECN HYSEP, Hydrogen Separation Modules. http://www.hysep.com. Accessed 16 Sept 2013.
30. Fuel Cells Handbook. EG&G Technical Services Inc. U.S. Department of Energy, Office of
Fossil Energy. National Energy Technology Laboratories.
31. ECN Phyllis, database for biomass and waste. http://www.ecn.nl/phyllis. Accessed 16 Sept
2013.
32. Chuntanapum A, Matsumura Y. Char formation mechanism in supercritical water gasi-
fication process: a study of model compounds. Ind Eng Chem Res. 2010;49:4055–62.
doi:10.1021/ie901346h.
33. Goodwin AK, Rorrer GL. Conversion of xylose and xylose-phenol mixtures to hydrogen-
rich gas by supercritical water in an isothermal microtube flow reactor. Energy Fuel.
2009;23:3818–25. doi:10.1021/ef900227u.
34. Lu Y, Guo L, Ji C, Zhang X, Hao X, Yan Q. Hydrogen production by biomass gasifi-
cation in supercritical water: a parametric study. Int J Hydrog Energy. 2006;31(7):822–31.
doi:10.1016/j.ijhydene.2005.08.011.
35. Castello D, Fiori L. Kinetics modeling and main reaction schemes for the supercritical water
gasification of methanol. J Supercrit Fluid. 2012;69:64–74. doi:10.1016/j.supflu.2012.05.008.
Part II
Reactor Design
Chapter 5
Optical Reactors for Microscopic Visualization
of Chemical Processes in Sub- and Supercritical
Water
Shigeru Deguchi and Sada-atsu Mukai
Abstract Direct visualization is a straightforward and yet powerful approach to

study chemical processes in sub- and supercritical water, where phase behaviors
of aqueous systems significantly differ from those at ambient conditions due to
largely different properties of water under such extreme conditions. This chapter
mainly reviews flow-through optical reactors that are designed to allow microscopic
observations of chemical processes in sub- and supercritical water with high optical
resolution. Applications of the reactors include studying crystalline-to-amorphous
transformation of cellulose in subcritical water, which is analogous to gelatinization
of starch in hot water, and bottom-up formation of nano-sized oil droplets in
subcritical water.
Keywords Optical microscopy • Cellulose • Crystalline-to-amorphous transfor-

mation • Gelatinization • Acid catalyst • Nanoemulsion
S. Deguchi ()
R&D Center for Marine Biosciences, Japan Agency for Marine-Earth Science
and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan
S. Mukai
R&D Center for Marine Biosciences, Japan Agency for Marine-Earth Science
and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan
Akiyoshi Bio-Nanotransporter Project, JST, ERATO, Akiyoshi Bio-Nanotransporter Project,
Katsura Int’tec Center, Kyoto daigaku-Katsura, Nishikyo-ku, Kyoto 615-8530, Japan
Department of Polymer Chemistry, Graduate School of Engineering, Kyoto University,
Kyoto daigaku-Katsura, Nishikyo-ku, Kyoto 615-8510, Japan
134 S. Deguchi and S. Mukai
5.1 Introduction
In 1869, Thomas Andrews reported his observation on a peculiar behavior of carbon

dioxide [1]. He wrote “On partially liquefying carbonic acid by pressure alone,
and gradually raising at the same time the temperature to 88ı Fahr., the surface of
demarcation between the liquid and gas became fainter, lost its curvature, and at
last disappeared.” This is the first scientific account of supercritical fluids (in fact,
the term “supercritical” was coined by himself to describe the newly found state of
matter). It is evident from his description that visual inspection was the technique
of choice behind this landmark discovery in physical chemistry. Up till now, direct
observation has widely been used to study chemical reactions in supercritical water
(SCW), which has attracted considerable scientific and technical interests as an
environmentally benign medium for chemical reactions [2, 3].
SCW exhibits properties that are remarkably different from those of ambient
liquid water, which we all are familiar with. For example, the dielectric constant
of water is 78 at 25 ı C and 0.1 MPa, and is very high among liquid solvents.
However, it decreases substantially when heated under pressure. More specifically,
the dielectric constant decreases to 6 at the critical point [2], and this value is
comparable to that of 1-dodecanol at ambient conditions (5.8) [4]. Because of
the large reduction of the dielectric constant, hydrocarbons become highly soluble
in water, while electrolytes precipitate out [2]. Such tremendous change of the
phase behavior, change of solubility in particular, should have profound impacts
on the reactions that occur in sub- and supercritical water, and direct visualization
is a powerful and straightforward approach for such studies, especially when it is
combined with optical microscopy.
This chapter reviews designs and applications of optical reactors for microscopic
visualization of chemical processes in sub- and supercritical water. Optical cells for
spectroscopic applications are not reviewed. We will begin with reviews of reactors
that are designed for studying high-pressure systems, and basic design requirements
will be discussed. We will then review optical reactors that are designed to operate at
high temperature and high pressure up to the supercritical state of water. Attention
will be placed on specific issues associated with extending the temperature range
for the operation of high-pressure optical reactors. Finally, applications of optical
reactors for chemical processes in sub- and supercritical water will be presented
with emphasis on crystalline-to-amorphous transformation and formation of nano-
sized oil droplets.
5 Optical Reactors for Microscopic Visualization of Chemical Processes. . . 135
5.2 Optical Reactors for High-Pressure Systems
5.2.1 Design Considerations for High-Pressure Optical

Reactors for Microscopy
Usefulness of direct observation is significantly enhanced when it is combined

with optical microscopy. Development of such reactors has become increasingly
important as the interest in biological processes at high pressures has emerged.
In designing such reactors, one has to consider specific requirements that the
microscope imposes, particularly the design of optical windows.
It is obvious that a thicker window has higher mechanical strength and, therefore,
is preferable for the construction of a reactor to be operated at high pressures.
However, to obtain microscopic images with high optical resolution, the window
has to be made as thin as possible. This is because an objective lens with high
numerical aperture, which is a critical component for obtaining high-resolution
images, generally has a short working distance.
Thus, the high-pressure cell for optical microscopy has to balance two conflicting
requirements; “mechanical strength” and “short distance between objective lens
and sample”. Two types of designs have been proposed to construct high-pressure
reactors for microscopic observations; opposed-anvil type cells and flow-type cells
(Fig. 5.1).
5.2.2 Opposed-Anvil Type Cells
The opposed-anvil cell consists of two optically-transparent anvils made of single

crystalline diamond or sapphire, and a metal gasket with a bore-hole where a sample
Fig. 5.1 Schematic representation of a diamond anvil cell (DAC) (a) and a cell with tapered-
windows (b)
is contained (Fig. 5.1a). The sample is compressed by pushing the opposed anvils
toward each other. Diamond anvil cells (DAC) can generate extremely high pressure
up to 102 GPa despite their simple and compact construction, even though only a
small volume of sample can be used.
One major drawback of DAC is the fact that the optical resolution of images is
not very high due to the use of thick diamond anvils. To circumvent this problem, the
cell that combines a flat window and a diamond anvil can be used [5]. In this cell,
the anvil was used for passing illumination, while observation was made through
the thin flat window. The chamber can compress the sample up to 1.4 GPa, and an
objective lens having the working distance of 6 mm could be used.
Another drawback associated with DAC is the difficulty of measuring the
pressure of the sample. This is usually done by putting a pressure indicator together
with the sample, and small ruby and Sm:YAG crystals, whose fluorescent properties
depend on the pressure, are commonly used for this purpose [6, 7].
The pressure can also be measured using a small air bubble trapped in the sample
chamber. The pressure in the sample chamber can be calculated with an accuracy
on the order of 1 % using the equation of state of H2 O once the temperatures
of the appearance and disappearance of the air bubble have been determined [6].
Other indicators that can be used for measuring the pressure include the ’-“
quartz transition and the ferroelastic phase transitions in BaTiO3 (tetragonal/cubic),
Pb3 (PO4 )2 (monoclinic/trigonal), and PbTiO3 (tetragonal/cubic) [7].
5.2.3 Flow-Type Cells
Flow-type reactors are used to utilize thinner optical windows for achieving micro-
scopic observations with high optical resolution. Figure 5.1b shows the structure of
common high-pressure cells that balance optical resolution and operating pressure
at a high level. Taper-shaped windows are usually made of sapphire because its
mechanical strength is superior to crystalline quartz. When the sample inside the
cell is pressurized, the windows are pressed outward against the window holders to
make the seal.
Many different designs of flow-type cells have been reported. For example,
Hartmann et. al. reported a high-pressure cell to be operated at pressures up to
300 MPa using an objective lens having the working distance of 3 mm [8–10].
Nishiyama reported the cell that can operate at pressures up to 200 MPa. The cell
used a 1.5 mm thick crystalline quartz as optical windows, and an objective lens
having the working distance of 4.3 mm could be used for observations [11]. There
are also reactors that use diamond discs as optical windows. However, the use of
diamond windows is still limited, primarily due to their high price tag.
Unlike DAC, the pressure of the sample in the reactor can easily be measured
and the reactor volume can be made large. A major disadvantage of the flow-type
reactor is that operating pressure range is lower than that of DAC.
5.3 Optical Reactors for High-Temperature

and High-Pressure Systems
5.3.1 Design Considerations for High-Temperature

and High-Pressure Optical Reactors for Microscopy
The primary function that is necessary for both high-pressure optical reactors and
high-temperature and high-pressure (HTHP) optical reactors is containment of a
sample at high pressure. Thus, in principle, high-pressure optical reactors can also
be used as HTHP optical reactors when a heating mechanism is incorporated.
However, extra technical issues have to be addressed when extending the operating
temperature range of the former to very high temperature around Tc of water.
Like microscopic observations of high-pressure systems, early attempts to visu-
alize high-temperature and high-pressure systems with optical microscopy started
with DAC. A DAC for hydrothermal studies was first developed to study the phase
behavior of H2 O – N2 mixtures at pressures to 2.1 GPa and temperatures to 830 K
[12]. Similar cells have also been developed for the study of the dehydration,
complexation, speciation, phase transitions, solubility and reaction kinetics of
various minerals [13, 14], and the phase behavior of synthetic polymers in water
at high temperatures and pressures including SCW [15–17]. The systems allow
operation in the very wide temperature and pressure range up to 927 ı C and 30 GPa
[12] or from 463 to 1,473 ı C and to 2.5 GPa [13]. However, incorporation of the
heating mechanism made the DAC so large that it could only be used with a simple
arrangement of lenses [12] or a stereo microscope [13]. The optical resolution of
images obtained by such systems is poor.
To realize visualization with the high optical resolution, flow-type reactors were
developed. As described above, although the operating pressure and temperature
range are limited compared with those of the DAC, the flow-type reactor has a
distinct advantage over the DACs in that the precise control and measurement of
the pressure can be done in the pressure range close to the critical pressure of water,
20–40 MPa, where the properties of SCW depends very much on pressure [2].
Such reactors have been developed and used to study properties of SCW and
chemical reactions in SCW with various optical techniques [18–25]. These designs
focus primarily on holding the high-temperature and high-pressure fluids without
leakage. However, the HTHP optical reactors for microscopy should fulfill the
specific requirements that the microscope imposes. First, the reactor dimension is
restricted by the working distance of an objective lens, which is typically less than
10 mm for imaging with a high optical resolution. Second, part of the microscope,
especially the objective lens, inevitably locates so close to the hot sample that an
effective cooling system must be implemented to prevent damage to the microscope
due to radiant heat, and in the meantime without affecting the temperature stability
of the sample in the reactor.
Fig. 5.2 Top view (a) and cross-sectional view (b) of the HTHP optical reactor for microscopy.
1 Optical window, 2 reactor body, 3 stopper, 4 Belleville washer, 5 compression nut, 6 thermo-
couple. Cross-sectional top (c) and side (d) views of the reactor contained in the cooling jacket.
Bottom view (e), showing the cooling plate, is also shown. 7 Reactor body, 8 heater block, 9
cartridge heater, 10 cooling jacket, 11 ceramic insulator, 12 cooling plate, 13 cooling jacket for an
objective lens, 14 objective lens, 15 thermocouple (Reproduced with permission from Deguchi and
Tsujii [26]. Copyright © 2002, American Institute of Physics)
5.3.2 HTHP Optical Cell for High-Resolution Optical

Microscopy
We developed a flow-type HTHP optical reactor that can be used with a conventional
inverted optical microscope [26]. Figure 5.2a, b show top view and cross-sectional
side view of the reactor body. The reactor was designed to operate above the
critical point of water up to 400 ı C and 35 MPa. Unlike water at ambient
conditions, corrosion is a severe problem in SCW [2]. Thus, the reactor body was
machined from a block of corrosion-resistant Ni-based super-alloy, Inconel 600.
The outer dimension of the reactor body was 80 40 35 mm. A sample chamber
of 3.18 mm 3.18 mm ID was bored in the reactor body. The volume inside the
chamber was 0.6 mL.
For providing optical access, a channel (3.18 mm 3.18 mm ID) was bored
perpendicularly to the sample chamber, and optical windows were placed at both
ends of the channel. The optical windows were made of type-I diamond (2.50 mm
anvil face, 4.50 mm table face, 2.83 mm thick, 0.66 carats). Side of the window
Table 5.1 Specifications of Magnification Working distance/mm Numerical aperture

objective lenses (Nikon EPI
SLWD) [26] 10 20.3 0.21
20 20.5 0.35
had 16 facets. Both ends of the optical channel were tapered so that the diamond
windows fit the cell body well. The diamond windows were fit to the reactor body
by titanium gaskets (2 mm ID, 7 mm OD, 0.3 mm thick) and fixed by a compression
nut. The initially flat titanium gasket deformed as the compression nut was driven
to 25 Nm, and made effective seal between the window and the reactor body. Two
Belleville washers (10 mm ID, 20 mm OD, 9 mm thick) were placed between the
window and the compression nut to accommodate the thermal expansion mismatch
among Inconel 600, titanium, and diamond. Temperature of the sample inside the
reactor was monitored by a chromel–alumel thermocouple (1.59 mm 1.59 mm
OD, 6 in Fig. 5.1), inserted in the reactor body and located 10 mm away from
the sample. Tubes of 3.18 mm ID, also made of Hastelloy C22, were welded
at both ends of the sample chamber, and served as an inlet and outlet of the
reactor.
Figure 5.2c–e show cross-sectional top view, cross-sectional side view, and
bottom view of the reactor. The reactor body was placed between two heater blocks
made of brass. Two 250 W electric cartridge heaters were embedded in each block.
The sample was heated indirectly by the heat transfer from the heater blocks to the
reactor body. Temperature was controlled by a PID controller (DSM5 temperature
control unit, Shimaden Co., Ltd., Tokyo, Japan).
The reactor body and the heater blocks were contained in a cooling jacket
(150 150 62.5 mm) made of SUS-316, and water that was kept at 20 ı C by an
external cooling bath (RTE-110, Neslab Instruments, Inc., Portsmouth NH, USA)
was circulated through the hollow jacket walls. The space between the reactor body
and the jacket walls was filled with blocks of heat insulator made of alumina–
silicate ceramics for better temperature stability. Additional cooling mechanism
included a cooling plate at the lower side of the reactor and a cooling jacket for
the objective lens, both of which were also cooled by circulating water. The reactor
was manufactured by AKICO Co., Ltd. (Tokyo, Japan).
5.3.3 Experimental Set-Up
The reactor was mounted on a bracket that was built on an Eclipse TE300 inverted
optical microscope (Nikon, Tokyo, Japan) [26]. The reactor position was adjustable
in x-y-z directions. Due to the thickness of the reactor, a limited range of objective
lenses could be used. We used two objective lenses with long working distance:
CF IC EPI Plan SLWD (Nikon, Tokyo, Japan) of 10 and 20 magnification. The
specifications of the lenses are shown in Table 5.1.
Fig. 5.3 Schematic

illustration of experimental
set-up. Water is continuously
supplied by HPLC pump.
Pressure is controlled by a
back pressure regulator
(Reproduced with permission
from Mukai et al. [27].
Copyright © 2005, Elsevier
B.V.)
The reactor was connected to a flow-type pressurizing system (Fig. 5.3) con-
sisting of an HPLC pump (PU-1580, JASCO, Hachioji, Japan), a back pressure
regulator (Model 880-81, JASCO), a pressure transmitter (KH15, Nagano Keiki Co.,
Ltd., Tokyo, Japan) interfaced to a digital pressure display, and a stop valve [26]. A
relief valve, set at 40 MPa, was also inserted between the pressure transmitter and
the stop valve to protect the reactor from accidental high pressure. The components
were connected by tubes (1.59 or 3.18 mm) made of SUS316.
In a typical experiment, a sample solution was introduced into the reactor. The
sample was pressurized to 25 MPa at room temperature, and heated to 400 ı C while
maintaining the pressure constant. The reactor was able to heat the sample to 350 ı C
within 25 min, and the heating rate was nearly constant at around 10–20 ı C/min
at temperatures above 50 ı C. Images of the sample were taken by using a
chilled CCD color camera (C5810, Hamamatsu Photonics K. K., Hamamatsu,
Japan) and recorded with a VCR (GV-D300, SONY, Tokyo, Japan) during heat-
ing. The recorded images were transferred digitally to a computer for image
analysis.
5.3.4 Further Improvement of HTHP Optical Reactor
We newly developed an HTHP optical reactor with improved temperature uni-

formity [27]. Figure 5.4 shows the top and cross-sectional view of the new
flow-type reactor. The reactor body was made of a disk of titanium alloy and a flow
channel was bored. Optical access to the sample in the reactor body was provided by
two optical windows that were made of synthetic single-crystalline diamond disks
(diameter 2.8 mm, thickness 1 mm). The reactor body was surrounded by four
electric heaters, which were arranged symmetrically to minimize the temperature
gradient across the reactor body. Temperature of the sample in the reactor body
was measured directly by a thermocouple, which was inserted through a flow
Fig. 5.4 Top-view (a) and cross-sectional view (b) of the HTHP optical reactor for microscope. 1
Reactor body, 2 optical window, 3 sample inlet, 4 sample outlet, 5 electric heater, 6 heater block,
7 thermocouple, 8 heat insulator, 9 cooling jacket (Reproduced with permission from Mukai et al.
[27]. Copyright © 2005, Elsevier B.V.)
channel. The outermost surface of the reactor and the objective lens were cooled
by the circulating water from a temperature-controlled bath. This reactor could be
operated at temperatures up to 450 ı C and pressures up to 40 MPa. The new
reactor is smaller than the previous one, and can be set up on a sample stage of
a conventional inverted microscope without any modifications (IX-71, Olympus,
Tokyo, Japan) [27].
5.4 Applications of HTHP Optical Reactors
5.4.1 Direct Observation of Crystalline Cellulose

in Subcritical Water
5.4.1.1 Cellulose and Starch as Feedstock for Biorefinery
Starch and cellulose are major feedstock for biorefinery. These two polysaccharides
only differ in their connectivity of glucose (’-1-4 vs. “-1-4 glycosidic linkages,
respectively), but their properties are profoundly different.
Starch is semi-crystalline at room temperature, but it undergoes crystalline-to-
amorphous transformation in water when heated to 60–70 ı C (known as gela-
tinization) [28]. The starch chains become swollen upon gelatinization and can be
attacked readily by hydrolytic enzymes. Gelatinization is the preliminary process
necessary to render starch suitable for enzyme-catalyzed biomass conversion [29].
Cellulose is the most abundant biomass on Earth with an estimated annual
production of 100 billion dry tons [30]. It is a highly crystalline material, and its
utilization is hampered by the recalcitrance of the crystalline domains to chemical or
enzymatic hydrolysis. The recalcitrance, as well as other unique characteristics such
as insolubility in most solvents including water and structural rigidity, arises from
Fig. 5.5 In situ crossed-polar microscopic images of crystalline cellulose in water taken between
300 and 330 ı C and at constant pressure of 25 MPa. Temperature, (a) 300 ı C, (b) 310 ı C, (c)
320 ı C, (d) 330 ı C. Scale bars represent 50 m (Reproduced with permission from Deguchi et al.
[39]. Copyright © 2005, The Royal Society of Chemistry)
hydrogen bonding networks formed between the cellulose chains in the crystals [31,
32]. Thus, a similar change would be expected if cellulose was gelatinized.
We know from our experience that gelatinization of cellulose does not occur in
water at ambient conditions. However, indirect evidence suggested some sorts of
change of the crystalline structure of cellulose in water at high temperature and
high pressure. Transformation of crystalline structure was reported after treating
crystalline cellulose in hot and compressed water [33, 34]. An anomalous increase
of the hydrolysis rate of cellulose was also reported in hot and compressed water
near the critical point, and the kinetics of the reaction was studied in detail
[35–38]. Based on the results, some sort of change in crystalline structure was
inferred [38].
Gelatinization of starch can be studied conveniently by following loss of birefrin-
gence using polarizing optical microscopy [28]. We performed in situ crossed-polar
microscopic observation of crystalline cellulose in subcritical and supercritical
water [39].
5.4.1.2 Crystalline-to-Amorphous Transformation of Cellulose
A series of in situ crossed-polar optical microscopic images of crystalline cellulose

(CF1, Whatman) in hot and compressed water was taken during heating at 25 MPa
with the heating rate of 11–14 ı C/min (Fig. 5.5). Birefringence of the fibrous
Fig. 5.6 In situ crossed-polar micrographic images showing a cellulose fiber (shown by a green
triangle in a) during the crystalline-to-amorphous transformation. Temperature, (a) 324 ı C, (b)
325 ı C, (c) 326 ı C, (d) 327 ı C, (e) 328 ı C, (f) 329 ı C. Each image is 100 200 m (Reproduced
with permission from Deguchi et al. [39]. Copyright © 2005, The Royal Society of Chemistry)
cellulose, which is evident from the pseudo color under crossed polars, was retained
up to 310 ı C without any noticeable change, showing that cellulose remained
crystalline up to this high temperature. Hydrolysis of cellulose seems negligible,
as the size of the fibrous cellulose did not change noticeably. However, cellulose
became less birefringent at around 320 ı C, and the birefringence was completely
lost at 330 ı C. The micrographs clearly show that cellulose undergoes crystalline-
to-amorphous transformation in water at around 320 ı C and 25 MPa. Dissolution
of cellulose followed the transformation, and no cellulose remained at 340 ı C.
Recrystallization was not observed when the system was cooled, confirming the
previous observations that cellulose was hydrolyzed very rapidly under similar
experimental conditions [35–38].
Figure 5.6 shows a sequence of images of a single cellulose fiber (indicated by a
green triangle in Fig. 5.6a) near the transformation temperature. The fiber gradually
lost birefringence as it was heated (Fig. 5.6a–c), with no significant change in the
shape. The fiber started to deform when it almost completely lost birefringence
(Fig. 5.6d), and deformed further at higher temperatures (Fig. 5.6e, f). The large
deformation suggests that cellulose becomes plastic, and the mechanical properties
change dramatically upon the transformation. Twisting and bending were observed
for most of the fibrous cellulose upon the transformation.
Crystalline cellulose consists of highly ordered crystallites called fringed
micelles and less-ordered domains in between [40]. Under normal conditions, water
only interacts with the less ordered domains [41]. The present observations clearly
show that water also interacts with highly ordered domains at high temperatures
under pressure, leading to a crystalline-to-amorphous transformation. It seems that
water is necessary to induce the transformation because no such transformation was
observed in ethanol (Tc D 243 ı C, Pc D 6.4 MPa) at 7 MPa and at temperatures
up to 350 ı C. Birefringence was retained throughout the observation, and char
formation resulted above 330 ı C. The result indicates that the transformation is not
a simple thermal melting, but rather interaction between water and cellulose plays
an essential role.
Fig. 5.7 A series of in situ optical microscopic images showing starch granules in water at a
constant pressure of 25 MPa and different temperatures. Scale bars represent 100 m (Reproduced
with permission from Deguchi et al. [43]. Copyright © 2008, The Royal Society of Chemistry)
5.4.1.3 Comparison with Starch
Unsurpassed stability of crystalline cellulose in hot and compressed water was

also revealed by comparing behavior of cellulose and starch. Figure 5.7 shows a
sequence of microscopic images of starch granules when heated in water up to
200 ı C at 25 MPa. The starch granules were swollen dramatically between 80 and
90 ı C, which is one of the characteristic features of a crystalline-to-amorphous
transformation of starch [28]. Upon further heating, the texture of the swollen
starch granules became fainter with temperature, and was lost completely at around
200 ı C. Like cellulose, no precipitate appeared upon cooling the specimen to room
temperature, suggesting that the swollen starch chains were hydrolyzed in hot and
compressed water at around 200 ı C.
The observation suggests that the ’-1-4-glycosidic linkage, which connects
glucose residues to form the starch chain, is too labile in water at such high
temperature. It is likely that the “-1-4-glycosidic linkage that makes up the cellulose
chain should have similar susceptibility to hydrolysis in hot and compressed
water. Thus, the fact that cellulose remains unaffected at significantly higher
temperatures is ascribed to a protective role of the crystals, in which extensive
networks of hydrogen bonds among the cellulose chains are formed [31, 32, 42].
Upon transformation to an amorphous state, the network is broken up and the
cellulose chains become accessible to high-temperature water, leading to rapid
hydrolysis.
Table 5.2 Structural characteristics of cellulose samples [43]

Mwa DPb Crystallinity (%)c Crystalline formc Ref.
CF 3.8 104 2.3 102 73.4 Cellulose-I [39]
SF 3.6 104 2.2 102 58.2 Cellulose-I [44]
SF-II 4.2 104 2.6 102 44.5 Cellulose-II [44]
a
Measured by viscometry in copper ethylenediamine according to TAPPI standard T 230 su-66
b
Degree of polymerization
c
Measured by X-ray diffraction
5.4.1.4 Cellulose Having Different Structural Characteristics
Considering the crucial role of the crystalline structure for the stability of cellulose
in hot and compressed water, structural characteristics such as crystallinity should
affect the behavior of cellulose in subcritical water. We compared the behavior of
three cellulose samples of different structural characteristics [43]. The characteris-
tics of the samples are summarized in Table 5.2.
SF shares the same crystalline form (cellulose-I) with CF, but differs in crys-
tallinity. SF-II is a highly porous material consisting of very thin fibers (20–50 nm
thick) of crystalline cellulose, while the granules of SF have smooth surfaces
[44–46]. Consequently, the specific surface area of porous cellulose (200 m2 /g)
[46] is one or two orders of magnitude larger than that for typical crystalline
cellulose (1–10 m2 /g) [47]. Thus, comparison of SF-II with the others would
help to understand possible surface effects, such as hydrolysis on the surface or
diffusion of water molecules into the cellulose crystals, on the behavior of cellulose
in hot and compressed water. The crystalline form of SF-II (cellulose-II) also differs
from others (cellulose-I). As cellulose-II is thermodynamically more stable than
cellulose-I [40], the difference may have an impact on the crystalline-to-amorphous
transformation of cellulose.
Observations of the cellulose samples with different structural characteristics are
summarized in Fig. 5.8, and were quantified by calculating the brightness of the
images (Fig. 5.9). In this analysis, the average brightness of the whole image was
first calculated. We found that the brightness was also affected by thermal expansion
of the cell body that changes the alignment of the two opposing birefringent
windows made of diamond. Thus, this effect was compensated by measuring the
brightness of a part of the image that was not covered with cellulose, and dividing
the average brightness of the whole image by that of the uncovered area. The
analysis parallels a turbidity measurement.
All the samples became more transparent with increasing temperature, and
eventually dissolved completely (Figs. 5.8 and 5.9), indicating that they under-
went crystalline-to-amorphous transformation in hot and compressed water and
were hydrolyzed. It is clear from Fig. 5.9 that the relative brightness changes
in the narrow temperature range, suggesting that the transformation proceeds
in a phase-transition-like cooperative manner. Compared with CF, however, the
Fig. 5.8 Effect of crystallinity, crystalline structure, and morphology of cellulose on the dis-
solution behavior in water at high temperatures and at a constant pressure of 25 MPa. The
images demonstrate CF, SF, and SF-II in water at a constant pressure of 25 MPa and at different
temperatures. Scale bars represent 100 m (Reproduced with permission from Deguchi et al. [43].
Copyright © 2008, The Royal Society of Chemistry)
complete dissolutions of SF and SF-II were observed at lower temperatures. For

example, complete dissolution of SF was observed at around 320 ı C, a temperature
that was approximately 20 ı C lower than that for CF. The result indicates that
the crystallinity is an important factor that determines crystalline-to-amorphous
transformation of cellulose in hot and compressed water.
It was rather unexpected that the nanofibrous form of cellulose (SF-II) exhibited
stability comparable to the others (SF and CF), despite the huge difference in the
specific surface area. The observation clearly shows that the surface effect does not
affect the behavior of cellulose in hot and compressed water, and corroborate the
previous conclusion that, instead, the crystalline structure of cellulose is essential
Fig. 5.9 Effect of

crystallinity, crystalline
structure, and morphology of
cellulose on temperature
dependent change of relative
brightness (Reproduced with
permission from Deguchi
et al. [43]. Copyright © 2008,
The Royal Society of
Chemistry)
for the stability of cellulose in hot and compressed water. The difference in the
crystalline form between SF-II and others may also contribute to the stability of the
nanofibers.
We also observed that the fine crystals of SF moved around on the surface of the
lower optical window of the cell due to convective flow. Random reorientation of
the crystals due to thermal fluctuations was also evident. However, the movements
ceased completely by visual inspection at around 260–270 ı C, which was followed
by a crystalline-to-amorphous transformation. The observation may suggest that
there was partial swelling at the surface of cellulose crystals before the transfor-
mation, which made the crystals stick to the diamond surface.
5.4.1.5 Effect of Acid Catalyst
These observations described above clearly demonstrate that crystalline cellulose is

transformed to an amorphous state in water at high temperature and high pressure,
and the transformation plays a critical role in the hydrolysis of recalcitrant cellulose.
In practice, however, hydrolysis of cellulose was often performed in the presence
of catalysts [48–51], such as the acid-catalyzed hydrolysis, which has been used
for saccharification of cellulosic biomass for more than a century [52]. To avoid
issues associated with the acid-catalyzed hydrolysis, such as wastewater treatment
and corrosion of the reactor, a dilute acid process is favored, in which hydrolysis
is performed at high temperatures (170–240 ı C) in the presence of dilute acids
[53]. At such high reaction temperatures, the acid catalyst may affect the structural
characteristics of cellulose, thereby accelerating the hydrolysis [48, 49, 54]. Indeed,
Mok et al. reported that the hydrolysis rate of cellulose in dilute sulfuric acid
changed suddenly when the reaction temperature was raised from 212 ı C to 215 ı C,
and speculated that physical changes in the structure of cellulose were responsible
Fig. 5.10 A series of in situ optical microscopic images showing crystalline cellulose at various
temperatures in the presence of various concentrations of H2 SO4 . Scale bars represent 100 m.
Particles that appeared in the presence of 10 and 20 mol dm3 H2 SO4 at high temperatures are due
to corrosion of the reactor body (Reproduced with permission from Deguchi et al. [55]. Copyright
© 2008, The Royal Society of Chemistry)
for the change [48]. However, nothing is known as to how acid catalyst affects the
behavior of cellulose in hydrothermal conditions.
We performed in situ optical microscopic observation of crystalline cellulose
in water at high temperatures and high pressures in the presence of various
concentrations of H2 SO4 (Fig. 5.10). In the presence of 10 and 20 mol dm3 H2 SO4 ,
significant formation of particulate objects was observed at temperatures above
200 ı C. Such objects appeared even when 10 mol dm3 H2 SO4 was heated without
Fig. 5.11 (a) Comparison of dissolution behavior of crystalline cellulose in the presence of
various concentrations of H2 SO4 . The figure shows the change of relative brightness of the optical
microscopic images. The data obtained in the presence of 20 mol dm3 H2 SO4 are noisy due
to significant formation of particulate objects resulting from corrosion of the cell. (b) Change of
dissolution temperature (Tdissolution ) of crystalline cellulose as a function of the concentration of
added H2 SO4 (Reproduced with permission from Deguchi et al. [55]. Copyright © 2008, The
Royal Society of Chemistry)
cellulose, indicating that dilute sulfuric acid became so corrosive at such high
temperatures that even a Ni-based corrosion-resistant super-alloy that comprises the
reactor body was corroded.
Despite the harsh solution atmosphere, crystalline cellulose remained essentially
unchanged in water well over 200 ı C, even in the presence of H2 SO4 up to
20 mol dm3 . We did not observe any significant changes between 212 and 215 ı C,
where the sudden change in the reaction behavior was reported previously [48].
Transformation of crystalline cellulose to an amorphous state was observed only
at temperatures above 260 ı C. The transformation was immediately followed by
complete hydrolysis, and no precipitate appeared when the specimen was cooled to
room temperature.
It appears that complete dissolution of cellulose occurred at lower temperatures
as the H2 SO4 concentration was increased. This trend was quantified by a computer-
based image analysis, in which change of relative brightness of the images was
measured as a function of temperature (Fig. 5.11). In all of the cases, the relative
brightness was almost constant up to about 260 ı C, and started to increase gradually
with increasing temperature, due to increasing transparency of cellulose associated
with the crystalline-to-amorphous transformation. It then increased sharply, which
was mainly ascribed to dissolution of cellulose due to rapid hydrolysis. Finally,
the brightness became constant again after the complete dissolution. The analysis
revealed that the slope at which the relative brightness increased sharply with
temperature became steeper as the H2 SO4 concentration was increased, while the
temperature at which the sharp increase started was not affected very much. As a
result, the temperature at which cellulose dissolved completely decreased with the
H2 SO4 concentration.
To further pursue the effect of the H2 SO4 concentration, the dissolution temper-
ature, Tdissolution , was defined as the temperature at which the relative brightness
reached 0.5 in Fig. 5.11a. Dependence of Tdissolution on the added H2 SO4 con-
centration is shown in Fig. 5.11b. Addition of 5 mol dm3 of H2 SO4 decreased
the dissolution temperature by approximately 10 ı C, but the decrease became
less pronounced upon further addition of H2 SO4 . By fitting the data with an
exponentially decaying function, we found that Tdissolution reaches a limiting value of
271 ı C at high H2 SO4 concentrations, and cannot be lowered further no matter how
high the H2 SO4 concentration is made above 40 mol dm3 .
5.4.1.6 Possible Roles of Water on Crystalline-to-Amorphous

Transformation of Cellulose
Our polarized microscopic observation (Fig. 5.5) was the first direct evidence to
show that cellulose underwent crystalline-to-amorphous transformation in hot and
compressed water, and revealed the underlying process that was responsible for the
previous observations. Transformation of crystalline cellulose to an amorphous form
in water at high temperatures and high pressures is not simple thermal melting,
but water plays an important role because no such transformation was observed in
ethanol.
It is suggested that the transformation is rather elicited by changes in the solid
properties of cellulose [43], such as glass transition [56, 57], transformation of
cellulose crystals [33], and a drastic change in the hydrogen-bond structure [58, 59],
all of which occur above 200 ı C. Such changes affect cellulose–water interactions,
leading eventually to the transformation to an amorphous state. Interestingly
enough, the limiting temperature in Fig. 5.11b (271 ı C), below which the transition
temperature cannot be lowered any further with H2 SO4 , agrees well with the
temperature at which the steep increase of the relative brightness starts regardless
of the concentration of H2 SO4 (Fig. 5.11a). This may suggest that such changes in
the solid properties occur at the limiting temperature. The crystalline structure of
cellulose is so robust that it is not affected by H2 SO4 at low temperatures. However,
once cellulose is transformed to an amorphous state, the cellulose chains become
easily accessible and their hydrolysis is catalyzed by H2 SO4 .
Elucidating water–cellulose interactions under the present experimental
conditions is not straightforward, because the properties of water at high
temperatures and pressures are remarkably different from those at ambient
conditions. On the one hand, the dielectric constant of water, which is 78 at
25 ı C and 0.1 MPa, decreases to 21 at 300 ı C and 25 MPa [60], the value of
which is comparable to that of 1-propanol. The difference is ascribed to a large
change in the extent of hydrogen bonding formation of the water molecules [61].
It seems unlikely that such a nonpolar solvent interacts favorably with cellulose.
On the other hand, water between the crystallites might be in a supercritical state,
even though the transformation takes place well below the critical temperature of
water. Recent computer simulation revealed that critical parameters of water are
decreased significantly when it is confined by walls that interact strongly with water
[62]. Considering hydroxyl groups on the surfaces of the cellulose crystallites, it
seems that this is exactly the case for the water molecules between the cellulose
crystallites. Unique solvation properties of supercritical fluids such as formation of
a dense solvation shell may play an important role in the transformation [2].
5.4.2 Formation of Oil Nano-Droplets in Subcritical Water
Advantages of SCW as a unique medium to synthesize metal oxide nanoparticles

have been demonstrated [63]. When it comes to organic compounds such as
cellulose, however, the use of SCW has mostly been limited to their hydrolysis or
decomposition [3]. We recently demonstrated that the unique solvent properties of
SCW can also be used for fabricating nano-sized oil droplets in water (nanoemul-
sions [64–66]), and direct observation was used to demonstrate the formation
process of the droplets [67].
Emulsions are widely used in such industries as food, pharmaceutical, cosmetic,
chemical, agricultural, print/ink, and petroleum [68]. Nanoemulsions are those that
contain droplets in the size range between 20 and 200 nm [64–66]. Owing to
the small droplet size, nanoemulsions are transparent or translucent. Creaming or
sedimentation is significantly slower in nanoemulsions due to enhanced Brownian
motion of the small droplets. The small droplet size also has potential benefits
for developing a new realm of applications such as pharmaceutical and cos-
metic formulations [65, 69–71] as well as reactors for synthesizing nanomaterials
[65, 70].
Nanoemulsions are usually prepared by top-down processes, in which external
forces are applied to water/oil/surfactant mixtures to deform and disrupt large
droplets into smaller ones, but there are difficulties associated with down-sizing
liquid droplets to nanometer range [64, 66, 68, 72]. Nano-sized solid particles are
usually prepared by bottom-up processes [73]. Unlike the top-down process, the
bottom-up process starts with homogeneous solutions and solutes are allowed to
assemble to form nanoparticles. It is plain to see that the bottom-up approach is also
preferable for fabricating nano-sized liquid droplets, but homogeneous solutions of
oil and water need to be prepared. We successfully addressed the issue by using
SCW that freely mixes with various oils [74–76].
When coarse droplets of dodecane in water were heated in the reactor under a
constant pressure of 25 MPa, they started to shrink above approximately 330 ı C
(Fig. 5.12a–c). The system eventually entered into a one-phase regime at around
337 ı C and the droplets disappeared completely to give a homogeneous solution
(Fig. 5.12d). The phase separation occurred when the solution was cooled, and
the dodecane droplets having a fairly uniform size reappeared (Fig. 5.12e), which
increased in size as it was cooled further (Fig. 5.12f).
Fig. 5.12 (a–d) In situ optical microscopy images of droplets of dodecane in water taken at a
constant pressure of 25 MPa. Images were taken while heating the mixture at 11.6 ı C/min. (e and f)
Formation of dodecane droplets upon cooling from (d) at 4.5 ı C/min. Scale bars represent 50 m
(Reproduced with permission from Deguchi and Ifuku [67])
In our process called MAGIQ (Monodisperse nAnodroplet Generation In

Quenched hydrothermal solution), a fast quench, 200 ı C/s, was applied to the
homogeneous solutions, such that phase separation occurred rapidly to generate
nano-sized droplets. Emulsions containing nano-sized and monodisperse oil
droplets with an average diameter of 61 nm were obtained in just 10 s [67].
5.5 Summary
“Seeing is Believing” in any branches of sciences. We hope this review has

convinced the readers that in situ microscopic observation is a powerful tool
for studying chemical processes in sub- and supercritical water. The optical
reactors were used successfully to visualize other chemical processes in sub-
and supercritical water, including dissolution of silica [77], swelling and thermal
decomposition of polystyrene [77], and structural change of fungal cells [26].
They were also combined with Raman spectroscopy to study fluid inclusions [78,
79], hydrolysis of polycarbonate in water [80], solubility of 2,4-dichlorotoluene in
water [81], crystallization of spodumene [82], diffusion of carbon dioxide in water
[83], disproportionation and thermochemical sulfate reduction reactions [84], and
phase behavior in aqueous MgSO4 solutions [85], all at high temperatures and high
pressures. The reactors were also used to study physicochemical processes in sub-
and SCW such as Brownian motion of nanoparticles [27].
The technique can be used most effectively in prototyping reaction conditions.
Unlike reactions at ambient conditions, there are significantly wider spaces for
optimization in terms of temperature and pressure for reactions in sub- and
supercritical water. Direct visualization allows scanning wide ranges of temperature
and pressure rapidly in the initial stage of prototyping reaction conditions. Indeed,
this is where we have been approached most frequently for collaboration.
References
1. Andrews T. On the continuity of the gaseous and liquid states of matter. Philos Trans R Soc
Lond. 1869;159:575–90.
2. Shaw RW, Brill TB, Clifford AA, Eckert CA, Franck EU. Supercritical water: a medium for
chemistry. Chem Eng News. 1991;69:26–39.
3. Savage PE. Organic chemical reactions in supercritical water. Chem Rev. 1999;99:603–21.
4. Lide DR, editor. CRC handbook of chemistry and physics. 76th ed. Boca Raton: CRC Press;
1994.
5. Oger PM, Daniel I, Picard A. Development of a low-pressure diamond anvil cell and analytical
tools to monitor microbial activities in situ under controlled P and T. BBA-Proteins Proteomics.
2006;1764(3):434–42.
6. Shen AH, Bassett WA, Chou I-M. Hydrothermal studies in a diamond anvil cell: pressure
determination using the equation of state of H2 O. In: Syono Y, Manghnani MH, editors. High-
pressure research: application to earth and planetary sciences, Geophysical monograph series,
vol. 67. Washington, DC: AGU; 1992. p. 61–8.
7. Chou I-M. Pressure calibrants in the hydrothermal diamond-anvil cell. Int Geol Rev.
2007;49(4):289–300.
8. Hartmann M, Kreuss M, Sommer K. High pressure microscopy – a powerful tool for
monitoring cells and macromolecules under high hydrostatic pressure. Cell Mol Biol.
2004;50(4):479–84.
9. Hartmann M, Pfeifer F, Dornheim G, Sommer K. HPDS-Hochdruckzelle zur Beobachtung
mikroskopischer Phänomene unter Hochdruck. Chem Ing Tech. 2003;75(11):1763–7.
10. Frey B, Hartmann M, Herrmann M, Meyer-Pittroff R, Sommer K, Bluemelhuber G.
Microscopy under pressure – an optical chamber system for fluorescence microscopic analysis
of living cells under high hydrostatic pressure. Microsc Res Tech. 2006;69(2):65–72.
11. Nishiyama M, Kimura Y, Nishiyama Y, Terazima M. Pressure-induced changes in the structure
and function of the kinesin-microtubule complex. Biophys J. 2009;96(3):1142–50.
12. Costantino M, Rice SF. Supercritical phase separation in H2 O-N2 mixtures. J Phys Chem.
1991;95(23):9034–6.
13. Bassett WA, Shen AH, Bucknum M, Chou I-M. A new diamond anvil cell for hydrothermal
studies to 2.5 GPa and from –190 to 1200 ı C. Rev Sci Instrum. 1993;64(8):2340–5.
14. Schmidt C, Chou I-M. The hydrothermal diamond anvil cell (HDAC) for Raman spectroscopic
studies of geological fluids at high pressures and temperatures. In: Dubessy J, Caumon M-C,
Rull F, editors. Applications of Raman spectroscopy to earth sciences and cultural heritage,
EMU notes in mineralogy, vol. 12. Twickenham: EMU, European Mineralogical Union; 2012.
p. 248–78.
15. Fang Z, Smith Jr RL, Inomata H, Arai K. Phase behavior and reaction of polyethylene
terephthalate–water systems at pressures up to 173 MPa and temperatures up to 490 ı C.
J Supercrit Fluids. 1999;15(3):229–43.
16. Fang Z, Smith Jr RL, Inomata H, Arai K. Phase behavior and reaction of polyethylene in
supercritical water at pressures up to 2.6 GPa and temperatures up to 670 ı C. J Supercrit
Fluids. 2000;16(3):207–16.
17. Smith Jr RL, Fang Z, Inomata H, Arai K. Phase behavior and reaction of nylon 6/6 in water at
high temperatures and pressures. J Appl Polym Sci. 2000;76(7):1062–73.
18. Schilling W, Franck EU. Combustion and diffusion flames at high-pressures to 2000 bar.
Ber Bunsenges Phys Chem. 1988;92(5):631–6.
19. Kohl W, Lindner HA, Franck EU. Raman spectra of water to 400 ı C and 3000 bar.
Ber Bunsenges Phys Chem. 1991;95(12):1586–93.
20. Pfund DM, Darab JG, Fulton JL, Ma YJ. An XAFS study of strontium ions and krypton in
supercritical water. J Phys Chem. 1994;98(50):13102–7.
21. Armellini FJ, Tester JW, Hong GT. Precipitation of sodium chloride and sodium sulfate in
water from sub- to supercritical conditions: 150 to 550 ı C, 100 to 300 bar. J Supercrit Fluids.
1994;7(3):147–58.
22. Ikushima Y, Hatakeda K, Saito N, Arai M. An in situ Raman spectroscopy study of subcritical
and supercritical water: the peculiarity of hydrogen bonding near the critical point. J Chem
Phys. 1998;108(14):5855–60.
23. Morita T, Kusano K, Ochiai H, Saitow K, Nishikawa K. Study of inhomogeneity of supercriti-
cal water by small-angle x-ray scattering. J Chem Phys. 2000;112(9):4203–11.
24. Brill TB. Geothermal vents and chemical processing: the infrared spectroscopy of hydrother-
mal reactions. J Phys Chem A. 2000;104(19):4343–51.
25. Alargova RG, Deguchi S, Tsujii K. Dynamic light scattering study of polystyrene latex sus-
pended in water at high temperatures and high pressures. Colloids Surf A. 2001;183:303–12.
26. Deguchi S, Tsujii K. Flow cell for in situ optical microscopy in water at high temperatures and
pressures up to supercritical state. Rev Sci Instrum. 2002;73(11):3938–41.
27. Mukai S, Deguchi S, Tsujii K. A high-temperature and -pressure microscope cell to observe
colloidal behaviors in subcritical and supercritical water: Brownian motion of colloids near a
wall. Colloids Surf A. 2006;282–283:483–8.
28. Atwell WA, Hood LF, Lineback DR, Varriano-Marston E, Zobel HF. The terminol-
ogy and methodology associated with basic starch phenomena. Cereal Foods World.
1988;33(3):306–11.
29. Bigelis R. Carbohydrases. In: Nagodawithana T, Reed G, editors. Enzymes in food processing.
3rd ed. San Diego: Academic; 1993. p. 123–5.
30. Zhang Y-HP, Himmel ME, Mielenz JR. Outlook for cellulase improvement: screening and
selection strategies. Biotechnol Adv. 2006;24(5):452–81.
31. Nishiyama Y, Langan P, Chanzy H. Crystal structure and hydrogen-bonding system in
cellulose I“ from synchrotron x-ray and neutron fiber diffraction. J Am Chem Soc.
2002;124(31):9074–82.
32. Nishiyama Y, Sugiyama J, Chanzy H, Langan P. Crystal structure and hydrogen bonding
system in cellulose I’ from synchrotron x-ray and neutron fiber diffraction. J Am Chem Soc.
2003;125(47):14300–6.
33. Horii F, Yamamoto H, Kitamaru R, Tanahashi M, Higuchi T. Transformation of native
cellulose crystals induced by saturated steam at high temperatures. Macromolecules.
1987;20(11):2946–9.
34. Yamamoto H, Horii F, Odani H. Structural changes of native cellulose crystals induced by
annealing in aqueous alkaline and acidic solutions at high temperatures. Macromolecules.
1989;22(10):4130–2.
35. Adschiri T, Hirose S, Malaluan R, Arai K. Noncatalytic conversion of cellulose in supercritical
and subcritical water. J Chem Eng Jpn. 1993;26(6):676–80.
36. Sasaki M, Kabyemela B, Malaluan R, Hirose S, Takeda N, Adschiri T, Arai K. Cellulose
hydrolysis in subcritical and supercritical water. J Supercrit Fluids. 1998;13(1–3):261–8.
37. Sasaki M, Adschiri T, Arai K. Kinetics of cellulose conversion at 25 MPa in sub- and
supercritical water. AIChE J. 2004;50(1):192–202.
38. Sasaki M, Fang Z, Fukushima Y, Adschiri T, Arai K. Dissolution and hydrolysis of cellulose
in subcritical and supercritical water. Ind Eng Chem Res. 2000;39(8):2883–90.
39. Deguchi S, Tsujii K, Horikoshi K. Cooking cellulose in hot and compressed water. Chem Com-
mun. 2006;31:3293–5.
40. Klemm D, Heublein B, Fink H-P, Bohn A. Cellulose: fascinating biopolymer and sustainable
raw material. Angew Chem Int Ed. 2005;44(22):3358–93.
41. Zeronian SH. Intercrystalline swelling of cellulose. In: Nevell TP, Zeronian SH, editors.
Cellulose chemistry and its applications. Chichester: Ellis Horwood; 1985. p. 139–58.
42. Langan P, Nishiyama Y, Chanzy H. X-ray structure of mercerized cellulose II at 1 Å resolution.
Biomacromolecules. 2001;2(2):410–6.
43. Deguchi S, Tsujii K, Horikoshi K. Crystalline-to-amorphous transformation of cellulose in
hot and compressed water and its implications for hydrothermal conversion. Green Chem.
2008;10(2):191–6.
44. Deguchi S, Tsudome M, Shen Y, Konishi S, Tsujii K, Ito S, Horikoshi K. Preparation and
characterisation of nanofibrous cellulose plate as a new solid support for microbial culture.
Soft Matter. 2007;3(9):1170–5.
45. Kuga S. The porous structure of cellulose gel regenerated from calcium thiocyanate solution.
J Colloid Interface Sci. 1980;77(2):413–7.
46. Jin H, Nishiyama Y, Wada M, Kuga S. Nanofibrillar cellulose aerogels. Colloids Surf A.
2004;240(1–3):63–7.
47. Ardizzone S, Dioguardi FS, Mussini T, Mussini PR, Rondinini S, Vercelli B, Vertova A.
Microcrystalline cellulose powders: structure, surface features and water sorption capability.
Cellulose. 1999;6(1):57–69.
48. Mok WS-L, Antal Jr MJ, Varhegyi G. Productive and parasitic pathways in dilute acid-
catalyzed hydrolysis of cellulose. Ind Eng Chem Res. 1992;31(1):94–100.
49. Bobleter O. Hydrothermal degradation of polymers derived from plants. Prog Polym Sci.
1994;19(5):797–841.
50. Minowa T, Fang Z, Ogi T. Cellulose decomposition in hot-compressed water with alkali or
nickel catalyst. J Supercrit Fluids. 1998;13(1–3):253–9.
51. Fukuoka A, Dhepe PL. Catalytic conversion of cellulose into sugar alcohols. Angew Chem Int
Ed. 2006;45(31):5161–3.
52. Harris EE. Wood saccharification. Adv Carbohydr Chem. 1949;4:153–88.
53. Girisuta B, Janssen LPBM, Heeres HJ. Kinetic study on the acid-catalyzed hydrolysis of
cellulose to levulinic acid. Ind Eng Chem Res. 2007;46(6):1696–708.
54. Zhao H, Kwak JH, Wang Y, Franz JA, White JM, Holladay JE. Effects of crystallinity on dilute
acid hydrolysis of cellulose by cellulose ball-milling study. Energy Fuels. 2006;20(2):807–11.
55. Deguchi S, Tsujii K, Horikoshi K. Effect of acid catalyst on structural transformation of
cellulose in hydrothermal conditions. Green Chem. 2008;10(6):623–6.
56. Kamide K, Saito M. Thermal analysis of cellulose acetate solids with total degrees of
substitution of 0.49, 1.75, 2.46, and 2.92. Polym J. 1985;17(8):919–28.
57. Nishio Y, Roy SK, Manley RSJ. Blends of cellulose with polyacrylonitrile prepared from N,
N-dimethylacetamide-lithium chloride solutions. Polymer. 1987;28(8):1385–90.
58. Watanabe A, Morita S, Ozaki Y. Study on temperature-dependent changes in hydrogen bonds
in cellulose I“ by infrared spectroscopy with perturbation-correlation moving-window two-
dimensional correlation spectroscopy. Biomacromolecules. 2006;7(11):3164–70.
59. Watanabe A, Morita S, Ozaki Y. Temperature-dependent changes in hydrogen bonds in
cellulose I studied by infrared spectroscopy in combination with perturbation-correlation
moving-window two-dimensional correlation spectroscopy: comparison with cellulose I“.
Biomacromolecules. 2007;8(9):2969–75.
60. Harvey AH, Peskin AP, Klein SA. NIST/ASME STEAM properties database. 2.21th ed.
National Institute of Standards and Technology. Boulder; 2004.
61. Matubayasi N, Wakai C, Nakahara M. NMR study of water structure in super- and subcritical
conditions. Phys Rev Lett. 1997;78(13):2573–6.
62. Brovchenko I, Geiger A, Oleinikova A. Water in nanopores. I. Coexistence curves from Gibbs
ensemble Monte Carlo simulations. J Chem Phys. 2004;120(4):1958–72.
63. Adschiri T, Hakuta Y, Arai K. Hydrothermal synthesis of metal oxide fine particles at
supercritical conditions. Ind Eng Chem Res. 2000;39(12):4901–7.
64. Tadros T, Izquierdo P, Esquena J, Solans C. Formation and stability of nano-emulsions.
Adv Colloid Interface Sci. 2004;108–109:303–18.
65. Solans C, Izquierdo P, Nolla J, Azemar N, Garcia-Celma MJ. Nano-emulsions. Curr Opin
Colloid Interface Sci. 2005;10(3–4):102–10.
66. Solans C, Solé I. Nano-emulsions: formation by low-energy methods. Curr Opin Colloid
Interface Sci. 2012;17(5):246–54.
67. Deguchi S, Ifuku N. Bottom-up formation of dodecane-in-water nanoemulsions from
hydrothermal homogeneous solutions. Angew Chem Int Ed. 2013;52(25):6409–12.
68. Leal-Calderon F, Schmitt V, Bibette J. Emulsion science: basic principles. 2nd ed. New York:
Springer; 2007.
69. Tamilvanan S, Schmidt S, Müller RH, Benita S. In vitro adsorption of plasma proteins onto the
surface (charges) modified-submicron emulsions for intravenous administration. Eur J Pharm
Biopharm. 2005;59(1):1–7.
70. Anton N, Benoit J-P, Saulnier P. Design and production of nanoparticles formulated from nano-
emulsion templates – a review. J Controll Release. 2008;128(3):185–99.
71. Sonneville-Aubrun O, Simonnet J-T, L’Alloret F. Nanoemulsions: a new vehicle for skincare
products. Adv Colloid Interface Sci. 2004;108–109:145–9.
72. Henry JVL, Fryer PJ, Frith WJ, Norton IT. Emulsification mechanism and storage instabilities
of hydrocarbon-in-water sub-micron emulsions stabilised with Tweens (20 and 80), Brij 96v
and sucrose monoesters. J Colloid Interface Sci. 2009;338(1):201–6.
73. Schmid G. Nanoparticles: from theory to application. Weinheim: Wiley-VCH; 2004.
74. Brunner E. Fluid mixtures at high pressures IX. Phase separation and critical phenomena in 23
(n-alkane C water) mixtures. J Chem Thermodyn. 1990;22(4):335–53.
75. Stevenson RL, LaBracio DS, Beaton TA, Thies MC. Fluid phase equilibria and critical
phenomena for the dodecane-water and squalane-water systems at elevated temperatures and
pressures. Fluid Phase Equilib. 1994;93(11):317–36.
76. Weingärtner H, Franck EU. Supercritical water as a solvent. Angew Chem Int Ed.
2005;44(18):2672–92.
77. Deguchi S, Ghosh SK, Alargova RG, Tsujii K. Viscosity measurements of water
at high temperatures and pressures using dynamic light scattering. J Phys Chem B.
2006;110(37):18358–62.
78. Chou I-M, Song Y, Burruss RC. A new method for synthesizing fluid inclusions in fused
silica capillaries containing organic and inorganic material. Geochim Cosmochim Acta.
2008;72(21):5217–31.
79. Wang X, Hu W, Chou I-M. Raman spectroscopic characterization on the OH stretching bands
in NaCl–Na2 CO3 –Na2 SO4 –CO2 –H2 O systems: implications for the measurement of chloride
concentrations in fluid inclusions. J Geochem Explor. 2013;132:111–9.
80. Pan Z, Chou I-M, Burruss RC. Hydrolysis of polycarbonate in sub-critical water in fused silica
capillary reactor with in situ Raman spectroscopy. Green Chem. 2009;11(8):1105–7.
81. Pan Z, Ma Y, Chou I-M. Solubility of 2,4-dichlorotoluene in water determined in fused silica
capillary reactor by in situ Raman spectroscopy. AIChE J. 2013;59(8):2721–5.
82. Li J, Chou I-M, Yuan S, Burruss RC. Observations on the crystallization of spodumene from
aqueous solutions in a hydrothermal diamond-anvil cell. Geofluids. 2013;13(4):467–74.
83. Lu W, Guo H, Chou IM, Burruss RC, Li L. Determination of diffusion coefficients of carbon
dioxide in water between 268 and 473 K in a high-pressure capillary optical cell with in situ
Raman spectroscopic measurements. Geochim Cosmochim Acta. 2013;115:183–204.
84. Yuan S, Chou I-M, Burruss RC. Disproportionation and thermochemical sulfate reduction
reactions in S–H2 O–CH4 and S–D2 O–CH4 systems from 200 to 340 ı C at elevated pressures.
Geochim Cosmochim Acta. 2013;118:263–75.
85. Wang X, Chou I-M, Hu W, Burruss RC. In situ observations of liquid–liquid phase separation
in aqueous MgSO4 solutions: geological and geochemical implications. Geochim Cosmochim
Acta. 2013;105:1–10.
Chapter 6
Fused Silica Capillary Reactor
and Its Applications
I-Ming Chou and Zhiyan Pan
Abstract The “fused silica capillary reactor” (FSCR) was constructed from fused
silica capillary (normally 0.665 mm OD, 0.3 mm ID, and about 25 mm long) with
both ends sealed by fusion in an oxyhydrogen flame after the sample in solid, liquid,
or gaseous form (or their mixtures) was loaded. The fused silica is stable in aqueous
solutions with low pH and the FSCR can hold 100 MPa internal pressure or slightly
higher. Two types of customer-designed heating-cooling stages were used to control
sample temperatures between 196 and 600 ı C and also the heating and cooling
rates (0.1 ı C/min to 50 ı C/min) of the sample. The volume of the FSCR expands
elastically up to 75 % when heated to about 400 ı C from room temperature. The
FSCR is highly permeable to hydrogen and provides the potential for developing
experimental methods for redox control in reactions at room temperature or higher.
Because the fused silica is stable in aqueous solutions with low pH, and inert to
sulfur, FSCR is an ideal container for experiments of many chemical systems at the
pressure-temperature conditions near the critical point of water. The transparency of
the fused silica allows in situ optical observations of the sample under a microscope,
spectroscopic analyses (e.g., Raman spectroscopy), and continuous recording of the
sample during heating or cooling for later reviews and kinetic studies. Examples are
given for the applications of FSCR in the studies of material in water at conditions
near the critical point of water.
I-M. Chou ()

Laboratory for Experimental Study Under Deep-Sea Extreme Conditions, Sanya Institute
of Deep-Sea Science and Engineering, Chinese Academy of Sciences, Sanya 572000,
People’s Republic of China
Z. Pan
Department of Environmental Engineering, Zhejiang University of Technology, Hangzhou
310032, People’s Republic of China
158 I-M. Chou and Z. Pan
Keywords Fused silica capillary reactor • Supercritical water • Heating-cooling

stage • Chemical reaction • Optical cell • Raman spectroscopy
6.1 Introduction
There are several types of optical cells available for the study of material in
supercritical water, including those with diamond windows [1–3] for pressures (P)
up to several GPa and temperatures (T) from 190 to about 1,000 ı C, and those
with fused silica windows [4–6] for pressures up to 100 MPa and temperatures from
196 to 600 ı C. The advantages of these optical cells include: (1) wide applicable
P-T conditions; (2) the transparency of the windows allows visual observations of
the samples in the cell under a microscope and in situ sample analyses using spectro-
scopic methods, including synchrotron X-ray and Raman spectroscopy; and (3) the
images of the samples can be recorded continuously for reviews and kinetic studies.
In this Chapter, we will describe one type of optical cell with fused silica
windows, the fused silica capillary reactor (FSCR), which was modified from the
fused silica capillary capsule (FSCC) described by Chou et al. [5] for geological
applications; the modifications enhanced the applications in biorefineries [7–9].
6.2 The Construction of the Fused Silica Capillary Reactor

(FSCR)
The procedures for the construction of FSCR are similar to those for the FSCC [5].
The FSCC was constructed from fused silica capillaries purchased from Polymicro
Technologies, LLC (http://polymicro.com), and it had either a square or round cross-
section.
The available dimensions range from 0.15 to 0.80 mm in OD and 0.002 to
0.70 mm in ID. However, for the construction of the FSCR for applications in
biorefineries, fused silica capillaries with larger OD and ID were normally used. For
examples, the tubes used by Liu and Pan [8] and Chen et al. [10] were 0.665 mm
OD and 0.3 mm ID, and even those tubes with 4 mm OD and 2 mm ID were used
[8] to enhance quantitative analyses. Fused silica capillaries may also be purchased
from other suppliers, such as MicroQuartz (http://www.microquartz.de/com), but
we have no experience on their products.
The capillary tubing is coated externally with a protective polyimide layer and
supplied in coils with any specified length, normally several tens of meters. A
section of about 2–6 cm long of the fused silica capillary tube was cut from the coil
and the polyimide coating was removed by burning in an oxyhydrogen flame before
sealing one end of the tube in an oxyhydrogen flame, such that the transparency of
the fused silica allows optical observations under a microscope and also in situ spec-
troscopic analyses. This tube with one end still open is ready for sample loading.
6 Fused Silica Capillary Reactor and Its Applications 159
6.3 Sample Loading in the FSCR
The procedures for loading samples, in solid, liquid, or gaseous forms (or their
mixtures) are similar to those for the FSCC [5]. However, because silica tubes with
larger ID were used in the FSCR, the amounts of solid or liquid samples loaded
can be quantified either by mass or volume measurements. For example, in the
experiments for the determination of chlorobenzene solubilities in subcritical water
in FSCR, Pan and Dong [11] injected liquid chlorobenzene into a capillary tube
(0.665 mm OD, 0.3 mm ID, and 2 cm long with one end sealed) using a specially
designed syringe, which had a delivery tip made from a fused silica capillary
with 0.2 mm OD and 0.15 mm ID. The volume of the injected chlorobenzene
was measured with a micrometer under a microscope (accurate to ˙1 m, Leica,
DM2500P, Germany; the mass uncertainty was ˙8 105 mg). Similarly, the
volume of the subsequently injected water was measured at room temperature after
it was centrifuged to the enclosed end, having an uncertainty of ˙7 105 mg in
mass. The space above the liquid phases was partially filled with a silica rod to
reduce evaporation of chlorobenzene and water in the free space. The enclosed end
of the capillary was then immersed into liquid nitrogen and the open end of the tube
was sealed with an oxyhydrogen flame to form a FSCR (Fig. 6.1). To load a gas or a
mixture of gases in the FSCR with or without previously loaded solid and/or liquid
phases, a gas loading system (Fig. 6.2) was used.
Fig. 6.1 An image of a sample in a fused silica capillary reactor, showing chlorobenzene, water,
vapor, and silica rod in the FSCR
Fig. 6.2 A schematic diagram of the sample loading system. V-1 to V-8 are three-way/two-stem
combination taper-seal valves from High Pressure Equipment Co. (Cat. No. 15-15AF1) (Taken
from Fig. 1 of Chou et al. [5])
As shown in Fig. 6.2, the open end of the tube was connected to a pressure
line after centrifugation and air in the sample tube was evacuated before gaseous
samples were loaded cryogenically by immersing the sealed end of the tube in
liquid nitrogen and switching the pressure line from vacuum to sample gas at about
0.2 MPa pressure. The gaseous component was allowed to condense or freeze to a
solid for several minutes before the tube was evacuated and the open end was sealed
by fusion in an oxyhydrogen flame, while the sample at other end remained frozen
and the pressure line was under vacuum. Normally, FSCC or FSCR of less than
25 mm in length were prepared (Fig. 6.1) to minimize temperature gradient of the
sample while being heated or cooled in a heating–cooling stage.
To load solid phases into a fused silica capillary with both ends open, fine sample
powders were pushed into the tube from one open end with a thin wire. The inner
wall was cleaned with a wet tissue and this open end was then sealed with an
oxyhydrogen flame. The liquid or gas samples can then be loaded from the other
open end following the procedures described above. However, it was much easier
by loading the solid powder into a short capillary tube with a small OD, such that
it can then be inserted into the sample capillary tube without getting its inner wall
contaminated by the solids to make the flame-seal process much easier.
6.4 The Physical and Chemical Properties of the FSCR
The fused silica is stable in aqueous solutions with low pH and the sealed FSCR can
hold 100 MPa internal pressure or even higher [5]. Chou et al. [5] performed a series
of measurements between 22 and 405 ı C on the changes of densities of methane
sample in FSCC based on the Raman shift of the C–H symmetric stretching band
of methane [12]. Their results indicate that the sample chamber of the FSCC was
non-isochoric during heating or cooling with the volume changes up to 50 or 75 %
in the tested temperature range, depending on the type of tubing (round or square
cross section), wall thickness, and the original internal methane pressure at room
temperature (see their Fig. 16). Also these deformations are elastic, as indicated
by the similar results obtained from both heating and cooling cycles and also the
similar results obtained in the first and second heating cycles (see their Table 1 and
Fig. 16).
Shang et al. [13] determined the diffusion coefficients of hydrogen in fused silica
between 23 and 250 ı C by Raman spectroscopy. Their results show that the high
permeability of the fused silica to hydrogen promotes oxidation reactions in the
FSCC due to rapid loss of hydrogen, and also provides the potential for developing
experimental methods for redox control in reactions at temperatures below 400 ı C
in the FSCC, especially for experiments containing organic material.
The fused silica is inert to sulfur, and the fused silica tubes have been used as
containers for experiments containing S [14–18]. FSCC has been used to monitor
cracking of hydrocarbons and the reactions between hydrocarbons and water at
temperatures up to 500 ı C for a period ranging from few hours to few days [5]. It is
also ideal for studying the characteristics of in situ Raman signals during heating of
the samples in the capsules [5, 7, 18].
6.5 Heating-Cooling Stages (USGS, INSTEC, Linkam)
When the FSCC and FSCR techniques were developed, the only heating-cooling
stage available for effective temperature control of the sample capsules of about
25 mm in length, with acceptable temperature gradient, was the USGS gas-
flow stage [19]. Later, two types of custom-designed heating-cooling stages were
produced for the HPOC [5], FSCC, and FSCR experiments; one is from INSTEC
and the other from Linkam. Because the USGS gas-flow stage has been described
previously [19] and also it is rather difficult to use for FSCR applications, it will not
be discussed further here.
The INSTEC heating-cooling stage (INS0908051, INSTEC Inc., Boulder, CO,
USA) under a microscope of a JY LabRam 800 Raman system is shown in Fig. 6.3a,
and the screen shows the two co-existing fluids in a FSCR.
An mk1000 temperature controller and SN2-SYS liquid nitrogen cooling system
were used to control sample temperatures in the HPOC, FSCC, and FSCR. The
sample holder of this stage has a sample slot (1 mm wide, 2.5 mm deep, and 40 mm
long) located in the middle of a silver plate (43 mm 16 mm 2.5 mm), and a K-
type thermocouple was inserted along the slot to monitor the sample temperatures
(Fig. 6.3b). A thin plate of silica glass was mounted at the bottom part of the slot
and an FSCC or FSCR was placed in the slot and on top of the glass plate. The
sample holder, together with sample and thermocouple, were then inserted into the
sample chamber of the stage, which has a 60 mm 40 mm window (Fig. 6.3c).
The applicable temperature range for the INSTEC stage is from 196 to 600 ı C.
To investigate possible temperature gradients in the sample slot, target temperatures
were set at 15, 30, 100, 200, 300 and 400 ı C, and the temperatures at five positions
evenly distributed along the sample slot were measured with a K-type thermocouple
(Fig. 6.3b). The thermocouple was calibrated with the freezing and boiling points
of pure water at ambient pressure at 0 and 100 ı C, respectively. At the target
temperatures of 15 and 30 ı C, the temperature differences among different positions
are 0.5 ı C, which increased to 2 ı C at the target temperature of 200 ı C, and
2.7 ı C at the target temperature of 400 ı C.
The proto-type Linkam CAP 500 Capillary Pressure Stage (Fig. 6.4a) was
designed for the HPOC, FSCC, and FSCR applications and will soon be available on
the market after minor modifications. Sample temperatures were controlled using a
T95-PE system controller (http://www.linkam.co.uk/t95-system-controllers/) with a
T95 LinkPad and LNP95 cooling system (http://www.linkam.co.uk/lnp95/) through
Linksys32 temperature control and video capture software. The stage body can be
water cooled by using an ECP (external circumferential piston) pump when heating
Fig. 6.3 (a) Photograph of an INSTEC heating-cooling stage and HORIBA JY HR 800 LabRam
Raman system. The insert shows the stage under a microscope, and the screen shows the image
of a meniscus separating two coexisting fluid phases in a FSCC (0.3 mm OD and 0.1 mm ID)
(Modified from Fig. 4 of Chou [6]). (b) A sample holder in the INSTEC heating-cooling stage
with a slot (40 mm 1 2.5 mm) at the center of a silver plate (43 mm 16 mm 2.5 mm). A
FSCC up to 3 mm long can be placed in this slot. The enclosed end of fused silica capillary of a
HPOC can also be inserted into the slot together with a type-K thermocouple for direct sample-
temperature measurements (Taken from Fig. 2b of Wang et al. [20]). (c) INSTEC heating-cooling
stage with a chamber area of 60 mm 40 mm to show the whole silver plate of the sample holder
(Taken from Fig. 2c of Wang et al. [20])
Fig. 6.4 (a) Photograph of a Linkam CAP 500 heating-cooling stage system, showing the
temperature controller in the lower right corner, a dewar, and the stage (under the microscope)
(Taken from Fig. 8 of Chou [6]). (b) Photograph of the sample chamber of the Linkam CAP 500
heating-cooling stage. A fused silica capillary of an HPOC was inserted into the slot of a silver
block under the microscope with the sample being positioned above a small hole at the center of
the silver block (Taken from Fig. 9 of Chou [6])
samples above 300 ı C for more than 30 min. A capillary tube of HPOC being
inserted into a channel (1 mm wide and 0.6 mm deep) of a silver block (20 mm
wide and 50 mm long), is shown in Fig. 6.4b; the capillary can be moved 12.5 mm
from the center for a total 25 mm of movement. A small hole was drilled through
the center of the silver block and its cover, so that samples can be viewed with
transmitted light. The silver block can be heated and cooled rapidly in the range
196–500 ı C at a rate from 0.1 to 50 ı C/min. The temperature sensor mounted in
the silver block is a high-accuracy 100 platinum resistor and is accurate to 0.1 ı C.
The block has been designed to minimize the temperature gradient along its length;
within the sample movement range of 25 mm, the maximum temperature differences
are 0.2, 0.7, and 2.0 ı C at the set temperatures of 100, 300, and 500 ı C, respectively.
6.6 Applications of FSCR (Examples)
The geological applications of FSCC have been described previously [5, 6]. Here,
few examples for the applications of FSCR in the studies of reactions at pressure-
temperature conditions near the critical point of water are given.
6.6.1 Supercritical Water Oxidation (SCWO) of Chlorinated

Organic Compounds
Chlorinated organic compounds are one of the largest groups of anthropogenic

materials with high stability and tend to accumulate in the environment. SCWO has
proved to be one of the most effective and environmentally friendly methods for the
destruction of the hazardous organic wastes. The SCWO reaction typically proceeds
in a single fluid-phase consisting of organic compounds, an oxidizer, and water.
Supercritical water (Tc D 374 ı C, Pc D 22.1 MPa) has a weaker hydrogen-bond
network and lower dielectric constant than those of ambient water. This reaction
system, which has no interphase mass transfer limitations, leads to a fast reaction
rate and high oxidation-efficiency. The SCWO process for wastewater treatment has
been extensively studied by many researchers. However, almost all of the previous
studies were performed in large-scale stainless steel reactors, which were readily
corroded at high temperatures in supercritical water, particularly in the presence of
chloride ions.
To avoid corrosion, Liu and Pan [8] studied SCWO of Chlorobenzene (CB)
in FSCR combined with a polarization microscope recorder system and a Raman
spectroscopic system. In that study, CB was completely oxidized to CO2 and H2 O:
C6 H5 Cl C 14H2 O2 ! 6CO2 C 16H2 O C HCl (6.1)
The effects of operating parameters, including the stoichiometric amount of oxidizer

(H2 O2 ), temperature (350–450 ı C), and reaction time (120–600 s), on oxidation
behavior were investigated.
6.6.1.1 Experimental Procedure
Two FSCRs of different sizes were used; one with 665 m OD, 300 m ID, and
2 cm long for observations of phase changes and the other with 4 mm OD, 2 mm
Fig. 6.5 Photomicrographs of chlorobenzene in hydrogen peroxide in fused-silica capillary

reactor during heating process (a), and cooling process (b) (Modified from Fig. 2 of Liu and
Pan [8])
ID, and 5 cm long for quantitative analyses. CB and H2 O2 were injected into each
reactor in different proportions, depending on the reaction conditions. The closed
end of the capillary was then immersed in liquid nitrogen and the sealing procedure
was fast enough (within 5 s) to avoid the temperature increase of the sample in the
FSCR, thus preventing vapor loss.
The loaded smaller FSCR was inserted into the sample chamber of the INSTEC
heating–cooling stage. Images of the sample during the reaction were observed
under a polarization microscope (DM2500P, Leica, Wetzlar, Germany), and
recorded continuously in a computer through a digital camera (TK-C1481, JVC,
Yokohama, Japan). The phase changes are shown in Fig. 6.5.
The larger FSCR was placed in a hot-air oven (Fig. 6.6), which had been
preheated to the desired reaction temperature. After a specified reaction time, the
FSCR was removed from the hot-air oven and quenched rapidly in a cold water
bath, ensuring cessation of all reactions in the FSCR. After cooling, the gas-
phase product, CO2 , was determined using Raman spectroscopy. Subsequently,
unspent CB and liquid intermediate products were identified with GC, and GC-mass
spectrometry (GC-MS) after breaking of the FSCR.
Fig. 6.6 Schematic diagram

of the hot-air oven
experimental setup: 1
fused-silica capillary reactor
holder, 2 fused-silica
capillary reactor, 3 fluidized
bed (with about 6 cm long
isothermal region), 4 quartz
sand, 5 oven, 6 K-type
thermocouples, 7 digital
temperature indicators, 8
temperature controller for the
oven
6.6.1.2 Experimental Results and Quantitative Determination

of Gas-Product CO2
Quantitative analysis of CO2 is very important because SCWO of chlorinated

organic compounds proceeds through a series of steps that produce many products
of incomplete oxidation, such that the reactions do not proceed to CO2 in a single
step. In this study, the oxidation efficiency in different conditions was estimated by
monitoring the amount of CO2 produced, represented by CO2 yield, and the results
are shown in Figs. 6.7 and 6.8. The Raman spectrum of CO2 consists of an upper
band, a lower band, and two hot bands; the upper and lower bands are especially
pronounced. Note that the amount of CO2 dissolved in the aqueous phase under the
experimental conditions is less than 0.28 mol kg1 , and can be neglected.
When compared with the other methods, the advantages of the FSCR-Raman
spectroscopy-based method are visually accessible, low in energy and materials
consumptions, expeditiousness, absent of undesired catalytic effects of the reactor
wall, resistant to corrosion, and could be directly coupled with a Raman spectro-
scopic system to monitor the progress of chemical reaction in situ. Moreover, due
to the small size of the reactor, it minimizes the resistances in mass transfer and
heat transfer, such that the observed kinetics approaches to the intrinsic one, and the
kinetic results may be applied to optimize reactor design at throughputs in industrial
operations.
For future studies on SCWO of chlorinated organic compounds in FSCR, the
use of other liquid oxidizers, such as nitric acid and sodium nitrate solution,
will be considered in addition to hydrogen peroxide. Recently, SCWO has been
Fig. 6.7 Raman spectra of CO2 produced by oxidation of CB in supercritical water with 150 %
stoichiometric amounts of oxidizer at 450 ı C at different reaction times. The increase in CO2
signals indicates the progress of the oxidation process (Taken from Fig. 4 of Liu et al. [21])
Fig. 6.8 Raman peak areas of CO2 (scattered symbols) and CO2 yields (lines) vs reaction time
with 150 % stoichiometric amount of oxidizer at different temperatures (Modified from Fig. 5 of
Liu et al. [21])
successfully applied to guaiacol, a model compound of lignin, using the same

method described above, with H2 O2 as an oxidizer at temperatures between 180
and 300 ı C, and the progress of the reaction was monitored by the yield of CO2 .
6.6.2 Determination of Hydrophobic Organic Compounds

(HOC) Solubilities in Subcritical Water (SBCW)
Hydrophobic organic compounds (HOC) are common contaminants in soils and

sediments. HOC include aromatic compounds in petroleum and fuel residue,
chlorinated compounds in commercial solvents and chemicals no longer produced
in the United States, for example DDT [22]. Due to the extreme conditions of
elevated temperature and pressure used in SBCW processes, the chemical data of
such organic compounds at these conditions are scarce, even though these data
are important for the formulation of solubility models and evaluation of promising
technologies. Therefore, the research of the solubilities of hydrophobic organics in
SBCW had aroused great attention.
Many different methods are used to measure the HOC solubilities in SBCW,
such as continuous-flow methods, visual synthetic methods, analytical isothermal
methods, etc., and expressions like “static” or “dynamic” are commonly used in
connection with many different methods. However, in the static method, it is difficult
to keep the bulk composition of a sample unchanged while taking a small portion
of the sample out of the equilibrium cells for analyses. On the other hand, in the
dynamic method, elaborate analytical procedures are needed in most of cases to
ensure that system has reached equilibrium before sampling. In addition, these two
methods both require considerable devices for subsequent analyses.
Pan and Dong [11] determined the solubilities of chlorobenzene (CB) in SBCW
in the FSCR (665 m OD, 300 m ID, and 2 cm long). The FSCR sample was
heated to the preset temperature in a heating-cooling stage from 173.3 to 266.9 ı C,
and allowed to equilibrate for a period between 8 and 10 h in the presence of a
vapor phase. All the phase changes in FSCR were observed under a microscope
and recorded by using a digital camera and a computer. For example, the images
of a sample containing 58.0 mg of CB/g of H2 O, taken during heating from 30.1 to
222.4 ı C were shown in their Fig. 3. CB swelled at 150.0 ı C, and dissolved totally
at 222.4 ı C after gradual dissolving. The pressure in the FSCR was calculated by
the saturated vapor pressures of water at 222.4 ı C, which was about 2.2 MPa. The
results indicated that the solubility of CB increased from 43.5 to 71.4 mg/g in water
when temperature rose from 173.3 to 266.9 ı C (their Fig. 4), and the temperature
effect can be represented by the linear equation:
S D 0:3069T 10:188 (6.2)
where S is the solubility in mg g1 , and T is temperature in ı C.

This type of solubility measurements in the FSCR is very efficient. However, to

make sure the solubility data obtained are reliable, there is a need to demonstrate
that the entire aqueous solution is homogeneous in composition when the studied
material is totally dissolved. In the most recent research, Pan et al. [9] demonstrated
the homogeneities of dissolved benzene and 2,4-dichlorotoluene in water in the
FSCR using in-situ Raman spectroscopy. The Raman spectra of liquid benzene and
2,4-dichlorotoluene were collected from four different positions in the FSCR after
heating at 226.8 ı C for 28 h, and 220.0 ı C for 20 h, respectively, and their results
(their Figs. 2(b) and 2(d)) showed that the solute diffused uniformly and the system
reached phase equilibrium within the specified times.
The solubilities of benzene in water they measured were in good agreement
with previous data. Their Fig. 3 showed the dissolution of 2,4-dichlorotoluene in
HCW during heating process, and their results (their Fig. 4) indicated that the
solubility of 2,4-dichlorotoluene increased from 22.6 to 104.2 mg/(g H2 O) when
the temperature rose from 266.3 to 302.4 ı C, and can be represented by the linear
equation:
S D 2:137T 545:725 (6.3)
where S is the solubility in mg g1 , and T is temperature in ı C.
6.6.3 Depolymerization of Waste Polymers in Hot Compressed

Water (HCW)
Polymers generally do not decompose naturally and techniques need to be devel-

oped for the recycling of plastic material. It has been demonstrated that the
depolymerization of polymer in hot compressed water (HCW, subcritical and
supercritical water above 200 ı C and at sufficiently high pressure) systems was a
feasible and environment friendly method for the recycling of monomers. This was
further supported by the results of our recent studies described below.
6.6.3.1 Depolymerization of Bis-Phenol A Poly(Carbonate) (PC)
Pan et al. [7] studied the depolymerization of Bis-phenol A poly-carbonate in sub-

critical water by using a newly developed method for studying hydrolysis of PC in
sub-critical water. At 553 K and about 6.4 MPa, in situ Raman analysis of the vapor
phase showed the presence of carbon dioxide (Fig. 6.9), which was produced by the
hydrolysis of PC.
Also, temperature was found to be the key factor for the hydrolysis of PC,
as indicated by the fast reaction they observed at 573 K in different runs. The
hydrolysis yield can be estimated by the amount of CO2 being produced. Analyses
of quenched products by both GC and GC-MS indicated that the only product in the
Fig. 6.9 Raman spectra of CO2 produced by the hydrolysis of PC in sub-critical water at 553 K
and at different reaction times. The spectra were collected under similar conditions, and the
increase of CO2 signals (lower and upper bands and hot bands) indicates the progress of hydrolysis
(Taken from Fig. 3 of Pan et al. [7])
Scheme 6.1 Reaction equation of PC hydrolysis in sub-critical water (Taken from Scheme 1 of
Pan et al. [7])
Fig. 6.10 Relationships of

Raman peak areas of CO2
(squares) and hydrolysis
yields (triangles) vs reaction
time at 553 K (Taken from
Fig. 4 of Pan et al. [7])
liquid phase was bisphenol A (BPA). Also, Raman spectra of the gas phase (Fig. 6.9)
indicated that CO2 was the only gas product in the vapor phase. Thus, in water at
553 K, PC was hydrolyzed to bisphenol A and CO2 according to Scheme 6.1.
It has been known that CO2 pressure (or density) is related to its Raman peak
area by a linear relation in a calibrated Raman spectrometer. As shown in Fig. 6.10,
the peak area increases with the reaction time at 553 K, and it levels off (with
peak area D S*) at about 46 min, indicating the completion of PC hydrolysis.
Therefore, the hydrolysis yield at time t, having a peak area of S, can be calculated
from:
S
Hydrolysis yield .%/ D 100 (6.4)
S
During hydrolysis at 553 K, the slopes of the hydrolysis yield curve increase
with the reaction time, indicating the reaction rates increase. In the hydrolysis, the
polymer chain starts to break up into shorter chains, which were more soluble and
reacted faster in the near-critical water than the long chains, thus the reaction rate
increased as the reaction proceeded.
6.6.3.2 Depolymerization of Polyimide (PI)
Huang et al. [23] studied phase behavior during the depolymerization reactions of
polyimide, synthesized from 4,40 -oxidiphthalic anhydride and 4,40 -diaminodiphenyl
ether monomers (ODPA/ODA PI), in a FSCR (665 m OD, 300 m ID) from 25
to 330 ı C in HCW. The sample was heated to a maximum temperature of 330 ı C at
a rate of 10 ı C min1 (their Fig. 4a). After that, the sample was held at 330 ı C for
30 min (their Fig. 4b), and then the temperature of the sample was gradually cooled
to ambient conditions (their Fig. 4c). All phase changes of the depolymerization
process are given in their Fig. 4. The only gas product detected in the vapor phase
by Raman spectroscopy in the FSCR was carbon dioxide.
6.6.3.3 Depolymerization of Poly (Butylene Terephthalate) (PBT)
The depolymerization of PBT in sub- and supercritical ethanol in a FSCR or a

batch autoclave reactor were studied by Pan et al. [24]. The main liquid products
of depolymerization were identified and quantified as 1,4-butanediol (1,4-BD)
and diethyl terephthalate (DET) by liquid chromatography mass spectrometry, gas
chromatography mass spectrometry, and gas chromatography. The phase behaviors
of PBT with ethanol in water during heating and cooling in a FSCR were shown in
their Fig. 1.
During the heating process, there were three phases (PBT in solid form, ethanol
solution, and vapor phase) coexisting in the FSCR initially. The liquid PBT spherule
gradually dissolved in the subcritical ethanol solution and small droplets appeared
during this process at 240.7 ı C. PBT completely dissolved in supercritical ethanol
solution at 330 ı C to form a homogeneous solution.
To investigate the effect of the reaction temperature and time on the PBT
depolymerization, reactions in sub- and supercritical ethanol were performed at
temperatures between 200 and 280 ı C with 20 ı C intervals. The reaction time
was between 5 and 60 min with the ethanol/PBT mass ratio of 10:1 (20 g/2 g).
The depolymerization yields of PBT are shown in their Fig. 5, the effects of
temperature and time on the yields of 1,4-BD and DET were shown in their Figs.
7 and 6, respectively. In the batch autoclave reactor, depolymerization of PBT was
greatly influenced by the temperature and time. Under the optimal conditions of
an ethanol/PBT mass ratio of 10:1, temperature of 240 ı C, and reaction time of
60 min, PBT was completely depolymerized, and the yields of 1,4-BD and DET
reached 97.7 % and 89.4 %, respectively.
6.6.3.4 Depolymerization of Polyethylene Terephthalate (PET)
Zinc acetate (Zn(Ac)2) catalyzed depolymerization of PET in hot compressed

water (HCW) was carried out in a batch autoclave reactor, as a potential method
for chemical recycling of waste PET [21]. The effects of the ratio of Zn(Ac)2
to PET (0–3 %), temperature (220–300 ı C), and reaction time (5–60 min) were
investigated. The optimal condition for complete catalytic depolymerization of PET
was at temperature 240 ı C, pressure 3.2 MPa, with reaction time of 30 min and
Zn(Ac)2/PET ratio of 1.5 %. The main products of catalytic depolymerization were
identified and quantified as terephthalic acid and ethylene glycol. The maximum
yield of terephthalic acid reached 90.5 %. The phase behavior of PET with or
without a catalyst in water during heating and cooling was studied in a FSCR
(their Fig. 2). According to the observed phase changes, the temperature of PET
dissolution in HCW to form a homogeneous aqueous solution decreased with the
addition of Zn(Ac)2 . To study the influence of reaction temperature on the catalytic
depolymerization, experiments were carried out in the temperature range 220–
300 ı C, with the Zn(Ac)2 /PET ratio fixed at 1.5 %, and reaction times of 15–60 min.
The results were shown in their Figs. 5, 6, and 7.
They observed that PET started to melt at about 220 ı C, dissolved completely
above 228 ı C, and eventually produced an aqueous solution coexisting with a vapor
phase in the presence of Zn(Ac)2 . In this situation, the –O– of the ester linkage
in the repeated unit of PET reacted with proton donors to form a protonated ester
group, which was hydrolyzed by water, just as other esters were. PET was thus
depolymerized to form monomers, i.e., terephthalic acid and ethylene glycol, in
HCW. The EG in turn generated acetaldehyde by protonation, and 1,4-dioxane
by cyclodehydration. Two molecules of acetaldehyde generated 3-hydroxybutanal
by aldol condensation, and then 3-hydroxybutanal was rapidly converted to 2-
butenal by dehydration in the presence of Zn(Ac)2 catalyst. The experimental results
showed that the addition of Zn(Ac)2 could accelerate the depolymerization of PET.
The acidic catalyst dissociated and produced a large number of hydrogen ions to
promote the hydrolysis of PET in the water. A suggested reaction mechanism for the
formation of the observed products from PET under their experimental conditions
was shown in their Fig. 11.
Fig. 6.11 Photomicrographs

of CCl4 with water in FSCR
during the heating process:
(a) CCl4 swells up between
34.1 and 231.0 ı C and
(b) gasifies between
231.3 and 260.0 ı C.
(c) Photomicrographs before
and (d) after reaction taken at
room temperature (Modified
from Fig. 2 of Chen et al.
[10])
6.6.4 Hydrolysis of Chlorinated Organic Compounds in HCW
Some low molecular weight chlorinated organic compounds in EPA Priority

Pollutant List, such as carbon tetrachloride and 1,1,1-trichloroethane, can be
hydrolyzed [10, 25] without any oxidizing agent in HCW. The method, apparatus,
and procedures they used are much the same as those in the research in SCWO.
6.6.4.1 Hydrolysis of Carbon Tetrachloride (CCl4 )
Chen et al. [10] heated the CCl4 sample in FSCR from 34.1 to 260.0 ı C (20 ı C/min)
in the INSTEC heating-cooling stage. As shown in Fig. 6.11, CCl4 swelled gradually
from 34.1 to 231.0 ı C, and began to gasify at 231.3 ı C. From 231.3 to 260.0 ı C, the
CCl4 gasified gradually, and gasified completely at 260.0 ı C after 37 s. The three
initial phases were thus converted into only two phases – an aqueous and a vapor
phase.
The in situ analysis by Raman spectroscopy (Fig. 3 of [10]), and subsequent

GC-MS, and IC analyses, as well as the changes in phase relations and images at
room temperature for the sample before and after reaction (Fig. 6.11c, d), confirmed
that CCl4 was hydrolyzed totally to CO2 and HCl at 260 ı C after 58 min:
CCl4 C 2H2 O ! CO2 C 4HCl (6.5)
The effect of different CCl4 initial concentration (0.076–0.145 g/cm3 ),

CCl4 (l):H2 O(l) volume ratio (1:2, 1:6, 1:10), and reaction temperature (240–300 ı C)
on the hydrolysis behavior were also investigated (their Figs. 6, 7, and 8). The
results demonstrated that the times needed for full CCl4 hydrolysis at 260 ı C were
35, 48 and 64 min for initial CCl4 concentrations of 0.076, 0.111 and 0.145 g/cm3 ,
respectively. Temperature was the prime determining factor for the CCl4 hydrolysis
rate; the times required for the complete CCl4 hydrolysis were 104, 64, 34, and
17 min at the reaction temperatures of 240, 260, 280, and 300 ı C, respectively.
Kinetics analysis of this research indicated that the formation of CO2 followed
the first order reaction kinetics, with an activation energy of 95.06 ˙ 6.71 kJ mol1
(Fig. 10 of [10]).
6.6.4.2 Hydrolysis of 1,1,1-Trichloroethane (TCA)
Unlike the hydrolysis of CCl4 , no CO2 was detected as a product in the hydrolysis
of TCA [24]. The main and side reactions can be written as follows:
Main reaction W C2 H3 Cl3 C 2H2 O ! CH3 COOH C 3HCl (6.6)
Side reaction W C2 H3 Cl3 ! 2C C 3HCl (6.7)
To study the phase behavior of this hydrolysis reaction, a mixture of deionized

water and TCA (10.6 wt %) in the FSCR was heated at a rate of 10 ı C min1
to a maximum temperature of 400 ı C [25]. As shown in Fig. 6.12, TCA was
gradually gasified when the temperature was increased to above 93.3 ı C. Then it
gasified completely and formed a vapor phase at a temperature of 257.2 ı C. As the
temperature increased further, some dark by-products (free carbon) were gradually
produced in the FSCR, the vapor phase gradually dissolved into the aqueous phase,
and a homogeneous solution formed above 364.4 ı C. After being held at 400 ı C for
about half an hour, the FSCR cooled to room temperature. It was noticed that there
was some free carbon remained and no other organic phase could be observed in the
FSCR at that time.
The quantitative hydrolysis of TCA in the larger FSCR was performed in
the temperature range of 210–270 ı C and reaction time range of 10–50 min,
using deionized water containing a 0.0244 M fraction of TCA [25]. Their results
indicated the conversion of TCA increased with increasing temperature in the range
Fig. 6.12 Photomicrographs of TCA in deionized water in FSCR. (a) Heating mixture with
deionized water. (b) Cooling (Modified from Fig. 2a of He et al. [25])
Fig. 6.13 Major reaction pathway of TCA for the products detected in the hydrolysis experiment
(Taken from Fig. 7 of Chen et al. [10])
10–50 min, and reaching a peak value at 270 ı C for 50 min. Subsequent analysis
confirmed the presence of acetic acid, 1,1-dichloroethylene, hydrogen chloride and
free carbon in the FSCR during the reaction process. A major reaction pathway from
TCA to the detected products was proposed, as illustrated schematically in Fig. 6.13.
6.7 Conclusion
The “fused silica capillary reactor” (FSCR) was constructed from fused silica
capillary normally with the dimension of 0.67 mm OD, 0.3 mm ID, and about
25 mm long. The temperatures of the sample in the FSCR as well as the heating
or cooling rates were controlled by using a heating-cooling stage, and there were
two types of stages designed specifically for the FSCR applications available
commercially. The sample temperatures can be set between 196 to 500 or 600 ı C
and the FSCR can withstand an internal pressure of 100 MPa or slightly higher.
Examples were given for quantitative and kinetic studies of reactions between solid
material and water at P-T conditions near the critical point of water using Raman
spectroscopic method. These examples include: (1) supercritical water oxidation

(SCWO) of chlorinated organic compounds; (2) determination of hydrophobic
organic compounds solubilities in subcritical water; (3) depolymerization of waste
polymers in hot compressed water; and (4) hydrolysis of chlorinated organic
compounds in hot compressed water.
The advantages of using FSCR for biorefinery applications include: (1) the
samples in the FSCR in solid, liquid, or gaseous states (or their mixtures) are
easy to prepare, and the quantities of sample material used are minimal and
therefore environmentally friendly; (2) the FSCR is inert to most of samples
containing neutral or acidic solutions, hydrocarbons, and sulfur; (3) the temperature
of the sample in the FSCR can be easily controlled with well-designed heating-
cooling stages, and the sample pressure can be estimated based on the equation
of state of the fluid in the reactor; (4) the FSCR can withstand the pressures and
temperatures near the critical point of water; (5) the transparency of the fused silica
allows in situ observations and sample characterizations, especially with advanced
spectroscopic techniques; and (6) quantitative Raman spectroscopic analyses of
reactants or products together with the images of the sample recorded continuously
during experiment make it possible for kinetic studies. Even though no biorefinery
applications have been reported so far using the FSCR, the examples given in
this report and also in Chou et al. [5] and Chou [6] show the great potential, as
demonstrated by our recent experimental results for the supercritical water oxidation
(SCWO) of guaiacol.
Acknowledgments We would like to thank Robert C. Burruss of U.S. Geological Survey, Yucai
Song, Shunda Yuan, and Jiankang Li of Chinese Academy of Geological Science, Wanjun Lu
of Chinese University of Geological Sciences (Wuhan), Linbo Shang of Inst. of Geochemistry,
Chinese Academy of Sciences, Xiaolin Wang of Nanjing University, Xiaochun Xu of Hefei
University of Science and Technology, Kai Li, Zhichao Hu, and Jiaojiao Jin of Zhejiang University
of Technology for the development and applications of the fused silica capillary reactor technique.
This work was supported by the Mineral Program and Energy program of U.S. Geological Survey,
the Knowledge Innovation Program (SIDSSE-201302) and the Hadal-trench Research Program
(XDB06060100) of Chinese Academy of Sciences, and the National Natural Foundation of China
(Grants 20677052, 20777070, 21077092).
References
1. Bsaaett W, Shen A, Bucknum M, Chou I. A new diamond anvil cell for hydrothermal studies
to 2.5 GPa and from -190 to 1200 ı C. Rev Sci Instrum. 1993;64:2340–5.
2. Schmidt C, Chou I. Chapter 7: The hydrothermal diamond anvil cell (HDAC) for Raman
spectroscopic studies of geological fluids at high pressures and temperatures. In: Dubessy J,
Rull F, editors. Applications of Raman spectroscopy to earth sciences and cultural heritage,
EMU notes in mineralogy, vol. 12. Vandoeuvre-Lès-Nancy: EMU International School; 2012.
p. 248–78.
3. Du Z, Miyagi L, Amulele G, Lee K. Efficient graphite ring heater suitable for diamond-anvil
cells to 1300 K. Rev Sci Instrum. 2013;84:024502–6. doi:10.1063/1.4792395.
4. Chou I, Burruss R, Lu W. Chapter 24: A new optical cell for spectroscopic studies of geologic
fluids at pressures up to 100 MPa. In: Chen J et al., editors. Advances in high-pressure
technology for geophysical applications. Amsterdam/Boston: Elsevier; 2005. p. 475–85.
5. Chou I, Song Y, Burruss R. A new method for synthesizing fluid inclusions in fused
silica capillary containing organic and inorganic material. Geochim Cosmochim Acta.
2008;72(21):5217–31.
6. Chou I. Optical cells with fused silica windows for the study of geological fluids. In: Dubessy
J, Rull F, editors. Applications of Raman spectroscopy to earth sciences and cultural heritage,
EMU notes in mineralogy, vol. 12. Vandoeuvre-Lès-Nancy: EMU International School; 2012.
p. 227–47.
7. Pan Z, Chou I, Burruss R. Hydrolysis of polycarbonate in sub-critical water in fused silica
capillary reactor with in situ Raman spectroscopy. Green Chem. 2009;11:1105–7.
8. Liu H, Pan Z. Visual observations and Raman spectroscopic studies of supercritical water
oxidation of chlorobenzene in an anticorrosive fused-silica capillary reactor. Environ Sci
Technol. 2012;46(8):3384–9.
9. Pan Z, Ma Y, Chou I. Solubility of 2, 4-dichlorotoluene in water determined in fused
silica capillary reactor by in-situ Raman spectroscopy. AIChE J. 2013;59(8):2721–5.
doi:10.1002/aic.14163.
10. Chen Y, Jin Z, Pan Z. In situ Raman spectroscopic study of hydrolysis of carbon tetrachloride
in hot compressed water in a fused silica capillary reactor. J Supercrit Fluids. 2012;72:22–7.
11. Pan Z, Dong Z. Determination of chlorobenzene solubilities in subcritical water in a fused
silica capillary reactor from 173 to 267 ı C. Ind Eng Chem Res. 2011;50(20):11724–7.
12. Lu W, Chou I, Burruss R, Song Y. A unified equation for calculating methane vapor
pressures in CH4 -H2 O system with measured Raman shifts. Geochim Cosmochim Acta.
2007;71(16):3969–78.
13. Shang L, Chou I, Lu W, Burruss R, Zhang Y. Determination of diffusion coefficients of
hydrogen in fused silica between 296 and 523 K by Raman spectroscopy and applica-
tion of fused silica capillaries in studying redox reactions. Geochim Cosmochim Acta.
2009;73(18):5435–43.
14. Kullerud G. Research techniques for high pressure and high temperature. In: Ulmer GC, editor.
Experimental techniques in dry sulfide research. New York: Springer; 1971. p. 288–315.
15. Barton P. Solid solutions in the system Cu-Fe-S, Part I; the Cu-S and CuFe-S joins. Econ Geol.
1973;68(4):455–65. doi:10.2113/gsecongeo.68.4.455.
16. Barton P, Toulmin P. Experimental determination of the reaction chalcopy-
rite C sulfur D pyrite C bornite from 350 ı C to 500 ı C. Econ Geol. 1964;59(5):747–52.
doi:10.2113/gsecongeo.59.5.747.
17. Barton P, Toulmin P. The electrum-tarnish method for the determination of the fugacity of
sulfur in laboratory sulfide systems. Geochim Cosmochim Acta. 1964;2(5):619–40.
18. Yuan S, Chou I, Burruss R. Disproportionation and thermochemical sulfate reduction reactions
in S-H2 O-CH4 and S-D2 O-CH4 systems from 200 to 340 ı C at elevated pressures. Geochim
Cosmochim Acta. 2013;118:263–75.
19. Werre RW Jr, Bodnar RJ, Bethke PM, Barton PB Jr (1980) Novel heating-freezing fluid
inclusion stage. U. S. Geological Survey Professional Paper, 1175, 190pp.
20. Wang X, Chou I, Hu W, Burruss R, Sun Q, Song Y. Raman spectroscopic measurements
of CO2 density: experimental calibration with high-pressure optical cell (HPOC) and fused
silica capillary capsule (FSCC) with application to fluid inclusion observations. Geochim
Cosmochim Acta. 2011;75:4080–93.
21. Liu Y, Wang M, Pan Z. Catalytic depolymerization of polyethylene terephthalate in hot
compressed water. J Supercrit Fluids. 2012;62:226–31.
22. Luthy R, Aiken G, Brusseau M, Cunningham S, Gschwend P, Pignatello J, Reinhard M, Traina
S, Westall J. Sequestration of hydrophobic organic contaminants by geosorbents. Environ Sci
Technol. 1997;31(12):3341–7.
23. Huang F, Huang Y, Pan Z. Depolymerization of ODPA/ODA polyimide in a fused silica

capillary reactor and batch autoclave reactor from 320 to 350 ı C in hot compressed water.
Ind Eng Chem Res. 2012;51(20):7001–6.
24. Pan Z, Shi Y, Liu L, Jin Z. Depolymerization of poly (butylene terephthalate) in sub- and
supercritical ethanol in a fused silica capillary reactor or autoclave reactor. Polym Degrad Stab.
2013;98(7):1287–92.
25. He W, Jin Z, Wang J, Pan Z. Decomposition of 1,1,1-trichloroethane in hot compressed water
in anti-corrosive fused silica capillary reactor and Raman spectroscopic measurement of CO2
product. Chem Eng Sci. 2013;94:185–91.
Chapter 7
Reactors for Supercritical Water
Oxidation Processes
Pablo Cabeza, Joao Paulo Silva Queiroz, M. Dolores Bermejo, Angel Martín,
Fidel Mato, and M. José Cocero
Abstract Supercritical water oxidation (SCWO) process can utilize different

reactors to reduce the operation problems related with solids precipitation and
corrosion. Tubular reactors are one of the most frequently used but its use is limited
by plugging problems. Industrial plants operate with two parallel reactors. While
one is under operation, the other is being cleaned to avoid obstruction due to salt
precipitation. Transpiring wall reactor avoids plugging problems but dilution water
reduces effluent temperature, reducing in this way the energy quality content of the
effluent. Vessel reactors are under application to minimize corrosion by the control
the wall temperature and the materials of construction. The implementation of the
hydrothermal flame as internal heat source reduces the residence time and opens the
way to reduce the reactor volume. Other reactors offer specific solution for some
SCWO processes.
Process intensification to reduce the reactor volume, net energy production,
solids separation and new construction materials are the challenges to improve the
design of the reactor and open the way for the industrial development of this process.
Keywords Supercritical water oxidation • Reactors • Kinetics • Corrosion •

Salt deposition • Tubular reactor • Transpiring wall reactor • Vessel reactor •
Cold wall reactor • Hydrothermal flames • Energy • Process intensification
P. Cabeza • J.P.S. Queiroz • M.D. Bermejo • A. Martín • F. Mato • M.J. Cocero ()
High Pressure Process Group, Department of Chemical Engineering and Environmental
Technology, University of Valladolid, C/Doctor Mergelina s/n, 47011 Valladolid, Spain
e-mail: [email protected]; [email protected]
180 P. Cabeza et al.
7.1 Introduction
7.1.1 Supercritical Water (SCW) as Solvent
Water offers many favourable properties at elevated temperatures and pressures

that make it therefore an excellent solvent and reaction medium for numerous
applications. In the vicinity of the critical point the thermodynamic properties
of water change drastically compared to those of liquid water. The density for
example, with a value of 378 kg/m3 , at the supercritical point, lies in between
those of water vapour (1 kg/m3 ) and liquid water (1,000 kg/m3 ). If increasing
simultaneously pressure and temperature of a mixture consisting of gas and liquid
at equilibrium conditions, the liquid becomes less dense, while the density of the
gas increases due to compression. At the critical point both densities are identical
and the distinction between gas and liquid disappears. The degree of dissolution at
supercritical conditions also differs compared to ambient conditions. An estimation
of the solvation power can be obtained by having a look at the dielectric constant
and the dissociation constant of water. The dielectric constant is a function of
temperature and pressure. Water presents a constant drop of the dielectric constant
from a value about 78.5 As/Vm at ambient conditions to 6 As/Vm at the critical
point [1].
The significant drop in the dielectric strength leads to a considerable increase in
the solubility of hydrocarbons in supercritical water. In correlation, the solubility of
inorganic salts reduces drastically. Supercritical water acts like a non-polar dense
gas with solvation properties equivalent to those of low polar organic sol-vents.
Thus, the solubility of inorganic salts in supercritical water is falling to values from
1 to 100 ppm [2].
The precipitation of inorganic salts causing plugging, fouling and scaling
problems is considered the most limiting factor in the massive commercialization
of this technology. In addition, the presence of these salts in the SCW oxidation
(SCWO) mixture cause different phase behaviour of the system with respect to when
they are not present in the mixture, producing the apparition of solid particles and/or
increasing the temperature at which an homogeneous supercritical phase is formed
retaining a fraction of water as a liquid. Specific equations of state as the Anderko-
Pitzer Eos must be used to calculate the phase behaviour of the system when salts
are present [3].
In general there are two generalized salts thermodynamic behaviours: type 1
and type 2 represented by two models salts NaCl and Na2 SO4 respectively. Type 1
systems have high salt solubility in the range of water’s critical temperature, while
Type 2 systems have low salt solubility in this region. The phase diagram of the
aqueous solutions of NaCl at 25 MPa presents a LV region between 390 and 450 ı C,
and the solid–vapor (SV) equilibrium region for temperatures higher than 450 ı C.
This means that in the LV area, the salt will precipitate, forming liquid-salt droplets
that will become solids at higher temperatures. In the case of Na2 SO4 there is no
LV area, so the salt precipitates, forming nuclei from the homogeneous fluid phase.
7 Reactors for Supercritical Water Oxidation Processes 181
Fig. 7.1 SCWO process diagram
Thus, NaCl particles will be porous and larger (up to 100 m), whereas the sulphate
rapidly forms a great number of smaller particles (1–3 m). A comprehensive
revision on this subject was performed by Hodes and coworkers [4].
7.1.2 SCW as Reaction Media
The dissociation of water leads to the formation of hydronium and hydroxyl ions
and determines the pH of pure water. The ionic product changes drastically from
ambient to supercritical conditions: it increases from a value of 1014 at ambient
conditions to values of about 1011 at 300 ı C to drop to values of 1018 at the critical
point and up to 1023 at supercritical conditions [5]. These physical properties are
the reason to consider supercritical water as an ideal media for the oxidation of
organic compounds [2, 4]. The reaction takes place in a single phase, avoiding
interfacial mass transfer resistances. The physical properties of the SCW allow
to have a reaction media where the reaction participate of the fast velocity of the
radically reactions, as the reactions that take place in the gas phase, and the high
selectivity; as the ionic reactions that take place in the liquid phase.
7.1.3 Process of Supercritical Water Oxidation
The process known as supercritical water oxidation or hydrothermal oxidation

(HTO) consists of the homogeneous oxidation of chemical compounds in an
aqueous medium using air or oxygen as oxidizing agent, at temperatures and
pressures above the critical point of water (374 ı C and 22.4 MPa). Figure 7.1 shows
a general diagram of the SCWO process.
The SCWO takes place in three steps [2]:

• Feed preparation and pressurization.
• Reaction and solids separation.
• Depressurization and heat recovery.
7.1.3.1 Feed Preparation and Pressurization
The feed flow of the supercritical water oxidation process consists of the wastewater
and the oxidant stream. The efficiency of the process is independent of the chosen
oxidant [6], as there can be used: liquid oxygen, air or hydrogen peroxide and there-
fore the choice depends on economic factors. Pure oxygen is the simplest option, but
requires added expensive cost of transporting and storing a cryogenic liquid. On the
other hand, the use of air requires air compressors and also requires bigger reactors
due to the significant amount feed flow due to the nitrogen (contained in the air) that
has to pass through the system. The use of hydrogen peroxide may be considered in
bench scale applications, but it would result in very high costs in industrial scale.
The wastewater or sludge stream is pressurized up to working pressure
(P > 22.1 MPa) and fed into the reactor. Feed streams with high organic load
can be diluted with water or mixed with streams of lower organic load to stay inside
the limits of the process (upper oxidation temperature limited to 700–750 ı C). The
feed flow can be preheated by means of a heat exchanger using the heat of the
effluent leaving the reactor.
7.1.3.2 Reactor
Once compressed, the oxidant may be added to feed before reaching the reactor or
directly into the reactor. The energy advantages of both options are discussed in
Chap. 16 of this book. Some substances can be added to improve the control of the
reactor conditions. For example, kerosene or other fuel can be used as auxiliary fuel
to reach a temperature high enough in the reactor when dealing with low calorific
waste. On the other hand, a stream of water can be added to keep low the reactor
temperature. Sometimes instead cold water, bases are added to neutralize the acids
formed during the oxidation.
The operating conditions depend of the reactor type and the waste to be oxidised.
In general, the efficiency of the process increased with temperature, so operating
temperatures between 600 and 650 ı C are usual. The efficiency of the process is
not improved by increasing the pressure, so usual operating pressures are closed to
critical pressure [2].
7.1.3.3 Depressurization and Heat Recovery
The product stream, after leaving the reactor, is cooled, and finally depressurized.
These steps can also be done in reverse way. The product flows through a heat
exchanger, decreasing its temperature. For dilute aqueous waste, with low heat
of reaction, it is possible to use the heat content of the products to preheat the
waste up to the operation conditions, using a heat exchanger. The minimum organic
matter content necessary depends on the heat exchanged and the required injection
temperature.
The effluent produced can preheat the feed from room temperature to the required
injection temperature, or can be expanded in a turbine producing electricity that can
be used to supply the energy requirements of the plant [7].
To decompress the product stream a pressure control valve or capillary tubes can
be used, in one or more stages, with the total flow of effluent or with gas and liquid
phases separated, in the case that the separation is carried out before the complete
decompression. If the effluent has high concentration of solids, pressure reducing
components can be damaged. To avoid this, a microfiltration system can be used [8].
On a larger scale system, heat recovery may be carried out by expansion of the
reaction products with a turbine. A system of this type could produce large amounts
of electrical supply for air compression, or oxygen and feed pumping. This topic is
widely discussed in Chap. 16 of this book.
7.1.3.4 Separation
The phase separation is performed after the depressurization, but can also occur at
high pressure. When conditions change from supercritical to subcritical, gaseous
products (mainly CO2 , N2 and O2 ) contained in the gas flow can be separated from
the liquid flow by a single separator or a flash device. Depending on the number of
existing depressurization steps there can be more than one gas-liquid separation.
The vapour phase is emitted directly to atmosphere or to a ventilation system or
sent to a system to recover the CO2 .
When the feed has little amount of solids, liquid effluent generally has an aspect
similar to mineral water, also may occur as a brine (i.e., high concentration of
dissolved salt) or sludge (i.e., high concentration of suspended solids).
Technically, the SCWO has the advantage of simple, fast and homogeneous
reactions without mass transfer limitations. It also has some limitations related to
extreme operating conditions and their effect on the materials equipment. These
must be overcome before the widespread industrial use of the process. Therefore,
the main challenges of SCWO are corrosion and deposition of salts, which are being
solved by the use of special construction materials and the development of new
reactor designs able to soften the conditions that the materials must resist [9].
7.1.4 Operational Problems: Corrosion in SCWO Systems
The two main difficulties of the process are corrosion and solids handling. They
take place when the organic compounds treated contain heteroatoms. Acids formed
from the degradation of these substances are aggressive to the system, especially in
the temperature range from 250 to 350 ı C, normally in the heating and cooling area,
where there are still subcritical conditions, and particularly, the dissociation constant
of water is at its maximum. When the temperature is supercritical no dissociation
of acids occurs in their corresponding ions, which can cause corrosion. Also, this
phenomenon can occur when dense or molten salt solutions are injected.
At room temperature the corrosion process is controlled by the kinetics of the
oxidation of materials. In contrast, the corrosion at high temperatures is controlled
by the process of dissolution of the protective oxide layer which contains metal.
When temperatures are higher than 600–700 ı C it has to be taken in account other
corrosion mechanism: corrosion at high temperature, also called creeping. Under
these conditions, metals such as iron, nickel and chromium create volatile corrosion
products, being removed from the metal surface or melted fast on the wall, which
generates a fast corrosion.
Respect to the construction materials of these reactors, most used alloys with high
content of nickel, Inconel or Hastelloy for their components and also for the pipes
under supercritical conditions [10]. These alloys are more resistant to corrosion
and temperature compared with carbon steel or stainless steel [11]. Also, when
temperatures are below 400 ı C, titanium should be used when the feed contains
high concentrations of chloride [12].
7.1.5 Operational Problems: Solids Handling

in SCWO Systems
Due to the low salt solubility of water at supercritical state, solid particles (both
sticky precipitated salts and non-sticky solids completely insoluble in SCW) are
present in SCWO processes. These particles can cause equipment fouling, plugging,
and erosion. Nevertheless, the low solubility of salts in supercritical water allows,
in theory, removing them by a solid–fluid separation, for instance, by means of
hydrocyclons or filtration systems [8]. These methods of recovering solids at the
outlet of the reactor are effective only when the solids do not tend to stick to the
wall of the reactor. In general, solids precipitation has different behavior depending
on the characteristics of the solids. For instance, sodium chloride presents a phase
diagram where the change from liquid to supercritical phase goes through a vapor-
liquid phase. At these conditions the salts precipitation is controlled by evaporation
and particles of 100 m are produced. Instead, Sodium sulphate precipitation from
liquid to supercritical conditions takes place by the fast change of the solubility
with the pressure, so the particles size is below 10 m. Also other solids could
be melted at the operating process condition [2]. The waste is a complex mixture.
Consequently non simple solutions could be applied; the solid precipitation has to be
taken into account in the reactor design. The temperature and pressure, the density
composition, time, geometry and fluids dynamics should be taken into account to
integrate de solid separation in the reactor.
Some industrial applications of SCWO developed so far are oxidation of sludge

[13] or the destruction of chemical weapons [14]. To develop new industrial
applications of supercritical water, research should focus on developing equipment
that resists the operating conditions, also trying to reduce costs. This way, it could
allow the joint of advantages in respect to the environmental benefits with the
economic viability of the process [2].
7.2 Kinetic Considerations for Reactor Design
In literature numerous kinetic studies of the oxidation of different compounds can

be found. Thus, there are available kinetics models to describe the supercritical
water oxidation of many pure substances as ammonia, methanol, ethanol etc. as
well as other kinetic models for the oxidation of specific waste. Kinetic data for
the oxidation of different organic compounds in supercritical water are important
as such to validate reaction models or to design commercial SCWO reactors.
Nevertheless, these models are not always consistent with each others in the same
range of conditions [15, 16].
Kinetics of the SCWO of organics are usually considered as a first order or
pseudo first-order with respect to the concentration of the organic compounds
and frequently considered independent or weakly dependent on the oxidant
concentration [17, 18].
Nevertheless it must be taken into account that real oxidation processes normally
consists of three main steps:
• Initiation phase in which conversion of reagent occurs and a critical concentration
of radicals is built up.
• Propagation phase in which the reagents are rapidly and quasi-steadily consumed
by reactions with these radicals.
• Termination phase, when most of the reagent and oxidant have been converted,
and termination reactions between radicals, which yield stable products, start to
dominate slowing down the conversion.
While the propagation regime alone can be satisfactorily described by first-
order kinetics, the whole regime including the initiation phase and the termination
phase cannot. But considering that the induction period is very short, when the
concentrations of reagents are high, pseudo-first order reactions are a good approach
to the description of the process, but it must not be expected an accurate prediction
of the conversion.
7.2.1 Adapting Mechanism for High Pressure
Webley and Tester [19] did the pioneering work in this field by adapting a
gas-phase combustion mechanism with 56 reversible elementary reactions to
SCWO conditions. They found that the model predicts methanol oxidation to be
very much faster than it was observed experimentally, and this mismatch led them,
and others, to speculate on how SCW might be influencing the kinetics. More
recent work in the field, however, which consistently finds good agreement between
experimental and predicted kinetics, suggests that the chief reason for this earlier
discrepancy was the authors’ use of a value for the rate constant for the reaction
H2 O2 $ OH C OH that was about two orders of magnitude too high. This high
value appeared to result from Webley and Tester erroneously extrapolating the
low-pressure limit rate constant to the high-pressure encountered in SCWO.
Dagaut et al. [20] used a mechanism derived from that published for the
combustion of natural gas in the gas phase. The pressure dependencies, relevant
to the SCWO conditions were included as well as the collision efficiency of water
when appropriate using literature values. Troe’s formalism [21] was used to derive
the modified Arrhenius expressions used at high pressure.
Alkam et al. [22] investigated oxidation kinetics of methanol and hydrogen in a
supercritical water medium. They suggest that kinetic models developed for low-
pressure applications, even models that contain pressure-dependent reaction rates,
should be corrected to account for high-pressure encountered in SCWO conditions.
According to this work, the decomposition of H2 O2 into OH radical is the dominant
reaction controlling the destruction of methanol at high pressures. The high-pressure
reaction rate is a factor of 20 higher than at atmospheric pressure.
These kinetic studies have been used for the more conventional reactor designs.
Many pilot plant facilities use this tubular reactor and same industrial plant as the
ones for sludge treatment.
7.2.2 Designing Reactors Using Kinetic Models
When reliable supercritical water oxidation kinetics of a compound or a given waste

are available, it is easy to apply them to design a reactor, calculating its volume to
obtain whole conversion of the waste at a given temperature following the general
equations of reactors design [23].
The most simple cases are that of isothermal plug flow reactors or stirred batch
mixed reactors normally used for kinetic determination at laboratory scale [24, 25].
Most frequently used reactors in industry are adiabatic tubular reactors normally
used in industry. A simple simulation of this kind of reactors can be performed in
order to know elimination. Normally a steady state simulation considering adiabatic
plug flow reactor would be enough to have initial data to design the reactor [26].
When dealing with this kind of reactor for the SCWO it must be taken into
account that a great amount of heat is released in the reaction, producing sharp
changes in the temperature along the reactors. This is important to be considered
because the properties of water (density, enthalpy) vary sharply with temperature
around the critical point. Thus, they never can be considered as a constant.
Fig. 7.2 Scheme of

premixed flames and
diffusion flames
Even when the behaviour of water and aqueous mixtures around of the critical point
are difficult to be reproduced several authors have reported satisfactory results using
cubic equations as Peng Robinson [27].
When dealing with tank and other complex reactor designs (such as transpiring
water reactors) normally the flow pattern is not simple but sometimes it can be a
successfully assimilated to a combination of flow patterns [28, 29]. Nevertheless,
sometimes it is desirable to use Computational Fluids Dynamics tools to provide a
more specific prediction of the reactor performance such as the formation of a cool
protecting film in a transpiring wall reactor [30, 31].
7.2.3 Hydrothermal Flames
Hydrothermal flames are combustion flames produced in aqueous environments at

conditions above the critical point of water (P > 22.1 MPa and T > 374 ı C).
The flame is defined as the visible part of a highly exothermic supercritical water
reaction zone. Geometrically consist of a surface where combustion is produced.
This surface separates the oxidant from the fuel in the case of diffusion or non-
premixed flames, in which fuel is injected into the oxidant. In the case of premixed
flames, that is, when the fuel and oxidant are injected already mixed, the flame is the
surface separating the reagents from the reaction products. In premixed flames, the
surface is moving towards the reagents with a flame front velocity. If this velocity is
the same as the fluid velocity the flame will remain still in a fixed position. If flow
velocity is higher than flame front velocity the flame is blown away from the tube.
If flow velocity is lower than flame front velocity, the flame will move against the
flow, resulting in backfire. A scheme of each kind of flame is show in Fig. 7.2
The term “hydrothermal combustion” was first used by Franck to describe
oxidation processes taking place in dense aqueous environments [32].
The presence of a flame in a SCWO system would be expected to enhance the
destructive abilities of the SCW medium [33, 34]. However, only a few works about
hydrothermal flames have been published, among them the works of Steeper et al.
[35]; Serikawa et al. [36] and Sobhy et al. [34].
Fig. 7.3 Ignition and extinction of hydrothermal flame (Reprinted with permission from Serikawa
et al. [36]. Copyright © 2002, Elsevier)
First experiments mostly carried out by Franck and coworkers consisted in

studying the ignition of fuel C water mixtures. In general, the device used consisted
of reaction chambers in which an oxidant stream (air, or pure oxygen) was
introduced into a chamber containing a mixture of a fuel (generally methane,
methanol or isopropyl alcohol) with supercritical water [32, 33, 37].
In general, flames ignited spontaneously beyond a certain temperature, normally
between 400 and 500 ı C [37]. This auto ignition temperature was decreased for
higher pressures and fuel concentrations. There was a minimum temperature and
a concentration under which flames were not produced. As the fuel concentration
decreased the flame lost temperature and luminosity, but even when the luminosity
was gone, the flame structure was maintained. In general, the conditions of the flame
ignition depended on the fuel, the oxidant, the ratio fuel/oxidant and the geometry
of the injection system. Serikawa and coworkers [36] developed a continuous
refrigerated facility for observing hydrothermal flames oxidizing isopropanol. In
Fig. 7.3 it can be observed how a flame is ignited and extinct inside a visual cell [36].
The first reactor probably working with a premixed hydrothermal flame inside
was the MODAR reactor, working in conditions of concentration, temperature and
pressure above the ignition conditions of methanol being able to work with injection
temperatures of 25 ı C and injecting the air at 220 ı C [38]. In the reactors used in the
University of Valladolid, operational conditions were above the ignition conditions
of IPA according to Serikawa et al. [36], thus they can be described as working at
hydrothermal flame regime with a premixed flame [39–41]. These reactors present
the advantage that the cold feed can be directly injected in the flame without
preheating.
The ETH Zurich has been developing different continuous hydrothermal burners
working with non premixed fuels since 1992 [42, 43]. They used the hydrothermal
flame as an internal heat source in a transpiring wall reactor. The direct injection of
the waste into a hydrothermal flame generated inside the reactor was developed as a
solution to avoid the external preheating of the waste up to supercritical conditions
[44–46]. Recently, some authors have proposed using them in the breaking of rocks
by thermal drilling spallation to make wells of more than 3 km deep, where water
or sludge is used to maintain the structure of the well [47, 48].
SCWO with a hydrothermal flame has a number of advantages over the flameless
process. Some of these advantages permit overcoming the traditional challenges
that make the successful and profitable commercialization of SCWO technology
difficult. The advantages include the following [49]:
• It allows the destruction of the pollutants in residence times of a few milliseconds,

which permits the construction of smaller reactors.
• It is possible to initiate the reaction with feed injection temperatures near to room
temperature when using vessel reactors [40, 41, 44–46]. This avoids problems
such as plugging and corrosion in a preheating system, having an advantage from
the operational and energy integration perspective.
• Higher operation temperatures improve the energy recovery.
The design of hydrothermal flame reactors is simple. Normally most authors
considered that the kinetics of the hydrothermal flame are very fast, being the
mixture the limiting factor [50–52]. This is especially important when dealing with
non premixed hydrothermal flames such as the ones used in the ETH Zurich [53].
The key element in the use of these flames is the design of the mixer. These designs
are frequently modeled by CFD simulation [51, 52].
When dealing with premixed flames, the mixing is not a key factor in the design
of the reactor. Our research group has found that the kinetic is a limiting factor in the
design. However, obtaining hydrothermal flame regime kinetics is not easy and only
the kinetics of isopropyl alcohol is found in literature [54]. Nevertheless, together
with kinetics, the flame front velocity is an important point to take into account
when designing premixed hydrothermal flame reactors. Flow velocity determines
the extinction temperature of the hydrothermal flame [55]. Hydrothermal flames are
stabilized in vessels in which the flow sections are large enough to keep a low fluid
flow velocity making possible the feed injection at subcritical temperatures. If the
fluid flow velocity was higher, the injection temperature necessary to keep a steady
flame was also higher. In order to go deep into this phenomenon the flame front
velocity was estimated [56]. Flame front velocities vary between 0.01 and 0.1 m/s,
much lower than the typical flame front velocities in air combustion (0.4–3 m/s). It
is observed that when the temperature of the mixture before ignition is higher, the
flame front velocity is also higher. This behaviour is qualitatively consistent with our
observations of the results in the tubular reactors, where at lower feed flows lower
injection temperatures were necessary to have ignition. As an indicative rule for the
design of vessel reactor with a flame inside it can be established that the velocity of
the fluid inside of the reaction chamber has to be between 0.1 and 0.01 m/s while the
velocity in the injector must be higher than 1 m/s to avoid ignition in the injector.
7.3 Reactors for SCWO Process
SCWO can be carried out in different reactor types. It is important to note that
depending on the design of the reactor main problems of this process, corrosion and
deposition of salts, can be avoided in greater or less way. The main classification
of reactors was developed by Schmieder and Abeln [57], which classified them by
tubular reactors, vessel reactors, transpiring wall and cooled wall.
Fig. 7.4 Tubular reactor with multiple injections
7.3.1 Tubular Reactor
The tubular reactor is the simplest design of existing ones, and therefore, the most
used. It is mainly used for testing in small laboratory plants new applications of
SCWO [58, 59], to obtain kinetic parameters and reaction heats [60, 61]. In addition
most big and industrial plants constructed so far use this kind of reactor [62, 63].
The main disadvantages of these reactors are easy plugging due to precipitation
of salts and the possible formation of hot spots when uncontrollable exothermic
reactions are produced at high speeds.
In this type of reactor the effects of pressure and temperature cannot be isolated,
this way tubular reactors are generally expensive and heavy due to the needs of a
considerable thickness reactor wall and expensive material (such a nickel alloys)
to make it possible to resist the high temperatures and pressures present inside the
reactor.
The diameters of the tubular reactors have to be small enough for obtaining a
high speed of the circulating fluid in order to prevent the deposition of salts. Even
that, the precipitated salts can stick to the reactor walls. Several solutions have been
proposed. Some authors patented scrapers to remove salt deposits from the wall of
the reactor [64, 65]. This solution is similar to that adopted by the Superwater®
process based on the design of Modell M [63, 66]. Others proposed an alternative
strategy is to periodically wash the equipments with room-temperature water to
clean salt deposits. To do so, more than one reactor must be connected in parallel, so
when one reactor was being cleaned the remaining reactors could stay operational
[67]. A similar solution was adopted in industrial applications such as AquaCritox®
and AQUACAT® processes.
In cases where the concentration of organic matter in the feed is considerably
high, there is even greater possibility of existence of hot spots within the reactor.
Since the reaction itself is extremely fast and difficult to control, an alternative
control system discussed in several patents consists of carefully dosing feed, oxidant
and/or quenching water through multiple-injection schemes as shown in Fig. 7.4.
The first patent proposing a multiple injection was filed by McBrayer and coworkers
from ECO-WASTE technologies in 1995 [68]. This patent described a tubular
reaction in which the waste water was fed together with oxidant, in a proportion
too low for completely oxidizing the organic material in the feed, but high enough
for reaching a certain supercritical temperature with the heat released by the
oxidation. Down-flow in the tubular reactor, additional inlets were installed, by
which additional amounts of oxidants were introduced, together with an additional
Fig. 7.5 MODAR vessel

reactor (Adapted from Huang
et al. [73])
flow of cold water to compensate the increase in temperature caused by the progress
of oxidation. In a second patent, the same authors described the dosing of organic
material instead of oxidant through the lateral injection points [69]. A few other
variations of the multiple-injection scheme have been patented. In 2005, Cansell
patented a tubular reaction with multiple injection ports, through which oxygen or
hydrogen peroxide were introduced (without water) [70]. In 2006, Gidner [71] from
Chematur proposed a similar reactor with side injections of both oxidant and organic
material diluted in cold water. This cold feed was preheated by the contact with
the reaction mixture, and an additional flow of oxidant was introduced to complete
the oxidation. This method also allows preheating the feeds introduced through the
lateral inlets by direct contact with the reaction products, having the additional
advantage of avoiding feed preheating, improving in this way energy integration
and avoiding salt precipitation problems during preheating.
7.3.2 MODAR Vessel Reactor
The vessel reactor consists of a container in which two zones can be distinguished:
the upper zone, at high temperature reaction, and the lower zone, at low temperature
(in subcritical conditions) where the salts are dissolved [72]. In a later design it was
improved by adding a film of water falling down the wall that prevents the salt depo-
sition on the wall of the reactor [73]. The outline of this reactor is shown in Fig. 7.5.
Fig. 7.6 Examples of cooled wall reactor. (a) Cooled wall reactor refrigerated with the feed flow
(Reprinted with permission from Bermejo et al. [39]. Copyright © 2009 American Chemical
Society). (b) Cooled wall reactor refrigerated with cold water
7.3.3 Cooled Wall Reactor
The cooled wall reactor has a design that makes it possible the independence of the
effects of temperature and pressure in the SCWO. The external wall, which holds
the pressure, is maintained at temperatures lower than 400 ı C by the action of a
cooling water flow pumped downward between the external and internal walls. This
external pressure vessel can be constructed of stainless steel because it does not
suffer oxidizing atmosphere and nor temperatures higher than 400 ı C. The internal
wall, also called reaction chamber, is where the reactants are mixed and takes place
the oxidation reaction. It is built with a special material capable of resisting the
oxidizing atmosphere at temperatures up to 700 ı C.
In the first design patented by our group, shown in Fig. 7.6a, the fluid circulating
between the pressure vessel internal wall and the reaction chamber external wall
is the aqueous feed that is preheated up to supercritical temperatures with the
heat released in the reaction. To avoid precipitation of salts inside of the reaction
chamber, some of the reactors developed posses a salt precipitation chamber
before the injector that conducts the feed to the reaction camber [39]. This design
was successfully tested with a number of wastes with low salt content [74–76],
nevertheless when dealing with feeds with high salt content, sometimes a blockage
Fig. 7.7 Example of

transpiring wall reactor
of the injector occurred and the increase of pressure of the feed caused collapse of
the reaction chamber wall that was not designed to stand pressure [39].
A new cooled wall reactor working with a hydrothermal flame as a heat source
was patented in 2009. The scheme of this reactor was shown in Fig. 7.6b. The
main difference with the old cooled wall reactor is that aqueous feed and air can be
injected cold in the reactor over the hydrothermal flame, while the stream circulating
between the pressure vessel and the reaction chamber is simply cold pressurized
water. This water stream is entering in the reaction chamber by its lower part,
accumulating as a subcritical water “pool” that dissolves the salt precipitated in the
upper area of the reactor. It has been tested successfully with salty feeds and with
sewage sludge [41, 77]. This reactor presents the additional advantage that part of
the products can be extracted of the reactor by its upper part, without mixing with
the cold water, allowing better energy integration because an effluent at temperatures
between 600 and 700 ı C can be obtained.
7.3.4 Transpiring Wall Reactor
The transpiring wall reactor (TWR) consists of a reaction chamber surrounded by

a wall through which clean water circulates for cooling and protection, avoiding
the action of corrosive agents, salt deposition and high temperatures. A schematic
diagram of the typical TWR is shown in Fig. 7.7.
There are different types of transpiring wall reactors; the main and common
element in all of them is a porous wall, a porous tube which constitutes the
reaction chamber, made of sintered metal or ceramic. Alternative designs have
been developed, as proposed by Ahluwalia [78, 79] consisting of a concentric
plate reactor being the reaction zone an intermediate annular section between them,
Bermejo et al. [80] have carried out and extensive research for the development of
TWR.
The main disadvantage to be considered for the TWR is the heat recovery. Hot
products introduced are cooled and diluted by mixing with the cold water that
transpires through the wall. As this water used to dissolve the salts as much as
possible must be at temperatures below the critical point, it mixes with the hot
products, reducing their the outlet temperature of the products and it is not high
enough for a good heat recovery.
With regard to the operational results, the data obtained with this kind of reactors
have been successful in contaminants removal. The biggest problem is the recovery
of existing salts introduced with the feed, although the problem of plugging is
improved [79].
7.3.5 Other Reactor Designs
7.3.5.1 Reactor Tank Reverse Flow with a Brine Pool
This type of reactor consists of an elongated cylinder, closed at its ends, which
constitutes the inner reaction chamber. Inside there are two different zones: an
upper zone at supercritical temperature, and a lower zone temperature subcritical.
The oxidation reaction takes place in the zone in which supercritical conditions are
achieved.
Inorganic salts or other dense material introduced with the feed or formed
by chemical reactions are insoluble in supercritical fluids, thus, they precipitate
and they pass to the subcritical zone where they are dissolved in the brine pool.
Furthermore, to avoid the deposition of these substances inside the wall reactor, it
has a water film which covers the wall.
MODAR Inc. (Natick, MA) developed and patented [81] the first reactor of
this type. An application of that is the commercial plant Nittetsu (Japan), for the
destruction of waste from the manufacture of semiconductors (63 kg/h of capacity)
[82]. Stone and Webster designed a compact automated plant to keep on board [83],
using also this design of reactor.
7.3.5.2 Fluidized Bed Reactor
It was developed by SRI International. This reactor consists of a variation of the

SCWO process that uses a bed of fluidized solids functioning as both reactant
and adsorptive surface for salt control. The process is referred to as “assisted
hydrothermal oxidation” (AHO), and works at lower temperatures and pressures
than the conventional SCWO process (T D 380–420 ı C and P < 22.1 MPa) [82].
7.3.5.3 Reverse-Flow Tubular Reactor
Patented by Abitibi-Price, the inverse tubular flow reactor consists of a tube which
only differs in two thermal zones, so that the process fluid can be fed in both
directions. Thus, when the reagents flow through the interior of the reactor, it can re-
dissolve salts layer formed when the reactor was operating in the reverse direction
(when the inlet was the outlet and vice versa) [82].
7.3.5.4 Double-Walled Tank Reactor
This design of a stirred reactor has a double titanium wall, through which prevents
corrosion and a mechanical stirrer which provides a turbulent flow, preventing the
precipitation of salts and favoring heat transmission. It was developed by the CEA
(Commissariat à l’Energie Atomique, Commission for Atomic Energy, France) [84].
7.3.5.5 SUWOX Process
The reactor used for the process SUWOX consists on a cooled wall reactor type,
with its two corresponding chambers: the external chamber where pressurized water
flows, and the internal chamber where the reaction takes place at 420–490 ı C and
pressures above 70.0 MPa. This way it is achieved a high enough density to maintain
the salts dissolved (Fig. 7.8).
It was developed after the creation of a process in the IKET (Institut für Kern-und
Energietechnik, Institute of Nuclear and Energetic Technology, Karlsruhe) and FZK
(Focusing Zentrum Karlsruhe, Germany) [85, 86].
7.3.6 Laboratory SCWO Reactors
The simplest way to perform SCWO experiments on SCWO in a lab can be using
batch type reactors. These reactors can be an autoclave type reactor consisting in a
reactor vessel with a heating system such as an electric resistance or furnace [87],
that sometimes can be provided with a stirrer, or can consist in the more simple
small closed tubes that can be introduced in an oven [88] or in a sand bath [89]
for heating and when finishing the experiment rapidly cooled in cooled water. The
main inconvenience of this kind of reactors is that in some cases, the preheating-
cooling times can be much longer than the reaction time itself. So they can be used
Fig. 7.8 SUWOX reactor

(Reprinted with permission
from Baur et al. [86].
for determining if a certain substance is susceptible to be oxidized in supercritical

water media, but when the kinetics is very fast a lot of information can be lost in the
heating and cooling process.
For the kinetics determination the most used configuration is a small isothermal
reactor. Sometimes with tube diameters a low as 1/1600 in order to increase heat
transmission and introduced in an oven or sand bath. Current reactor designs
used for SCWO are generally based on closed-loop flow reactors [90]. With
these designs, the reactor effluent is sampled only after being quenched by a heat
exchange, to reduce temperature, followed by expansion through a throttling valve,
to reduce pressure. This results in a flow separation and the subsequent recovery of
a two-phase effluent stream. The recovered liquid phase is generally sampled and
analyzed by HPLC, while GC is used for the gas phase.
The HPLC and GC techniques for analyzing the reactor effluent are limited in
that they can only detect remaining reactants and stable product species – unstable
and short-lived intermediates and radicals are not observed with these techniques.
Also, the temperature and pressure reduction processes have typical residence times
on the order of seconds, not necessarily favorable conditions for quenching the
oxidation reaction.
These techniques are useful for global (apparent) rate reaction studies, but
not for mechanism determination. As it was discussed in Sect. 7.2, a first-order
representation of apparent kinetics is adequate only when induction times are
negligible and only up to a certain conversion level. Induction times increase as
temperatures or initial fuel and oxygen concentrations are decreased. For high
conversions, e.g. >99 % at 500 ı C or >70 % at 450 ı C, a deviation from first-order

behavior is expected even in the absence of induction times and a mechanism based
reaction rate should be used [13]. To determine mechanism other type of analysis
techniques, frequently associated with optical devices must be used.
Sometimes batch of flow transparent reactors or re actors provided with a window
can be used to visually follow a reaction (i.e. the oxidation processes of a solid
particle [91, 92]). Windows inserted in cell use to be made of sapphire that presents
the highest resistance to SCWO [91, 92]. Other possible transparent materials using
in SCW environment are diamond anvil cells [93] or quartz tubes, being this last
soluble in SCWO after some hours, thus this type of device must be disposable [94].
Optical devices will be the subject of Chap. 5, thus the structure and characteristics
of this kind of device will not be discussed here.
Maharrey and Miller [94] used one of this quartz capillary reactors to build a
sampling mass spectrometry (DSMS) system to probe the reaction species present
in the supercritical water, for acetic acid oxidation. This work remarks the advantage
that DSMS is one of the few that can monitor several species at once, has very good
sensitivity, and allows for the use of isotopic substitution to elucidate individual
mechanistic steps. A disadvantage is that it must be operate under high vacuum.
Problems associated with the requirement of operation in a vacuum are the design
of the reactor/detector interface and the mass flow into the vacuum system that must
be handled by the vacuum pumps.
Cells with windows allows the direct observation of the reaction with laser
optical techniques such as Raman can incorporate optical diagnostics directly into
the heated zone of the reactor and have been able to observe the rates of change of
many species in solution simultaneously [95–97].
Rice et al. [95, 96] used in situ Raman spectroscopy to investigate the oxidation
rate of methanol and isopropyl alcohol in a tubular reactor. Because the optical
access is conveniently translated along the length of the reactor, the entire reac-
tion history, including production and destruction of key intermediates, is easily
obtained. Koda et al. [97] employed Raman spectroscopy in a CSTR-type cell, also
for investigation of methanol oxidation in supercritical water.
7.4 Reactors in Industrial Applications
Here is presented a summary of some industrial application focusing in the SCWO

reactors.
7.4.1 SuperWater Solutions
Co-founded by Dr Michael Modell, SuperWater started in 2006 in Wellington

(Florida). Its main work has been focused on the treatment of non-corrosive
sludge [63].
Fig. 7.9 SCWO reactor of Bluegrass Chemical Agent Pilot Plant System (GA) [100]
Working with systems based on similar designs of Modell’s previous company,

MODEC, SuperWater Solutions features a tubular reactor system with mechanical
brushes for control/removal of salts/solids accumulation [66, 98]. Currently a full
scale SCWO plant for sludge oxidation with a capacity of 5 t/day in dry basis, is in
trials since 2009 in Iron Bridge Regional Water Treatment Facility (Florida, USA),
since February 2011 is operating with sludge [99].
7.4.2 General Atomic
Created in San Diego, General Atomic (GA) has the longest tenure in SCWO
of all active SCWO companies (The subdivision of SCWO was created in 1991)
[100]. GA has typically utilized a vessel type reactor design and has had exten-
sive experience with several different methods for controlling corrosion and salt
precipitation/accumulation, such as the use of liners, coatings, feed additives, and
mechanical scrapers [1, 12] (Fig. 7.9).
A number of SCWO systems have been or are being built and delivered. Most
of GA’s work has been for government/military entities but they have also worked
with industrial clients:
A system to treat pink and red water from TNT operations was provided to DAC
for a facility in Korea (Korea System, 4.5 L/min). A system to treat hazardous
wastes and sewage sludge was provided to partner commercial companies in Japan
for evaluation in the Japanese hazardous waste destruction market (Japan System).
The Bluegrass chemical munitions demilitarization plant will use SCWO systems
to destroy hydrolysate from chemical agent and energetic neutralization operations
(BGCAPP System).
A simplified SCWO system utilizing over 20 years of development experience is
being supplied to Tooele Army Depot (TEAD System) for destruction of energetic
hydrolysate from the CAD demilitarization plant (11.4 L/min). The Bluegrass
Army Depot will use SCWO to destroy hydrolysate from the hydrolysis of excess
explosives and propellants (BGAD System with a capacity of 38 L/min).
7.4.3 SuperCritical Fluids International (SCFI)
In 1995 the Swedish company Chematur AB bought a license for the EWT (Eco
Waste Technologies) SCWO process in Europe and then in 1999 it bought the
worldwide rights to EWT SCWO, finally, Chematur developed their version of
SCWO process under the name Aquacritox®. In 2007, Chematur sold their super-
critical fluids division and equipment to SCFI after developed and named different
customized versions of the Aquacritox® process in collaboration with various
clients, in the way, SCFI was founded in 2007 to commercialize AquaCritox® in
the industrial and municipal market [62].
SCFI utilizes a tubular reactor design and has chosen to focus primarily on
sewage sludge and digestate feed applications [101].
SCFI has installed one of their models: the Aqua Critox® A10, in the European
Validation Centre (EVC) in Ringaskiddy, Cork, Ireland. The Aquacritox A10 has
been used to process waste streams that include precious metal catalysts, municipal
sewage and drinking water sludge and high strength pharmaceutical waste streams
(Fig. 7.10).
7.4.4 Hanwha Chemical Corporation
Created in Seoul, South Korea, they have been working with supercritical water-
based technologies since 1994. Hanwha Chemical Corporation is one of the most
versatile companies involved with hydrothermal technologies.
They have utilized both vessel and tubular reactor types in their systems and have
several pilot scale systems on which they have performed testing of various feeds
by hydrothermal treatment.
About their work in SCWO process, they have built two full-scale SCWO plants;
a 2,000 kg/h system for dinitrotoluene (DNT) wastewater and a 5,500 kg/h system
for terephthalic acid (TPA) wastewater [102]. Both of these plants are no longer
operating.
Fig. 7.10 AquaCritox’s sewage-to-energy demonstration plant in Cork, Ireland [62]
7.4.5 SRI International
Located in Menlo Park The scientific research institute SRI International developed
the AHO version of SCWO. This is the technology that is being utilized in the
full-scale facility built for JESCO and currently in operation in Tokyo Japan for
destruction of PCBs [101].
7.5 Future Prospects and Research Needs
The SCWO process needs to be competitive with other conventional processes

that do not require such high investments, so the reduction of the reactor cost is
a challenge. Implementation of hydrothermal flame operation conditions will allow
the process intensification to reduce the reactor volume. Studies about hydrothermal
flame from biomass and other solid materials as well as studies to implement the
hydrothermal flames in the reactor design are needed.
Net energy production by SCWO process is a challenge that needs research
studies about reactor design and the energetic integration of the process.
New reactor designs that achieve solid separation without dilution in the effluent
will improve the energy recovery and the valorisation of the solid waste.
Another improvement will be the application of new construction materials in
the reactor design.
References
1. Deul R, Franck EU. The static dielectric-constant of the water-benzene mixture system to
400-degrees-c and 2800-bar. Ber Bunsen-Ges Phys Chem. 1991;95:847–53.
2. Bermejo MD, Cocero MJ. Supercritical water oxidation: a technical review. AIChE J.
2006;52:3933–51.
3. Anderko A, Pitzer KS. EOS representation of phase equilibria and volumetric properties of
the system NaCl–H2O above 573 K. Geochim Cosmochim Acta. 1993;57:1657–80.
4. Hodes M, Marrone PA, Hong GT, Smith KA, Tester JW. Salt precipitation and scale control
in supercritical water oxidation—Part A: fundamentals and research. J Supercrit Fluids.
2004;29:265–88.
5. Marshall WM, Franck EU. Ion product of water substance, 0–1000 ı C, 1–10,000 bars: new
international formulation and its background. J Phys Chem Ref Data. 1981;10:295–304.
6. Phenix BD, DiNaro JL, Tester JW, Howard JB, Smith KA. The effects of mixing and oxidant
choice on laboratory-scale measurements of supercritical water oxidation kinetics. Ind Eng
Chem Res. 2002;41:624–31.
7. Cocero MJ, Alonso E, Sanz MT, Fdz-Polanco F. Supercritical water oxidation process under
energetically self-sufficient operation. J Supercrit Fluids. 2001;24:37–46.
8. Goemans MGE, Tiller FM, Li L, Gloyna EF. Separation of metal oxides from supercritical
water by crossflow microfiltration. J Membr Sci. 1997;124:129–45.
9. Kritzer P, Dinjus E. An assessment of supercritical water oxidation (SCWO): existing
problems, possible solutions and new reactor concepts. Chem Eng J. 2001;83:207–14.
10. Kim H, Mitton DB, Latanision RM. Stress corrosion cracking of alloy 625 in pH 2 aqueous
solution at high temperature and pressure. Corrosion. 2011;67:035002-1–8.
11. Asselin E, Alfantazi A, Rogak S. Thermodynamics of the corrosion of alloy 625 supercritical
water oxidation reactor tubing in ammoniacal sulfate solution. Corrosion. 2008;64:301–14.
12. Kritzer P. Corrosion in high-temperature and supercritical water and aqueous solutions: a
review. J Supercrit Fluids. 2004;29:1–29.
13. Griffith JW, Raymond DH. The first commercial supercritical water oxidation sludge
processing plant. Waste Manag. 2002;22:453–9.
14. Marrone PA, Cantwell SD, Dalton DW. SCWO system designs for waste treatment: applica-
tion to chemical weapons destruction. Ind Eng Chem Res. 2005;44:9030–9.
15. Vogel F, Blanchard JLD, Marrone PA, Rice SF, Webley PA, Peters WA, Smith KA, Tester JW.
Critical review of kinetic data for the oxidation of methanol in supercritical water. J Supercrit
Fluids. 2005;34:249–86.
16. Bermejo MD, Cantero F, Cocero MJ. Supercritical water oxidation of feeds with high
ammonia concentrations: pilot plant experimental results and modelling. Chem Eng J.
2008;137:542–9.
17. Portela JR, Nebot E, Martínez de la Ossa E. Kinetic comparison between subcritical and
supercritical water oxidation of phenol. Chem Eng J. 2001;81:287–99.
18. Rice SF, Steeper RR. Oxidation rates of common organic compounds in supercritical water.
J Hazard Mater. 1998;59:261–78.
19. Webley PA, Tester JW. Fundamental kinetics of methanol oxidation in supercritical water, in:
supercritical fluid science and technology. Am Chem Soc. 1989;1989:259–75.
20. Dagaut P, Cathonnet M, Boettner J-C. Chemical kinetic modeling of the supercritical-water
oxidation of methanol. J Supercrit Fluids. 1996;9:33–42.
21. Troe J. Predictive possibilities of unimolecular rate theory. J Phys Chem. 1979;83:114–26.
22. Alkam MK, Pai VM, Butler PB, Pitz WJ. Methanol and hydrogen oxidation kinetics in water
at supercritical states. Combust Flame. 1996;106:110–30.
23. Froment GF, Bischoff KB, De Wilde J. Chemical reactor analysis and design. 3rd ed.
Hoboken: Wiley; 2011.
24. García-Jarana MB, Kings I, Sánchez-Oneto J, Portela JR, Al-Duri B. Supercritical water
oxidation of nitrogen compounds with multi-injection of oxygen. J Supercrit Fluids.
2013;80:23–9.
25. Goto M, Nada T, Kodama A, Hirose T. Kinetic analysis for destruction of municipal sewage
sludge and alcohol distillery wastewater by supercritical water oxidation. Ind Eng Chem Res.
1999;38:1863–5.
26. Ermakova A, Anikeev VI. Modeling of the oxidation of organic compounds in supercritical
water. Theor Found Chem Eng. 2004;38:333–40.
27. Bermejo MD, Rincón D, Vazquez V, Cocero MJ. Supercritical water oxidation: fundamentals
and reactor modeling. CI Ceq. 2007;13:79–87.
28. Fauvel E, Joussot-Dubien C, Pomier E, Guichardon P, Charbit G, Charbit F, Sarrade S.
Modeling of a porous reactor for supercritical water oxidation by a residence time distribution
study. Ind Eng Chem Res. 2003;42:2122–30.
29. Bermejo MD, Fernández-Polanco F, Cocero MJ. Modeling of a transpiring wall reactor for the
supercritical water oxidation using simple flow patterns: comparison to experimental results.
Ind Eng Chem Res. 2005;44:3835–45.
30. Bermejo MD, Martín A, Queiroz JPS, Bielsa I, Ríos V, Cocero MJ. Computational
fluid dynamics simulation of a transpiring wall reactor for supercritical water oxidation.
Chem Eng J. 2010;158:431–40.
31. Zhou N, Krishnan A, Vogel F, Peters WA. A computational model for supercritical water
oxidation of organic toxic wastes. Adv Environ Res. 2000;4:79–95.
32. Schilling W, Franck EU. Combustion and diffusion flames at high-pressures to 2000 bar.
Ber Bunsen-Ges Phys Chem. 1988;92:631–6.
33. Pohsner GM, Franck EU. Spectra and temperatures of diffusion flames at high-pressures to
1000 bar. Ber Bunsen-Ges Phys Chem. 1994;98:1082–90.
34. Sobhy A, Butler IS, Kozinski JA. Selected profiles of high-pressure methanol-air flames in
supercritical water. Proc Combust Inst. 2007;31:3369–76.
35. Steeper RR, Rice SF, Brown MS, Johnston SC. Methane and methanol diffusion flames in
36. Serikawa RM, Usui T, Nishimura T, Sato H, Hamada S, Sekino H. Hydrothermal flames
in supercritical water oxidation: investigation in a pilot scale continuous reactor. Fuel.
2002;81:1147–59.
37. Hirth T, Franck EU. Oxidation and hydrothermolysis of hydrocarbons in supercritical water
at high-pressures. Ber Bunsen-Ges Phys Chem. 1993;97:1091–8.
38. Oh CH, Kochan RJ, Charlton TR, Bourhis AL. Thermal-hydraulic modeling of supercritical
water oxidation of ethanol. Energy Fuel. 1996;10:326–32.
39. Bermejo MD, Rincon D, Martin A, Cocero MJ. Experimental performance and modeling
of a new cooled-wall reactor for the supercritical water oxidation. Ind Eng Chem Res.
2009;48:6262–72.
40. Bermejo MD, Fdez-Polanco F, Cocero MJ. Experimental study of the operational param-
eters of a transpiring wall reactor for supercritical water oxidation. J Supercrit Fluids.
2006;39:70–9.
41. Bermejo MD, Jiménez C, Cabeza P, Matías-Gago A, Cocero MJ. Experimental study of
hydrothermal flames formation using a tubular injector in a refrigerated reaction chamber.
Influence of the operational and geometrical parameters. J Supercrit Fluids. 2011;59:140–8.
42. La Roche HL, Weber M, Trepp C. Rationale for the film cooled coaxial hydrothermal burner
(FCHB) for supercritical water oxidation (SCWO). In: Briefing book of the first international
workshop on supercritical water oxidation, Jacksonville, FL, USA, 6–9 Feb 1995, Session VI;
1995.
43. LaRoche HL, Weber M, Trepp C. Design rules for the wall-cooled hydrothermal burner
(WHB). Chem Eng Technol. 1997;20:208–11.
44. Wellig B, Lieball K, Rudolf Von Rohr P. Operating characteristics of a transpiring-wall
SCWO reactor with a hydrothermal flame as internal heat source. J Supercrit Fluids.
2005;34:35–50.
45. Príkopský K, Wellig B, Rudolph von Rohr P. SCWO of salt containing artificial wastewater
using a transpiring-wall reactor: experimental results. J Supercrit Fluids. 2007;40:246–57.
46. Wellig B, Weber M, Lieball K, Príkopský K, Rudolf von Rohr P. Hydrothermal methanol
diffusion flame as internal heat source in a SCWO reactor. J Supercrit Fluids. 2009;49:59–70.
47. Wideman T, Potter J, Potter R, Dreesen D. Methods and apparatus for thermal drilling. Patent
US20100089576; 2010.
48. Rudolf von Rorh P, Rothenfluh T, Schuler M. Rock drilling in great depths by thermal
fragmentation using highly exothermic reactions evolving in the environment of a water-based
drilling fluid. Patent EP2379835; 2010.
49. Augustine C, Tester JW. Hydrothermal flames: from phenomenological experimental demon-
strations to quantitative understanding. J Supercrit Fluids. 2009;47:415–30.
50. Moussiere S, Roubaud A, Boutin O, Guichardon P, Fournel B, Joussot-Dubien C. 2D and 3D
CFD modelling of a reactive turbulent flow in a double shell supercritical water oxidation
reactor. J Supercrit Fluids. 2012;65:25–31.
51. Narayanan C, Frouzakis C, Boulouchos K, Prikopsky K, Wellig B, Rudolf von Rohr P.
Numerical modelling of a supercritical water oxidation reactor containing a hydrothermal
flame. J Supercrit Fluids. 2008;46:149–55.
52. Sierra-Pallares J, Parra-Santos M, Garcia-Serna J, Castro F, Cocero MJ. Numerical modelling
of hydrothermal flames. Micromixing effects over turbulent reaction rates. J Supercrit Fluids.
2009;50:146–54.
53. Prikopsky K. Characterization of continuous diffusion flames in supercritical water. Doctoral
thesis no. 17’374, ETH Zurich, Switzerland; 2007.
54. Queiroz JPS, Bermejo MD, Cocero MJ. Kinetic model for isopropanol oxidation in supercrit-
ical water in hydrothermal flame regime and analysis. J Supercrit Fluids. 2013;76:41–7.
55. Bermejo MD, Cabeza P, Bahr M, Fernández R, Ríos V, Jiménez C, Cocero MJ. Experimental
study of hydrothermal flames initiation using different static mixer configurations. J Supercrit
Fluids. 2009;50:240–9.
56. Bermejo MD, Cabeza P, Queiroz JPS, Jiménez C, Cocero MJ. Analysis of the scale up of a
transpiring wall reactor with a hydrothermal flame as a heat source for the supercritical water
oxidation. J Supercrit Fluids. 2011;56:21–32.
57. Schmieder H, Abeln J. Supercritical water oxidation: state of the art. Chem Eng Technol.
1999;22:903–8.
58. Veriansyah B, Park T-J, Lim J-S, Lee Y-W. Supercritical water oxidation of wastewater
from LCD manufacturing process: kinetic and formation of chromium oxide nanoparticles.
J Supercrit Fluids. 2005;34:51–61.
59. Lee HC, Kim JH, In JH, Lee CH. NaFeEDTA decomposition and hematite nanoparticle
formation in supercritical water oxidation. Ind Eng Chem Res. 2005;44:6615–21.
60. Aymonier C, Gratias A, Mercadier J, Cansell F. Global reaction heat of acetic acid oxidation
in supercritical water. J Supercrit Fluids. 2001;21:219–26.
61. Benjamin KM, Savage PE. Supercritical water oxidation of methylamine. Ind Eng Chem Res.
2005;44:5318–24.
62. Aquacitrox. http://www.scfi.eu/. Last accessed 26 Aug 2013; 2008.
63. SuperWater Solutions. http://www.superwatersolutions.com/technology.html. Last accessed
14 Aug 2013.
64. Huang CY. Apparatus and method for supercritical water oxidation. US5100560; 1992.
65. Hazlebeck DA, Spritzer MH, Downey KW, Martinez MR, Isoya T, Suzuki K, Nakayama S.
System and method for solids transport in hydrothermal processes. US6773581; 2004.
66. Modell M, Kuharich E, Rooney MS. Supercritical water oxidations process and apparatus of
organics with inorganics. WO/1993/000304; 1993.
67. Bond LD, Mills C, Whiting P, Koutz S, Hazlebeck D, Downey K. Method and apparatus to
remove inorganic scale from a supercritical water oxidation reactor. WO9533693; 1995.
68. McBrayer R, Eller JM, Swan JG, Deaton JE, Gloyna RR. Method and apparatus for treating
waste water streams. WO9526929; 1995.
69. Eller JM, McBrayer RN, Swan JG. Method for controlling reaction temperature. US5770174;
1998.
70. Cansell F. Method for treating waste by hydrothermal oxidation. US6929752; 2005.
71. Gidner A. Method and system for supercritical water oxidation of a stream containing
oxidizable material. WO2006052207; 2006.
72. Hong GT, Killilea WR, Thomason TB. Method and apparatus for solids separation in a wet
oxidation type process. WO8902874; 1989.
73. Huang CY, Barner HE, Albano JV, Killilea WR, Hong GT. Method for supercritical water
oxidation. WO9221621; 1992.
74. Cocero MJ, Alonso E, Torío R, Vallelado D, Fdz-Polanco F. Supercritical water oxidation
in pilot plant of nitrogenous compounds-isopropanol mixtures in the temperature range
500–750 ı C. Ind Eng Chem Res. 2000;39:3707–16.
75. Cocero MJ, Alonso E, Vallelado D, Torío R, Fdz-Polanco F. Optimization of operational vari-
ables of a supercritical water oxidation SCWO process. Water Sci Technol. 2000;42:107–13.
76. Cocero MJ, Alonso E, Torío R, Vallelado D, Sanz T, Fdz-Polanco F. Supercritical water
oxidation (SCWO) for polyethylene terephthalate (PET) industry effluents. Ind Eng Chem
Res. 2000;39:4652–7.
77. Cabeza P, Queiroz JPS, Arca S, Jiménez C, Gutiérrez A, Bermejo MD, Cocero MJ.
Sludge destruction by means of a hydrothermal flame. Optimization of ammonia destruction
conditions. Chem Eng J. 2013;232:1–9.
78. Ahluwalia K. Internal platelet heat source and method for use in a supercritical water
oxidation reactor. U.S. Patent 5,571,424; 1996.
79. Ahluwalia K. Internal platelet heat source and method for use on a supercritical water
oxidation reactor. U.S. Patent 5,670,040; 1997.
80. Bermejo MD, Fdez-Polanco F, Cocero MJ. Effect of the transpiring wall on the behavior of a
supercritical water oxidation reactor: modeling and experimental results. Ind Eng Chem Res.
2006;45:3438–46.
81. Hong GT, Killilea WR, Thomason TB. Method for solids separation in a wet oxidation type
process. US Patent 4,822,497; 1989.
82. Marrone PA, Hodes M, Smith KA, Tester JW. Salt precipitation and scale control in
supercritical water oxidation – Part B: commercial/full-scale applications. J Supercrit Fluids.
2004;29:289–312.
83. Cohen LS, Jensen D, Lee G, Ordway DW. Hydrothermal oxidation of Navy excess hazardous
materials. Waste Manag. 1998;18:539–46.
84. Calzavara Y, Joussot-Dubien C, Turc HA, Fauvel E, Sarrade S. A new reactor concept for
hydrothermal oxidation. J Supercrit Fluids. 2004;31:195–206.
85. Casal V, Schmidt H. SUWOX – a facility for the destruction of chlorinated hydrocarbons.
86. Baur S, Schmidt H, Kramer A, Gerber J. The destruction of industrial aqueous waste
containing biocides in supercritical water – development of the SUWOX process for the
technical application. J Supercrit Fluids. 2005;33:149–57.
87. Cui B, Cui F, Jing G, Xu S, Huob W, Liu S. Oxidation of oily sludge in supercritical water.
J Hazard Mater. 2009;165:511–7.
88. Goto M, Nada T, Ogata A, Kodama A, Hirose T. Supercritical water oxidation for the
destruction of municipal excess sludge and alcohol distillery wastewater of molasses.
89. Jin F, Kishita A, Moriya T, Enomoto H. Kinetics of oxidation of food wastes with H2 O2 in
90. Pinto LDS, Freitas dos Santos LMF, Al-Duri B, Santos RCD. Supercritical water oxidation
of quinoline in a continuous plug flow reactor – Part 2: kinetics. J Chem Technol Biotechnol.
2006;81:919–26.
91. Shoji D, Sugimoto K, Uchida H, Itatani K, Fujie M, Koda S. Visualized kinetic aspects
of decomposition of a wood block in sub- and supercritical water. Ind Eng Chem Res.
2005;44:2975–81.
92. Sugiyama M, Kataoka M, Ohmura H, Fujiwara H, Koda S. Oxidation of carbon particles in

supercritical water: rate and mechanism. Ind Eng Chem Res. 2004;43:690–9.
93. Fang Z, Xu SK, Smith RL, Arai K, Kozinski JA. Destruction of deca-chlorobiphenyl in
supercritical water under oxidizing conditions with and without Na2 CO3 . J Supercrit Fluids.
2005;33:247–58.
94. Maharrey SP, Miller DR. A direct sampling mass spectrometer investigation of oxidation
mechanisms for acetic acid in supercritical water. J Phys Chem A. 2001;105:5860–7.
95. Hunter TB, Rice SF, Hanush RG. Raman spectroscopic measurement of oxidation in
supercritical water. 2. Conversion of isopropyl alcohol to acetone. Ind Eng Chem Res.
1996;35:3984–90.
96. Rice SF, Hunter TB, Ryden AC, Hanush RG. Raman spectroscopic measurement of oxidation
in supercritical water. 1. Conversion of methanol to formaldehyde. Ind Eng Chem Res.
1996;35:2161–71.
97. Koda S, Kanno N, Fujiwara H. Kinetics of supercritical water oxidation of methanol studied
in a CSTR by means of Raman spectroscopy. Ind Eng Chem Res. 2001;40:3861–8.
98. Sloan DS, Modell M, Pelletie RA. Sludge management in the city of Orlando- it’s supercriti-
cal! Fla Water Resour J. 2008;60:46–54.
99. Supercritical Water Oxidation Update. City council workshops. http://www.cityoforlando.net/
cityclerk/citycouncil/workshop_files/presentations/2011-04-25_oxidation.pdf. Last accessed
26 Aug 2013; 2011.
100. General Atomic. Bluegrass chemical agent pilot plant system. http://www.ga.com/bluegrass-
chemical-agent-pilot-plant-system. Last accessed 26 Aug 2013; 2014.
101. Marrone PA. Supercritical water oxidation—current status of full-scale commercial activity
for waste destruction. J Supercrit Fluids. 2013;79:283–8.
102. H. CHEMICAL. www.hanwhachemical.co.kr/english/pro/psu_panc_idx.jsp. Last accessed
30 June 2012.
Chapter 8
Effects of Reactor Wall Properties, Operating
Conditions and Challenges for SCWG of Real
Wet Biomass
Mohammad S.H.K. Tushar, Animesh Dutta, and Chunbao (Charles) Xu
Abstract Hydrogen is considered the cleanest fuel as it does not create any
pollution during combustion. However, hydrogen is not readily available in nature
as a primary energy source, but a secondary energy generated from the primary
energy sources via various conversion processes. Supercritical water gasification
(SCWG) process in contrast to the conventional gasification process does not require
biomass drying, and rather it uses the moisture and external water in the reaction.
Much research has been performed on SCWG of various types of biomass for
hydrogen production, where numerous reactors made from different materials were
used. Various operating conditions were also widely studied, such as catalysts,
temperature, pressure, feed concentration etc. Researchers have tried quartz reactors
to avoid corrosion of metallic reactors due to the supercritical water, but the overall
performance was not satisfactory as the total gas yield and H2 yield were less
compared to any metallic reactor. Using catalysts improved the overall gas and
hydrogen yield but the common challenge is the supported catalysts becoming
deactivated over a longer operating time. Temperature has the most prominent
and positive effect on H2 yields, while pressure does not have any significant
effect rather it has complex effects on the process. A higher concentration of the
feed increases CO2 production, and a lower concentration increases H2 and CH4
yields. Moreover, the major challenge of SCWG is feeding real biomass into the
reactor. Most of researchers used model biomass or fractions of biomass (such as
M.S.H.K. Tushar • A. Dutta ()

Mechanical Engineering Program, School of Engineering, University of Guelph,
Guelph, ON N1G 2W1, Canada
C. Xu ()
Institute for Chemicals and Fuels from Alternative Resources, Department of Chemical
and Biochemical Engineering, Western University, London, ON N6A 5B9, Canada
208 M.S.H.K. Tushar et al.
cellulose or lignin) for conversion. Little literature work was reported on SCWG
of real biomass in continuous flow reactors, mostly in batch reactors. Therefore,
this chapter aims to discuss about the effects of the reactor wall properties,
operating parameters on supercritical water gasification of real wet biomasses and
the associated operating challenges.
Keywords Hydrogen • Supercritical water gasification • Real biomass •

Operating parameters • Operating challenges • Batch process • Continuous
process • Feeding technologies • Reactor types
8.1 Introduction
Due to ever increasing energy demand, increasing trend of fossil fuel price and need
to reduce the dependency on fossil fuels, researchers are in search for renewable
but environmentally benign energy sources. Biomass is one of the options that have
been studied by the researchers for its various advantages, e.g., its renewability and
wide availability. However, biomass has disadvantages like low bulk density, high
alkali contents particularly for some agricultural feedstock, and a high moisture
content. Although indigenous alkali metal contents of biomass have positive effect
on hydrogen production by gasification leading to less char/coke production [1, 2],
it is well known that the presence of alkali metals creates agglomerates which
significantly reduces the performance of conventional fluidized bed gasifiers. A
high moisture content of biomass makes conventional thermochemical processes
less energy efficient, while in supercritical water gasification (SCWG) process, high
moisture content is an advantage as the process employs water, at its critical point
(374 ı C and 22.1 MPa), as a reactive medium.
Hydrogen production plays a very important role in the creation of hydrogen
economy. Hydrogen production is mainly from steam reforming methane, but
producing hydrogen from renewable and inexpensive sources is certainly more
environmentally friendly and promising. A large number of different processes have
been studied for hydrogen production from biomass. The methods of hydrogen
production from biomass can be divided to two main groups: thermochemical
processes (pyrolysis, gasification and reforming) and biochemical processes (bio-
photolysis, gas fermentation). Biomass consists of three main chemical components:
cellulose, hemicellulose, and lignin. Among these components, lignin cannot be
easily converted to gaseous product via biochemical conversion, whereas it can
be converted completely in thermochemical conversion [3]. Thus, thermochemical
gasification processes for hydrogen production are more advantageous as they are
feedstock insensitive [4].
While typical biomass such as rice husk has a low energy content of 15 MJ/kg
(dry-and-ash-free or daf), conventional biomass gasification process requiring
biomass drying, where a major portion of heat is used for drying, leads to a
negative net energy production [5, 6]. Therefore, it is not economically viable
8 Effects of Reactor Wall Properties, Operating Conditions and Challenges. . . 209
to convert wet lignocellulosic biomass by traditional techniques like pyrolysis,

combustion, and gasification due to high cost and energy requirements for water
evaporation. Wet biomass can be converted into fuel gas rich in either hydrogen
or methane by hydrothermal gasifying in hot compressed water. In a hydrothermal
process at pressure higher than 20 MPa, compared to ambient condition, the heat of
evaporation of water is very marginal, so high thermal efficiencies can be obtained
although the feedstock has low dry matter content. In addition, ionic reactions are
promoted in hot compressed water through radical routes which leads to lesser char
formation [7, 8]. This provides a major impetus for the use of hot compressed water
as a reaction medium for energy recovery from organic wastes. Hot compressed
water at above water’s critical points (374 ı C and 22.1 MPa) proved to be more
reactive for biomass gasification. SCWG of wet biomass produces hydrogen-rich
gas products of a high efficiency and lower char and tar formation. Table 8.1 presents
some detailed comparison between various thermochemical processes [9–11].
Solvolysis of hemicellulose and lignin macromolecules starts at temperatures
above 190 ı C when biomass is heated in liquid water [13–17]. All of the hemi-
cellulose and much of the lignin can dissolve in hot-compressed water at 220 ı C
after only 2 min. The remaining lignocellulosic solids undergo solvolysis and
pyrolysis at higher temperatures [18], accompanied by a phenomenal mixture
of hydrothermolysis reactions such as isomerization, dehydration, fragmentation,
and condensation of intermediate products, etc. Besides the hydrothermolysis
reactions, steam reforming (Eq. 8.1), water-gas shift reactions (WGS) (Eq. 8.2), and
methanation reactions (Eq. 8.3) also play an important role in SCWG, responsible
for the generation of hydrogen-rich syngas [19–22].
Steam reforming:

x
CHx Oy C .1 y/ H2 O ! CO C 1 y C H2 (8.1)
2
Water-gas shift:
CO C H2 O ! CO2 C H2 (8.2)
Methanation:
CO C 3H2 ! CH4 C H2 O (8.3)
Modell [23] found that wood (maple sawdust) can be completely solubilized in
supercritical water, but less than 40 % of carbon in the feedstock was converted to
gases due to limited gasification at temperatures below 380 ı C. In addition, catalysts
(including Ni, Co/Mo and Pt on alumina) did not show significant effect on the
reaction chemistry under the conditions of study. Amin et al. [24] did some related
experiments using glucose as a model compound for SCWG, producing 100 %
conversion but with a very low gas yield. The gas produced contained less than
20 wt% of the carbon in the glucose feedstock.
Table 8.1 Comparison of various thermochemical technologies for biomass conversion [9–11]
210
Energy or fuel
Technology Major biomass feedstocks produced Advantages Disadvantages
Direct Combustion Wood, forestry residues, Heat 1. The simplest form of biomass 1. High investment and operating costs
agricultural waste, and conversion process to obtain heat
municipal solid waste Steam 2. Low emission of NOx and SOx, [12] 2. High particulate emission
Electricity 3. High efficiency in terms of heat 3. High CO and intermediate aromatic
(>90 %) products generation
4. Agglomeration, sintering because of
higher alkali content in some biomass
such as crop residue, and corrosion in
heat transfer surfaces because of a
relatively higher Cl content
Conventional Wood, forestry residues, Low or 1. Possible for small scale power 1. Requiring special technical and
steam/air agricultural waste, and medium-Btu generation. managerial skills for plant operation
gasification municipal solid waste producer gas 2. Relatively high conversion 2. Sensitive to the quality of fuel
efficiencies
3. Modest investment costs 3. Ash agglomeration
4. Severe tar and char formation
5. Depending on types of biomass
(turndown ratio)
Supercritical water Wood, forestry residues, Hydrogen 1. Suitable for wet biomass (moisture 1. High reaction pressure (>22 MPa)
gasification agricultural waste, content > 70 %) leading to high reactor costs
(SCWG) municipal solid waste, Methane 2. High H2 gas yield 2. Continuous feeding of real biomass to a
sewage sludge, and model SCWG reactor being very challenging
biomass 3. Unique properties of supercritical 3. SCW is highly corrosive in nature,
water (non-polarity, high diffusivity, posing challenge in materials for reactor
absence of mass transfer, etc.) manufacture
favoring gasification of biomass
4. Higher yield of gas products due to
high reaction rates
M.S.H.K. Tushar et al.
Pyrolysis Wood, forestry residues, Pyrolysis oil 1. Relatively mature technology with 1. Markets are yet to be developed for the
agricultural waste, and (bio-oil), large scale production demonstrated bio-oil and bio-char products
municipal solid waste biochar 2. Produces valuable products (bio-oil 2. Less capable to handle wet biomass, as
and bio-char) wet biomass requires a costly drying
process
Carbonization Wood, forestry residues, Biochar 1. Improved physical characteristics of 1. Low enhancement in volumetric density
(Torrefaction) agricultural waste, and biomass
municipal solid waste 2. A higher energy content than the 2. No significant improvement in energy
original biomass (per unit volume) density
3. Reduced moisture content 3. Unable to reduce alkali content from ash
4. Making biomass friable
Hydrothermal Wood, forestry residues, Liquid oil 1. Suitable for wet biomass (moisture 1. Not yet demonstrated in a commercial
liquefaction (HTL) and agricultural waste (biocrude), content > 70 %) scale
biochar, and 2. Feedstock insensitive 2. Lower overall yield of oil (biocrude)
some gases 3. Biocrude has higher heating values 3. Continuous feeding of real biomass in a
than pyrolysis oils reactor being challenging
4. Yielding an energy efficiency of
85–90 %
5. Biocrude can recover more than
70 % of the feedstock carbon
content
8 Effects of Reactor Wall Properties, Operating Conditions and Challenges. . .
211
SCW is a green solvent and highly reactive for biomass thermochemical

conversion reactions. Non-polar organic compounds can dissolve in SCW while
they cannot normally dissolve in liquid water or steam. The kinetic limitations
associated with mass transfer between different phases in a reactor can be effectively
eliminated in SCW due to the absence of the phase boundaries and high diffusivity
of gas in SCW, leading to rapid and complete reactions [25, 26]. As such, higher
conversions of the solid material to fuel gas, and less solid residue can be achieved
by SCWG relative to conventional direct or indirect air- or steam-blown gasification
[27, 28]. Also, SCW gasification process can operate in a directly heated gasifier, or
an indirectly heated reactor [29].
The objectives of this work are to discuss the effects of reactor types and
operating conditions on SCWG of real wet biomass. The key operating conditions
for SCWG include temperature, pressure, residence time, initial concentration, types
of catalysts and feedstock. In addition, challenges of SCWG of real biomass are
summarized.
8.2 SCWG with Different Types of Reactors
Various types of batch and continuous reactors have been employed for SCWG, as
summarized as follows.
8.2.1 Batch Reactors
One of the many reasons for using batch reactors for SCWG over continuous
processing is that it allows for the collection of significant data sets in a reasonably
short period of time. In addition to this, it avoids the challenge of pumping feedstock
into a SCWG process under very high pressure, and in batch reactors solid biomass
can be used directly which however is extremely difficult in continuous operation
[30]. As an inherent feature of batch reactors, the feedstock is heated up at a
relatively low rate, when it starts to decompose during heating. Usually tar and
char are generated during heating due to hydrothermal pyrolysis prior to gasification
reactions, although they can further gasify to form gaseous product. However,
amount of tar and char production strongly depends on the heating rate and the
type and substrate concentration of the feedstock in SCWG. As such, it is of a
great interest understand the performance of batch reactors for SCWG of various
feedstocks.
Experiments for liquefying cellulose in hot-compressed water in a batch reactor
produced 70 % conversion of biomass at a residence time under 60 min and
temperatures ranging from 200 to 400 ı C under pressures ranging from 8 to 22 MPa
using nickel catalysts and alkali salts [31, 32]. The use of a catalyst significantly
reduced residual chars and tars. Another important study of gasification of organic
wastes in hot compressed water (350 ı C and 20 MPa, 2 h) resulted in over 85 %

biomass conversion [33]. A high methane yield was observed from the gasification
of aliphatic hydrocarbon in SCW at lower temperatures, while much less time is
required to obtain similar biomass conversion with a continuous reactor [34].
Batch reactor systems build using different materials have been studied by
researchers. Most of the systems were purchased commercially rather than con-
structed in-house. Most commonly, autoclave batch reactors were used in various
studies [27, 28, 35–41]. Besides these autoclave reactors, uses of various other
types of batch reactors, e.g., conventional batch reactors [30, 42–46], tumbling
batch reactor [47], small unstirred batch reactor [48], micro volume batch reactor
[49], mini batch reactor [50] and in-house made reaction vessel [51] were reported.
Most of these batch reactors are made of stainless steel, mainly 316 L [27, 30, 35,
36, 38–40, 44, 46, 50], some are made of Inconel 625 [28, 41, 47, 51] and a few
are made of quartz [43, 49]. Catalytic effects of reactor wall on SCWG efficiency
in batch reactors are discussed in a separate section (Sect. 8.8) of this chapter. It
shall however be noted that the batch reactor wall could be passivated by carbon
deposition after a few runs of experiments in the batch reactor [42, 52].
8.2.2 Continuous Flow Reactors
Continuous flow reactors (predominantly tubular reactors) proved to be more

efficient and scalable than batch reactors for SCWG of biomass. Almost all
continuous flow SCWG setups are currently on a bench scale. A typical setup
includes a tubular reactor [42, 48–51, 53–66] heated inside an electric furnace,
and a high-pressure pump (e.g., HPLC pump) for feeding liquid biomass or model
compounds. Some researchers also tested fluidized bed reactors for SCWG [36,
67] as well as continuous-stirred tank reactor (CSTR) [52, 68]. Continuous SCWG
systems allow for high feed throughput which is a must for any potential scale-up
of the SCWG process. Moreover, compared to batch operation the heating rate in
a continuous reactor can be much faster. The benefit of employing fast heating rate
is to eliminate undesired byproducts (tar and char) from the feedstock, which can
cause clogging of the reactor system – the major challenge of SCWG in continuous
flow tubular reactors [52, 69].
8.3 Effects of Temperature on SCWG
Temperature has dominating effects on hydrogen yield. With an increase in tem-

perature the hydrogen and carbon dioxide yields increase while the methane yield
decreases. Thermodynamically at lower reaction temperatures, H2 and CO2 readily
react to form methane and water via the methanation reaction. A higher temperature
could limit the methanation reaction and promote water gas shift reaction, leading
Fig. 8.1 Effects of temperature on hydrogen yield
to low CH4 and CO formation. In a SCWG process, the presence of excess water
leads to a preference for the formation of H2 and CO2 instead of CO. Supercritical
water reforming also reduces the formation of various intermediate and complex
tarry compounds such as organic acids, phenols, ketones, cresols, furfurals and
aldehydes which may otherwise form in appreciable concentrations in conventional
gasification processes [27, 30, 36, 38, 40, 43, 46, 49–51]. Figure 8.1 shows the
effects of temperature on hydrogen yield, with data extracted from literatures cited
in the legend. As clearly shown in this Figure, hydrogen yield increases with
increasing temperature.
In SCWG, formation of hydrogen or methane are competing each other. Antal
[70] suggested the following equations for the generation of hydrogen and methane
by using glucose as a model compound to simplify the stoichiometry:
Hydrogen formation:
C6 H12 O6 C 6H2 O ! 6CO2 C 12H2 Ho 298 D 158KJ =mol (8.4)
Methane formation:
C6 H12 O6 ! 3CO2 C 3CH4 Ho 298 D 134:1KJ =mol (8.5)
The formation reactions of hydrogen and methane have opposite characteristics.

The formation of hydrogen is endothermic whereas the formation of methane is
exothermic [42], which implies that with the increase in temperature, hydrogen
and carbon dioxide yields increase as per the Le Chatelier’s principle. On the
contrary, methane yield decreases with increasing temperature based on the same
principle. Temperature could also affect the formation of char and tar – byproducts
Fig. 8.2 Effects of temperature on gas yield from SCWG of corn cob. Pressure: 25 MPa;
feedstock: 5 wt% corn cob C 2 wt% carboxymethyl cellulose (CMC); flow rate of feedstock:
25.05 g/min; flow rate of pre-heated water: 126.7 g/min. [69] (Reprinted with permission from
International Association of Hydrogen Energy, Copyright © 2008 International Association for
Hydrogen Energy. Published by Elsevier Ltd.)
of gasification reactions. It was reported by Kruse et al. [71] that char formation
could be eliminated by increasing the heating rate. Usually below the critical point,
furfurals are formed upon heating of biomass (particularly glucose). However,
when the temperature reaches the supercritical range furfurals would form char
by condensation/polymerization with other products such as phenolic compounds
derived from lignin [72]. Similarly, tar could be promoted at slower heating rates. It
has been reported that temperature has a strong impact to reduction of tar formation.
For instance, most of the tars would convert into gaseous product at a temperature
above 700 ı C [73].
Lu et al. [69] studies SCWG of corn cob at 25 MPa and the gas yield vs.
temperature is shown in Fig. 8.2. It shows that H2 and CO2 yields increase
sharply while CO decreases with the increase in temperature. Gasification efficiency
(or carbon efficiency) also increases with the increase in temperature as well as
residence time [65, 66, 69]. A higher temperature favors free-radical reactions, and
hence enhances reaction rates, which improves gas yield [69].
8.4 Effects of Pressure on SCWG
Pressure shows complex effects on biomass gasification in SCW. With the increase
in pressure, the properties of water, such as density, static dielectric constant and
ion product also increase. This in turn increases the rates of the ion reactions and
restrains free-radical reactions [1]. Also hydrolysis reaction plays a significant role
in biomass gasification in SCW, which is promoted by the presence of HC or OH .
As increasing pressure favors the ion product generation, the hydrolysis rate could
Fig. 8.3 Effect of pressure on hydrogen yield (#c D continuous reactor system, *b D batch reactor
system)
be promoted. Moreover, high pressure favors water–gas shift reaction, but reduces
decomposition reaction rate as per the Le Chatelier’s principle. From some research
work at lower temperatures [27, 38, 49], increasing pressure slightly increased the
hydrogen yield, but pressure has complex effects on gas yield at higher temperatures
as discussed below.
The effects of pressure on the mechanism of supercritical gasification of biomass
are however very complicated. With the increase in pressure while the other
parameters such as temperature, substrate concentration, and flow rate are remained
constant, the density and ion product of water increase. An increase in the rates at
higher pressure accelerates ion reactions that favor gas production via the hydrolysis
reaction and water gas shift reaction. On the other hand, higher pressure restrains
free-radical reactions, which in turn inhibits gas formation reactions via free radicals
[74]. From Le Chatelier’s principle, a reaction that produces more molecules is
inhibited at high pressure region. Thus, the gasification process is generally favored
at lower pressure [75]. The combined effects of pressure result in the complicated
effects of pressure on SCWG [42]. Figure 8.3 shows that the reactor pressure does
not have significant effect on SCWG of biomass in either batch or continuous
reactors. The special physical and chemical properties of SCW disappear when
the pressure is below the critical point, which could inhibit hydrogen production.
However, operation at high pressure greatly increased operating cost. As a result,
25 MPa was very commonly used as the operating pressure for a SCWG process to
balance the effects of pressure on hydrogen yield and the operating costs [76].
In summary, Temperature has predominant effect on SCWG of biomass if
compared to pressure. The variation in polarity (i.e. dielectric constant) of SCW
could account for the complex effects of pressure in both batch and continuous
processes of SCWG. For instance, when the pressure was increased from 27.5
to 42.5 MPa at 500 ı C, the dielectric constant of water increased from 1.59 to
2.6 [77]. Nevertheless, with the increase in temperature, the dielectric constant of
Fig. 8.4 Two-way Table illustrating the interaction of reaction temperature and time (Reprinted
with permission from [30]. Copyright © 2011, Elsevier)
water decreases considerably. For example, the dielectric constant of water reduced
from 4.84 to 1.59 when the temperature was increased from 400 to 500 ı C at
27.5 MPa [78].
8.5 Effects of Residence Time on SCWG
From the thermodynamics perspective, complete gasification of biomass can be

achieved in SCW to produce H2 and CO2 as the only gaseous products for a
sufficiently long residence time. However, the SCWG process is usually controlled
by kinetics and may take up to several hours to complete. Figure 8.4 shows the two-
way interaction Table between residence time and reaction temperature, where it
is clearly shown that a longer residence time favors gaseous production and hence
increases the hydrogen yield with less amount of tars [30, 36, 42, 49–51].
Generally, reaction kinetics controls the gasification reactions. In continuous
reactors, Sato et al. [75] found that methane and carbon dioxide production
increased accompanied by decrease in tar with increasing in reaction time until
180 min but hydrogen production remained almost constant. Miller et al. [79]
reported positive and significant effects of residence time on gas yield: the gasi-
fication rate increased by increasing either temperature or residence time.
8.6 Effects of Feed Concentration on SCWG
Feed concentration directly affects the gas yield. Very commonly a lower biomass
concentration led to a higher gas yield (rich in H2 ) [30, 36, 42, 49, 50]. Usually water
is present in excess during SCWG of any biomass as water acts as both solvent and
reactant in the process. At a lower biomass to water ratio, solvation during heat-up
can proceed more efficiently. This in turns reduces the possibility of formation of
polymerized carbonaceous material – precursors for char/tar. Guan et al. [50] found
that the yield of H2 was very sensitive to feed concentration although the total yield
of carbon-containing gases was largely insensitive to this variable. For instance,
hydrogen yield in SCWG of algae reduced by more than twofold as the algae
loading increased from 1 to 15 wt%. Guo et al. [36] also obtained the same findings
and recommended a higher reaction temperature to complete SCWG of higher
concentration feedstock, which is consistent with the thermodynamic calculation.
In the case of continuous SCWG processes, feed concentration has significant
effect on gasification efficiencies and yields too, although it does not show a
significant effect on the gasification rate. Increased feed concentration resulted
in decreased gasification and carbon efficiencies and H2 yield but an increased
CH4 yield [36, 43, 66, 69]. From kinetics, a higher concentration leads to a
higher gasification rate, which thus seems to contradict that observed in the
SCWG experiments at high substrate concentration (the efficiencies and gas yield
decreased). A plausible explanation is that polymerization reactions are promoted
at higher feed concentration due to the formation of defiant species from the feed
materials, which reduces the number of available feed molecules for gasification.
Usually CH4 yield increases at higher feed concentrations, which is likely due to
the lack of water that restricts the methane reforming reaction [48–54, 62, 66].
8.7 Effects of Catalyst on SCWG
A lot of catalysts have been used for conventional thermal gasification of biomass
[80]. Nonetheless, there are differences between the catalysts used for conventional
thermal gasification and SCWG as the operating conditions of these two processes
are different, e.g. pressure and temperature. Four types of catalysts were commonly
used by the researchers which include activated carbon, metal, metal-oxide and
alkali [81, 82]. Elliot [81], Calzavara et al. [82] and Guo et al. [83] reported in their
reviews that to improve hydrogen formation and reduce char or tar produced inside
the reactor, catalysts proved to be very effective. Alkali compounds [30, 35, 42]
such as K2 CO3 , KOH, Trona (NaHCO3 .Na2 CO3 .2H2 O) were used in batch reactors
to facilitate H2 and CO2 formation while reducing CO by catalyzing water–gas shift
reaction. These catalysts become base when dissolved in water and affect the water–
gas shift reaction to improve the gas quality, i.e., H2 yield. Lu et al. [42] studied
SCWG of some biomass in the presence of various metal/metal-oxide catalysts,
including CeO2 , nCeO2 , n(CeZr)xO2 , Pd/C and Ru/C, among which Ru/C catalyst
demonstrated to be the most effective for hydrogen production.
Nickel is a metal catalyst that is usually employed not only in the conventional
gasification [80], but also for SCWG. Kersten et al. [43] and Azadi et al. [45] tested
several different forms of nickel catalysts in a batch reactor and obtained superior
hydrogen selectivity. It is well known fact that Ni catalysts promote water–gas shift
reaction. Some noble metal catalysts were also experimented for biomass SCWG.
Yamamura et al. [51] found that with RuO2 catalyst, cellulose and glucose were
almost completely gasified at 450 ı C and 47.1 MPa in a reaction vessel (batch
reactor). Similarly, Kersten et al. [43] observed that 1 wt% glucose solution could
completely gasify with 3 wt% Ru/TiO2 at 800 ı C and 30 MPa in a quartz batch
reactor. Nevertheless, nickel is much less expensive compared to the noble metal
and hence nickel shows more promise for large-scale hydrogen production from
biomass SCWG.
In continuous SCWG processes, similar effects as those of batch processes were
observed. In a continuous SCWG process, the catalyst (both the support and the
active metal) must be stable enough in supercritical water environment, not to
mention it should have reasonable activity and selectivity to the target gas products
(H2 , CH4 and CO). Although numerous studies have been reported but only a few
are dealing with SCWG of real biomasses in continuous reactors. In a flow reactor,
addition of K2 CO3 increases the gasification efficiency but decreases CO yield [84].
Jarana et al. [67] reported that hydrogen production was at maximum in SCWG
using KOH catalyst. They concluded that the rate of the water-gas shift reaction
increases by alkali salts (most probably as an acid–base catalyst). Pei et al. [62]
tested a variety of catalysts: Raney-Ni, Ca(OH)2 , Na2 CO3 , K2 CO3 , NaOH, KOH,
LiOH, ZnCl2 , dolomite (CaMg(CO3 )2 ) and olivine ((Mg.Fe)2SiO4 ). In presence of
Raney-Ni catalyst, complete gasification was obtained with the maximum hydrogen
yield up to 28.03 g/kg biomass at 400 ı C and 24 MPa [62]. Chen et al. [63]
used NaOH as a catalyst for gasification of wheat stalk, and found that hydrogen
yield was increased. Byrd et al. [64] investigated hydrogen production in catalytic
gasification of switchgrass biocrude in supercritical water employing nickel, cobalt,
and ruthenium catalysts supported on titania, zirconia, and magnesium aluminum
spinel supports. Their findings can be summarized as follows:
1. MgAl2 O4 was not a suitable catalyst support for biocrude reforming as catalysts
supported on MgAl2 O4 were charred immediately;
2. Higher conversion of biocrude was achieved with ZrO2 -supported catalysts than
TiO2 -supported ones. However, TiO2 supports produced the smallest amount of
char;
3. Ni/ZrO2 produced the highest hydrogen yield whereas Ru/ZrO2 yielded the
lowest hydrogen;
4. When using Ru/ZrO2 in SCWG, a great amount of char was formed accounting
for its lowest gasification efficiency.
Rönnlund et al. [65] studied effects of KOH, K2 CO3 , NaOH and black liquor on
SCWG of paper mill sludge. Similar catalytic effects were observed when adding
black liquor to the sludge. This finding paved the way of improving gasification
yield without expensive catalysts.
Figure 8.5 shows different reaction conditions depending on the desired gas
product (H2 or CH4 ) in biomass SCWG. Minowa et al. [32] demonstrated that
methane was the main product when the process temperature was between 374 and
500 ı C at above the critical pressure, whereas hydrogen was the main product when
the process temperature was over 500 ı C under pressure above the critical pressure.
Fig. 8.5 Reaction conditions

for different main products of
SCWG
It is a well-known fact that alkali salts could aid in splitting of C–C bonds, which
in turns assist to produce more H2 during the SCWG. For instance, the presence of
K2 CO3 can catalyze WGS reaction in the following mechanism, leading to a higher
yield of H2 and CO2 [85].
K2 CO3 C H2 O ! KHCO3 C KOH (8.6)
KOH C CO ! HCOOK (8.7)
HCOOK C H2 O ! KHCO3 C H2 (8.8)
2KHCO3 ! CO2 C K2 CO3 C H2 O (8.9)
8.8 Effects of Reactor Wall Properties on SCWG
Most of the reactors used for batch and continuous flow SCWG processes were
made of metal alloys: such as stainless steel 316L [27, 30, 35, 36, 38–40, 42, 44–46,
50] and Inconel 625 [41, 42, 47, 51]. Only a few were made of quartz reactor [43,
49, 86]. SS 316L is an alloy of iron, chromium (16.00–18.00 wt%), nickel (10.00–
14.00 wt%) and molybdenum (2.00–3.00 wt%) [87]. As most of the supercritical
water reactor materials are alloy, the inner wall of the reactor might catalyze the
SCWG reactions. It is well known that water becomes corrosive at its supercritical
state. As a result, any nickel present in the reaction process, either as a catalyst
loaded or part of reactor material will affect the reaction process. Kersten et al.
[43] showed that metal reactors had higher carbon conversion and water-gas shift
activity due to the catalytic effects of the metallic reactor wall, compared with quartz
reactors. Resende and Savage [88] reported that the total gas yield and the H2 mole
fraction from SCWG of biomass were lower in a quartz reactor than in a stainless
steel reactor at moderate water densities (0.07 g/cm3 ), which proved the catalytic
Fig. 8.6 Backscattered electron image of corrosion product on inner diameter of used Hastelloy
reactor, at 150 magnification, showing analysis points 1–8. The unaltered tube metal is the
brightest layer at the top of the image (Reprinted with permission from [55]. Copyright © 2000,
effects of the reactor wall on both the formation rate and composition of the gas
products. However, the influence of reactor metal surfaces was found to be relatively
less significant in the process that is more affected by other reaction conditions, in
particular reaction temperature.
Boukis et al. [89] studied methanol reforming in supercritical water in a flow-
reactor made of Inconel 625. They found chromium (Cr) and molybdenum (Mo)
dissolved in the effluent and the concentration of Cr and Mo in the effluent increased
with increasing reaction temperature. With further experimentation, only Ni was
left on the inner wall. Antal et al. [55] studied SCWG with a reactor made of
Hastelloy C-276 which is composed of mainly Ni and Fe, as well as Mo, Cr and
traces of Co (cobalt), W (tungsten) and Mn (manganese). They observed that the
experiments after the second run produced less amount of hydrogen but more
methane. It was found out that the deposition of carbon on the reactor inner wall
made the difference in the products because it reduced the catalytic effect of the
reactor wall. The catalytic effect was restored by using hydrogen peroxide (H2 O2 )
washing after each reaction. They analyzed ash from the reactor residues after
gasification, and realized that ash from the reactor residues contains almost all types
of the elements of the Hastelloy reactor, which suggests that metals of the Hastelloy
alloy reactor were leached out during the SCWG. SEM analysis of the Hastelloy
reactor tubing after gasification process showed heavily corroded surface. Figure 8.6
shows backscattered electron microscope image displaying two distinct layers. The
darkest and porous part represents the outer most part of the inner wall which was
the nearest to the reactants, and was rich in feedstock ash. The farthest part can be
attributed to the solid residues from the reactants, and the brightest part at the inner
side of the inner wall is the un-reacted base metal of Hastelloy. Due to the corrosion
at the innermost wall, all metals except iron contained in the Hastelloy alloy reactor
wall could undergo depletion by leaching after a long time-on-stream.
8.9 SCWG Operating Challenges and Possible Solutions
8.9.1 Batch Reactors
There is a concern about the high operating temperature and pressure in SCWG.
The major drawback of a batch reactor is poor upscalability. Batch reactors are
less economical for commercial/industrial production of bulk products. Another
problem with batch reactors for SCWG of biomass is associated with reactor
cleaning due to the polymerization of the decomposed products particularly at
higher substrate concentrations [90]. Capillary quartz tubes as high-pressure batch
reactors used for SCWG reactions are cheap and safe, but such reactor is only
suitable for bench-scale testing due to its extremely small size. Moreover, the small
diameter of a capillary quartz tube reactor led to non-uniform distribution of the
catalyst inside the reactor thus affecting the product yield [91]. Another major
disadvantage of batch reactors compared with continuous reactors is their inherent
slower heating rate that could reduce the gasification efficiency due to enhanced
polymerization or charring reactions at a low heating rate [30].
8.9.2 Continuous Reactors
SCWG in continuous flow reactors is a promising biomass to energy processing

technology. However, due to its high operating temperature and pressure, no
commercial-scale process has been built and operated so far. However, a pilot plant
called “VERENA” was built in Germany. The pilot plant VERENA (a German
acronym for “experimental facility for the energetic exploitation of agricultural
matter”) is in operation since 2003 with a total throughput of 100 kg/h and
can handle biomass-water slurry containing a maximum of 20 wt% dry biomass.
The maximum designed operating pressure is up to 35 MPa and a maximum
designed temperature is 700 ı C. Usually the common operating conditions for the
“VERENA” plant are: 50 or 100 kg/h feeding rate, up to 660 ı C and 28 MPa
[92, 93].
Some key challenges for SCWG processes are summarized as follows:
(a) In a supercritical water reactor, a biomass feed would take a relatively long
time for being heated up. The lower heating rate would promote formation of
furfurals and other unsaturated intermediate complex products, which tend to
polymerize at higher temperatures to generate char, tar, and coke. This degrades
the product gas quality and reduces the overall efficiency of the process [71, 94].
(b) Matsumura et al. [95] suggested for the economics of the SCWG process, it
is very important to have heat exchange between the reactor effluent and the
reactor feed by considering the heat balance. However, heat recovery from the
high-pressure corrosive effluent and use it for the preheating of the feed material
is another design problem as it would pose challenges for heat exchangers

design with respect to materials selection and pressure rating for the heat
exchangers.
(c) Continuous processes are commonly facing severe reactor plugging problem
[55, 60]. This plugging causes an acute problem that would require forcefully
shutting down the whole system [95]. Reactor plugging was mainly caused
by the formation of char and tars in a flow reactor. Due to remarkably low
solubility of the polar inorganic salts present in the SCWG reactors, either from
the feedstock or from the reactor, they precipitate inside the reactor, forming
agglomerates and coating on the inner reactor wall, which hinder heat transfer
through the reactor wall. If the deposition/coating is overlooked and allowed to
grow inside, there is a possibility of plugging of the reactor. These in turn lead
to tedious and hectic cleaning process, and hence reduce the overall efficiency
of the SCWG process [1, 96, 97].
(d) Feeding real biomass into a SCWG reactor is very important for a continuous
process and has been a long-standing technological challenge for the develop-
ment of SCWG technology. Dry biomasses are hard to dissolve into water to
prepare gelatinous solution for the feeding purpose. A biomass pretreatment
method using hot compressed water treatment (at 150 ı C for 30 min), so-
called ‘softening’ was successfully applied by Matsumura et al. [88] for feeding
cabbage to a SCWG reactor. Antal et al. [55] used corn starch gel to mix sawdust
and successfully delivered the mixture into a SCWG reactor by using a ‘cement
pump’. Hao et al. [68] successfully fed solid biomass feedstock continuously
into a SCWG reactor using sodium carboxymethylcellulose (CMC) as surfac-
tant in mixing the solid biomass feedstock and water. D’Jesus et al. [60] used
xanthan successfully to feed biomass continuously in to a SCWG reactor.
8.10 Conclusions
The following conclusions can be drawn from this literature analysis, focused on
effects of reactor types, operating conditions and challenges for SCWG of real wet
biomass:
1. The process parameters need to optimize in order to achieve better process
efficiency and economics for various types of feed and concentration.
2. There is a need to develop inexpensive, long lasting, more active, more selective
and durable catalysts for the gasification process.
3. Reactor plugging is one of the major operating issues that needs sufficient amount
of attention. This issue can be addressed by using proper feeding mechanism and
employing some novel biomass pre-treatment methods.
4. There is very limited literature work available on kinetics studies of a SCWG
process [82, 83, 98–100]. The development of kinetic model requires a good
formulation of the elementary steps from the feed materials to the products.
However, a more elaborate scheme of reactions describing the steps is necessary

to explain the complex reactions that occur during the gasification process.
Hence, a significant amount of research in this regard is urgently needed.
5. Reactor corrosion is another major issue of SCWG processes.
Acknowledgements The authors are grateful for the financial support from NSERC/FPInnova-
tions Industrial Research Chair Program in Forest Biorefinery and the Ontario Research Fund-
Research Excellence (ORF-RE) from Ministry of Economic Development and Innovation. Support
from the industrial partners including FPInnovations, Arclin Canada, BioIndustrial Innovation
Centre is also acknowledged. The authors (C. Xu and A. Dutta) also acknowledge the funding
from NSERC via the Discovery Grant Program.
References
1. Guo L, Cao C, Lu Y. Supercritical water gasification of biomass and organic wastes.

In: Momba MNB, editors. Biomass. ISBN: 978-953-307-113-8, InTech; 2010. Avail-
able from: http://www.intechopen.com/books/biomass/supercritical-water-gasification-of-
biomass-and-organic-wastes
2. Kannan MP, Richards GN. Gasification of biomass chars in carbon dioxide: dependence of
gasification rate on the indigenous metal content. Fuel. 1990;69(6):747–53.
3. Hemmes K, de Groot A, den Uil H. BIO-H2 . Application potential of biomass related
hydrogen production technologies to the Dutch energy infrastructure of 2020–2050. ECN-
C-03-028, April 2003.
4. Ni M, Leung DYC, Leung MKH, Sumathy K. An overview of hydrogen production from
biomass. Fuel Process Technol. 2006;87(5):461–72.
5. Buhler W, Dinjus E, Ederer HJ, Kruse A, Mas C. Ionic reactions and pyrolysis of glyc-
erol as competing reaction pathways in near- and supercritical water. J Supercrit Fluid.
2002;22(1):37–53.
6. Penninger JML, Kersten RJA, Baur HCL. Reactions of diphenylether in supercritical water-
mechanism and kinetics. J Supercrit Fluid. 1999;16:119–32.
7. Leible L, Kälber S, Kappler G, Lange S, Nieke E, Wintzer D Fürniss B. Competitiveness
and CO2 mitigation costs of biogenic residues and waste for heat and power production. In:
Bridgwater AV, Boocock DGB, editors. Science in thermal and chemical biomass conversion.
UK: CPL press; vol. 2. 2006. p. 1730.
8. ECN. Energy Research Centre of the Netherlands, General information, PHYLLIS, the
composition of biomass and waste. 2007. http://www.ecn.nl/phyllis/DataTable.asp
9. Demirbaş A. Biomass resource facilities and biomass conversion processing for fuels and
chemicals. Energ Convers Manage. 2001;42(11):1357–78.
10. McKendry P. Energy production from biomass (part 2): conversion technologies. Bioresour
Technol. 2002;83(1):47–54.
11. Faaij A. Modern biomass conversion technologies. Mitig Adapt Strat Glob. 2006;11(2):
335–67.
12. Jenkins BM, Baxter LL, Miles Jr TR, Miles TR. Combustion properties of biomass. Fuel
Process Technol. 1998;54(1):17–46.
13. Bobleter O, Binder H. Dynamic hydrothermal degradation of wood. Holzforschung.
1980;34:48–51.
14. Bouchard J, Nguyen TS, Chornet E, Overend RP. Analytical methodology for biomass
pretreatment. Part 1: solid residues. Biomass. 1990;23(4):243–61.
15. Bouchard J, Nguyen TS, Chornet E, Overend RP. Analytical methodology for biomass
pretreatment. Part 2: characterization of the filtrates and cumulative distribution as a function
of treatment severity. Bioresour Technol. 1991;36(2):121–31.
16. Mok WSL, Antal Jr MJ. Uncatalyzed solvolysis of whole biomass hemicellulose by hot
compressed liquid water. Ind Eng Chem Res. 1992;31(4):1157–61.
17. Allen S, Kam LC, Zemann AJ, Antal MJ. Fractionation of sugar cane with hot, compressed,
liquid water. Ind Eng Chem Res. 1996;35(8):2709–15.
18. Bobleter O, Concin R. Degradation of poplar lignin by hydrothermal treatment. Cell Chem
Technol. 1979;13:583–93.
19. Antal MJ, Mok WS, Richards GN. Mechanism of formation of 5-(hydroxymethyl)-2-
furaldehyde from D-fructose and sucrose. Carbohydr Res. 1990;199:91–109.
20. Antal MJ, Mok WS, Richards GN. Four-carbon model compounds for the reactions of sugars
in water at high temperature. Carbohydr Res. 1990;199:111–5.
21. Antal MJ, Leesomboon T, Mok WS, Richards GN. Mechanism of formation of 2-furaldehyde
from D-xylose. Carbohydr Res. 1991;217:71–85.
22. Holgate HR, Meyer JC, Tester WJ. Glucose hydrolysis and oxidation in supercritical water.
AIChE J. 1995;41(3):637–48.
23. Modell M. Gasification and liquefaction of forest products in supercritical water. In: Overand
RP, Milne TA, Mudge LK, editors. Fundamentals of thermochemical biomass conversion.
London: Elsevier Applied Science Publishers, Ltd.; 1985. p. 95–119.
24. Amin S, Reid RC, Modell M. Reforming and decomposition of glucose in an aqueous phase.
In: Intersociety conference on environmental systems, San Francisco, CA, 21–24 July; 1975.
ASME 75-ENAs-21.
25. Dinjus E, Kruse A. Hot compressed water – a suitable and sustainable solvent and reaction
medium? J Phys Condens Matter. 2004;16(14):161–9.
26. Bazargan M, Fraser D, Chatoorgan V. Effect of buoyancy on heat transfer in supercritical
water flow in a horizontal round tube. J Heat Transf. 2005;127(8):897–902.
27. Demirbas A. Hydrogen production from biomass via supercritical extraction. Energ Source.
2005;27(15):1409–17.
28. Yanik J, Ebale S, Kruse A, Saglam M, Yüksel M. Biomass gasification in supercritical water:
part 1. Effect of the nature of biomass. Fuel. 2007;86(15):2410–5.
29. Hrabovsky M, Konrad M, Kopecky V, Hlina M, Kavka T. Gasification of biomass in
water/gas-stabilized plasma for syngas production. Czech J Phys. 2006;56(2):B1199–206.
30. Venkitasamy C, Hendry D, Wilkinson N, Fernando L, Jacoby W. Investigation of ther-
mochemical conversion of biomass in supercritical water using a batch reactor. Fuel.
2011;90(8):2662–70.
31. Minowa T, Ogi T, Dote Y, Yokoyama S. Methane production from cellulose by catalytic
gasification. Renew Energy. 1994;5(5):813–5.
32. Minowa T, Fang Z, Ogi T. Cellulose decomposition in hot – compressed water with alkali or
nickel catalyst. J Supercrit Fluid. 1998;13:253–9.
33. Elliott D, Sealock L, Baker E. Chemical processing in high pressure aqueous environment. 3.
Batch reactor process development experiments for organic destruction. Ind Eng Chem Res.
1994;33(3):558–65.
34. Elliott D, Phelps M, Sealock L, Baker E. Chemical processing in high – pressure aqueous
environment 4. Continuous – flow reactor process development experiments for organics
destruction. Ind Eng Chem Res. 1994;33(3):566–74.
35. Schmieder H, Abeln J, Boukis N, Dinjus E, Kruse A, Kluth M, Petrich G, Sadri E, Schacht
M. Hydrothermal gasification of biomass and organic wastes. J Supercrit Fluid. 2000;17(2):
145–53.
36. Guo LJ, Lu YJ, Zhang XM, Ji CM, Guan Y, Pei AX. Hydrogen production by biomass
gasification in supercritical water: a systematic experimental and analytical study. Catal
Today. 2007;129(3):275–86.
37. Yanagida T, Minowa T, Nakamura A, Matsumura Y, Noda Y. Behavior of inorganic elements
in poultry manure during supercritical water gasification. 日本エネルギー学会誌 (J Jpn Inst
Energ). 2008;87(9):731–6.
38. Demirbas A. Hydrogen-rich gas from fruit shells via supercritical water extraction. Int
J Hydrogen Energ. 2004;29(12):1237–43.
39. Hao X, Guo L, Zhang X, Guan Y. Hydrogen production from catalytic gasification of cellulose
in supercritical water. Chem Eng J. 2005;110(1–3):57–65.
40. Xu ZR, Zhu W, Htar SH. Partial oxidative gasification of municipal sludge in subcritical and
supercritical water. Environ Technol. 2012;33(11):1217–23.
41. Yanik J, Ebale S, Kruse A, Saglam M, Yüksel M. Biomass gasification in supercritical water:
II. Effect of catalyst. Int J Hydrogen Energ. 2008;33(17):4520–6.
42. Lu YJ, Guo LJ, Ji CM, Zhang XM, Hao XH, Yan QH. Hydrogen production by biomass
gasification in supercritical water: a parametric study. Int J Hydrogen Energ. 2006;31(7):
822–31.
43. Kersten SR, Potic B, Prins W, Van Swaaij WP. Gasification of model compounds and wood
in hot compressed water. Ind Eng Chem Res. 2006;45(12):4169–77.
44. Xu ZR, Zhu W, Li M. Influence of moisture content on the direct gasification of dewatered
sludge via supercritical water. Int J Hydrogen Energ. 2012;37(8):6527–35.
45. Azadi P, Khan S, Strobel F, Azadi F, Farnood R. Hydrogen production from cellulose, lignin,
bark and model carbohydrates in supercritical water using nickel and ruthenium catalysts.
Appl Catal B Environ. 2012;117–118:330–8.
46. Madenoğlu TG, Kurt S, Sağlam M, Yüksel M, Gökkaya D, Ballice L. Hydrogen production
from some agricultural residues by catalytic subcritical and supercritical water gasification.
J Supercrit Fluid. 2012;67:22–8.
47. Kruse A, Maniam P, Spieler F. Influence of proteins on the hydrothermal gasification and
liquefaction of biomass. 2. Model compounds. Ind Eng Chem Res. 2007;46(1):87–96.
48. Stucki S, Vogel F, Ludwig C, Haiduc AG, Brandenberger M. Catalytic gasification of
algae in supercritical water for biofuel production and carbon capture. Energy Environ Sci.
2009;2(5):535–41.
49. Sricharoenchaikul V. Assessment of black liquor gasification in supercritical water. Bioresour
Technol. 2009;100(2):638–43.
50. Guan Q, Savage PE, Wei C. Gasification of alga Nannochloropsis in supercritical water.
J Supercrit Fluid. 2012;61:139–45.
51. Yamamura T, Mori T, Park KC, Fujii Y, Tomiyasu H. Ruthenium (IV) dioxide-catalyzed
reductive gasification of intractable biomass including cellulose, heterocyclic compounds,
and sludge in supercritical water. J Supercrit Fluid. 2009;51(1):43–9.
52. Matsumura Y, Minowa T. Fundamental design of a continuous biomass gasification process
using a supercritical water fluidized bed. Int J Hydrogen Energ. 2004;29(7):701–7.
53. Manarungson S, Mok WS, Antal MJ. Hydrogen production by gasification of glucose and
wet biomass in supercritical water. In: Veriziroglu T, Takahashi PK, editors. Hydrogen energy
progress VIII. Hawaii: Pergamon Press; 1990. p. 345–55.
54. Xu X, Antal MJ. Gasification of sewage sludge and other biomass for hydrogen production
in supercritical water. Environ Prog. 1998;17(4):215–20.
55. Antal MJ, Allen SG, Schulman D, Xu XD, Divilio RJ. Biomass gasification in supercritical
56. Matsumura Y. Evaluation of supercritical water gasification and biomethanation for wet
biomass utilization in Japan. Energ Convers Manage. 2002;43(9–12):1301–10.
57. Matsumura Y, Harada M, Li D, Komiyama H, Yoshdda Y, Ishttani H. Biomass gasification
in supercritical water with partial oxidation. Nihon Enerugi Gakkaishi (J Jpn Inst Energ).
2003;82(12):919–25.
58. Lu YJ, Ji CM, Guo LJ. Experimental investigation on hydrogen production by agricultural
biomass gasification in supercritical water. J Xi’an Jiaotong Univ. 2005;39(3):239–42.
59. Lu YJ, Guo LJ, Hao XH, Ji CM. Influence of main parameters on wood sawdust gasification
in supercritical water [J]. CIESC J. 2004;55(12):2060–6.
60. D’Jesús P, Boukis N, Kraushaar-Czarnetzki B, Dinjus E. Gasification of corn and clover grass
in supercritical water. Fuel. 2006;85(7):1032–8.
61. Matsumura Y, Harada M, Nagata K, Kikuchi Y. Effect of heating rate of biomass feedstock
on carbon gasification efficiency in supercritical water gasification. Chem Eng Commun.
2006;193(5):649–59.
62. Pei A, Zhang R, Jin H, Guo L. Research on catalysts and their catalytic characteristics for
hydrogen production by gasification of peanut shell in supercritical water. J Xi’an Jiaotong
Univ. 2008;7:027.
63. Chen J, Lu Y, Guo L, Zhang X, Xiao P. Hydrogen production by biomass gasification
in supercritical water using concentrated solar energy: system development and proof of
concept. Int J Hydrogen Energ. 2010;35(13):7134–41.
64. Byrd AJ, Kumar S, Kong L, Ramsurn H, Gupta RB. Hydrogen production from cat-
alytic gasification of switchgrass biocrude in supercritical water. Int J Hydrogen Energ.
2011;36(5):3426–33.
65. Rönnlund I, Myréen L, Lundqvist K, Ahlbeck J, Westerlund T. Waste to energy by industrially
integrated supercritical water gasification–Effects of alkali salts in residual by-products from
the pulp and paper industry. Energy. 2011;36(4):2151–63.
66. Lu Y, Guo L, Zhang X, Ji C. Hydrogen production by supercritical water gasification of
biomass: explore the way to maximum hydrogen yield and high carbon gasification efficiency.
Int J Hydrogen Energ. 2011;37(4):3177–85.
67. Jarana G, Sánchez-Oneto J, Portela JR, Nebot Sanz E, Martínez de la Ossa EJ. Supercritical
water gasification of industrial organic wastes. J Supercrit Fluid. 2008;46(3):329–34.
68. Hao XH, Guo LJ, Mao X, Zhang XM, Chen XJ. Hydrogen production from glucose used
as a model compound of biomass gasified in supercritical water. Int J Hydrogen Energ.
2003;28(1):55–64.
69. Lu YJ, Jin H, Guo LJ, Zhang XM, Cao CQ, Guo X. Hydrogen production by biomass
gasification in supercritical water with a fluidized bed reactor. Int J Hydrogen Energ.
2008;33(21):6066–75.
70. Antal MJ. Hydrogen and food production from nuclear heat and municipal wastes. In:
Veziroğlu TN, editor. Hydrogen energy. New York: Plenum Press; 1975. p. 331–8.
71. Kruse A, Henningsen T, Sinag A, Pfeiffer J. Biomass gasification in supercritical water;
influence of the dry matter content and the formation of phenols. Ind Eng Chem Res.
2003;42(16):3711–7.
72. Chuntanapum A, Matsumura Y. Char formation mechanism in supercritical water gasification
process: a study of model compounds. Ind Eng Chem Res. 2010;49(9):4055–62.
73. Antal MJ. The effects of reactor severity on the gas phase pyrolysis of cellulose and kraft
lignin derived volatile matter. Ind Eng Chem Res. 1983;22:366–75.
74. Bühler W, Dinjus E, Ederer HJ, Kruse A, Mas C. Ionic reactions and pyrolysis of
glycerol as competing reaction pathways in near-and supercritical water. J Supercrit Fluid.
2002;22(1):37–53.
75. Sato T, Furusawa T, Ishiyama Y, Sugito H, Miura Y, Sato M, Suzuki N, Itoh N. Effect of
water density on the gasification of lignin with magnesium oxide supported nickel catalysts
in supercritical water. Ind Eng Chem Res. 2006;45(2):615–22.
76. Jin H, Lu Y, Guo L, Cao C, Zhang X. Hydrogen production by partial oxidative gasifi-
cation of biomass and its model compounds in supercritical water. Int J Hydrogen Energ.
2010;35(7):3001–10.
77. Madenoğlu TG, Sağlam M, Yüksel M, Ballice L. Simultaneous effect of temperature and
pressure on catalytic hydrothermal gasification of glucose. J Supercrit Fluid. 2012;73:
151–60.
78. Uematsu M, Frank EU. Static dielectric constant of water and steam. J Phys Chem Ref Data.
1980;9(4):1291–306.
79. Miller A, Hendry D, Wilkinson N, Venkitasamy C, Jacoby W. Exploration of the gasification
of Spirulina algae in supercritical water. Bioresour Technol. 2012;119:41–7.
80. Sutton D, Kelleher B, Ross JRH. Review of literature on catalysts for biomass gasification.
Fuel Process Technol. 2001;73(3):155–73.
81. Elliott DC. Catalytic hydrothermal gasification of biomass. Biofuel Bioprod Biorefin.
2008;2(3):254–65.
82. Calzavara Y, Joussot-Dubien C, Boissonnet G, Sarrade S. Evaluation of biomass gasi-
fication in supercritical water process for hydrogen production. Energ Convers Manage.
2005;46(4):615–31.
83. Guo Y, Wang SZ, Xu DH, Gong YM, Ma HH, Tang XY. Review of catalytic super-
critical water gasification for hydrogen production from biomass. Renew Sust Energ Rev.
2010;14(1):334–43.
84. Kruse A, Krupka A, Schwarzkopf V, Gamard C, Henningsen T. Influence of proteins on the
hydrothermal gasification and liquefaction of biomass. 1. Comparison of different feedstocks.
Ind Eng Chem Res. 2005;44(9):3013–20.
85. Sinag A, Kruse A, Schwarzkopf V. Key compounds of the hydropyrolysis of glucose in
supercritical water in the presence of K2 CO3 . Ind Eng Chem Res. 2003;42(15):3516–21.
86. Chakinala AG, Brilman DW, van Swaaij WP, Kersten SR. Catalytic and non-catalytic
supercritical water gasification of microalgae and glycerol. Ind Eng Chem Res. 2009;49(3):
1113–22.
87. Bringas JE. Handbook of comparative world steel standards. 3rd ed. West Conshohocken:
ASTM International; 2004. p. 4. ISBN 0-8031-3362-6.
88. Resende FL, Savage PE. Expanded and updated results for supercritical water gasification of
cellulose and lignin in metal-free reactors. Energ Fuel. 2009;23(12):6213–21.
89. Boukis N, Diem V, Habicht W, Dinjus E. Methanol reforming in supercritical water. Ind Eng
Chem Res. 2003;42(4):728–35.
90. Yu D, Aihara M, Antal Jr MJ. Hydrogen production by steam reforming glucose in
supercritical water. Energ Fuel. 1993;7(5):574–7.
91. Manzour PA. Hydrogen production using catalytic supercritical water gasification of ligno-
cellulosic biomass. PhD thesis, University of Toronto; 2012.
92. Boukis N, Galla U, D’Jesus P, Muller H, Dinjus E. Gasification of wet biomass in supercritical
water: results of pilot plant experiments. In: 14th European biomass conference, 17–21 Oct,
Paris, France; 2005. pp. 964–7.
93. Boukis N, Galla U, Müller H, Dinjus E. Biomass gasification in supercritical water.
Experimental progress achieved with the VERENA pilot plant. In: 15th European conference
& exhibition, 7–11 May, Berlin, Germany 7; 2007. pp. 1013–6
94. Kruse A. Supercritical water gasification. Biofuel Bioprod Biorefin. 2008;2(5):415–37.
95. Matsumura Y, Minowa T, Potic B, Kersten SRA, Prins W, van Swaaij WPM, van de Beld
B, Elliott DC, Neuenschwander GG, Kruse A, Antal MJ. Biomass gasification in near- and
super-critical water: status and prospects. Biomass Bioenerg. 2005;29(4):269–92.
96. Hodes M, Marrone PA, Hong GT, Smith KA, Tester JW. Salt precipitation and scale control
in supercritical water oxidation—Part A: fundamentals and research. J Supercrit Fluid.
2004;29(3):265–88.
supercritical water oxidation—Part B: commercial/full-scale applications. J Supercrit Fluid.
2004;29(3):289–312.
98. Loppinet-Serani A, Aymonier C, Cansell F. Current and foreseeable applications of supercrit-
ical water for energy and the environment. ChemSusChem. 2008;1(6):486–503.
99. Peterson AA, Vogel F, Lachance RP, Fröling M, Antal Jr MJ, Tester JW. Thermochemical
biofuel production in hydrothermal media: a review of sub-and supercritical water technolo-
gies. Energ Environ Sci. 2008;1(1):32–65.
100. Waldner MH, Vogel F. Renewable production of methane from woody biomass by catalytic
hydrothermal gasification. Ind Eng Chem Res. 2005;44(13):4543–51.
Part III
Near-Critical and Supercritical
Water Applications
Chapter 9
Production of Renewable Solid Fuel Hydrochar
from Waste Biomass by Sub- and Supercritical
Water Treatment
Zhengang Liu, Rajasekhar Balasubramanian, and S. Kent Hoekman
Abstract Raw biomass feedstocks are not ideal fuels because of their inherent
properties. Raw biomass combustion alone, or co-combustion with low rank coal,
encounters serious problems in existing power plants. This chapter discusses sub-
and supercritical water treatment (SSCWT) to upgrade biomass feedstocks, and
presents a systematic characterization of fuel properties of the resultant hydrochars.
Hydrochars from SSCWT of biomass have significantly improved fuel quality
compared to raw biomass; the quality of some hydrochars is even similar to or higher
than that of lignite. Compared to parent biomass, the hydrochars have increased
carbon content, elevated heating value and reduced ash content. The hydrochars
have increased ignition temperatures and higher combustion temperature regions
compared to raw biomass feedstocks. In addition, due to the significantly reduced
ash content, ash-related problems from hydrochar combustion are expected to be
mitigated. In comparison to raw biomass, a higher percentage of nitrogen is retained
in the char, and less NH3 and HCN are formed during pyrolysis, indicating that an
additional benefit of reducing emissions of nitrogen pollutants can be achieved by
replacing raw biomass with hydrochar during energy production.
Keywords Biomass energy • Fuel quality • Combustion characteristics • Ash

problem • Kinetics • Pellet fuel • Pollutant emission
Z. Liu • R. Balasubramanian ()

Department of Civil and Environmental Engineering, National University of Singapore,
1 Engineering Drive 2, E1A 07-03, Singapore 117576, Singapore
S.K. Hoekman
Division of Atmospheric Sciences, Desert Research Institute, 2215 Raggio Parkway,
Reno, NV 89512, USA
232 Z. Liu et al.
9.1 Introduction
Lignocellulosic biomass, the most abundant natural organic material on earth, has
enormous potential as a feedstock for the production of fuels, heat and electrical
power. The attractiveness of lignocellulosic biomass has increased in recent years
due to rapid depletion of conventional fossil fuels and growing concerns of
environmental pollution and climate change. Among various utilization methods,
biomass combustion alone or co-combustion of biomass with low rank coals for
heat and energy production is considered to be a promising short-term solution for
biomass energy recovery and substantial reduction of air pollutants (CO2 , NOx , CO
and SO2 ) per unit energy generated [1]. Much work has been done on co-combustion
of biomass and coal [2, 3]. However, several inherent properties of biomass limit its
widespread applications in energy generation. For example, impurities in biomass,
particularly alkali metals and halogens, can cause operational problems with regard
to slagging, fouling and corrosion [4, 5]. It has also been reported that addition
of biomass to coal impacts flame characteristics, particularly flame front and
brightness, and lowers flame stability [6]. These effects vary with the physical and
chemical properties of the biomass feedstock. Other disadvantages such as seasonal
availability, poor grindability, and high degree of heterogeneity in physical shape,
size and density among different feedstocks also largely limit the use of biomass as
a direct fuel [7]. The physical and chemical differences between biomass and coal,
coupled with a wide diversity of biomass properties, result in a range of techno-
logical challenges for biomass combustion, or co-combustion with coal in existing
coal-fired power plants. To overcome these problems, it is imperative to pre-treat
the biomass to upgrade its fuel properties such that it more closely resembles coal,
and can be utilized within existing power plants and fuel handling infrastructure.
Sub- and supercritical water treatment (SSCWT) has exhibited superior char-
acteristics for the production of functional carbonaceous material from biomass.
Several reviews have appeared on this research topic [8, 9]. SSCWT has been mostly
studied on a limited number of feedstocks, ranging from pure substances to slightly
more complex biomass types such as wood, with an emphasis on nanostructure
generation. Now, there is a considerable interest to increase the quality of the solid
product (hydrochar) and to enhance its suitability as a solid biofuel.
SSCWT is an environmentally friendly route using water as solvent, and carried
out in a sealed reaction container under controlled temperature. This technology has
been in use for simulating natural coalification in coal petrology for nearly a century.
SSCWT is one of the important upgrading processes for biomass feedstocks,
offering significant advantages for biomass conversion including avoidance of an
energy-intensive drying process, high conversion efficiency, and relatively low
operation temperatures. The key advantage of SSCWT is that it can convert wet
input material into carbonaceous solids at relatively high yield without the need
for energy-intensive drying before or during the process. This opens up the field
of potential feedstocks to a variety of nontraditional sources: wet animal manures,
human wastes, sewage sludge, municipal solid waste, as well as aquaculture and
algal residues.
9 Production of Renewable Solid Fuel Hydrochar from Waste Biomass. . . 233
SSCWT of biomass generates three product streams: liquid (bio-oil), gaseous

and solid products (hydrochar). The distributions and properties of these products
are strongly dependent on treatment conditions. Temperature plays a vital role in
biomass conversion under sub- and supercritical water conditions. When process
temperatures of around 220 ı C and corresponding pressures of up to 20 bar are
maintained, little gas (1–5 %) is generated, and most organics remain as water-
soluble products or are transformed into solids. At increased temperatures, more
liquid hydrocarbon is formed and more gas is produced, commonly referred to as
hydrothermal liquefaction and hydrothermal gasification. If the supercritical state
for water is reached, the primary product is gaseous. Currently much attention is
paid to liquid and gaseous biofuel production from waste biomass. However, the
production of bio-oil and gaseous biofuel faces several technical challenges and
is not economically competitive with fossil energy. Therefore, the production of
hydrochar to be used as solid fuel seems a feasible option for energy recovery from
renewable biomass. However, there has been little work done exploring the SSCWT
of waste biomass to produce hydrochar. This may arise partly from the fact that
even coal is not an ideal fuel form. Currently, coal is still a dominant fuel, therefore,
upgrading of biomass feedstocks to coal-like products is a more feasible option due
to their low fuel quality compared to coal.
To maximize the hydrochar yield, subcritical water treatment of biomass is
generally used. Such a process is known by several names, including hydrothermal
pre-treatment, hot compressed water treatment, subcritical water treatment and
hydrothermal carbonization (HTC). In this chapter, the name HTC is used because
this appears to be the most widely accepted terminology. HTC was firstly introduced
by Bergius as early as 1913 and was employed to transform cellulose into coal-like
materials from biomass [10, 11].
Several reviews have been published about the HTC of waste biomass from
the viewpoint of chemical mechanisms [10, 12]. This chapter is a brief summary
of production of the solid fuel hydrochar from waste biomass, mainly based
on our recent work on this topic. This work highlights the fuel qualities of
the hydrochars resulting from HTC of waste biomass including energy density,
combustion characteristics and ash-related problems. In addition, the environmental
impacts of replacing raw biomass with hydrochar during pyrolysis are demonstrated
in this chapter.
9.2 Physicochemical Properties of Hydrochar
9.2.1 Surface Area
Surface area is a significant parameter, since it may strongly influence the

hydrochar’s combustion reactivity [13]. Due to decomposition of constituents in the
biomass, the hydrochars produced from HTC have increased surface area compared
234 Z. Liu et al.
to raw material, although they are still classified as non-porous or macroporous

solids. In general, the surface area of hydrochar increases with increasing process
temperature, reaching a maximum around 300 ı C, and then decreases with further
increasing temperature. The decreased surface area at high temperatures is due
to coalescence of the pores and formation of plastic-like hydrochar as the result
of melting. As an example, the hydrochar from pinewood under HTC has been
determined systematically. The surface area of the hydrochar was small (9.65 m2 /g)
at 250 ı C, but increased dramatically to 20.43 m2 /g when the temperature increased
from 250 ı C to 300 ı C. However, the surface area decreased to 15.02 m2 /g
when temperature was further increased beyond 300 ı C. In the case of palm
empty fruit bunches, the surface areas were 6.08, 8.03 and 2.04 m2 /g for the
hydrochars produced from HTC at 150, 250 and 350 ı C [14]. Because of the variety
of substrates and operating conditions, direct surface area comparison between
hydrochar and pyrolytic biochar is not possible. However, the BET surface areas
of hydrochars obtained by HTC processing exceed those of biochars obtained from
pyrolysis that are reported in literature [15–17]. The different porous structures are
ascribed to the different decomposition/carbonization processes of biomass under
HTC and pyrolysis. In addition, reduced ash content of hydrochar contributes to its
higher surface area compared to pyrolytic biochars [18, 19].
9.2.2 Surface Morphology
Hydrochar has a coarser surface than raw biomass, and shows unique surface
micrographs due to the decomposition/carbonization of biomass’s constituents.
Most of the cellulose and hemicellulose decompose into small fragments, even
into monomers of these polymers. At higher temperatures, these small fragments
and monomers are melted and carbonized, resulting in formation of carbonaceous
spheres. In contrast to cellulose and hemicellulose, the decomposition temperature
of lignin is higher than the carbonization temperature; therefore, the scaffold
structure is retained during the HTC process. Figure 9.1 shows the amorphous and
heterogeneous structure of hydrochar formed from pinewood under HTC (250, 300
and 350 ı C) and supercritical water conditions (400 ı C).
Pinewood has a smooth surface, although some slit-like fragments can be
observed. Only small vesicle structures are formed covering the external surface
at 250 ı C, while more significant changes are observed after treatment at 300 ı C.
At 350 ı C, the surface exposed to water is etched significantly and some rice-like
structure is formed and covers most of the surface. Larger porosity and honeycomb-
like structures are observed after treatment at 400 ı C, due to further thermal
decomposition of the pinewood. Some of the carbonaceous spheres originating
from the low meting point decomposition intermediates are also seen in the 400 ı C
hydrochar.
Fig. 9.1 Micrographs of hydrochar produced from hydrothermal carbonization of pinewood at

different temperatures
9.2.3 Chemical Composition
Characterization of the chemical and phase composition of a given solid fuel is

an important initial step during the investigation and application of such a fuel.
This composition determines the properties, quality, potential applications and
environmental problems related to the fuel. Biomass is a complex heterogeneous
236 Z. Liu et al.
mixture of organic matter and, to a lesser extent, inorganic matter, containing

various solids and fluids that are intimately associated. In general, hydrochar has
similar composition to its parent biomass, especially for hydrochars produced at
low temperature. Due to the decomposition/carbonization of organic constituents,
and the dilution of inorganics during HTC, the hydrochars are rich in amorphous
carbon. The inorganics in hydrochar are relatively low and are also in amorphous
phase compared to the parent biomass, although some crystalline structures of
inorganics can be formed under certain HTC conditions. For example, crystalline
phase SiO2 was formed in water hyacinth-derived hydrochars that were generated
at long residence times [20]. In addition, hydrochar surfaces are rich in functional
groups, which differ from pyrolytic biochars. For example, the amounts of oxygen-
containing groups (hydroxyl, carboxyl and carbonyl groups) were reported to have
increased by 98 % and 62 % upon HTC treatment (at 250 ı C) of pinewood and rice
husk, respectively [21]. Due to the presence of these functional groups, hydrochars
have been used as adsorbents for heavy metal removal from wastewater, and for soil
amendment [21–23].
9.3 Structure Evolution of Biomass Under HTC Conditions
Several reviews have appeared concerning the reaction mechanisms for biomass
feedstocks under HTC. Due to the complex composition of these feedstocks and
the variations in conditions of HTC, many reactions take place simultaneously, and
the process is very complicated. However, since lignocellulosic biomass is mostly
composed of hemicellulose, cellulose and lignin, investigation of the evolution of
these three main constituents can provide mechanistic insights into the HTC process.
FT-IR and 13 C NMR are two useful and widely used techniques for eval-
uating the chemical conversion of raw biomass under HTC. The spectroscopic
results show that hemicellulose, cellulose and lignin exhibit different decompo-
sition/carbonization temperatures. In general, the decomposition of hemicellulose
begins at low temperature, with nearly complete decomposition at temperatures
below 200 ı C. The decomposition of cellulose occurs over the temperature range
of 200–300 ı C depending on the specific biomass feedstocks. Lignin shows the
highest thermal resistance, and only decomposes at temperatures above 250 ı C.
As an example, Figs. 9.2 and 9.3 show the FT-IR spectra and cross polariza-
tion/magic angle spinning (CP/MAS) 13 C NMR spectra of coconut fiber and its
corresponding hydrochars. In Fig. 9.2, the presence of alcohol and phenol structures
are indicated by the peaks at 2,927 and 2,854 cm1 which are ascribed to the
C–H stretching vibration and deforming vibration, respectively. These two peaks in
the hydrochars weaken with increasing temperatures, suggesting that dehydration
reactions are enhanced with increasing temperatures. The peak at 1,064 cm1 is
ascribed to the “-glycosidic bond in cellulose and hemicellulose. The intensity
of this peak decreases significantly with increasing temperatures, indicating that
nearly all hemicellulose and cellulose decompose at temperatures of 250 ı C
Fig. 9.2 FT-IR spectra of

coconut fiber and its derived
hydrochars (Reprinted with
permission from [33].
Copyright 2013, Elsevier)
and greater. The strong C–O band at 1,103 cm1 is assigned to –OCH3 groups
in lignin. This band also weakens with increasing temperatures, indicating a loss
of –OCH3 by deoxygenation reactions. The peaks around 1,612 and 1,449 cm1
in the raw biomass correspond to the C D C stretching of aromatic groups in
lignin. These peaks are present in all the hydrochars analyzed, implying that some
lignin fragments and intermediate structures remain in the resulting hydrochars.
Thus, lignin does not appear to decompose completely under HTC conditions
investigated. The absorption band around 3,375 cm1 indicates the existence of free
and intermolecular bonded hydroxyl groups –OH in raw biomass. The hydroxyl
content in hydrochar decreases with increasing temperatures and almost disappears
at 375 ı C treatment, implying an increase in hydrophobicity of hydrochar compared
to raw biomass. High hydrophobicity is important for fuel storage and handling, as
this affords higher resistance to humidity.
CP/MAS 13 C NMR spectra provide valuable complementary data to those from
the FT-IR analysis in describing biomass conversion under HTC. The NMR spec-
trum of coconut fiber gives sharp characteristic peaks of hemicellulose, cellulose
and lignin as shown in Fig. 9.3 [16, 24–26]. The peaks appearing in the range from
60 to 105 ppm are ascribed to cellulose and hemicellulose. The peak at 105 ppm
238 Z. Liu et al.
Fig. 9.3 13 C NMR spectra of coconut fiber and its derived hydrochars (Reprinted with permission
from [33]. Copyright 2013, Elsevier)
is ascribed to cellulose C-1 and the peak around 102 ppm, which overlaps with
105 ppm, is ascribed to hemicellulose. The strong peak appearing at 73 ppm (the
C2, C3 and C5 of carbohydrate) can be seen as an indicator of both cellulose
and hemicellulose. Figure 9.3 shows that the decomposition of such carbohydrates
occurs at low temperatures. At 220 ı C, the intensities of the peaks decrease
significantly; they are undetectable above 250 ı C. This observation is consistent
with the disappearance of the resonance peak at 88 ppm, which is ascribed to the C-
4 of crystal-interior cellulose. Peaks in the range from 116 to 154 ppm and 55 ppm
(methoxyl group) are ascribed to lignin. The peaks appearing at 144 and 154 ppm
are ascribed to the C3/C5 of syringyl unit (not ether-linked) and C3/C5 of ether-
linked syringyl units, respectively. The peaks at 55 ppm and 116 ppm (C-5 guaiacyl
or C3 and C5 of 4-hydroxyphenyl unit) are present in the spectrum of coconut fiber.
Their intensities remain unchanged until the HTC temperature exceeds 300 ı C,
indicating that lignin degrades after hemicellulose and cellulose are degraded.
Therefore, hydrochar obtained at 300 ı C treatment contains considerable amounts
of lignin segments or unreacted lignin. The decomposition temperature of lignin
is higher than its melting temperature. The melting of lignin forms droplets that
are immiscible in hot compressed water, and which precipitate before degradation.
Similar results have also been observed during the hydrothermal dissolution of
willow in hot compressed water [27].
Fig. 9.4 FT-IR spectra of raw biomass and corresponding biochars produced from hydrothermal
carbonization and low temperature pyrolysis at 250 ı C (Reprinted with permission from [40].
Copyright 2013, Elsevier)
Hydrochar composition exhibits a progressive trend of functional group deple-

tion in hemicellulose, cellulose and lignin substructures. Above 300 ı C, hydrochar
carbon is mainly in the form of aromatics, with the degree of aromaticity increasing
with increasing temperatures. After treatment at 375 ı C, the peaks at 145 (guaiacyl
C3, C4) and broad peak at 35 ppm (CHx ) nearly disappear, indicating that the
hydrochar carbon is almost entirely aromatic, with very little organic oxygen or
hydrogen substituents.
Low temperature pyrolysis (LTP) is also widely employed for upgrading waste
biomass to solid fuels. It is interesting to compare the chemical evolution of biomass
feedstocks under HTC and LTP conditions. Figure 9.4 provides FT-IR spectra of
biochars produced by these two processes from coconut fiber and eucalyptus leaves.
These spectra differ mainly in the IR band at 1,064 cm1 , which is ascribed to
the “-glycosidic bond in cellulose and hemicellulose. The intensity of this peak
is nearly the same for the raw biomass and the pyrolytic biochar, but decreases
significantly in hydrochars, indicating that nearly all hemicellulose and cellulose
decomposed under HTC. This result implies that only a small part of cellulose
decomposes under LTP at 250 ı C [28]. The peaks around 1,513 cm1 (C D C
aromatic ring vibration) and 1,262 cm1 (aromatic C–O stretching of methoxyl and
phenyl propane structure) are present in both hydrochars and pyrolytic biochars,
indicating that lignin is not degraded significantly in either process. The peak
240 Z. Liu et al.
corresponding to hemicellulose (around 1,735 cm1 , attributed to C D O stretching

vibration of carboxylic acids in hemicellulose), disappears in both hydrochar and
pyrolytic biochars, while the peak around 1,699 cm1 is ascribed to the degradation
product of hemicellulose. These observations indicate that hemicellulose degrades
under both HTC and LTP conditions at 250 ı C. These results also suggest that more
extensive decomposition of biomass took place under HTC than LTP, when carried
out at the same temperature.
9.4 Kinetic Analysis of HTC of Biomass
The knowledge of reaction kinetics is vital to design and operate reactor units
efficiently. Therefore kinetic analysis should be performed for better understanding
of reaction characteristics of HTC. However, there are very few reports on the
kinetics of HTC of biomass [29–31].
First-order kinetics is most frequently assumed for solid fuel decomposition, and
hence is used to evaluate the kinetics of biomass HTC. The kinetic equation can be
described by the following formula:

ln .1 a/ AR 2RT E
ln D ln 1 (9.1)
T2 ˇE E RT
where E is apparent activation energy; A is the pre-exponential factor; T is absolute

temperature; R is the gas constant; ’ and “ are mass loss and the heating rate,
respectively.
If the kinetics follow a first-order reaction mechanism, the plot of ln[ln(1a)/T2]
versus 1/T is expected to give a straight line. From the slope, E/R, the apparent
activation energy can be obtained and by taking the absolute temperature at which
half conversion rate is achieved, the pre-exponential factor A can be calculated.
Considering different reaction rates above or below 300 ı C, kinetic parameters of
HTC of coconut fiber and eucalyptus leaves were calculated separately within these
two distinct temperature ranges (temperature range I (150–300 ı C) and range II
(300–375 ı C)). In addition, the kinetics of LTP were also determined under the
same temperature ranges for the comparison.
Plots of ln[(ln(1a)/T2)] against 1/T for HTC of coconut fiber and eucalyptus
leaves are shown in Fig. 9.5a, b and the regression functions were obtained
by a linear fitting method. The high correlation coefficients (R2 , presented in
Table 9.1) indicate that first-order reaction mechanisms well describe HTC of
these feedstocks within both temperature ranges. There are significant differences
for the apparent activation energies of HTC in the different temperature ranges
for both coconut fiber and eucalyptus leaves. The coconut fiber had activation
energies of 67.5 and 179.5 kJ/mol while the eucalyptus leaves had 59.2 and
173.7 kJ/mol for temperature ranges I and II, respectively. From the above chemical
structural analyses, hemicellulose decomposed first at lower temperatures (lower
Fig. 9.5 Linear fitted plots for HTC and LTP of coconut fiber and eucalyptus leaves using Coats-
ı
Redfern method within two separated temperature ranges (range I: 150–300 C and range II: 300–
375 ı C) (Reprinted with permission from [40]. Copyright 2013, Elsevier)
Table 9.1 Kinetic parameters for HTC and LTP of coconut fiber and eucalyptus leaves
Temperature
ı
Sample ( C) Equation R2 E (kJ/mol) A
Coconut HTC 150–300 Y D 8,107.5x C 4.27 0.991 67:5 4.08E C 12
fiber HTC 300–375 Y D 21,595x C 22.64 0.955 179:5 1.23E C 21
LTP 150–300 Y D 8,598.5x C 2.99 0.985 71:5 1.15E C 12
LTP 300–375 Y D 17,662x C 15.93 0.978 146:8 1.44E C 18
Eucalyptus HTC 150–300 Y D 7,118.1x C 2.22 0.903 59:2 4.38E C 11
leaves HTC 300–375 Y D 20,893x C 21.37 0.992 173:7 2.05E C 20
LTP 150–300 Y D 7,982.6x C 1.86 0.945 66:4 1.01E C 12
LTP 300–375 Y D 18,090x C 16.46 0.982 150:4 2.53E C 18
than 250 ı C) followed by cellulose which decomposed completely at 300 ı C under

HTC. For the lignin, at temperatures lower than 300 ı C, only melting would occur
instead of degradation. In the present study, the decomposition of hemicellulose
and cellulose, and the decomposition/carbonization of lignin occurred primarily
at lower temperatures (temperature range I) and higher temperatures (temperature
range II), respectively. Therefore, the apparent activation energy of temperature
242 Z. Liu et al.
range I was the combined activation energies of decomposition of hemicellulose and

cellulose, while the activation energy of temperature range II was mainly applicable
to lignin. As a result of the large difference between carbohydrates (hemicellulose
and cellulose) and lignin, the reactions that occurred were quite different between
temperature ranges I and II, leading to different apparent activation energies. In
addition, from the viewpoint of chemical reactions, competing reactions such as
condensation, depolymerization, and polymerization occurred during the HTC of
biomass. Depolymerization reactions are thought to dominate at low temperatures,
while polymerization and aromatization reactions dominate at high temperatures
[12, 32]. As a consequence, different apparent activation energies resulted from the
different reactions in these two temperature ranges.
Two separated temperature ranges, namely, 150–300 ı C and 300–375 ı C, were
also used to evaluate the reaction kinetics to get a direct comparison between HTC
and LTP. Plots of ln[(ln(1a)/T2)] against 1/T for LTP treatment of coconut fiber
and eucalyptus leaves are shown in Fig. 9.5c, d. The high correlation coefficients
(R2 , presented in Table 9.1) indicate that first order reaction mechanisms also fit
well LTP processing of these feedstocks. Although the biomass underwent greater
decomposition under HTC than LTP, the apparent activation energy of HTC was
lower than that of LTP in the first temperature range. This lower activation energy
indicates the promising potential of HTC for biomass upgrading.
9.5 Fuel Quality of Hydrochar
9.5.1 Ultimate Analysis, Proximate Analysis and Energy

Densification
As mentioned above, most of the previous research reported in the literature was
carried out under subcritical water conditions to maximize the hydrochar yield. Only
one report is available that describes biomass treatment under supercritical water
conditions [33]. The properties of hydrochars obtained under different operating
conditions are summarized in Table 9.2. Hydrochar yield strongly depends on
the treatment conditions, particularly reaction temperature. For a given feedstock,
hydrochar yield decreases with increasing reaction severity.
Compared to raw biomass, hydrochars have increased carbon contents and
decreased oxygen contents. For sulfur and nitrogen contents, there is no major
difference between raw biomass and the corresponding hydrochars. Hydrochars
have lower H/C and O/C ratios than raw biomass; therefore, the hydrochars are
expected to have improved combustion characteristics compared to raw biomass
due to reduced energy loss, smoke and water vapor during the combustion process.
Because of reduced low energy O–C bonds, and increased high energy C–C bonds,
hydrochars have higher energy density compared to raw biomass. It is generally
considered that the higher heating value (HHV) of a solid fuel should exceed
Table 9.2 Proximate analysis, ultimate analysis and energy analysis of coconut fiber, eucalyptus leaves and their derived hydrochars (lignite for comparison)
Volatile Fixed Mass
matter carbon yield HHV
Sample (%) Ash (%) (%) N (%) C (%) H (%) S (%) O (%) (%) (kJ/mol) ED EY (%)
Coconut fiber 80.9 8.1 11.0 0.90 47.75 5.61 0.23 45.51 - 18.4 - -
CF-220 69.8 6.2 24.0 0.90 62.47 5.28 0.26 31.09 57.17 24.7 1.34 76.67
CF-250 67.9 5.0 27.1 0.98 67.10 5.20 0.29 26.43 45.27 26.7 1.45 65.70
CF-300 53.6 4.3 42.1 1.13 73.22 5.09 0.35 20.21 40.72 29.4 1.60 65.00
CF-350 56.6 4.9 38.5 1.17 73.37 4.52 0.36 20.58 35.76 28.7 1.56 55.78
CF-375 42.6 8.6 48.8 1.23 78.20 4.31 0.33 15.93 35.55 30.6 1.66 59.00
Eucalyptus leaves 79.2 10.5 10.3 1.23 46.96 6.22 0.77 44.82 - 18.9 - -
EL-200 72.5 7.3 20.2 1.37 61.11 6.13 0.65 30.74 65.43 25.3 1.33 87.34
EL-250 70.1 6.9 23.0 1.44 62.30 5.47 0.44 30.35 46.40 25.0 1.32 61.12
EL-300 61.2 7.1 31.7 1.62 68.87 6.00 0.72 22.79 40.51 28.7 1.51 61.32
EL-350 56.2 9.9 33.9 1.60 70.50 5.93 1.52 20.45 30.80 29.4 1.55 47.84
EL-375 43.2 14.2 42.6 1.64 72.19 4.81 1.51 19.85 28.27 28.7 1.51 42.78
Lignite 48.8 10.3 41.0 1.74 61.64 5.72 0.77 30.13 - 25.0 - -
- Not available
9 Production of Renewable Solid Fuel Hydrochar from Waste Biomass. . .
CF-xx: The hydrochar derived from coconut fiber and xx shows the temperature
EL-xx: The hydrochar derived from eucalyptus leaves and xx shows the temperature
243
244 Z. Liu et al.
20 MJ/kg to ensure auto-thermal combustion. Most hydrochars exceed this HHV

threshold, and thus are promising alternatives for solid fuel applications. In addition,
it is well known that about 80–90 % of raw biomass is combusted in the form
of volatile matter. Raw biomass is typically ignited at lower temperatures (around
250 ı C) and rapidly reaches a maximum weight loss during the combustion process.
In the case of hydrochars, the volatile matter content is decreased and the fixed
carbon content is increased compared to raw biomass. Therefore, hydrochars are
ignited at higher temperature and in a mild combustion mode, overcoming some of
the drawbacks of raw biomass as a direct fuel. With this improved fuel quality,
hydrochars have the potential to be a satisfactory solid fuel for either direct
combustion, or co-combustion with low rank coal in existing coal-fired plants.
The fuel quality of hydrochars obtained from coconut fiber and eucalyptus
leaves has been systematically evaluated. As expected, the carbon content increased
while the oxygen content decreased dramatically with increasing hydrothermal
temperature [33, 34]. A slight decrease of hydrogen content was also observed
with increasing temperature. The H/C and O/C ratios decreased with increasing
hydrothermal temperature. Even for the hydrochar obtained at a relatively low
temperature of 200 ı C, the values were far lower than those of raw biomass. It
is noteworthy that the values of H/C and O/C ratios of the hydrochar obtained from
250 ı C treatment were lower than those of the lignite sample used. Therefore, it is
expected that the hydrochars have better fuel qualities than either raw biomass or
lignite.
From the proximate analysis results, it can be seen that HTC significantly
decreased the fraction of volatile matter in combustible carbon (volatile matter/
(volatile matter C fixed carbon)). These fractions for hydrochars produced from
coconut fiber at 300 ı C and eucalyptus leaves at 350 ı C are 0.56 and 0.62, very close
to that of the lignite sample (0.56). High volatile matter content reduces combustion
efficiency and increases pollutant emissions when biomass is directly combusted
[5]. When biomass co-combusts with low rank coal, the large difference in the
volatiles ratio between biomass and coal leads to separated combustion regions,
with biomass being combusted at low temperatures while coal combusts at high
temperatures. The similar proximate composition and HHV of hydrochars to lignite
(Table 9.2) imply that the hydrochars can be used directly for coal-fueled boilers
without significant modifications. Moreover, the combination of decreased oxygen
and volatile matter content of hydrochars can potentially reduce the release of
inorganic vapors during combustion, compared to the parent biomass [5]. As for
the ash content, it decreased with increasing temperatures to 300 ı C for coconut
fiber and 350 ı C for eucalyptus leaves, and then increased up to 375 ı C. From
the decreased hydrochar yield and decreased ash content, it can be seen that some
fraction of ash was dissolved in the water medium at lower temperatures.
The energy densification and energy yield of the produced hydrochar were
also evaluated. Energy densification was determined by the energy content of the
hydrochars divided by the energy content of raw biomass. Energy yield is defined
as the hydrochar mass yield multiplied by the energy densification [35]. As shown
in Table 9.2, energy densification increased with increasing temperatures, from a
low value of 1.34 to a high value of 1.66 and from 1.34 to 1.56 for coconut fiber and
eucalyptus leaves derived hydrochars, respectively. The highest energy yield was
obtained at the lowest temperature used. Energy yield decreased with increasing
temperatures, with the lowest energy yields at 55.78 % and 42.78 % for coconut
fiber at 350 ı C and eucalyptus leaves at 375 ı C, respectively.
9.5.2 Combustion Characteristics
Figure 9.6 shows DTG curves for the coconut fiber, eucalyptus leaves and their
derived hydrochars, with the combustion parameters summarized in Table 9.3. From
the curves, it can be seen that the combustion behavior of the biomass changed
significantly after HTC, especially for the coconut fiber. For coconut fiber, a sharp
DTG peak was observed centered at about 296 ı C and a weight loss of 82 %
was measured at 327 ı C. The combustion of coconut fiber involves mainly volatile
matter combustion, which was initiated at a low temperature of 273 ı C due to the
high reactivity of volatile matter. The rapid weight loss of coconut fiber within
a short time at the lower temperature range implies that there was incomplete
combustion with low efficiency and high pollutant emissions (CO and PAH) [5].
Compared to coconut fiber, the reactivity of the hydrochar decreased, resulting
in a higher ignition temperature and combustion in a higher temperature range. This
elevated combustion temperature with high weight loss rate implies improved com-
bustion safety, increased combustion efficiency and decreased pollutant emission.
These combustion characteristics are significant improvements over raw biomass
feedstock as a fuel [5, 36, 37].
Among all coconut fiber-derived hydrochars, that produced from supercritical
water (375 ı C) was unique with the lowest maximum weight loss rate and burnout
and the highest ash content. In the DTG curve of hydrochar (375 ı C), a wide peak
with a maximum weight loss rate (0.47 %/ ı C) was observed at 506 ı C and a minor
weight loss was also seen at 728 ı C, which was attributed to decomposition of the
ash. Similar results were also observed for eucalyptus leaves and leaves-derived
hydrochars. For raw eucalyptus leaves, two separated peaks were observed due to
the large differences in reactivities of the components. Around 60 % weight loss
was observed to occur at 360 ı C due to the high reactivity. After HTC treatment, the
height of the first peak in the DTG curve decreased and the maximum weight loss
rate shifted to the second peak (0.82 %/ ı C for leaves and higher than 0.98 %/ ı C
for the chars). Among all eucalyptus leaf-derived hydrochars, the maximum weight
loss was achieved in hydrochars produced at 250 ı C (similar to coconut fiber).
Hydrochar at 375 ı C had the lowest burnout and widest combustion range beyond
temperature 605 ı C. For evaluation of the ignition performance, ignition index (Di )
of the biomass and hydrochar was calculated with the following equation [38]:
Di D Rmax = .tmax ti / (9.2)

246 Z. Liu et al.
Fig. 9.6 DTG curves of coconut fiber, eucalyptus leaves and their corresponding hydrochars
(Reprinted with permission from [33]. Copyright 2013, Elsevier)
where Rmax is the maximum weight loss rate, tmax and ti correspond to the time of
the maximum weight loss rate and ignition temperature, respectively.
As shown in Table 9.3, the coconut fiber-derived hydrochars have lower Di
values than those of raw coconut fiber, and the Di values decreased with increasing
temperatures. For the eucalyptus leaves, the ignition index increased up to a
Table 9.3 Combustion parameters of coconut fiber, eucalyptus leaves and their derived
hydrochars
ı ı ı ı
Fuel Ti ( C) Tm ( C) Tb ( C) Rmax ( %/ C) Di (102 )
Coconut fiber 273 295 326 2.39 1.48
CF-220 295 450 472 1.64 0.59
CF-250 372 436 472 1.71 0.49
CF-300 400 458 478 1.67 0.42
CF-350 410 451 512 0.94 0.24
CF-375 393 505 580 0.47 0.11
Eucalyptus leaves 253 312 456 0.82 0.52
EL-200 288 414 449 1.59 0.64
EL-250 288 423 445 1.97 0.78
EL-300 374 417 449 1.15 0.35
EL-350 369 410 470 0.98 0.30
EL-375 428 522 581 0.55 0.11
Ti D Ignition temperature
Tb D Burnout temperature
Di D Ignition index
temperature of 250 ı C and then decreased with further increase in temperature. The
highest index value was obtained at 250 ı C treatment of eucalyptus leaves.
Overall, the combustion behaviors of the hydrochars were similar, regardless of
feedstock. HTC has the effect of homogenizing different biomass feedstocks, which
increases the potential for biomass utilization. At supercritical conditions, the low
combustion reactivity, together with high ash content and low hydrochar yield, show
that very high temperatures and pressures are not favorable for solid fuel hydrochar
production from biomass [12].
9.5.3 Ash Related Problems
Biomass materials generally have lower ash contents than those of most types
of coal, but the nature of biomass ashes is very different from that of most coal
ashes. Raw biomass as a direct solid fuel suffers from serious ash related problems
including slagging and fouling due to the high content of alkali and alkaline earth
metals, and potentially high chlorine content compared to coal. For example, during
raw biomass combustion alone, and biomass/coal co-combustion, the high content
of alkali metals can cause severe slagging and fouling problems on boiler heat
transfer surfaces, and the high chlorine content leads to corrosion. The above
challenges are not significant in low co-firing ratios, or when high-quality biomass
is used. However, with increasing biomass/coal ratios, especially when low-quality
biomass is used, they may have a major influence on the economy of the plant.
Apart from combustion issues, the reutilization of blended ash is also a key
concern. Ash from coal combustion alone is mostly utilized in the construction
248 Z. Liu et al.
industry. However, in the case of biomass ash or mixed biomass and coal ashes,
they cannot be utilized in the same applications as pure coal ash. Thus, utilization
options for the ash residues from co-firing can be a constraint for co-firing.
The ash content and the chemical composition of solid fuels also have an
influence on the application of combustion technology and subsequent ash disposal.
Therefore, during pre-treatment, the fate of the metals contained in raw biomass,
especially those closely related to ash problems, should be determined as this can
control subsequent combustion performance of the resultant hydrochars. To date,
little information is available in the literature on the distribution of metal species
after HTC treatment of biomass feedstocks, especially those species related to ash
problems [39, 40]. It has recently been reported that some inorganic components
contained in raw biomass can be leached into the water medium during the HTC
process, leading to decreased ash contents in hydrochar relative to its parent biomass
[39]. In general, light metals and heavy metals show different removal rates during
HTC conditions: higher removal rates for light metals and lower removal rates for
the heavy metals. This can be ascribed to differences in the inherent properties of
light and heavy metals, and the existence of their different chemical forms in the
raw biomass.
In contrast to light metals, heavy metals have a strong tendency to hydrolyze.
Hydrolysis of these heavy metals is likely to occur under HTC conditions. The
hydrolysis products precipitate from the water medium and mix with the hydrochar.
Heavy metals that do not hydrolyze thoroughly still have a strong co-ordination
ability. These metals coordinate with the functional groups on the surface of the
hydrochars, resulting in high retention rates [21]. In addition, different chemical
forms of light and heavy metals in raw biomass can contribute to the different
retention rates during HTC. For example, substantial amounts of heavy metals such
as Fe and Ti are known to exist as oxyhydroxides and silicates. Such heavy metals
present in the form of crystalline inorganics generally have low water solubility
[41]. In contrast, light metals such as K and Na have high water solubility, and even
simple water washing can efficiently remove much of them from raw biomass. For
example, up to 92 % of Na and 62 % of K were removed from herbaceous biomass
by water washing [42]. The retention rate of metals is associated with ash-related
problems including slagging and fouling during the combustion of a solid fuel. This
has been systematically investigated for the hydrochars prepared from coconut fiber
and eucalyptus leaves.
Table 9.4 presents the metal concentrations of major ash species in raw biomass
and hydrochars. In both coconut fiber and eucalyptus leaves, Ca and K are the
dominant ash problem-related metals (Ca and K are 0.33 and 1.60 %, and 2.28 and
0.66 % for CF and EL, respectively). The contents of other metals such as Al and
Ti are very low. HTC treatment of biomass resulted in significantly decreased metal
contents in the hydrochars. For example, K and Na, which are among the highest
concentration species in biomass generally, were decreased by more than 80 % in
the hydrochars (CF-250/coconut fiber: K 11.50 %, Na 15.45 %; EL-250/eucalyptus
leaves: K 17.52 %, Na 19.23 %).
Table 9.4 Metals contents in ash from coconut fiber, eucalyptus leaves and their corresponding
ı
hydrochars (250 C)
Ca (%) K (%) Al (%) Na (%) Mg (%) Fe (%) Si (%) Ti (%)
Coconut 0.334 1.603 0.005 0.123 0.058 0.019 0.022 0.007
fiber
CF-250 0.191 0.184 0.001 0.019 0.013 0.016 0.017 0.004
Eucalyptus 2.282 0.663 0.007 0.026 0.097 0.011 0.019 0.005
leaves
EL-250 1.962 0.121 0.001 0.005 0.019 0.008 0.014 0.003
Fig. 9.7 Retention rates of major ash metals during hydrothermal carbonization of coconut fiber
ı
and eucalyptus leaves at 250 C (Reprinted with permission from [40]. Copyright 2013, Elsevier)
The retention rate of a particular metal is defined as the percentage of that

metal in hydrochar relative to the percentage in raw biomass. Figure 9.7 shows the
retention rates of the tested metals after HTC. It is seen that the retention rates for
all metals in the hydrochars were less than 40 %. The lowest retention rates were for
K (5.09 %) and Al (6.63 %) in coconut fiber and eucalyptus leaves, respectively. In
addition, it was confirmed that light and heavy metals exhibited noticeably different
retention rates during HTC. For the light metals (Al, K, Na and Mg), the retention
rates were in the range of 7–11 %, while for the heavy metals (Fe, Si and Ti),
retention rates were in the range of 27–38 % During HTC, the highest retention rates
were observed for Fe (38.1 %) and Ca (39.9 %) for coconut fiber and eucalyptus
leaves, respectively.
250 Z. Liu et al.
Currently, the indices used for evaluating ash-related problems of a solid fuel,
including fouling and slagging tendencies, come from the fuel ash analysis [43, 44].
For example, the slagging and fouling of a solid fuel can be predicted by the
calculations shown in Eqs. 9.3 and 9.4. To use these equations, it is necessary
to carry out various ash analyses. Due to the significant reductions in alkali and
alkaline earth metals with HTC treatment, it is expected that hydrochars would
exhibit reduced fouling and slagging tendencies compared to combustion of raw
biomass. In contrast, pyrolytic biochars have nearly 100 % metal retention rates,
and increased sulfur contents, thus they are expected to exhibit more serious ash
related problems than raw biomass.
It has been reported that LTP (low-temperature pyrolysis) combined with
washing pretreatment of biomass produced solid fuels with low ash contents and
improved fuel quality [42]. With the exception of the use of chemicals, extensive
energy was also consumed for the drying process during subsequent combustion.
Therefore, HTC is able to potentially mitigate ash-related problems and significantly
upgrade biomass feedstocks while maintaining relatively low energy consumption
in comparison with LTP.
Slagging index .SI/ D .B=A/ S% (9.3)
Fouling index .FI/ D .B=A/ .Na2 O C K2 O/ (9.4)
where B/A D (Fe2 O3 C CaO C MgO C Na2 O C K2 O)/(SiO2 C Al2 O3 C TiO2 ); S is

the percent of the sulfur in the dry fuel sample.
9.6 Pelletization Properties of Hydrochar
The inherent properties of biomass feedstocks such as high dust levels and low
bulk density together with a wide range of physical shapes create big challenges
for their handling logistics and combustion technology. In contrast, pellet fuels
have improved quality in terms of increased bulk density and uniform shape and
size [45–48]. Increased bulk density can reduce the storage and transportation
costs while uniform dimensions facilitate fuel handling and feeding. In addition,
the improved fuel quality makes biomass pellets suitable for many industrial and
residential applications, including combustion in grate furnaces and gasification in
fluidized bed furnaces. The global pellet market has experienced a rapid growth
during the last decade and is expected to have even a faster growth in the near future
[49].
As is known, raw biomass has poor pelletization capacity due to the presence of
extractives, which lead to weak H-bonding and van der Waal’s forces by preventing
a close contact between adjacent bonding sites. Considering the difference of
physicochemical properties between raw biomass and corresponding hydrochars,
the pelletization of raw biomass and corresponding hydrochar was investigated.

Pinewood was selected as a representative woody biomass and coconut fiber and
coconut shell for agricultural residues. Pellets were prepared using a single pelletizer
at room temperature, without any additional binder. The assembly consists of two
parts: (1) a piston with 13.2 mm diameter and 30 mm length, (2) a cylinder with
13.5 mm inside diameter and 40 mm length. The end of the die is closed using a
removable backstop. The biomass/hydrochar was loaded stepwise in same amounts
into the pelletizer, and then compressed at a maximum pressure of 280 MPa. After
holding 5 s, the backstop was removed and the pellet was pushed out by applying
pressure. The L/D of the pellets was around 0.9 and at least 6 pellets were made for
each sample [50].
Tensile strength of raw biomass pellets was in the following order: pinewood
(3.91 MPa) > coconut fiber (1.51 MPa) > coconut shell (0.96 MPa). The lignin
originally contained in raw biomass is not thought to act as a binder at this
low pelletization temperature (room temperature). Therefore, the bonding forces
involved in raw biomass pellets are mainly related to attractive forces including H-
bonding, van der Waal’s forces and mechanical interlocking. The tensile strength
of raw biomass was found to be correlated with the content of extractives in raw
biomass (the extractives contents of raw biomass were in the order: pinewood
(4.0 %) < coconut fiber (6.9 %) < coconut shell (8.4 %)). This observation implies
that the bonding forces within raw biomass pellets can be mainly ascribed to
attractive forces.
The hydrochar pellets had much smoother surfaces than the raw biomass pellets.
The surface micrographs of the pellets were examined at a magnification of 200
(shown in Fig. 9.8) to further investigate bonding mechanisms within the pellets.
From the micrographs, it can be seen that the surfaces of hydrochar pellets are very
uniform without obvious void spaces. In contrast, a substantial amount of voids and
gaps exist on the surface of raw biomass pellets. These spaces can reduce the pellet’s
resistance to deformation and promote the relative movement of particles within the
pellet matrix, leading to weak mechanical durability.
The raw biomass pellets deformed mainly in the axial direction (the biomass was
fed to the cylinder stepwise in amount until the pellets had the desired length and the
deformation implied heterogeneous structures within raw biomass pellet) under the
application of compressive force. In contrast to raw biomass pellets, the hydrochar
pellets reached a definite breaking force and broke into two halves in the radial
direction. The broken cross sections confirmed the homogenous structure within the
hydrochar pellets. The definite maximum resistant force and breaking pattern also
suggest that strong bonding occurs within hydrochar pellets.
The tensile strength of hydrochar pellets was significantly increased in compari-
son to that of raw biomass pellets, especially for coconut fiber pellet (1.51 MPa) and
its hydrochar pellet (7.50 MPa). As a direct comparison, the tensile strength of these
hydrochar pellets was much higher than those of raw biomass pellets and pretreated
biomass pellets that have been reported in the literature [51–53]. The extremely high
tensile strength suggests the presence of strong bonding forces between hydrochar
252 Z. Liu et al.
Fig. 9.8 Micrograph of surface of the pellets prepared from raw biomass and corresponding
hydrochars (Reprinted with permission from [50]. Copyright 2013, Elsevier)
particles within the hydrochar pellets. Unique properties of hydrochar contribute to

the strong bonding forces, as explained in detail below.
Firstly, high molecular weight organic compounds provide liquid bridges
between adjacent particles within hydrochar pellets. After HTC, a considerable
amount of polar organic compounds (so called bio-oil) deposits on the surface
of the hydrochar [54]. Similar to water, these organic compounds can connect
adjacent particles by forming liquid bridges during pelletization, resulting in high
mechanical strength of hydrochar pellets [55]. In addition, the liquid bridge is
affected by capillary pressure, surface tension and viscosity. The liquid bridges
within hydrochar pellets are expected to be stronger than the bridge with water
because of the higher viscosity of organic compounds in hydrochar [55].
Secondly, enhanced attractive forces within hydrochar pellets likely form due to
the increased polar functional groups on the hydrochar surface, higher contact sur-
face area between hydrochar particles, and the absence of extractives in hydrochars.
HTC treatment creates more functional groups on the surface of hydrochar relative
to those on raw biomass, and it completely removes extractives that were originally
contained in the raw biomass [21, 35]. For example, the contents of total oxygen-
containing groups (hydroxyl, carboxyl and carbonyl groups) were reported to have
been increased 98 % and 62 % for pinewood and rice husk by hydrothermal
treatment at 250 ı C, respectively [21]. Increased polar functional groups entail
strong electrostatic attraction including H-bonding and van der Waal’s forces. As
an example, the strength of H-bonding can reach 2.5–120 kJ/mol and the bond
lengths vary between 0.12 and 0.32 nm depending on the bond strength. Therefore,
by compressing the hydrochar particles to reach a distance of less than 0.32 nm,
increased H-bonding can occur due to increased oxygen-containing groups on the
surface. Furthermore, the absence of extractives in the hydrochars also facilitates the
formation of strong attractive forces between adjacent hydrochar particles [35, 54,
56]. As for the enlarged contact surface, it is ascribed to the highly friable property
of the hydrochars caused by the increased lignin content [33]. As a result of the high
friability, the hydrochar breaks into small particles under applied compressive force
during pelletization. Consequently, the contact area between adjacent hydrochar
particles was enlarged compared to that of raw biomass. The increased contact
surface area further enhanced liquid bridging and attractive forces between adjacent
particles [57]. Finally, mechanical interlocking possibly plays an increased role
within hydrochar pellets compared to raw biomass pellets, due to the increased
surface area and pore structure of hydrochar particles [21].
During 2-weeks’ storage, some particles were detached from the surface of raw
biomass pellets, and the surface became coarser, especially for coconut shell pellets.
This was probably caused by the elastic recovery of raw biomass particles; it also
indicates the weak bonding within raw biomass pellets. In addition, the length
expansions for the hydrochar pellets were far lower than those of raw biomass
pellets. For example, length expansions of pinewood pellets reached 4.9 and 6.9 %
after 1 and 2 days, respectively and reached the highest expansion 7.75 % after 4
days. In the case of pinewood hydrochar pellets, limited length expansions were
observed (less than 0.71 %) and reached their maximum within 2 days. The limited
length expansions of hydrochar pellets confirmed the strong bonding between
hydrochar particles and the excellent durability of hydrochar pellets.
It is noteworthy that agro-residues pellets suffer from weak mechanical durability
[58]. This is confirmed by the lower tensile strength of coconut fiber pellets and
coconut shell pellets in comparison to pinewood pellets. In the present study,
hydrochar pellets made from agricultural biomass had a comparable tensile strength
to woody biomass pellets, and the coconut fiber derived hydrochar pellet even had
higher tensile strength than pinewood-derived hydrochar pellets. Therefore, HTC
combined with pelletization narrows the difference between the agricultural and
woody biomass, overcoming an existing hurdle for agricultural residue pelletization.
254 Z. Liu et al.
In addition, high friability of the hydrochars decreases the influence of particle size
on pellet durability and is helpful in reducing the production cost of the pellets
compared to raw biomass [59].
9.7 Nitrogen Containing Pollutant Emissions

from Hydrochar Pyrolysis
In the context of developing advanced clean and renewable energy, pollutant emis-
sions are also a concern during hydrochar combustion. Pyrolysis or devolatilization
occurs at the primary stage of the combustion; therefore efficient conversion of
hydrochar nitrogen to inert substances such as nitrogen gas in the pyrolysis stage
can reduce NOx and N2 O emissions in the subsequent combustion process. In
addition, pyrolysis is not only an important process of combustion, but also an
important thermal conversion method. Great efforts have been made to study nitro-
gen conversion during biomass/coal pyrolysis [60–62]. For example, interactions
between biomass and coal have been reported to increase nitrogen partitioning in
favor of volatile compounds, and to decrease the generation of NH3 and HCN
at higher operating temperatures during rapid pyrolysis [60]. With improved fuel
quality, less pollutant emissions are expected during hydrochar pyrolysis compared
to raw biomass. To date, only one published report is available describing nitrogen-
containing pollutant emissions from pyrolysis of coconut fiber derived hydrochar
[63]. The nitrogen conversion of coconut fiber and its hydrochar (250 ı C) during
pyrolysis is summarized as follows.
Hydrochar and its parent biomass, coconut fiber, show quite different patterns of
nitrogen partitioning of HCN and NH3 during pyrolysis, as shown in Fig. 9.9. Of
these species, HCN is the dominant nitrogen pollutant from pyrolysis of coconut
fiber, especially at low pyrolysis temperatures. For example, HCN and NH3 account
for 8.5 and 2.8 % of total nitrogen at 600 ı C, respectively. In contrast, NH3 is
the dominant pollutant from hydrochar pyrolysis, which is similar to that of lignite
pyrolysis [63]. The yields of HCN and NH3 from hydrochar pyrolysis are 0.36 and
0.75 % at 600 ı C, and 0.13 and 3.00 % at 900 ı C. In addition, temperature plays an
important role in the nitrogen partitioning between HCN and NH3 for both coconut
fiber and hydrochar. For both coconut fiber and its hydrochar, the yield of NH3
increases with increasing temperature up to 800 ı C, and then decreases at 900 ı C.
For HCN yields, a consistent decrease with pyrolysis temperatures is observed for
coconut fiber, while in the case of the hydrochar, the yield reaches a maximum
(1.22 %) at 800 ı C, and then decreases at higher temperature. In comparison to
hydrochar, more HCN and NH3 are generated from coconut fiber pyrolysis at all
tested temperatures. The most important source of N2 O is believed to be oxidation
of gaseous HCN [64]. Therefore, more NOx and N2 O are expected from combustion
of coconut fiber than from combustion of hydrochar.
Nitrogen conversion during biomass pyrolysis is known to be closely related
to the chemical composition of biomass materials [60, 65]. During raw biomass
Fig. 9.9 Nitrogen partitioning of HCN and NH3 during coconut fiber and hydrochar pyrolysis
pyrolysis, the higher HCN yield than NH3 is ascribed to high contents of lignin
present in these biomass materials. As discussed earlier, hydrochar has a different
chemical composition than its parent biomass, with the hemicellulose and cellulose
originally contained in raw coconut fiber being nearly completely decomposed
during the hydrothermal process, and lignin only melting instead of degrading [33].
As a result, hydrochars have increased lignin content compared to raw coconut
fiber. However, NH3 is the dominant nitrogen pollutant from pyrolysis of hydrochar,
not HCN, despite the high lignin content of the hydrochar. This result implies
that besides total lignin content, the state of lignin is important in determining
nitrogen distributions between NH3 and HCN during pyrolysis. In addition, HCN is
believed to be generated primarily from thermal cracking of Unstable N-containing
compounds in the volatiles, while NH3 is formed from more Stable N-containing
compounds in the nascent char during pyrolysis [65, 66]. Due to sharply lower
volatile contents of hydrochar (80.69 and 67.92 % for coconut fiber and hydrochar,
respectively), less polymerization reactions among volatiles occurs during pyrolysis
of hydrochar, resulting in less HCN being formed during hydrochar pyrolysis as
compared to pyrolysis of raw coconut fiber. Increased cavity in hydrochar also
could lead to less polymerization, contributing to the low HCN yield from hydrochar
pyrolysis [67].
The total nitrogen in pyrolysis products is commonly divided into three groups:
char-N, (HCN C NH3 )-N, and (tar C N2 )-N. Due to difficulties in collecting and
quantifying tars, the amount of (tar C N2 )-N from each experiment is calculated by
256 Z. Liu et al.
Fig. 9.10 Nitrogen distribution from pyrolysis of coconut fiber and hydrochar at different
temperatures
difference between the total fuel-N and the sum of char-N and (HCN C NH3 )-N.
Total fuel-N partitioning of each product group from coconut fiber and hydrochar
pyrolysis are shown in Fig. 9.10. It is clear that more nitrogen remained in the char
and less (tar C N2 )-N was formed from pyrolysis of hydrochar than from pyrolysis
of coconut fiber. During hydrochar pyrolysis, char-N was the dominant form of N,
although it decreased with increasing temperature. The yields of (HCN C NH3 )-N
and (tar C N2 )-N were minimal at temperatures 700 ı C. The combined amounts
of these two groups only account for 2.98 and 3.25 % of total nitrogen from
hydrochar pyrolysis at 600 ı and 700 ı C, respectively. However, when the pyrolysis
temperature was increased to 800 ı C, a sharp decrease in char-N yield was observed,
with a corresponding increase in (tar C N2 )-N and (HCN C NH3 )-N. The yield
of (HCN C NH3 )-N increased to 5.22 % at 800 ı C, then decreased to 3.13 % at
900 ı C. The calculated yield of (tar C N2 )-N increased to 27.01 % at 800 ı C, with
a further increase to 51.63 % at 900 ı C. However, less tar is formed with increasing
temperatures. Therefore, more nitrogen in the hydrochar is converted into harmless
N2 at high temperatures.
In the case of coconut fiber pyrolysis, the yields of char-N and (HCN C NH3 )-
N decreased with increasing temperature. When compared to hydrochar pyrolysis,
the char-N yields are significantly lower while the yields of (HCN C NH3 )-N
are higher for coconut fiber pyrolysis over the entire temperature range, but
especially at low temperature. The (tar C N2 )-N yield from coconut fiber (51.5 %)
at 900 ı C is similar to that from hydrochar pyrolysis (51.63 %). However, less tar
is formed from hydrochar pyrolysis, as confirmed by visual observation. Therefore,
at high temperatures, more nitrogen in the hydrochar is converted into harmless N2
compared to nitrogen in coconut fiber.
9.8 Conclusions
Under HTC conditions, cellulose and hemicellulose are almost totally decomposed
at temperatures lower than 250 ı C; lignin starts to decompose around 300 ı C. Some
lignin fragments and decomposition intermediates remain in the hydrochar even
after supercritical water carbonization. The hydrochars have improved fuel quality
compared to raw biomass, including increased fixed carbon contents, elevated
heating values and reduced ash contents. The combustion behaviors of hydrochars
are distinct from raw biomass, with increased maximum weight loss rates, elevated
ignition temperatures and wide combustion ranges at higher temperatures. This
combustion behavior suggests that hydrochars produced from HTC of biomass are
appropriate for combustion or co-combustion with lignite in existing coal-fueled
boilers for heat generation.
During HTC, a high fraction of the alkali and alkaline earth metals originally
contained in raw biomass is dissolved in the water medium. Due to high removal
rates of these metals, ash-related problems, including slagging and fouling, are
expected to be mitigated when combusting hydrochar. As for the process kinetics,
HTC treatment of biomass follows first order reaction mechanisms. Compared to
low temperature pyrolysis, HTC shows lower activation energy in the temperature
range of 150–300 ı C, in spite of HTC’s greater degree of decomposition and
carbonization.
Upon pyrolysis, raw biomass and hydrochar give different nitrogen partitioning
of NH3 and HCN: HCN and NH3 are the dominant nitrogen pollutants from
pyrolysis of raw biomass and hydrochar, respectively. Compared to raw biomass,
hydrochar produces less HCN and NH3. Also, under HTC conditions, a higher
percent of nitrogen is converted into harmless N2 at high temperatures, indicating
that besides its higher thermal conversion efficiency, hydrochar combustion can
produce additional environmental benefits compared to raw biomass.
References
1. Kubacki ML, Ross AB, Jones JM, Williams A. Small-scale co-utilisation of coal and biomass.
Fuel. 2012;101:84–9.
2. Dai J, Sokhansnji S, Grace JR, Bi X, Kim CJ, Melin S. Overview and some issues related to
co-firing biomass and coal. Can J Chem Eng. 2008;86:367–86.
258 Z. Liu et al.
3. Sami M, Annamalai K, Wooldridge M. Co-firing of coal and biomass fuel blends. Prog Energy
Combust Sci. 2001;27:171–214.
4. Jensen PA, Frandsen FJ, Hansen J, Dam-Johensen K, Henriksen N, Horlyck S. SEM investiga-
tion of superheater deposits from biomass-fired boilers. Energy Fuel. 2004;18:378–84.
5. Khan AA, Jong W, Jansens PJ, Spliethoff H. Biomass combustion in fluidized bed boilers:
potential problems and remedies. Fuel Process Technol. 2009;90:21–50.
6. Lu G, Yan Y, Cornwell S, Whitehouse M, Riley G. Impact of co-firing coal and biomass on
flame characteristics and stability. Fuel. 2008;87:1133–40.
7. Hess JR, Wright CT, Kenney KL, Searcy EM. Uniform-format solid feedstock supply
system: a commodity-scale design to produce an infrastructure-compatible bulk solid from
lignocellulosic biomass. Idaho National Laboratory Report INL/EXT-09-15423. 2009. www.
inl.gov/bioenergy/uniform-feedstock
8. Titirici MM, Antonietti M. Chemistry and materials options of sustainable carbon materials
made by hydrothermal carbonization. Chem Soc Rev. 2010;39:103–16.
9. Hu B, Yu SH, Wang K, Liu L, Xu XW. Functional carbonaceous materials from hydrothermal
carbonization of biomass: an effective chemical process. Dalton Trans. 2008;5414–5423.
10. Libra JA, Ro KS, Kammann C, Funke A, Berge ND, Neubauer Y, Titirici M-M, Fuhner
C, Bens O, Kern J, Emmerlich K-H. Hydrothermal carbonization of biomass residuals: a
comparative review of the chemistry, processes and applications of wet and dry pyrolysis.
Biofuels. 2011;2:89–124.
11. Titirici MM, Thomas A, Antonietti M. Back in the black: hydrothermal carbonization of
plant material as an efficient chemical process to treat the CO2 problem? New J Chem.
2007;31:787–9.
12. Funke A, Ziegler F. Hydrothermal carbonization of biomass: a summary and discussion of
chemical mechanisms for process engineering. Biofuels Bioprod Biorefin. 2010;4:160–77.
13. Guerrero M, Ruiz MP, Millera M, Xed U, Bilbao R. Characterization of biomass chars formed
under different devolatilization conditions: differences between rice husk and eucalyptus.
Energy Fuel. 2008;22:1275–84.
14. Parshetti G, Hoekman SK, Balasubramanian R. Chemical, structural and combustion charac-
teristics of carbonaceous products obtained by hydrothermal carbonization of palm empty fruit
bunches. Bioresour Technol. 2013;135:683–9.
15. Sharma RK, Wooten JB, Baliga VL, Lin X, Geoffrey Chan W, Hajaligol MR. Characterization
of chars from pyrolysis of lignin. Fuel. 2004;83:1469–82.
16. Sharma RK, Wooten JB, Baliga VL, Martoglio-Smith PA, Hajaligol MR. Characterization of
char from the pyrolysis of tobacco. J Agric Food Chem. 2002;50:771–83.
17. Hu S, Xiang J, Sun L, Xu M, Qiu J, Fu P. Characterization of char from rapid pyrolysis of rice
husk. Fuel Process Technol. 2008;89:1096–105.
18. Demirbas A. Effects of temperature and particle size on bio-char yield from pyrolysis of
agricultural residues. J Anal Appl Pyrol. 2004;72:243–8.
19. Raveendran K, Ganesh A. Adsorption characteristics and pore-development of biomass-
pyrolysis char. Fuel. 1998;77:769–81.
20. Gao Y, Wang X, Wang J, Li X, Cheng J, Yang H, Chen H. Effect of residence time on
chemical and structural properties of hydrochar obtained by hydrothermal carbonization of
water hyacinth. Energy. 2013;58:376–83.
21. Liu Z, Zhang FS. Removal of lead from water using biochars prepared from hydrothermal
liquefaction of biomass. J Hazard Mater. 2009;167:933–9.
22. Abel S, Peters A, Trinks S, Schonsky H, Fachael M, Wessolek G. Impact of biochar
and hydrochar addition on water retention and water repellency of sandy soil. Geoderma.
2013;202–203:183–91.
23. Kumar S, Loganathan VA, Gupta RB, Barnett MO. An assessment of U(VI) removal from
groundwater using biochar produced from hydrothermal carbonization. J Environ Manage.
2011;92:2504–12.
24. Gilardi G, Abis L, Cass AEG. Carbon-13 CP/MAS solid-state NMR and FT-IR spectroscopy
of wood cell wall biodegradation. Enzyme Microb Tech. 1995;17:268–75.
25. Lahaye M, Rondeau-Mburo C, Deniaud E, Buleon A. Solid-state C-13 NMR spectroscopy

studies of xylans in the cell wall of Palmaria Palmate. Carbohyd Res. 2003;338:1559–69.
26. Love GD, Snape CE, Jarvis MC. Comparison of leaf and stem cell-wall component in barley
straw by solid state 13 C NMR. Phytochemistry. 1998;49:1191–4.
27. Hashaikeh R, Fang Z, Butler IS, Hawari J, Kozinski JA. Hydrothermal dissolution of willow in
hot compressed water as a model for biomass conversion. Fuel. 2007;86:1614–22.
28. White JE, Catallo WJ, Legendre BL. Biomass pyrolysis kinetics: a comparative critical review
with relevant agricultural residues case study. J Anal Appl Pyrol. 2011;91:1–33.
29. Luo G, Strong PJ, Wang H, Ni W, Shi W. Kinetics of the pyrolytic and hydrothermal
decomposition of water hyacinth. Bioresour Technol. 2011;102:6990–4.
30. Khajavi SH, Kimura Y, Oomori T, Matsuno R, Adachi S. Degradation kinetics of monosaccha-
rides in subcritical water. J Food Eng. 2005;68:309–13.
31. Mochidzuki K, Sakoda A. Liquid-phase thermogravimetric measurement of reaction kinetics
of the conversion of biomass wastes in pressurized hot water: a kinetic-study. Adv Environ
Res. 2003;7:421–8.
32. Savage PE, Levine RB, Huelsman CM. Hydrothermal processing of biomass. In: Crocker M,
editor. Thermochemical conversion of biomass to liquid fuels and chemicals. Cambridge: RSC
Publishing; 2010. p. 192–220.
33. Liu Z, Quek A, Balasubramanian R. Production of solid biochar fuel from waste biomass by
hydrothermal carbonization. Fuel. 2013;103:943–9.
34. Hoekman SK, Broch A, Robbins C, Zielinska B, Felix L. Hydrothermal carbonization
(HTC) of selected woody and herbaceous biomass feedstocks. Biomass Convers Biorefin.
2013;3:113–26.
35. Yan W, Acharjee TC, Coronella CJ, Vasquez VR. Thermal pretreatment of lignocellulosic
biomass. Environ Prog Sust Energ. 2009;28:435–40.
36. Ayhan D. Combustion characteristics of different biomass fuels. Prog Energy Combust Sci.
2004;30:219–30.
37. Hanzade H-A. Combustion characteristics of different biomass materials. Energ Convers
Manage. 2003;44:155–62.
38. Li X, Ma B, Xu L, Hu Z, Wang X. Thermogravimetric analysis of the co-combustion of the
blends with high ash coal and waste tyres. Thermochim Acta. 2006;441:79–83.
39. Reza MT, Lynam JG, Helal Uddin M, Coronella CJ. Hydrothermal carbonization: fate of
inorganics. Biomass Bioenergy. 2013;49:86–94.
40. Liu Z, Balasubramanian R. Upgrading of waste biomass by hydrothermal carbonization
(HTC) and low temperature pyrolysis (LTP): a comparative evaluation. Appl Energy.
2014;114:857–64.
41. Vassilev SV, Baxter D, Andersen LK, Vassileva CG, Morgan TJ. An overview of the organic
and inorganic phase composition of biomass. Fuel. 2012;94:1–33.
42. Saddawi A, Jones JM, Williams A, Coeur CL. Commodity fuels from biomass through
pretreatment and torrefaction: effects of mineral content on torrefied fuel characteristics and
quality. Energy Fuel. 2012;26:6466–74.
43. Baxter LL, Miles TR, Miles JTR, Jenkins BM, Milne T, Dayton D, Bryers RW, Oden LL.
The behavior of inorganic material in biomass-fired power boilers: field and laboratory
experiences. Fuel Process Technol. 1998;54:47–78.
44. Monti A, Virgilio ND, Venturi G. Mineral composition and ash content of six major energy
crops. Biomass Bioenergy. 2008;32:216–23.
45. Saidur R, Abdelaziz EA, Demirbas A, Hossain MS, Mekhilef S. A review on biomass as a fuel
for boilers. Renew Sustain Energy Rev. 2011;15:2262–89.
46. Li H, Liu X, Legros R, Bi XT, Lim CJ, Sokhansanj S. Pelletization of torrefied sawdust and
properties of torrefied pellets. Appl Energy. 2012;93:680–5.
47. Gil MV, Olego P, Casal MD, Pevida C, Pis JJ, Rubiera F. Mechanical durability and combustion
characteristics of pellets from biomass blends. Bioresour Technol. 2010;101:8859–67.
48. Obernberger I, Thek G. Physical characterization and chemical composition of densified
biomass fuels with regard to their combustion behavior. Biomass Bioenergy. 2004;27:653–69.
260 Z. Liu et al.
49. Junginger M, Sikkema R, Faaij A. Analysis of the global pellet market. Utrecht: Copernicus
Institute, Utrecht University; 2009.
50. Liu Z, Balasubramanian R. Preparation and characterization of fuel pellets from woody
biomass, agro-residues and their corresponding hydrochars. Appl Energy. 2014;113:1315–22.
51. Shaw MD, Karunakaran C, Tabil LG. Physicochemical characteristics of densified untreated
and steam exploded polar wood and wheat straw grind. Biosyst Eng. 2009;103:198–207.
52. Kong L, Tian SH, He C, Du C, Tu YT, Xiong Y. Effect of waste wrapping paper fiber as a
“solid bridge” on physical characteristics of biomass pellets made from wood sawdust. Appl
Energy. 2012;98:33–9.
53. Stelte W, Holm JK, Sannadi AR, Barsberg S, Ahrenfeldt J, Henriksen UB. A study of bonding
and failure mechanisms in fuel pellets from different biomass resources. Biomass Bioenergy.
2011;35:910–8.
54. Liu Z, Zhang FS. Effects of various solvents on the liquefaction of biomass to produce fuels
and chemical feedstocks. Energ Convers Manage. 2008;49:3498–504.
55. Simons SJR. Liquid bridge in granules. In: William JC, Allen T, editors. Handbook of power
technology. Netherlands: Elsevier; 2007. p. 1257–316.
56. Nielsen NPK, Gardner DJ, Felby C. Effect of extractives and storage on the pelletizing process
of sawdust. Fuel. 2010;89:94–8.
57. Kalyan N, Money RV. Natural binders and solid bridge type binding mechanisms in briquettes
and pellets made from corn stover and switchgrass. Bioresour Technol. 2010;101:1082–90.
58. Lam PS, Sokhansanj S, Bi X, Lim CJ, Melin S. Energy input and quality of pellets made from
steam-exploded douglas fir (Pseudotsuga Menziesii). Energy Fuel. 2011;25:1521–8.
59. Shankar J, Wright C, Kenney K, Hess R. A technical review on biomass processing: densifica-
tion, preprocessing, modeling, and optimization. Biofuels and Renewable Energy Technologies
Department, Energy Systems and Technologies Division, Idaho National Laboratory, ASABE
meeting presentation; 2010.
60. Yuan S, Chen XL, Li W, Liu H, Wang F. Nitrogen conversion under rapid pyrolysis
of two types of aquatic biomass and corresponding blends with coal. Bioresour Technol.
2011;102:10124–30.
61. Yuan S, Zhou Z, Li J, Chen X, Wang F. HCN and NH3 (NOx precursors) released under rapid
pyrolysis of biomass/coal blends. J Anal Appl Pyrol. 2011;92:463–9.
62. Chang L, Xie Z, Xie KC, Pratt KC, Hayashi JI, Chiba T, Li CZ. Formation of NOx precursors
during the pyrolysis of coal and biomass. Part VI. Effect of gas atmosphere on the formation
of NH3 and HCN. Fuel. 2003;82:1159–66.
63. Liu Z, Quek A, Parshetti G, Jain A, Srinivasan MP, Hoekman SK, Balasubramanian R. A
study of nitrogen conversion and polycyclic aromatic hydrocarbon (PAH) emissions during
hydrochar-lignite co-pyrolysis. Appl Energy. 2013;108:74–81.
64. Kilpinen P, Hupa M. Homogeneous N2 O chemistry at fluidized bed combustion conditions: a
kinetic modeling study. Combust Flame. 1991;85xai:94–104.
65. Yuan S, Zhou Z, Li J, Chen X, Wang F. HCN and NH3 released from biomass and soybean
cake under rapid pyrolysis. Energy Fuels. 2010;24:6166–71.
66. Hansson KM, Samuelsson J, Tullin C, Amand LE. Formation of HNCO, HCN, and NH3
from the pyrolysis of bark and nitrogen-containing model compounds. Combust Flame.
2004;137:265–77.
67. Glarborg P, Jensen AD, Johnsson JE. Fuel nitrogen conversion in solid fuel systems.
Prog Energy Combust Sci. 2003;29:89–113.
Chapter 10
Supercritical Water Oxidation (SCWO)
for Wastewater Treatment
Mesut Akgün and Onur Ömer Söğüt
Abstract Water behaves as an acidic and alkaline precursor for acidic or basic
reactions, since the formation of both H3 OC and OH– ions takes place in accor-
dance with the self-dissociation of water at near-critical and above supercritical
(Tc D 374 ı C, Pc D 22.1 MPa) conditions. Therefore, supercritical water is consid-
ered both as a solvent for organic materials and as a reactant at processes such as the
oxidative treatment of wastewaters, the gasification of aqueous organic solutions and
the production of fine metal oxide particles. Supercritical water oxidation is a very
efficient method for wastewater treatment, which is based on oxidation of organic
compounds in aqueous media above critical temperature and pressure conditions of
pure water.
In this chapter, general information on supercritical water oxidation processes
is given, effects of operational parameters such as temperature, pressure, reaction
time, waste and oxidant concentration on waste treatment are discussed, and applied
scientific methods and practical solutions to possible operational problems such
as corrosion, salt deposition and carbonization are compiled in detail. Treatment
of industrial wastewaters such as olive mill, textile dyehouse, cheese whey and
commercial pesticide with high environmental hazard potential are subjected and
evaluated in order to understand the effects of reactor temperature, organics
and oxidant concentrations, residence time and reactor pressure. Hydrothermal
degradation and oxidation kinetics of wastewater are stracked in terms of total
organic carbon (TOC) and analyzed, followed by discussion of supercritical water
oxidation rate models.
Keywords SCWO • Industrial wastewater treatment • Textile wastewater •

Cheese whey • Pesticide • Reaction kinetics
M. Akgün () • O.Ö. Söğüt

SCFT—Supercritical Fluid Technologies Research Group, Chemical Engineering Department,
Yıldız Technical University, Davutpasa Campus, 34210 Istanbul, Turkey
262 M. Akgün and O.Ö. SöMgüt
10.1 Introduction
Consumption rates of the sources and the industrial products increase in parallel
with welfare and development levels of societies [1]. Increased energy, paper,
water, etc. production in order to satisfy required public consumption inevitably
brings along generation of polluting side products. Maximized consumption which
indicates the development rankings of countries causes vast amounts of waste
products.
The diversity of organic molecules is much higher in comparison to inorganic
molecules [2, 3]. This fact is valid for compounds in solid, liquid and gaseous
waste mixtures as well, which may contain hazardous compounds e.g. phenols, het-
erocyclic compounds comprising heteroatoms, polyaromatics, chlorinated organic
compounds and carboxylic acids. In order to prevent the medium and long term
hazards of these compounds on biosphere, the mentioned compounds should be
effectively and safely removed before they contaminate the soil, underground
and surface water basins with wastewater leakage and drainage, as well as the
atmosphere through the evaporation of hazardous volatile organics.
Treatment of concentrated hazardous organic wastes is one of the highest
priorities in our day. The concentrated hazardous organic wastes production of
only the armed forces of the USA is several thousand tons, and the waste sludge
production of paper industry in the same country is in magnitude of several million
tons [4]. Every year, mankind constantly generates huge amounts of hazardous
organic wastes, biological sludge and domestic waste water which need to get
dealed with. To remove the organic contents of wastewaters, several methods are
known e.g. sanitary landfilling, controlled incineration, carbon adsorption and wet
air oxidation and supercritical water oxidation (SCWO). None of them, including
SCWO, is perfectly suitable for low-cost treatment of all kinds of wastewater, since
each of these methods have their own advantages and constraints, which are briefly
evaluated below.
By sanitary landfilling, organic wastes buried underground are metabolized and
decomposed by bacteria; but various poisonous metabolites like dioxins and furans
kill the bacteria before complete destruction of hazardous organics content of the
buried wastes. Therefore, the landfilling areas are required to extend consistently, in
the long term, landfilling is not completely suitable to be considered as a sustainable
waste treatment method [5].
Incineration is one of the methods which provides complete destruction of
organic wastes in very short time. Incineration processes, which reach to temper-
atures around 900–1,100 ı C and utilize 100–200 % of excess air, characteristically
require high start-up and operational costs [6]. Requirement for economical opera-
tion of incineration processes is a minimum of 30 % organics content in wastewa-
ters, and incineration of wastewaters with organics content less than the mentioned
ratio requires utilization of additional fuel [7], thus increasing the operational
costs. Additionally, incomplete oxidation of various organic compounds results
in generation of unwanted by-products e.g. halogenated volatile or polyaromatic
10 Supercritical Water Oxidation (SCWO) for Wastewater Treatment 263
hydrocarbons, oxides of nitrogen, oxides of sulphur, carbon monoxide, dioxines,

dibenzo furan and solid particles which contaminate the atmosphere [8]. Additional
facilities for preventing unwanted by-products from contaminating the atmosphere
increase the start-up and operational costs of incineration processes.
Activated carbon adsorption is used by the treatment of wastewaters having
diluted organics content. In this method, used up activated carbon which has lost
its adsorptive capacity should be regenerated using organic solvents, or handled as
a solid waste by land filling or incineration; where each one of these solutions can
be considered as generation of new waste, instead of disposal of the same.
Biological treatment combined with chemical treatment is a common method
for treatment of wastewaters with diluted organics content. Yet, not every organic
compound is suitable for treatment with biological methods [9]. The selection of
microbiological species for biological treatment is of vital importance for these
processes. Microorganisms used for wastewater treatment may be sensitive against
some organic compounds present in the bioreactor. Wastewaters are usually complex
mixtures comprising various organics, hence it is rather difficult to specify the
microorganisms suitable for use at the treatment of every wastewater sample.
Additionally, biological treatment is a slow process which usually takes days, which
translates into huge bioreactor volumes thus bio-treatment plant areas for plants with
high wastewater generation capacities.
Another method for treatment of organic content is wet air oxidation (WAO).
This process is usually operated under 150–350 ı C of temperature and 1–20 MPa
of pressure, in which conditions two phases – liquid and gas – coexist in the same
reactor. Whilst absorption of air into water under standard conditions, gas bubbles
migrate in the opposite direction of the gravitation; thus the mass transfer is under
several limitations such as short contact time, bubble surface area and saturation
concentration of the gas in water. Furthermore, the contact of a liquid hydrocarbon
with liquid water results with formation of two different phases which results in
coexistence of three liquid phases in a reaction environment. To overcome the
mass transfer restrictions, the residence times are in order of magnitude of hours,
which requires high reactor volumes for plants with high wastewater generation
capacities. Especially the start-up and operational costs tend to be enormous for
high volume WAO reactors which commonly are operated under pressures of around
10 MPa. Moreover, since the process conditions are not harsh enough to complete
the oxidation of refractory organics, the organics conversion rates reaches up to 75–
90 % and several refractory compounds such as acetic acid or ammonia leave the
reactor without complete degradation [7, 10], which pose necessity for additional
facilities thus increased costs.
SCWO is another organics removal method which requires high setup costs,
since it requires reactor systems suitable to be operated under harsh conditions:
at least 22.1 MPa pressure and temperatures over 374 ı C. This method is capable of
oxidizing any type of organics content into pure water and carbon dioxide in very
short residence times from seconds to minutes and accordingly small reactor vol-
umes. In turn, the required reactor conditions bear several difficulties to solve such
as plugging and corrosion, which are also discussed in this chapter. The wastewater
should preferably have a certain lower limit of organics concentration to work

under autothermal conditions for minimizing the operational costs. SCWO can be
considered as a favorable method for wastes requiring fast and total removal of
especially toxic and non-biodegradable organics.
10.2 Physical Properties of Supercritical Water
Critical temperature (Tc ) and pressure (Pc ) of water are 374.14 ı C and 22.064 MPa,
respectively. Both its temperature and pressure are higher than these values at
supercritical state. Compressibility (Zc ) and acentric (¨) factors of water at the
critical point are 0.229 and 0.344, respectively [11]. The occasions to encounter
with supercritical water are very rare in the nature. For instance, a magma leakage
at the ocean bottom 3 km below sea level was viewed as it turns the seawater into its
critical phase [12]. Likewise at a drilling operation within the scope of geothermal
energy investigations near Iceland, supercritical water found 2,500 m below the sea
level, where the required heat was provided by magma again [13].
Supercritical water can be considered as a high density vapor which is miscible
with other gases regardless of their proportions. Transition from ambient to
supercritical conditions increases both the kinetic energy values and intermolecular
distances, thus; hydrogen bonds between water molecules get highly difficult to
maintain under supercritical conditions, where because of these reasons water shows
a solvent behavior comparable with that of medium-polarity fluids [7, 14]. Some
properties of water under standard and supercritical conditions are comparatively
shown in Table 10.1. Hydrocarbons are easily miscible with supercritical water,
while the solubility of inorganics which have high polarity dramatically decrease
to ppm levels. Aqueous mixtures of non-polar organic substances heated and
compressed into sufficient temperature and pressure levels, become fit for treatment
with supercritical water oxidation thanks to SCW’s suitable properties e.g. dielectric
constant, viscosity, diffusivity, and hydrocarbon solubility.
Physical properties of supercritical water offer an ideal medium for oxidation
of organic substances. Oxygen and the organics dissolved in a single supercritical
water phase can rapidly react with each other because of high velocity molecular
Table 10.1 Comparison of water properties under standard and supercritical conditions
Under standard
Feature conditions [15] Supercritical fluid [15, 16]
ı
Temperature ( C) 25 400 500
Pressure (MPa) 0.1 25 25
Density (kg m3 ) 1,000 170 78
Static dielectric constant 78.5 5.9 1.46
pKw 14 19.4 23
Dynamic viscosity (kg m1 s1 ) 89 10 5 3 10 5 3 10 5
Table 10.2 Relative Organic compound Dielectric constant (©)

dielectric constants of several
organic substances at 25 ı C Propane 1.6
[4] Hexane 1.8
Heptane 1.9
Carbon tetrachloride 2.2
Benzene 2.3
Acetone 20.7
Ethanol 24.5
Methanol 32.6
movements and high frequency effective collisions in correlation with elevated

system temperatures. Additionally, the reactions are facilitated by minimized mass
transfer limitations thanks to the absence of phase boundaries in supercritical
medium.
10.2.1 Ion Product and Dielectric Constant
The ratio of relative static permittivity of a matter (dielectric) to that of vacuum

(practically air) is called “relative dielectric constant”, where relative static permit-
tivity means the capacitance of a capacitor comprising two conductive plates with
the dielectric between each other, when under voltage [17]. This concept is used by
characterization of molecular structure of a substance in terms of dipoles [18].
Polar water molecules under standard conditions position close to each other
and build hydrogen bridges, thus having a high dielectric constant of around 78.5
and resulting in high solubility of polar or ionic inorganic substances in water.
This high dielectric constant also means that water is around 78.5 times more
conductive compared to the air. The distance between water molecules rise with
higher temperatures, thus number of available hydrogen bridges and the value of
relative dielectric constant decrease consequently [19]. At the critical point, the
number hydrogen bridges becomes around one third of that at standard conditions,
and the value of relative dielectric constant falls to around 6, which further
decreases to 2–3 at conditions of 25 MPa and 500–600 ı C where supercritical
water oxidation processes are usually operated. The relative dielectric constant
values of supercritical water are close to that of organic solvents e.g. hexane and
heptane under standard conditions, which are 1.8 and 1.9, respectively. Thus, the
solvent behavior of water changes in favor of non-polar organics solubility, and
the solubility of polar and ionic inorganic substances minimize and they leave the
aqueous mixture accordingly.
Relative dielectric constants of several organics which have various polarities are
shown for at 25 ı C in Table 10.2 [4]. High relative dielectric constant values show
that the molecule polarity of a substance, thus its solubility in water under standard
conditions is accordingly high as well. The substances which have closer the relative
dielectric constant dissolve in each other easily.
The solubility of benzene in water under standard conditions is 0.07 % (w/w),
which could be practically considered as insolubility. At 260 ı C, that solubility rises
to 7–8 %, independent from pressure. Starting from 287 ı C it starts to be dependent
to pressure and under about 20–25 MPa it rises to 18 %, and to even 35 % at 295 ı C.
At 300 ı C, past the critical temperature of benzene-water binary mixture, it builds an
only homogeneous phase; thus benzene and water can be miscible in all proportions
[4, 19].
The self-dissociation of water into its ions is a reversible reaction shown in
Eq. 10.1, where its ion product is given in Eq. 10.2.
2 H2 O $ H3 OC C OH (10.1)

Kw D H3 OC ŒOH (10.2)
Here, Kw is the ion product of water, [H3 OC ] and [OH ] represent the ion
concentrations of hydronium and hydroxyl, respectively. The ion product of
water under standard conditions is Kw D 1014 mol2 L2 and its value increases
with temperature. In supercritical water oxidation process, during the preheating
Kw D 1011.4 mol2 L2 at 320 ı C and 25 MPa which indicates a rise in self
dissociation of water compared to that at standard conditions. At 400 ı C 25 MPa
and 420 ı C 25 MPa, the self-dissociation of water decreases and Kw becomes
1019.4 mol2 L2 and 1020.9 mol2 L2 , respectively.
It is possible to govern the chemical reactions occur in the aqueous media in favor
of ionic or free radical reactions by adjusting the ion product of water with changing
the temperature and pressure [15, 19]. In a typical supercritical water conditions,
which is 25 MPa of pressure and 500–600 ı C of temperature, the ion product of
water has small values (Kw < 1022 ) [20]. At such small ion product and relative
dielectric constant values of supercritical water, radical reactions predominates
the ionic reactions, and also non-polar organic substances disperse in the water
homogeneously, thus the chemical reactions occur with high rate constants without
getting influenced of mass transfer limitations [14].
10.2.2 Thermal Conductivity, Mass Transfer Coefficient

and Viscosity
The viscosity of supercritical water is very low thanks to rapid movements of

particles with high kinetic energy, resulting in increased diffusivity in supercritical
water compared to that in water at standard conditions, thus increased organics mass
transfer rates, also the frequency of intermolecular collisions and the reaction rates
[4, 7]. Under standard conditions, water has a viscosity of 1.002 10 5 kg m 1 s 1 ,
thermal conductivity of 0.6 W m1 K1 , and diffusivity of 2.3 10 9 m2 s 1 ; and
benzene diffusivity in water is 1.1 10 9 m2 s 1 . The magnitudes of these properties

differ dramatically in supercritical water.
Because of increased particle kinetic energies in elevated temperatures, thermal
conductivity of water under 400 ı C – 25 MPa and under 500 ı C – 25 MPa are
2,700 and 1,658 times higher than that under standard conditions [15, 21]. Also
the viscosity of supercritical water at 400 ı C – 25 MPa and 500 ı C – 50 MPa is
around 3 and 4 % compared to that at standard conditions. The diffusivity of water
and benzene diffusivity in water at critical point of water is 24 times higher and 13
times higher than at standard conditions, respectively. Thus, low viscosity and high
diffusivity in supercritical water media combined with high particle kinetic energies
in heated media facilitate effective collisions between molecules, thus tremendous
increments in reaction rates.
10.3 Supercritical Water Oxidation
Elimination [22], hydrogenation [23, 24], dehydration [25, 26], condensation [27],
partial and total oxidation [28, 29] in supercritical water are being investigated by
several research groups throughout the world. Especially monomer synthesis by
hydrolytic de-polymerization of several plastic wastes, and wastewater treatment
by total oxidation in supercritical water attracted intensive scientific interest [14].
Physical properties of supercritical water offer an ideal medium for oxidation of
organic substances. Oxygen and the organics dissolved in a single supercritical
water phase can rapidly react with each other because of high velocity molecular
movements and high frequency effective collisions in correlation with elevated
system temperatures. Additionally, the reactions are facilitated by minimized mass
transfer limitations thanks to the absence of phase boundaries in supercritical
medium.
Oxidation of organic contents of aqueous streams by mixing with oxygen
generally is called hydrothermal oxidation (HTO), which occurs under 350–
650 ı C of temperature and 14–70 MPa of pressure [30]. This process is called
supercritical water oxidation (SCWO) when operated over 374 ı C and 22.064 MPa
[7, 31]. In WAO, the operating conditions of 150–300 ı C and 1–20 MPa are
milder compared to SCWO, but the corresponding reaction times are around 100
times longer compared to that at SCWO, thus requiring huge reactor volumes.
Even the pressure levels at WAO is lower than that at SCWO, maintaining high
pressures in WAO reactor volumes means considerable increases in setup costs.
Additionally, the WAO reactor conditions isn’t sufficiently effective for eliminating
refractory contaminants and intermediate products [7, 10]. Incineration processes
have their own shortcomings including incomplete oxidation and inducing air
pollution by generating nitric oxides in high temperatures. Removal of water in
aqueous wastes to be incinerated is a significant contribution to operating costs. The
high operating incineration temperatures cause severe corrosion of the incinerators
Table 10.3 System conditions at several oxidation processes [32]

System System Residence time
Process temperature (ı C) pressure (MPa) (min) Phases
Controlled 800–1,100 1 – Three phases:
incineration sClCg
WAO 150–300 1–20 10–100 Three phases:
sClCg
HTO 350–650 14–70 0.1–10 s C l C g or scf
SCWO 400–650 25–100 0.1–1 One phase: scf
s solid, l liquid, g gas, scf supercritical fluid
which therefore frequently should be replaced [4, 7]. In these respects, SCWO has
its superiorities over incineration processes as well. A comparative table showing
operating conditions of WAO, incineration and SCWO is provided in Table 10.3.
Since the temperature in SCWO reactor is not as high as that in incinerators, the
oxidation steps where the nitrous and sulphur oxides generate do not occur, hence
SCWO process do not produce typical unwanted gases such as SOX and NOX . Thus,
the typical effluents of SCWO reactors basically consist of clean water which is
nearly pure and ready to be released into the nature; clean solids in form of metal
oxides and salts, and non-poisonous gases such as CO2 and N2 [19]. The gaseous,
liquid and solid products generated by SCWO of wastes are usually considered as
non-toxic.
Another potential application area of SCWO is the recovery of agricultural
soil contaminated with toxic organics, where they can be extracted and oxidized
in supercritical water. Fine mud comprising 1–20 % of organic wastes can also
be fed into SCWO reactors without subjecting to extraction. The reactor effluent
consists of water, carbon dioxide and nitrogen dioxide, and the resulting soil is
sterilized and free from organic contaminants. Also concentrated organic wastes can
be mixed with dilute wastewaters to have adequate burning heat, which corresponds
to 1–20 % organics content, and fed into SCWO process [7].
The SCWO processes are highly flexible: they can easily be scaled up; be sta-
tionary or mobile by mounting on trailers; be run at variable capacities ranging from
liters to cubic meters, by employing adjustable pumps [7]. 40–400 cubic meters/day
of toxic waste with organics content of around 10 % is economically applicable in
large scale stationary facilities [31]. In spite of the advantages mentioned above,
the treatment of industrial wastewaters with SCWO is not widely known and
implemented worldwide [6, 32, 33]. High investment costs due to high working
pressures and temperatures, potential cloggings due to ignoring the formation of
salts, corrosion costs based on wrong choices of reactor materials intimidate the
investors and make SCWO difficult to increase its popularity. Therefore, studies
of real wastewaters by SCWO at pilot plant scale are scarce, and application
at commercial scale is very rare. Notwithstanding, the current process designs
minimize corrosion and cloggings due to salt formation [34–39].
High operational costs are an important item in SCWO process. However, SCWO
of wastes with high flammable organics content e.g. waste oils, solvents, cleansing
mixtures, paints, paper and sewage sludge, refinery wastes provide considerable
reaction heat in the reactor. Therefore, they provide economical provident reducing
heating costs [4, 7, 31]. On the other hand, hydrothermal energy generated during
SCWO and a portion of energy used for preheating of the feed streams can be
recovered with proper heat integration and minimizing heat losses by improving
the thermal isolation of the reactor and process lines [7, 39].
Eco Waste Technologies, subordinated to Huntsman Co. from Austin, Texas, US
implemented SCWO treatment facility with capacity of 19 L/min for wastewaters
containing alcohols and amines; and reported treatment costs estimation of $0.03–
0.05 per liters for a plant run under capacity of 380 L/min. These values are close
to operational cost of incineration in a cement furnace with setup cost of around ten
folds compared to that of the corresponding SCWO plant [31].
Vadillo et al. evaluated the SCWO of flammable industrial wastewaters from an
economical point of view, and obtained an estimated overall cost of A
C109 per ton for
the SCWO treatment of wastewater in a 1 m3 h1 plant with tubular reactor system,
with a depreciation time over 10 years. Here, the composition of the wastewater is
assumed to be suitable for treatment with continuous SCWO. Additionally, it is also
reported that a such plant can be competitive with an optimized energy recovery,
and is favorable from an environmental point of view [39].
10.4 SCWO Processes
Basic flowchart of a typical SCWO process is shown in Fig. 10.1. The feed stream
is a wastewater rich with organic substances. Typically air, oxygen or hydrogen
peroxide is fed into the system, in case of lack of oxidants in the feed stream. If the
organics content of the feed stream is insufficient, fuel can be added for supporting
the oxidation. The streams entering to the SCWO reactor are heated and pressurized
to the supercritical conditions. Industrial SCWO processes generally are designed to
be run at temperatures below 700 ı C [31]. SCWO reactor systems can be operated
either as batch or continuous systems. At industrial scale, since batch operations are
not economically feasible, discussions in this section are predominantly focused on
continuous SCWO systems.
For maintaining the system pressure around 25 MPa at continuous systems,
usually back pressure regulators are employed. The system pressure is kept basically
constant from feed entrance to effluent exit. The feed streams are preheated to
subcritical temperatures and they reach the reaction temperatures owing to the
reaction heat released during the progression of organics oxidation reactions.
Maximum exergy can be seek by organizing the heat integration systems so that
the temperature differences between contacting streams are as low as possible, thus
the entropy generation and operating costs can be minimized.
Wastewater with 10 % organics content preheated to 350 ı C is expected to
provide reactor temperatures of around 550 ı C upon undergoing SCWO. Gaseous
intermediate products are formed by decomposition of larger organic molecules
Fig. 10.1 Basic flow scheme of a continuous SCWO process
present in wastewaters under high pressures exceeding 22 MPa and temperatures

around 375–450 ı C. Decomposition reactions presumably occur in preheating stage
possibly result in carbonaceous solid particles, i.e. carbonization products which can
clog the pipelines. In order to prevent this phenomenon, a rapid preheating should
be performed. Preheating step may further be accelerated by combining the feed
with overheated supercritical aqueous stream, thus the presumable carbonaceous
solids can be dissolved and carried by the supercritical stream into the reactor where
oxidation occurs [7].
Among the intermediate products, hydrogen, carbon monoxide and hydrocarbons
containing more than two carbon atoms are easy to get ignited around 200–250 ı C,
where more refractory molecules like methane require higher temperatures of
around 300–400 ı C [7]. The organic compounds instantly undergo oxidation upon
arriving the reactor which is operated under harsh conditions of typically 450–
550 ı C and 25 MPa. The reaction heat reaches to around 600 ı C with higher
organics concentrations. Acetic acid and ammonia, which are of most refractory
intermediate products generated in SCWO systems [4, 7, 10], rapidly oxidize along
with other present organics in harsh reactor conditions and with oxidants, which
are usually fed in excess. Organics conversion ratios of 99.99 % and above can be
reached within short reaction times ranging from seconds to minutes.
As briefly mentioned in the previous section, the heteroatoms present in organic
compounds, i.e. the elements apart from C, H, O and N form their acids. Thus,
hydrogen chloride, sulphuric acid and phosphoric acid form from chlorine, sulphur
and phosphorus present in organic content of the wastewater, respectively. These
acids can be neutralized by adding base solutions comprising alkaline and earth
alkaline metal ions to the feed stream, and separated from the reaction mixture in
form of corresponding inorganic salts [4, 31] by means of a separator e.g. cyclone
placed in the reactor exit, so that the reactor effluent will be maintained free of salts.
Afterwards, an extent of the energy transferred to the system can be recovered
by contacting the reactor effluent with the feed stream in a heat exchanger. In the
heat exchanger, the feed stream can be preheated to around 350 ı C and the effluent
is cooled to some extent in exchange. Following the heat exchanger, the effluent is
depressurized and subjected to phase separation where the fluid mixture separates
into liquid and gaseous streams.
Liquid wastes, halogenated and non-halogenated aliphatic and aromatic hydro-
carbons present in sludge and aqueous wastes, aldehydes, ketones, esters, carbo-
hydrates, nitrous organics, polychlorinated biphenyls (PCBs), phenols, benzene,
aliphatic and aromatic alcohols, pathogen bacteria and viruses, mercaptans, sul-
phides and other sulphurous compounds, dioxins and furans, dissolved metals,
various military wastes including low explosive compounds and smoke grenades,
paints and chemical weapons can be oxidized by SCWO into mainly water, carbon
dioxide, oxygen and nitrogen [4, 7, 31, 32].
10.5 Effects of System Parameters on SCWO Process
The main system parameters affecting SCWO process are temperature, pressure,
residence time, oxidant type and organics concentration in feed stream [31].
Generally, reaction orders for organics and oxidant in rate equations are around
one and zero, respectively [40]. It is probable to get a false idea such as the oxidant
concentration is not very influential on the reaction, but the actual reason of this
phenomenon is the fact that the SCWO processes are usually conducted in presence
of 200 % of excess oxidant [31]. At a certain temperature, especially around the
critical point of water, the system pressure dictates the component concentrations
and properties of water, the predominant species in the SCWO reaction medium.
10.5.1 Choosing the Oxidant
The reaction mechanisms emerging using hydrogen peroxide, oxygen or air differ
from each other, where it is reported that the most effective amongst them is
hydrogen peroxide [41]. Additionally, pumping of aqueous hydrogen peroxide and
mixing it with the feed stream homogeneously is easier compared to with other
oxidant options; and this can be achieved with a rather low-cost pumping system.
Fig. 10.2 Influence of oxygen concentration and excess oxygen on TOC conversion during the
treatment of Olive mill wastewater by SCWO at 600 ºC and 25 MPa ([TOC]0 D 21.61 g L1 ,
£D10 s) (Reprinted with permission from Erkonak et al. [42]. Copyright © 2008, Elsevier)
However, the purchase cost of hydrogen peroxide is higher than that of oxygen.
And when air is considered to be employed as oxidant instead of oxygen, even if
air is free of charge, since it comprises only 21 % of oxygen and the rest is mainly
inert, the compression efforts must be about five times higher compared to that by
feeding pure oxygen. At such medium capacity system, the compressor constitutes
the main cost item [7]. The optimum oxidant type can be decided by calculations
considering all of the factors including capacity, organics concentration in the feed,
and the specific requirements in regard with various oxidanttypes.
10.5.2 Initial Oxidant Concentration
Oxidants are used in stoichiometric excess by SCWO, where the main target is the
removal of organics with highest possible conversion rates. However, the initial
oxidant concentrations have their limits considering the oxidant purchasing costs
and pumping costs especially when a gaseous oxidant is selected.
Experiments [42] conducted at various initial oxygen concentrations (all in
excess) show that the initial oxygen concentrations are effective on organics con-
version, which is shown here in terms of total organic carbon concentrations (TOC).
Figure 10.2 shows that for this set of experiments, excess oxidant concentrations
greater than a certain value (here, 400 %) has no significant effect on organics
Fig. 10.3 Influence of organic concentration in Textile wastewater on TOC conversion during the
treatment by SCWO at 450 ºC and 25 MPa ([TOC]0 D 856.9 mg L1 , [O2 ]0 D 16.41 mmol L1 ,
£D10 s) (Reprinted with permission from Söğüt and Akgün [47]. Copyright © 2010, Wiley)
conversion ratio, which is already brought to 1 here; thus further increasing the
excess oxygen results in unnecessary costs without bringing any technical benefits.
As expected regarding to excess oxygen use, the rate expressions obtained through
kinetic experiments show that the reaction orders for oxidant have positive values
close to zero [42–45].
10.5.3 Initial Organics Concentration
The most important parameters for SCWO are the presence and concentration
the organic pollutants. To express the bulk concentrations of various organic
compounds, types and concentrations of which continuously change by thermal and
oxidative reactions in the course of SCWO process, the terms ‘total organic carbon
concentration’ (TOC) and ‘chemical oxygen demand’ (COD) are widely used [46].
An example showing the initial TOC concentration of the feed on TOC conversion
at SCWO of textile dyehouse wastewater is given in Fig. 10.3.
The SCWO reaction orders for organic pollutants in terms of TOC or COD are
found to be around 1, and this finding is widely accepted [42–44, 48, 49]. Hence,
the chemical reaction rates in a SCWO reactor strictly depend on initial organics
bulk concentration in terms of TOC or COD, which is however not the predominant
factor affecting the SCWO process [4].
Fig. 10.4 Influence of reaction temperature on TOC conversion at different O2 concentrations

during the treatment of Olive mill wastewater by SCWO at 25 MPa ([TOC]0 D 21.61 g L1 ,
£D10 s) (Reprinted with permission from Erkonak et al. [42], Copyright © 2008, Elsevier)
10.5.4 Temperature
Temperature in a reactor determines the kinetic energy, linear velocity and effective
collision probabilities of molecules, thus the rate constants; and the most affordable
reaction mechanism amongst various other possible paths according the activation
energies, and in consequence the variety and concentrations of ultimate products
[41, 42].
Oxidation, hydrolysis and thermal decomposition reactions occur simultaneously
in a SCWO reactor. At insufficient oxidant concentrations, hydrolysis and thermal
decomposition reactions are of great importance by removal of organics content
from liquid effluent. In which cases, considerable fractions of organic pollutants
break down into smaller molecules without being completely oxidized, and leave
the reactor system in gas phase instead of remaining in the liquid effluent. The
positive effect of temperature is notable for wastewater treatment both oxidation
and non-oxidative decomposition reactions in SCWO reactors [42–44]. TOC con-
versions achieved by two different initial oxidant concentrations, and various reactor
temperatures at olive mill wastewater treatment with SCWO [42] is shown in
Fig. 10.4.
Figure 10.5 shows total organics concentration and conversion ratio by treatment
of textile dyehouse wastewater in presence (SCWO) and absence (thermal or
hydrothermal degradation) of oxidant within the SCWO reactor, which is run several
times at different reactor temperatures. Note that high conversion ratios can be
achieved even in lower temperatures when oxidant is present. The TOC content
Fig. 10.5 Influence of reaction temperature for thermal and oxidative treatment by SCWO
at 25 MPa (Thermal: [TOC]0 D 5.6 g L1 , [O2 ]0 D 0 mmol L1 £D16 s; Oxidative:
[TOC]0 D 5.6 g L1 , [O2 ]0 D 591.2 mmol L1 £D7.8 s) (Reprinted with permission from Söğüt
et al. [45]. Copyright © 2011, IWA Publishing)
of the effluent is measured for only the liquid phase. The reason of high apparent
organics conversion ratios in absence of oxidant is formation of low molecular
weight organic products which leave the liquid effluent. Both the oxidative and
non-oxidative operations result in greater TOC removal from the fluid at higher
temperatures [45]. However, reactor temperatures have their own limits considering
the costs related to energy consumption for heating, and increased setup costs related
to selected corrosion resistant reactor building materials.
10.5.5 Pressure
Reactor pressure destines the component concentrations by determining the fluid

mixture density. The component concentrations increase with pressure. Accord-
ingly, it is possible to expect higher reaction rates at higher pressures at first
glance; but on the contrary, observations show that increments in pressure has
negative effects on treatment yield [19, 41, 45]. On the other hand, the regulation
of the system pressure for high conversions should not be performed at high
temperatures. Because, when the pressure is decreased, solids are formed as a result
of carbonization due to decreasing fluid density in the system, and cause system
cloggages with time. Actually, even though carbon conversion increase decreasing
Fig. 10.6 Effect of system pressure on TOC conversion during the treatment of olive mill
wastewater by SCWO at 600 ºC ([TOC]0 D 21.61 g L1 , ([O2 ]0 D 500 mmol L1 , £D10 s)
(Reprinted with permission from Erkonak et al. [42]. Copyright © 2008, Elsevier)
pressure or solvent density, formed and precipitated this solid carbon may cause that
the conversion is determined inaccurately.
The reaction rate constant is affected by pressure at high pressure reactions
in gas phase, according to ‘transition state theory’, which proposes a reversible
equilibrium between the hypothetical transition state (X¤ ) and the reactants (A
and B). The atoms are considered to gradually change their relative positions on
the corresponding molecules, from their starting positions on reactants towards
their final positions on products. As the molecule structures and related chemical
bonds change at every consecutive step, the value of internal energy of the system
changes accordingly. The mean internal energy within the reactor increases up to its
maximum value at the transition state, and decreases again while the products are
formed. Activation volume (V¤ ) is obtained by taking the partial derivative with
respect to pressure. V¤ , which is given in Eq. 10.3, is a residual function between
partial volumes of reactants and transition state, and gives a correlation between rate
constant and pressure [19, 50–52].

@ ln k V¤
D (10.3)
@P T RT
Organics removal by SCWO of olive mill wastewater and cheese whey under
various reactor pressures is shown in Figs. 10.6 and 10.7, respectively, in terms of
TOC conversion ratio [42, 45].
Fig. 10.7 Effect of system pressure on TOC conversion during the treatment of cheese whey
wastewater by SCWO at 450 ºC and 25 MPa ([TOC]0 D 5.6 g L1 , £D7.8 s) (Reprinted with
permission from Söğüt et al. [45]. Copyright © 2011, IWA Publishing)
10.5.6 Residence Time
The main target of SCWO processes is minimizing the organic pollutant concentra-
tions. With a suitable residence time, the reactor effluent leaves the SCWO system
with acceptably lowered organics concentrations. The residence time at continuous
reactor systems is set by deciding the reactor volume and arranging the fluid flow
rates. Total pollutant conversion (x D 1) is achievable at SCWO reactors with short
residence times changing from several seconds to minutes [7, 31, 42–45]. An
example is shown in Fig. 10.8 where total organics conversion is achieved within
15 s with sufficient oxidant supply, at treatment of olive mill wastewater by SCWO
in a tubular reactor [42].
10.5.7 Catalyst Use
SCWO reactions potentially result in formation of aromatic dimers and other

condensation products along with several refractory compounds e.g. acetic acid.
Complete oxidation of such refractory compounds require longer residence times
and high reactor temperatures over 600 ı C. Employing catalysts enables to capacity
increase without increasing the reactor volume, economize setup and operational
costs by reducing the residence time i.e. reactor volume by alternating the reaction
mechanism and modifying the reaction conditions into relatively moderate temper-
ature and pressure values [53].
Fig. 10.8 Effect of reaction time on TOC conversion during the treatment of olive mill wastewater
by SCWO at 600 ºC and 25 MPa ([TOC]0 D 21.61 g L1 , £D10 s) (Reprinted with permission
from Erkonak et al. [42]. Copyright © 2008, Elsevier)
The catalyst activities of several catalysts in descending order are KMnO4 ,

MnSO4 , Cu2C and Fe2C ; however it is reported that highest conversion rate is
observed using cupper salts as catalyst at acetic acid oxidation using H2 O2 as
oxidant [54, 55]. Mn-Ce and Cu-Zn catalysts are used for SCWO of short-chain
carboxylic acids. Cu-Zn catalyst is preferred by oxidation of phenolic compounds
and p-cumaric acid; and Ru/CeO2 catalyst shows much higher performance com-
pared to cupper salts in oxidation of formaldehyde and formic acid [4].
Heterogeneous and homogeneous catalysts have their own limitations regarding
to each other. Homogeneous catalysts leave the reactor in solution with the
effluent stream, and should be separated from the effluent, since the catalyst ions
(mainly heavy metal ions) harm the nature by contaminating the water sources,
and consistent catalyst loss turns into an important cost item. On the other hand,
heterogeneous catalysis is only suitable for use with homogeneous wastewaters,
since the catalysis shouldn’t be expected without maintaining contact between
the inner catalyst surfaces and organic pollutants. Even though corrosion is an
unwanted phenomenon, its products such as metal oxides and ions increase the
SCWO efficiency as catalysts [56].
10.5.8 Deciding the Operation Type
SCWO reactors can be operated in batch, semi-batch and continuous manners.

Continuous flow pipe reactor was employed at MODAR, the first patented SCWO
system [57]. Continuous SCWO reactors are not suitable for roughly ground organic
solids, since even if they get fluidized, they may cause clogging and blockages
around back pressure regulators [58, 59]. Therefor batch or semi-batch operated
reactor systems are more suitable for waste streams comprising solid matter.
Continuous operation is considered to be the most favorable for waste streams
free of solid matter [7], since other options has various shortcomings including
variable product quality, increased operational costs due to loading and discharging
by interrupting the reaction, reheating, re-pressurizing and cooling of the vessel at
every run.
10.5.9 Salt Precipitation and Corrosion
The presence of corrosive compounds and heteroatoms in feed stream cause two
main problems, first of which is the contamination of effluent and ash with metals
e.g. chromium dissolved from the reactor walls; and the latter, increased clogging
possibility because of solid corrosion products such as metal oxides, especially
around pressure regulators [60]. Ideally, the salts content in the wastewater should
not exceed several ppms in terms of order of magnitude [39].
Inorganic substances and heteroatoms e.g. halogens, phosphorus and sulphur
present in organic compounds convert into their acids, salts or oxides. These
inorganic salts and oxides are immiscible in supercritical water and leave the
phase where the reactions occur [31]. If acid formation is expected, the wastewater
stream should be fed with stoichiometric amount of base for neutralization of acids
into corresponding salts to be separated by subsequent precipitation. Phosphorus
converts into phosphate, sulphur into sulphate and nitrous compounds into N2 and
N2 O which can be catalytically reformed into nitrogen gas.
The minimization of these incidents by developing corrosion-resistant alloys,
new reactor concepts and investigation of corrosion kinetics is arousing scientific
interest. Main corrosion types encountered at SCWO systems can be listed [35] as
general corrosion, dealloying, pitting, stress corrosion cracking, crevice corrosion,
under-deposit corrosion, galvanic corrosion, hydriding, intergranular corrosion and
non-coupled corrosion, cross-section photos of examples for several of these
types are available in the literature [61]. As mentioned before, solubility of ionic
inorganic compounds in supercritical water is very low compared to in water under
standard conditions. NaCl hydrolysis observed at pressures around 10 MPa, which
is expressed in Eq. 10.4, basically disappears around 25 MPa, thus above the critical
pressure of water. In supercritical water, it is understood that NaCl is dissolved
physically according to Eq. 10.5, instead of chemical solvation [20].
NaCl.s/ C H2 O.g/ $ NaOH.s/ C HCl.g/ (10.4)
NaCl.s/ C nH2 O.g/ $ NaCl nH2 O.g/ (10.5)

Other alkali metal salts entering SCWO reactor systems dissolved in water
may also undergo hydrolysis and turn into their hydroxides, some of which form
insoluble crystals which cause scaling and erosion [62]. Several approaches for
minimization of corrosion at SCWO systems can be compiled [35] as follows:
• Forming a corrosion-resistant barrier:
– Use of high corrosion-resistant materials
– Liners of corrosion-resistant materials
– Coatings
• Prevention of corrosive species from reaching a solid surface:
– Transpiring wall/film-cooled wall reactors (explained below)
– Adsorption/reaction on fluidized solid phase
– Vortex/circulating flow reactor
• Managing/minimizing corrosion:
– Liners of sacrifice materials
– Use of adequate corrosion-resistant materials at short-term applications
• Adjusting process conditions to avoid or minimize corrosion:
– Pre-neutralization
– Cold feed injection
– Feed dilution
– Avoidance of corrosive feeds
– Effluent dilution/cooling
– Optimization of process operating conditions
• Prevention of corrosive species from reaching a solid surface:
– Transpiring wall/film-cooled wall reactors (explained below)
– Adsorption/reaction on fluidized solid phase
– Vortex/circulating flow reactor
It is important to secure the separation of salts from the main liquid stream
without problems. Under supercritical conditions, salts are sticky due to high
attractive forces between them, and tend to adhere to the reactor inner walls. Various
reactor concepts are developed in order to prevent this phenomenon. One of them
is based on continuous removal of high density brine solution from the lower-
temperature near-critical bottom of a vertical reactor, where the feed stream is fed
from a nozzle located at the top-center without touching the walls before the newly
formed salt crystals reach the bottom of the reactor, thus brine [19, 34]. A simple
drawing of the reactor is given at Fig. 10.9.
Another approach is the permeable-wall reactor, which comprises two coaxial
tubes, where the flow is also vertical and downwards. The outer tube is solid
and the inner one is made of porous metal or ceramic; center of which is the
reaction zone where the highest temperature of the system appears. Relatively low
Fig. 10.9 Continuous tank

reactor with two regions
from Marrone et al. [34].
temperature (near-critical temperature) water flow through the annulus between the
both tubes, and with higher pressure compared to the inside of the porous inner
tube so that it can transfer through the pores towards the tube axis. Here, water
forms a film carrying the salt continuously to the bottom of the reactor preventing
it from adhering to and causing corrosion on the reactor walls. If the water is under
critical temperature, it does so by dissolving the salts, otherwise the salt is dragged
with supercritical water, and the brine leaves the reactor system from the bottom. A
drawing of permeable-wall reactor is given at Fig. 10.10.
Another reactor concept with downwards flow is the “film-cooled reactor”, where
reactor inner wall is consistently washed with liquid water covering the wall by
forming a film. This liquid film prevents salts from adhering to the wall surface
by washing them away, moreover it minimizes corrosion thanks to low ion product
value of liquid water compared to that at supercritical water [19, 34, 35].
The adhesion of solids onto the reactor surfaces can also be minimized by
subjecting the reactor body to ultrasound waves [63]. Another way of achieving the
same target is maintaining the linear velocities of fluid streams within pipe reactors
at high values as in plug flow; so that the formed solid salts are dragged by high
velocity fluid stream, until they dissolve in liquid water by cooling upon exiting the
reactor. For example, Modell et al. have estimated minimum fluid velocities of about
1–5 ms1 in a horizontal reactor with an inner diameter of 6.7 mm for suspension
flow of solids under supercritical conditions. For scraping accumulating solids off
a tubular reactor, a solid body with a suitable shape can periodically be passed
Fig. 10.10 Film formation at

permeable wall reactor
from Marrone et al. [34].
through the reactor and took out from a dedicated exit. Also a periodical chemical
maintenance method comprising washing the tubular reactor by passing a solvent
liquid or solution through the reactor for dissolving the presumably deposited solids
on the reactor walls [34, 35].
Also clogging due to carbonization during preheating of organics containing
stream is one of the main problems encountered in SCWO systems. A reactor
concept is developed for minimizing this issue. According to this concept, a non-
preheated nonaqueous or concentrated organic waste stream with a reduced flowrate
is directly injected into a continuous SCWO reactor, as the reactor is already fed
with preheated diluted aqueous stream and a preheated oxidant stream (e.g. air) with
higher mass flowrates compared to the waste stream; so that the temperature of the
mixture of the streams is near-critical, and exceeds the critical temperature of water
as the organics start to burn. Thus, SCWO reactions start to occur. This concept is
reported to be applicable to CSTR, semi-batch and transpiring wall reactors as well
as tubular reactors [64].
10.6 Generalized Chemistry of SCWO
As explained in Sect. 10.2.1, SCWO reactions are mainly based upon radical
reaction mechanisms where the ion product of water, the most abundant species
in the SCWO medium, is at very low values. The reactions start and continue with
attacks of free radicals to organic molecules. The main radical reactions taking place
during SCWO can be listed as following equations [65]:
Initiation of the radical chain reaction:
RH C O2 ! R C HO2 (10.6)
Propagation, spreading of the reactions and alternative branches of the radical

chain reactions:
R C O2 ! RO2 (10.7)
RO2 C R0 H ! R00 OOH (10.8)
R00 OOH ! R00 O C HO (10.9)
2 RO2 ! O2 C 2RO (10.10)
RO2 ! HOOR0 (10.11)
HOOR0 ! HO C O D R00 (10.12)
Termination of the radical chain reaction:
RO2 C RO2 ! Products (10.13)
SCWO of methanol can be considered as a basic example for radical chain

reactions, proposed equations for which are given in Eqs. 10.14, 10.15, 10.16, 10.17,
10.18, 10.19, 10.20, and 10.21. M here represents any given radical existing in the
reaction medium [66]:
CH3 OH C HO ! CH2 OH C H2 O (10.14)
CH3 OH C HO ! CH3 O C H2 O (10.15)
CH3 O C M ! CH2 O C M (10.16)
CH2 OH C O2 ! CH2 O C HO2 (10.17)
CH2 O C HO ! HCO C H2 O (10.18)

HCO C O2 ! CO C HO2 (10.19)
CO C HO ! HCO2 C H2 O (10.20)
HCO2 C O2 ! CO2 C HO2 (10.21)
It is difficult to provide detailed reaction mechanisms for SCWO of high

molecular weight organic compound and complex mixtures, since the variety of
species is very high because of the capability of carbon atom to make four covalent
bonds. The presence and concentration of organic pollutants are monitored in bulk,
in terms of TOC or COD.
Some of the low molecular weight intermediate products formed during SCWO,
such as acetic and formic acids, methanol and carbon dioxide are refractory
against oxidation; kit is also obvious regarding to the activation energy values:
170–350 kJ mol1 for these low molecular weight refractory compounds, and 20–
100 kJ mol1 for higher molecular weight organics. Therefore, degradation of
low molecular weight organics which are basically the intermediate degradation
products of higher molecular weight organic pollutants; can be considered as rate-
limiting steps of SCWO [4, 67]. A SCWO product of nitrous organic compounds
is nitrogen gas; and in case of incomplete oxidation, also ammonia can also be
expected in some extent [7, 10].
Simplification of SCWO mechanisms result in Eqs. 10.22, 10.23, and 10.24 [67]:
1: A C O2 ! CI r1 D k1 ŒA/1 (10.22)
2: A C O2 ! BI r2 D k2 ŒA/2 (10.23)
3: B C O2 ! CI r3 D k3 ŒB/3 (10.24)
A, B and C represent initial organics which are readily oxidized, relatively refrac-
tory rate-limiting intermediate products and final oxidation products, respectively.
The A and B content in the mixture can be expressed in bulk concentrations as
TOC and COD. It is deducible that there is an inverse correlation between value of
k2 /k1 ratio and facility of complete oxidation of initial products, since low molecular
weight refractory intermediate products form in a negligible extent at low values of
this ratio, and it is possible to use k1 as the sole rate constant at rate expressions in
terms of bulk concentrations.
Selectivity, which indicates the formation tendency of refractory intermediate
products can be expressed as seen in Eqs. 10.25 and 10.26, by simplification that all
of the reactions during SCWO are of first order (’1 D ’2 D 1):
r2 k2 ŒA/2 k2
Sintermediate product D D /1 D (10.25)
r1 k1 ŒA k1
The expression takes the following form:

Ea;2
k2 k0 e RT k0 .Ea;2 Ea;1 /
Sintermediate product D D 2 Ea;1 D 20 e RT (10.26)
k1 k01 e RT k1
The selectivity in favor of formation of refractory intermediate products increases

with higher values of k2 . Value of k1 can be determined with kinetic investigation
based on initial reaction rates.
In a set of reactions comprising multiple steps, the reaction rates are dependent
on the values of activation energies. Since the bulk degradation of refractory
intermediate products has a greater activation energy (Ea,3 ) compared to that of other
reactions taking place in a SCWO medium, elevated reactor temperatures should be
provided in order to achieve an effective degradation of the intermediate products.
The difference between Ea,1 and Ea,2 are expected to be small, since B and C are
usually products of similar reactions started with similar initial species. Therefore,
the value of the term in the power of Euler number is near zero, and it is deducible
that the value of Sintermediate product should be around the value of the ratio between
frequency factors of both reactions:
k02
Ea;2 Š Ea;1 ) Sintermediate product Š (10.27)
k01
For example, the value of Sintermediate product is expected to be low by treatment of

wastewaters containing low molecular weight, except refractory species like acetic
acid, which directly undergoes the reaction nr. 3. Hence, it is possible to minimize
the heating costs by running the SCWO reactors at lowest possible temperatures,
i.e. around 400 ı C. According to these facts, selectivity can be considered as a
useful tool for determining the optimum reactor temperatures and characterizing
the wastewaters to be treated with SCWO.
10.7 Kinetics of Supercritical Water Oxidation
10.7.1 Calculations for Kinetics
Density data for complex mixtures under SCWO process conditions are difficult
to predict. Since water is the main component of the SCWO reaction mixtures,
the density of SCWO reaction mixtures is presumed to be that of pure water at
supercritical conditions [16]. Steam tables provide density data for pure water [68],
and it can be also estimated using equations of state e.g. Peng-Robinson Equation of
State (PR-EoS), which is suitable for estimation of pure water densities in elevated
temperatures and pressures. PR-EoS and the same in terms of density are given in
Eqs. 10.28 and 10.31, respectively [69].
RT a ’ .T/
PD (10.28)
V b V .V C b/ C b .V b/
0:45724 .RTc /2 0:07780 RTc h

p i2
aD ; bD and ’ .T/ D 1CK 1 T=Tc
Pc Pc
(10.29)
K D 0:37464 C 1:54226 ¨ 0:26992 ¨2 (10.30)

b3 P C b2 RT ab ¡3 3b2 P C 2bRT a ¡2 C .bP RT/ ¡ C P D 0 (10.31)
where a and b are generalized functions of critical temperature, critical pressure

and acentric factor (¨). The real root of obtained by solving Eq. 10.31 gives water
density at reactor conditions.
As shown Fig. 10.1, feed streams quickly reach to the reactor conditions whilst
passing through the preheating section. Since fluid density at reactor pressure
and temperature is different from those at feed conditions, fluid flow rate and
reaction time are required to be calculated for reactor conditions. Fluid flow rate
at supercritical conditions is calculated with Eq. 10.32.
¡feed
Freactor D Ffeed (10.32)
¡SCW .T; P/
where Freactor is total volumetric flow rate of waste and oxidant combined stream in
the reactor at supercritical conditions, Ffeed and ¡feed represent flow rate and fluid
density at feed conditions, respectively. ¡SCW is water density which depends on the
temperature and pressure at reactor conditions. Reaction time or residence time is
calculated using the below equation;
Vreactor ¡SCW .T; P/

£D (10.33)
Ffeed ¡feed
Here, £, Vreactor , and Ffeed represent residence time (s), inner volume of the
continuous reactor (mL) and the cumulative volumetric flow rate of the united
stream (mL s1 ), respectively.
As explained before, by removal of organic pollutants from wastewaters, the
organics concentrations are widely monitored in bulk in terms of chemical oxygen
demand (COD, mg(O2) L1 ) or total organic carbon concentrations (TOC, mg(C)
L1 ). At kinetic studies, particle concentration units are easier to conceive compared
to the above units; and their calculations are made using Eqs. 10.34 and 10.35,
respectively.
¡SCW .T; P/ Fwastewater

ŒCCOD D ŒCOD (10.34)
¡feed Ffeed
¡SCW .T; P/ Fwastewater
ŒCTOC D ŒTOC (10.35)
¡feed Ffeed
Where [COD] and [TOC] are contaminant concentrations in wastewater at ambi-

ent conditions, respectively. ¡SCW (T,P) and ¡feed are fluid densities in reactor and
ambient conditions, respectively. Fwastewater and Ffeed are wastewater and cumulative
stream volumetric flow rates towards the continuous SCWO reactor under ambient
conditions. Generally, wastewater and oxidant matter are pumped separately into
the reactor passing through a preheater. H2 O2 is the most preferred oxygen source
especially in the lab-scale applications, each mmol of which is completely thermally
degraded into 1 mmol of water and 0.5 mmol of oxygen during the preheating period
according to Eq. 10.36. For example, it is reported [70] that H2 O2 conversion is
about 98.5 % at 400 ı C and about 99.9 % at 440 ı C, for a residence time of 1 s.
1
H2 O2 ! H2 O C O2 (10.36)
2
Oxygen concentration in reactor conditions is calculated using Eq. 10.37. Here,
CO2 , [H2 O2 ] and FH2 O2 represent oxygen concentration in reactor conditions,
hydrogen peroxide concentration in oxidant storage vessel (with regard to Eq. 10.36)
and volumetric flow rate of the oxidant stream.
ŒH2 O2 ¡SCW .T; P/ FH2 O2

ŒCO2 D (10.37)
2 ¡feed Ffeed
10.7.2 Kinetic Models
Removal of organic contaminants from wastewater with supercritical water oxida-

tion is carried out simultaneously via thermal degradation and oxidation reactions.
These reactions can affect each other because of taking place at the same medium,
however, to define the magnitude of this synergy is not easy. Therefore, by
simplification with assuming that thermal decomposition and oxidation reactions
do not interfere each other, both rate expressions can be combined as cumulative
SCWO rate expression.
10.7.2.1 Kinetics of Thermal Degradation
There are several other terms used to refer to thermal decomposition in absence
of oxidants e.g. hydrolysis, pyrolysis and thermal degradation. Even though the
thermal degradation is performed in absence of oxidants, the dissociation of water
into H3 OC and OH– under supercritical conditions results in a reaction with the
organics of the wastewater. Therefore, supercritical water is considered as a solvent
for organic materials and as a reactant at the same time [44]. However, since the
reaction medium always consists of more than 98 % of water, its impact on organics
degradation rate is omitted, and the reaction order for water is considered as zero.
The thermal decomposition taking place in supercritical water can be expressed with
first order reaction kinetics given with Eq. 10.38 [45, 47].
d ŒC
D kthermal ŒC (10.38)
dt
Here [C], t and kthermal represent organics bulk concentration at reactor con-
ditions (mmol L1 ), time (s) and rate constant for thermal decomposition (s1 ),
respectively. The organics bulk concentration can be described in terms of organics
conversion ratio;
ŒC D ŒC0 .1 x/ (10.39)
Here, x, [C] and [C]0 represent the organic pollutants conversion ratio, cor-
responding organics concentration at reactor conditions, and the initial organics
concentration at reactor conditions, respectively. The reaction rate constant for
thermal degradation rate can be modified to Arrhenius equation;
Ea;thermal
kthermal D k0;thermal e RT (10.40)
Where k0,thermal and Ea,thermal represent the Arrhenius frequency factor and
the activation energy, respectively. Equation 10.38 can be rearranged and solved
analytically to provide Eq. 10.41 with respect to organic pollutants conversion ratio,
for the initial conditions x D 0 and t D 0;
Ea;thermal
x D 1 ek0;thermal e R T t
(10.41)
Here, residence time (£) can be used in place of time (t). Best-fit values of
frequency factor (k0,thermal ) and activation energy (Ea,thermal ) for thermal degradation
can be determined by non-linear multiple regression analysis. The values can be
obtained by minimizing the sum of the squared differences of the experimental and
the predicted decomposition ratio for all data points.
X
Nexp
2
xExp xPred (10.42)
i
Figure 10.11 shows experimental and predicted data obtained for several types
of wastewaters [45, 47, 71]. The results confirm that the first order kinetic model
Fig. 10.11 The accordance of experimental results with first order kinetics for thermal degradation
of several types of industrial wastewater
fits satisfactorily for thermal degradation of several industrial wastewater types in

supercritical water.
10.7.2.2 Kinetics of Oxidation Reaction
Oxidation of organics in wastewaters at supercritical conditions can be expressed

with Eq. 10.43. As explained before, supercritical water is considered as a solvent
as well as a reactant for organic materials [44, 67]. However, since the SCWO
reaction media usually consist of more than 98 % of water, its impact on the organic
disappearance rate is omitted, and the reaction order for water is considered as zero.
d ŒC
D koxi ŒCa ŒO2 b (10.43)
dt
[C] represents the bulk organics concentration in reaction media, in terms of
COD or TOC; [O2 ] oxygen concentrations; koxi the rate constant for oxidation; and
finally a and b the reaction orders for organics and oxygen, respectively. Moreover,
since oxidants are usually used in excess, and oxidant concentrations are open to
be considered as remaining constant, so that the Eq. 10.43 can be simplified into
Eq. 10.44.
d ŒC
D koxi ŒCa ŒO2 b0 (10.44)
dt
Fig. 10.12 Accordance between model equation and experimental results for textile dyehouse
wastewater, olive mill wastewater and cheese whey
Here, [O2 ]b0 represents the initial concentration of oxygen at reactor conditions.
Equation 10.45 is obtained with rearranging the Eq. 10.44 for organic contaminants
bulk conversion ratio and solved analytically using the initial conditions.
1a
1
Ea;oxi

x D 1 1 C .a 1/ k0;oxi e R T ŒCa1
0 ŒO2 b0 t ; .a ¤ 0/ (10.45)
where k0,oxi and Ea,oxi represent the Arrhenius frequency constant and activation
energy of the oxidation reaction, respectively. Here, residence time (£) can be used
instead of time (t). The kinetic parameters of oxidation reaction can be obtained by
minimizing the sum of the squared differences of the experimental and the predicted
decomposition ratio for all data points using Eq. 10.42.
As explained above, thermal and oxidative degradation reactions are considered
as simultaneous and independent from each other. Accordingly, the cumulative
reaction rates of both types of reactions give the overall organics degradation rate
of wastewaters by SCWO. The overall SCWO rate expression is given in Eq. 10.46.
The equation is regressed using all together of the thermal degradation and oxidation
data for the estimation of kinetic parameters.
d ŒC Ea;thermal Ea;oxi

D k0;thermal e R T ŒC C k0;oxi e R T ŒCa ŒO2 b (10.46)
dt
Figure 10.12 shows a comparison of experimental data and predicted data
obtained using Eq. 10.46 for some wastewaters. The results confirm that the
combined kinetic model for the thermal degradation and oxidation reaction is
satisfactory for the removal of organics from industrial and domestic wastewaters
at the supercritical conditions of water. Kinetic model parameters proposed for
SCWO of several aqueous organic contaminants mixtures including industrial
wastewaters are compiled and summarized in Table 10.4. As seen in Table 10.4,
some researchers propose merely oxidative kinetic models, where some others
provide model parameters considering both the oxidative and thermal degradation
reactions.
10.7.2.3 Kinetic Parameters and Discussion
SCWO kinetic parameters for various types of industrial wastewaters are provided
in Table 10.4, where k0,thermal and k0,oxi thermal and oxidative degradation frequency
factors, Ea,thermal and Ea,oxi activation energy values for thermal and oxidative
degradation, a and b the reaction orders for bulk organic pollutants and oxidants,
respectively. The results given in line 5 calculated for a textile wastewater sample
containing acetic acid are in accordance with the theory given in Sect. 10.6, since
the calculated thermal degradation activation energy of 104.12 kJ mol1 is much
higher than the other activation energies given in lines 1–4 and 6–8, and is rather
close to the activation energy values given in lines 10 and 11 dedicated to acetic acid
removal with SCWO, which are 131 and 172.7 kJ mol1 respectively [42–44, 49,
71, 72]. Since acetic acid present in textile wastewater with parameters given in line
5 is resistant against thermal degradation, thermal degradation can be considered as
the rate limiting step at SCWO of the wastewater.
On the other hand, C.I. Disperse Orange 25 (lines 1–3), C.I. Basic Blue 41
(line 4) and pesticide (o,o-dimethyl-2,2-dichloro-vinyl phosphate), (line 7) are
larger compounds high molecular weight. Their activation energy values for thermal
and oxidative degradation reactions have moderate magnitudes ranging between
18.88–36.084 kJ mol1 which means that these organic compounds decompose
and fully oxidize easily compared with acetic acid containing wastewaters. These
conclusions are in accordance with various other in literature [4, 7, 10, 67].
Additionally, for each wastewater and pollutant type mentioned in Table 10.4,
the effect of oxidant concentrations is less than that of organics concentration;
although the pollutants shown in lines 4 and 6, where the reaction orders for oxidant
concentrations are 0.4 and 0.32, respectively, with a possible reason that lower
oxidant excess was kept around 100 %. As mentioned before, SCWO processes
are generally run with about 200 % excess of oxidant [31], and it is not surprising
that a moderate effect of oxidant concentration was observed with lower values of
oxidant excess such as 100 %. Hence, oxidant-independent kinetic model equations
can only be considered as suitable for SCWO processes where at least 200 % of
excess oxidant is employed.
292
Table 10.4 Kinetic parameter values for use in rate expressions for SCWO treatment of wastewaters containing various organic pollutants
Wastewater type
Line # or pollutant k0,thermal (s1 ) k0,oxi (*) Ea,thermal (kJ mol1 ) Ea,oki (kJ mol1 ) a b Ref.
1 C.I. Disperse N/A 3.43 (˙1.5) N/A 27.8 1 0 [43]
Orange 25**
2 C.I. Disperse N/A 218.7 (˙2.1) N/A 40.181 1 0 [72]
Orange 25**
3 C.I. Disperse N/A 169.9 (˙1.6) N/A 37.441 (˙0.5) 0.964 0.064 [72]
Orange 25**
4 C.I. Basic Blue 41 0.84 (˙0.15) 2.8 (˙0.5) 25.89 (˙3.1) 18.88 (˙0.9) 0.84 (˙0.03) 0.32 (˙0.05) [44]
5 Textile dyehouse 1.59 (˙0.5) 105 5.181 (˙1.3) 104.12 (˙2.6) 18.194 (˙1.09) 1.169 (˙0.3) 0.075 (˙0.04) [47]
wastewater
6 Cheese whey 107.72 (˙4.1) 1.86 (˙0.5) 50.022 (˙1.7) 20.337 (˙0.9) 1.2 (˙0.4) 0.4 (˙0.1) [45]
7 Pesticide 1.954 (˙0.964) 0.628 (˙0.12) 19.425 (˙3.203) 36.084 (˙5.8) 1.15 (˙0.25) 0.2 (˙0.05) [71]
8 Olive mill 14.09 (˙1.05) 0.214 (˙0.5) 40.36 (˙0.46) 33.24 (˙0.9) 1.02 (˙0.031) 0.89 (˙0.054) [42]
wastewater
9 Olive mill – 15–30 – 35 1 – [73]
wastewater**
10 Olive mill – 200 – 35 1 – [74]
wastewater**
11 Oily sludge** – 8.99 1014 – 213.13 103 1.405 – [75]
12 Oily wastes** – 35 – 63 1 0.579 [76]
13 Cutting oil – 4.851 105 – 90.3 1 – [77]
wastes** 9.257 103 69.1
14 Biosolids 51,280 – 9.522 104 – 1 – [78]
15 LCD Manufact. – 2.78(˙0.71) 102 – 47.79 (˙1.52) 1.01 (˙0.01) 0.065 (˙0.01) [79]
Wastewater**
M. Akgün and O.Ö. SöMgüt
16 Landfill – 34.86 – 32.1 1 0 [80]
leachate**
17 Phenol** – 101.34 ˙ 0.77 – 39.2 ˙ 10.7 1 0 [30]
18 Hydrogen** – 3.95 1010 – 390 (˙60) 1.1 (˙0.25) 0.02 (˙0.29) [81]
19 Carbon – 4.91 103 – 164 (˙32) 0.96 (˙0.3) 0.34 (˙0.24) [82]
monoxide**
20 Carbon – 1.59 1011 – 167 1 0 [83]
monoxide**
21 Methanol** – 1.59 1026 – 97 (˙20.4) 1 0 [84]
22 Methanol** – 3.16 1026 – 408.4 1.1 0.02 [10]
23 Methanol** – 6.31 1028 – 107 (˙30) 0.89 (˙0.69) 0.12 (˙0.66) [84]
24 C3-C5 polyolsc 2 1013 –7 1013 – 173–177 – – – [85]
25 Acrylonitrile plant – 5.22 (˙1.74) 102 – 53.48 (˙33.57) 1 0 [86]
wastewater**
26 Acrylonitrile plant – 6.07 (˙6.89) 103 – 66.33 (˙5.87) 1.26 (˙0.15) 0 (˙0.15) [86]
wastewater**
27 Acetic acid** – 1.72 1015 – 219 1 0 [87]
28 Acetic acid** – 19.8 1010 – 308 1 0 [88]
29 Acetic acid** – 9.3 1010 – 172.7 1 0 [88]
30 Ethanol** – 794.4 (˙2.5) – 53.8 (˙4.6) 1 0 [89]
31 Ethanol** – 1.7 1017 – 214 (˙18) 1.34 (˙0.11) 0.55 (˙0.19) [90]
10 Supercritical Water Oxidation (SCWO) for Wastewater Treatment
32 Amonia** – 1019 (˙4.5) – 83 (˙19) 1 0.44 (˙0.3) [91]

33 Isopropyl amine** – 2.46 (˙0.65) 103 – 64.12 (˙1.94) 1.13 (˙0.02) 0.24 (˙0.01) [92]
*Frequency constants for oxidative degradation have various forms in accordance to corresponding rate expressions
**Parameter values are calculated only for oxidative degradation reactions
***Parameter values are calculated only for thermal degradation reactions in absence of oxidants
293
10.8 Conclusions
Despite the fact that homogeneous or heterogeneous aqueous organics mixtures

including solutions, emulsions and micro-particle suspensions are suitable for
treatment with continuous SCWO systems; suspensions with rough particles are
preferred to be treated in semi-batch or batch SCWO systems, for minimizing the
probability of mechanical difficulties. It is suggested that SCWO processes should
be designed and implemented in accordance with the following considerations
obtained from literature and our own research:
• Using at least 10 % of excess oxidant,
• Operating with system pressures not higher than 25 MPa,
• Oxidant selection (H2 O2 , O2 , air etc.) with optimization calculations regarding
to operational and capacity-dependent setup costs,
• Seeking minimum reactor temperature allowing desired pollutant conversion
ratio within chosen residence time,
• Heat integration for utilizing the enthalpy of reactor effluent.
Further research concerning SCWO is expected to focus on following subjects:
• Economical evaluation of intermediate products of hydrothermal degradation and
incomplete oxidation,
• Innovative heat integration systems concept designs for hydrothermal degrada-
tion and incomplete oxidation,
• Novel reactor inner wall coating materials and methods for prevention of
corrosion, scaling and intermediate products adhesion,
• Further kinetic investigations aiming for classification of system types and
conditions with regard to wastewater compositions.
Acknowledgments Financial support provided by The Scientific and Technological Research

Council of Turkey (TUBITAK) through project 104M214 is gratefully acknowledged.
References
1. Bolla V, Hauschild W, Hoffmeister O, Jung D, Lock G, Pavlovic A, Scheller A, Tronet V.

Sustainable development in the European Union, 2009 monitoring of the EU sustainable
development strategy. Luxemburg: Eurostat, European Commission; 2009.
2. McMurry J. Organic chemistry. 5th ed. Mason: Thomson Learning; 1999.
3. Carey FA. Organic chemistry. 4th ed. Boston: McGraw-Hill Higher Education; 2000.
4. Tang WZ. Physicochemical treatment of hazardous wastes. London: CRC Press; 2004.
5. Weber R, Watson A, Forter M, Oliaei F. Persistent organic pollutants and landfills – a review
of past experiences and future challenges. Waste Manag Res. 2011;29:107–21.
6. Lin H, Ma X. Simulation of co-incineration of sewage sludge with municipal solid waste in a
grate furnace incinerator. Waste Manag. 2012;32:561–7.
7. Modell M. Detoxification and disposal of hazardous organic chemicals by processing in super-

critical water. Research Report, US Army Medical Research and Development Command,
Maryland, USA; 1985.
8. Ledesma EB, Kalish MA, Nelson PF, Wornat MJ, Mackie JC. Formation and fate of PAH
during the pyrolysis and fuel-rich combustion of coal primary tar. Fuel. 2000;79:801–14.
9. Donlagic J, Levec J. Oxidation of an azo dye in subcritical aqueous solutions. Ind Eng Chem
Res. 1997;36:3480–6.
10. Shende RV, Levec J. Wet oxidation of refractory low molecular mass carboxylic acids. Ind Eng
Chem Res. 1999;38:3830–7.
11. Poling BE, Prausnitz JM, O’Connell JP. The properties of gases and liquids. 5th ed. New York:
McGraw-Hill Higher Education; 2001.
12. New Scientist, Environment, Found: The hottest water on Earth, http://www.newscientist.com/
article/dn14456-found-the-hottest-water-on-earth.html
13. BBS News, Science/Nature, Drilling into a hot volcano, http://news.bbc.co.uk/2/hi/sci/tech/
4846574.stm
14. Oshima Y. Oxidation and hydrolysis reactions in supercritical water. In: Arai Y, Sako T,
Takebayashi Y, editors. Supercritical fluids: molecular interactions, physical properties, and
new applications. Berlin: Springer; 2002.
15. Krammer P, Vogel H. Hydrolysis of esters in subcritical and supercritical water. J Supercrit
Fluids. 2000;16:189–206.
16. Zhou N, Krishnan A, Vogel F, Peters WA. A computational model for supercritical water
oxidation of toxic organic wastes. Adv Environ Res. 2000;4:79–95.
17. Rose A, Rose E. The condensed chemical dictionary. New York: Reinhold Publishing Co.;
1961.
18. Hoechst AG. Kleineswörterbuch der anwendungstechnik. Stuttgart: Klett; 1975.
19. Loppinet-Serani A, Aymonier C, Cansell F. Supercritical water for environmental technologies.
J Chem Technol Biotechnol. 2010;85:583–9.
20. Armellini FJ, Tester JW. Solubility of sodium chloride and sulfate in sub- and supercritical
water vapor from 450-550 ı C and 100–250 bar. Fluid Phase Equilib. 1993;84:123–42.
21. Franks F. Water: a comprehensive treatise. In: Franks F, editor. The physics and physical
chemistry of water. New York: Plenum Press; 1972.
22. Wang T, Yang M, Xiang B, Shen Z. Investigation on the elimination of organic substances in
urine by supercritical water oxidation. Space Med Med Eng. 1997;10:370–2.
23. Kobiro K, Sumoto K, Okimoto Y, Wang P. Saccharides as new hydrogen sources for one-pot
and single-step reduction of alcohols and catalytic hydrogenation of olefins in supercritical
water. J Supercrit Fluids. 2013;77:63–9.
24. Fedyaeva ON, Vostrikov AA. Hydrogenation of bitumen in situ in supercritical water flow with
and without addition of zinc and aluminum. J Supercrit Fluids. 2012;72:100–10.
25. Lehr V, Sarlea M, Ott L, Vogel H. Catalytic dehydration of biomass-derived polyols in sub-
and supercritical water. Catal Today. 2007;121:121–9.
26. Anikeev VI, Yermakova A, Manion J, Huie R. Kinetics and thermodynamics of 2-propanol
dehydration in supercritical water. J Supercrit Fluids. 2004;32:123–35.
27. Sasaki M, Furukawa M, Minami K, Adschiri T, Arai K. Kinetics and mechanism of cellobiose
hydrolysis and retro-aldol condensation in subcritical and supercritical water. Ind Eng Chem
Res. 2002;41:6642–9.
28. Brunner G. Near and supercritical water. Part II: Oxidative processes. J Supercrit Fluids.
2009;47:382–90.
29. Kıpçak E, Akgün M. Oxidative gasification of olive mill wastewater as a biomass source in
supercritical water: effects on gasification yield and biofuel composition. J Supercrit Fluids.
2012;69:57–63.
30. Portela JR, Nebot E, de la Ossa EM. Kinetic comparison between subcritical and supercritical
water oxidation of phenol. Chem Eng J. 2001;81:287–99.
31. Kroschwitz JI, Seidel A. Kirk-Othmer encyclopedia of chemical technology. 5th ed. Michigan:
Wiley-Interscience; 2004.
32. Yesodharan S. Supercritical water oxidation: an environmentally safe method for the disposal
of organic wastes. Curr Sci. 2002;82:1112–22.
33. Shaw RW, Dahmen N. Destruction of toxic organic materials using super-critical water
oxidation: current state of the technology. In: Kıran E, Debenedetti PG, Peters CJ, editors.
Supercritical fluids–fundamentals and applications. Dordrecht: Kluwer Academic Publishers;
2000.
supercritical water oxidation-part B: commercial/full-scale applications. J Supercrit Fluids.
2004;29:289–312.
35. Marrone PA, Hong GT. Corrosion control methods in supercritical water oxidation and
gasification processes. J Supercrit Fluids. 2009;51:83–103.
36. Lee HC, In JH, Lee SY, Kim JH, Lee CH. An anti-corrosive reactor for the decompo-
sition of halogenated hydrocarbons with supercritical water oxidation. J Supercrit Fluids.
2005;36:59–69.
37. Prikopsky K, Wellig B, Von Rohr PR. SCWO of salt containing artificial wastewater using a
transpiring-wall reactor: experimental results. J Supercrit Fluids. 2007;40:246–57.
38. Xu DH, Wang SZ, Gong YM, Guo Y, Tang XY, Ma HH. A novel concept reactor design for
preventing salt deposition in supercritical water. Chem Eng Res Des. 2010;88:1515–22.
39. Vadillo V, García-Jarana MB, Sánchez-Oneto J, Portela JR, de la Ossa EM. Supercritical water
oxidation of flammable industrial wastewaters: economic perspectives of an industrial plant.
J Chem Technol Biotechnol. 2011;86:1049–57.
40. Vogel F, DiNaro Blanchard JL, Marrone PA, Rice SF, Webley PA, Peters WA, Smith KA,
Tester JW. Critical review of kinetic data for the oxidation of methanol in supercritical water.
41. Lin KS, Wang HP, Li MC. Oxidation of 2,4-dichlorophenol in supercritical water. Chemo-
sphere. 1998;36:2075–83.
42. Erkonak H, Söğüt OÖ, Akgün M. Treatment of olive mill wastewater by supercritical water
43. Söğüt OÖ, Akgün M. Treatment of textile wastewater by SCWO in a tube reactor. J Supercrit
Fluids. 2007;43:106–11.
44. Söğüt OÖ, Akgün M. Removal of C.I. Basic Blue 41 from aqueous solution by supercritical
water oxidation in continuous-flow reactor. J Ind Eng Chem. 2009;15:803–8.
45. Söğüt OÖ, Kıpçak E, Akgün M. Treatment of whey wastewater by supercritical water
oxidation. Water Sci Technol. 2011;63:908–16.
46. Shu HY, Chang MC. Pre-ozonation coupled with UV/H2 O2 process for the decolorization
and mineralization of cotton dyeing effluent and synthesized C.I. Direct Black 22 wastewater.
J Hazard Mater. 2005;B121:127–33.
47. Söğüt OÖ, Akgün M. Treatment of dyehouse waste-water by supercritical water oxidation: a
case study. J Chem Technol Biotechnol. 2010;85:640–7.
48. Sato T, Adschiri T, Arai K. Decomposition kinetics of 2-propylphenol in supercritical water.
J Anal Appl Pyrolysis. 2003;70:735–46.
49. Chen G, Lei L, Hu X, Yue PL. Kinetic study into the wet air oxidation of printing and dyeing
wastewater. Sep Purif Technol. 2003;31:71–6.
50. Akiya N, Savage PE. Roles of water for chemical reactions in high-temperature water. Chem
Rev. 2002;102:2725–50.
51. McNaught AD, Wilkinson A. IUPAC. Compendium of chemical terminology: the gold book.
2nd ed. Oxford: Blackwell Science; 1997.
52. Tiltscher H, Hoffmann H. Trends in high pressure chemical reaction engineering. Chem Eng
Sci. 1987;42:959–77.
53. Fogler HS. Elements of chemical reaction engineering. 2nd ed. Upper Saddle River: Prentice
Hall International; 2006.
54. Chang KC, Li L, Gloyna EF. Supercritical water oxidation of acetic acid bu potassium
permanganate. J Hazard Mater. 1993;33:51–62.
55. Imamura SI, Hirano A, Kawabata N. Wet oxidation of acetic acid catalyzed by Co-Bi complex
oxides. Ind Eng Chem Prod Res Dev. 1982;21:570–5.
56. Son SH, Lee JH, Lee CH. Corrosion phenomena of alloys by subcritical and supercritical water
oxidation of 2-chlophenol. J Supercrit Fluids. 2008;44:370–8.
57. Comynis AE. Encyclopaedic dictionary of named processes in chemical technology. 3rd ed.
Boca Raton: CRC Press; 2007.
58. Green LA, Akgerman A. Supercritical CO2 extraction of soil-water slurries. J Supercrit Fluids.
1996;9:177–84.
59. Park S, Gloyna EF. Statistical study of the liquefaction of used rubber tyre in supercritical
water. Fuel. 1997;76:999–1003.
60. Kriksunov LB, MacDonald DD. Corrosion testing and prediction in Supercritical water
oxidation environments. ASME Heat Transf Div. 1995;317:281–8.
62. Leusbrock I, Metz SJ, Rexwinkel G, Versteeg GF. The solubility of magnesium chloride and
calcium chloride in near critical and supercritical water. J Supercrit Fluids. 2010;53:17–24.
63. Aymonier C, Bottreau M, Berdeu B, Cansell F. Ultrasound for hydrothermal treatments of
aqueous wastes: solution for overcoming salt precipitation and corrosion. Ind Eng Chem Res.
2000;39:4734–40.
64. Vadillo V, García-Jarana MB, Sánchez-Oneto J, Portela JR, de la Ossa EJM. New feed system
for water-insoluble organic and/or highly concentrated wastewaters in the supercritical water
oxidation process. J Supercrit Fluids. 2012;72:263–9.
65. Boock LT, Klein MT. Lumping strategy for modeling the oxidation of C1-C3 alcohols and
acetic acid in high-temperature water. Ind Eng Chem Res. 1993;32:2464–73.
66. Brock EE, Oshima Y, Savage PE, Barker JR. Kinetics and mechanism of methanol oxidation
in supercritical water. J Phys Chem. 1996;100:15834–42.
67. Li L, Chen P, Gloyna EF. Generalized kinetic model for wet oxidation of organic compounds.
AIChE J. 1991;37:1687–97.
68. Wagner W, Prub A. The IAPWS formulation 1995 for the thermodynamic properties of ordi-
nary water substance for general and scientific use. J Phys Chem Ref Data. 2002;31:387–535.
69. Peng D, Robinson DB. A new two-constant equation of state. Ind Eng Chem Fundam.
1976;15(1):59–64.
70. Croiset E, Rice SF, Hanush RG. Hydrogen peroxide decomposition in supercritical water.
AIChE J. 1997;43:2343–52.
71. Sögüt OÖ, Yıldırım E, Akgün M. The treatment of wastewaters by supercritical water
oxidation. Desalination Water Treat. 2011;26:131–8.
72. Söğüt OÖ, Akgün M. Degradation of aqueous disperse orange 25 by supercritical water
oxidation. Fresenius Environ Bul. 2008;17:864–71.
73. Rivas FJ, Gimeno O, Portela JR, de la Ossa EM, Beltrán FJ. Supercritical water oxidation of
olive oil mill wastewater. Ind Eng Chem Res. 2001;40:3670–4.
74. Chkoundali S, Alaya S, Launay JC, Gabsi S, Cansell F. Hydrothermal oxidation of olive oil
mill wastewater with multi-injection of oxygen: simulation and experimental data. Environ
Eng Sci. 2008;25:173–9.
75. Cui B, Cui F, Jing G, Xu S, Huo W, Liu S. Oxidation of oily sludge in supercritical water.
J Hazard Mater. 2009;165:511–7.
76. Jimenez-Espadafor F, Portela JR, Vadillo V, Saánchez-Oneto J, Villanueva JAB, Garcıá MT, de
la Ossa EJM. Supercritical water oxidation of oily wastes at pilot plant: simulation for energy
recovery. Ind Eng Chem Res. 2001;50:775–84.
77. Sánchez-Oneto J, Portela JR, Nebot E, de la Ossa EM. Hydrothermal oxidation: application to
the treatment of different cutting fluid wastes. J Hazard Mater. 2007;144:639–44.
78. Shanableha A. Generalized first-order kinetic model for biosolids decomposition and oxidation
during hydrothermal treatment. Environ Sci Technol. 2005;39:355–62.
79. Veriansyah B, Park TJ, Lim JS, Lee YW. Supercritical water oxidation of wastewater from LCD
manufacturing process: kinetic and formation of chromium oxide nanoparticles. J Supercrit
Fluids. 2005;34:51–61.
80. Weijin G, Xuejun D. Degradation of landfill leachate using transpiring-wall supercritical water
oxidation (SCWO) reactor. Waste Manag. 2010;30:2103–7.
81. Holgate HR, Tester JW. Fundamental kinetics and mechanisms of hydrogen oxidation in
supercritical water. Combust Sci Technol. 1993;88:369–97.
82. Holgate HR, Webley PA, Tester JW, Helling RK. Carbon monoxide oxidation in supercritical
water: the effects of heat transfer and the water-gas shift reaction on observed kinetics. Energy
Fuel. 1992;6:586–97.
83. Cui BC, Liu SZ, Cui FY, Jing GL, Liu XJ. Lumped kinetics for supercritical water oxidation
of oily sludge. Process Saf Environ Prot. 2011;89:198–203.
84. Tester JW, Webley PA, Holgate HR. Revised global kinetic measurements of methanol
oxidation in supercritical water. Ind Eng Chem Res. 1993;32(1):236–9.
85. Lehr V, Sarlea M, Ott L, Vogel H. Catalyticdehydration of biomass-derived polyols in sub- and
supercritical water. Catal Today. 2007;121:121–9.
86. Shin YH, Shin NC, Veriansyah B, Kim J, Lee YW. Supercritical water oxidation of wastewater
from acrylonitrile manufacturing plant. J Hazard Mater. 2009;163:1142–7.
87. Savage PE, Smith MA. Kinetics of acetic acid oxidation in supercritical water. Environ Sci
Technol. 1995;1:216–21.
88. Maharrey SP, Miller DR. Quartz capillary microreactor for studies of oxidation in supercritical
water. AIChE J. 2001;5:1203–11.
89. Koido K, Ishida Y, Kumabe K, Matsumoto K, Hasegawa T. Kinetics of ethanol oxidation in
subcritical water. J Supercrit Fluids. 2010;55:246–51.
90. Schanzenbacher J, Taylor JD, Tester JW. Ethanol oxidation and hydrolysis rates in supercritical
91. Ploeger JM, Madlinger AC, Tester JW. Revised global kinetic measurements of ammonia
oxidation in supercritical water. Ind Eng Chem Res. 2006;45:6842–5.
92. Veriansyah B, Kima JD, Lee JC, Lee YW. OPA oxidation rates in supercritical water. J Hazard
Mater. 2005;B124:119–24.
Chapter 11
Production of Hydrogen from Biomass via
Supercritical Water Gasification
Jude A. Onwudili and Paul T. Williams
Abstract The influence of process conditions on the yield of syngas and hydrogen
from the supercritical water gasification of biomass are reviewed. The yield and
composition of the products from the processing of model biomass compounds and
different types of biomass are discussed. The influence of the key process conditions
of temperature, pressure, residence time and feed concentration on product yield and
gas composition and hydrogen are presented. The influence of homogeneous alkali
catalysts and metal-based heterogeneous catalysts and their influence on hydrogen
and syngas production are also discussed.
Keywords Biomass • Hydrogen • Catalysts • Supercritical water • Syngas •

Cellulose • Hemicellulose • Lignin
11.1 Introduction
Hydrogen is seen as a key fuel for the low carbon energy systems of the future
because water is the only product of combustion and it has the potential to dra-
matically reduce the world’s dependence on fossil fuels. However, the sustainable
benefits of hydrogen energy lie with the source of the production process for
the hydrogen. At present, hydrogen is largely produced either from fossil fuel
sources such as natural gas, naphtha and coal or via water electrolysis, photolysis or
thermolysis [1]. Processes involving fossil fuels, use high temperatures, requiring
high inputs of energy and generate significant carbon dioxide during the process.
There is therefore growing interest in production processes for hydrogen based
around alternative feedstocks.
J.A. Onwudili • P.T. Williams ()

Energy Research Institute, Faculty of Engineering, University of Leeds, Leeds LS2 9JT, UK
300 J.A. Onwudili and P.T. Williams
Biomass and biomass wastes are regarded as a sustainable resource and there has
been much recent research into producing hydrogen from such sources. There has
been interest in the production of hydrogen from biomass via gasification by steam
reforming. However, steam does not readily react with biomass at atmospheric
pressure, resulting in char and tar formation [2]. In addition, thermochemical
processing of biomass through, for example, pyrolysis and gasification usually
requires moisture contents below 10 wt% [3]. However, certain biomass wastes
such as the crude glycerol from biodiesel production, waste distillers grain from
bioethanol production, sewage sludge pulp and paper mill sludge, food wastes etc.,
are very wet, with moisture contents >70 wt% [3–5]. Such very wet biowastes
are extremely difficult to treat by dry pyrolysis and gasification, which limits
the treatment options available to produce high value end products. For example,
the high energy input required to reduce the moisture content for conventional
thermochemical processes is prohibitive [5]. In addition, it has been shown that
high water contents (above 40 %) of feedstocks results in a drastic reduction of the
thermal efficiency of steam reforming gasification [6].
However, for supercritical water gasification, the fact that the biomass wastes
are very wet is particularly advantageous, since the reactions require water as
the reaction medium. Yanik et al. [2] have also reported that supercritical water
gasification of biomass results in high solid conversion, that is, low levels of char
and tar formation. In addition, because of the large amount of excess water, up to
50 % of the hydrogen formed in the process can originate from the reaction of carbon
monoxide and water via the water-gas shift reaction to produce hydrogen and carbon
dioxide [7, 8]. The hydrothermal environment (i.e. above and below the critical point
of water namely, 374.8 ı C and at 22.1 MPa) offers a reaction medium ideal for
the processing of very wet bio-waste feedstocks [9]. The properties of water under
these conditions are markedly different from that of ambient water [3]. Supercritical
water has unique features with respect to its density, dielectric constant, ionic
product, viscosity, diffusivity, electric conductance, and solvent ability. The water
exhibits a single dense fluid phase with gas-like diffusion rates along with liquid-
like collision rates so that organic compounds become highly soluble and gases
such as oxygen are completely miscible in supercritical water, thus minimising
mass transfer resistances and providing rapid reaction rates. Water below its critical
point (sub-critical water) also has excellent solvating and reactant properties with
respect to organic materials. Supercritical water gasification of biomass involves
the thermochemical degradation of the biomass in the presence of water to produce
a syngas rich in hydrogen, methane, carbon dioxide, carbon monoxide and C2 -C4
hydrocarbons [3, 9–11].
Lu et al. [12] undertook thermodynamic modelling for chemical equilibrium
for the supercritical water gasification of biomass in relation to an experimental
supercritical water biomass gasification reactor system. The chemical equilibrium
model was based on minimising Gibbs free energy. Their thermodynamic analysis
included the chemical equilibrium in the supercritical water reactor, gas-liquid
equilibrium in the high pressure separator and exergy and energy analysis of the
whole system. Figure 11.1 shows the variation of equilibrium gas yields in relation
11 Production of Hydrogen from Biomass via Supercritical Water Gasification 301
CO yield (10-3 mol/kg biomass)

80
Gas yield (mol/kg biomass)
H2 8
60 6
CO2
40 4
CO
20 2
CH4
0 0
427 527 627 727 827
Temperature°C
Fig. 11.1 Equilibrium gas yields as a function of temperature for biomass gasification at 25 MPa
pressure and 5 wt% dry biomass content [12] (Adapted with permission from Lu et al. [12].
to reactor temperature from 400 to 800 ı C (673 to 1,073 K). The data shows that
the yields of hydrogen and carbon dioxide increase with increasing temperature
and a consequent decrease in the yield of methane. Above about 650 ı C (923 K),
there is little increase in hydrogen yield. Carbon monoxide yield was small in
comparison to the other gases. The data illustrate the great potential of supercritical
water gasification of biomass for hydrogen production.
In this chapter, the supercritical water gasification of biomass for hydrogen
production is reviewed in relation to the influence of process conditions on the yield
and composition of the products. The influences of biomass type and composition,
as well as the addition of catalysts (homogeneous alkali catalysts and metal based
heterogeneous catalysts), on product yield and composition are also discussed.
11.2 Formation of Hydrogen in Supercritical Water

Gasification of Biomass Model Compounds
Biomass is chemically very complex, being composed of the major component

polymers, cellulose, hemicellulose and lignin, in addition to varying amounts of
inorganic materials. Basu and Mettanant [13], Kruse [3], Matsumura et al. [14] and
Peterson et al. [8] for example, have reviewed the influence of different biomass
feedstock composition and model biomass compounds on the supercritical water
gasification of biomass. To understand the reactions involved in the degradation
of biomass under supercritical water gasification conditions, the main components
of biomass have been investigated; for example, cellulose [15–19] hemicellulose

[18, 19] and lignin [18, 20, 21]. In addition, smaller monomers which make up
the three main biomass polymers have also been investigated. For example, there
have been several studies using glucose as the representative monomer of cellulose
[22–24] and studies using guaiacol as a representative model compound of lignin
[20, 25, 26].
In this section, the supercritical water gasification of the model compounds of
biomass, i.e. cellulose, hemicelluloses and lignin are reviewed. Investigations of
such biomass model components have shown that the formation of hydrogen during
the supercritical water gasification process involves a series of complex reactions,
involving several steps and intermediaries. However, the overall idealized reaction
using cellulose as the model compound can be represented by;
C6 H10 O5 C 7H2 O ! 12H2 C 6CO2 (11.1)
The overall reaction for supercritical water gasification of biomass is endother-

mic and therefore requires a heat input, and for hydrogen production higher
temperatures are favoured. For example, typical experimental temperature condi-
tions for maximising hydrogen from biomass are in the range of 300–600 ı C
[23, 27–30]. A range of other solid and liquid phase reactions are involved in the
formation of gas from the supercritical water gasification of biomass. These include,
cellulose hydrolysis to produce glucose, followed by glucose decomposition, steam
reforming, oxidation of species, char formation through intermediates and char from
pyrolysis of the feedstock, methanation, water gas shift and hydrogenation reactions,
etc. [16].
Several other intermediate reactions, sometimes competing reactions, are also
involved in the supercritical water gasification of biomass. For example, reforming
of biomass to form intermediate products including organic acids such as, acetic,
formic, lactic and levulinic acids, and aldehyde intermediates such as acetic and
formic aldehydes have been reported [27, 28, 31–34].
For example, formation of intermediates and hydrogen can be represented as
follows;
C6 H12 O6 C 6H2 O ! 6HCOOH C 6H2 (11.2)
C6 H12 O6 C 6H2 O ! 4HCOOH C 2CO2 C 8H2 (11.3)
C6 H12 O6 C 4H2 O ! 4HCOOH C CH3 COOH C 4H2 (11.4)
C6 H12 O6 C 2H2 O ! 2HCOOH C 2CH3 COOH C 2H2 (11.5)
C6 H12 O6 C 2H2 O ! 2CH3 COOH C 2CO2 C 4H2 (11.6)

The intermediate products are then subsequently reformed, for example;
HCOOH ! CO C H2 O (11.7)
CH3 COOH ! CH4 C CO2 (11.8)
CH4 C H2 O ! CO C 3H2 (11.9)
Hydrogen is also produced via the water-gas shift reaction;
CO C H2 O ! H2 C CO2 (11.10)
Cellulose is a polysaccharide composed of glucose units linked via “-(1-4)-

glycosidic bonds and with different properties depending on the biological source
[8]. Hemicellulose is a complex macromolecular structured component of biomass
composed of sugar monomers such as xylose, mannose, glucose, galactose etc.
Lignin is a complex, high molecular weight, aromatic macromolecular component
of biomass, composed mainly of monomeric phenylpropane derivatives such as
coumaryl, coniferyl and sinapyl alcohols [8]. Basu and Mettanant [13], have
suggested that the production of hydrogen from real biomass samples cannot be
easily predicted since the reactions are more complex and organic materials other
than cellulose, hemicelluloses and lignin may influence gas yield and composition.
Yoshida and Matsumura [18] investigated the supercritical water gasification
of cellulose, hemicellulose (xylan) and lignin mixtures at 350 ı C and 25 MPa
pressure in a batch reactor and in the presence of a nickel catalyst. They reported
that cellulose gave the highest yield of hydrogen, followed by hemicellulose with
lignin giving the lowest yield. The presence of lignin acted as an inhibitor to syngas
production and it was suggested that the intermediate products from cellulose and
hemicelluloses reacted with lignin and reduced hydrogen formation. Azadi et al.
[19] showed data for the supercritical water gasification of cellulose, hemicellulose
(xylan) and lignin using a batch autoclave reactor. They reported that the main gases
produced were carbon dioxide, carbon monoxide, hydrogen and methane. Minowa
and Inoue [35] reported the gasification of cellulose at 400 ı C and 13 MPa in a
stainless steel autoclave. The products were gas (32 wt% yield), oil (<5 wt%)
and char (25 wt%) and water (38 wt%). The gas was mainly composed of
carbon dioxide and also hydrogen and methane. Williams and Onwudili [15]
investigated the subcritical and supercritical water gasification of cellulose in
relation to temperature and residence time using a batch reactor. The cellulose
was mainly converted to carbon dioxide gas, ranging from 35.8 to 47.4 wt% CO2
depending on the reaction conditions. The hydrogen content of the cellulose was
reported to be mainly converted to water with only low conversion of the cellulose
hydrogen to hydrogen gas (between 11 and 14 wt%). Other product gases indentified
included carbon monoxide, and C1 –C4 hydrocarbons (mainly methane). Moving
from subcritical to supercritical water resulted in a decrease in the oil and char
Fig. 11.2 Schematic diagram of the proposed overall reaction scheme for the supercritical water
gasification of biomass
yield with a corresponding increase in the gas yield, mainly through an increase in
the yield of carbon dioxide. Resende et al. [16] investigated the supercritical water
gasification of cellulose for a range of parameters to determine their influence on gas
yield. The results showed that the main gases produced were hydrogen, methane,
carbon dioxide and carbon monoxide and that at higher temperatures the rate of
formation of all gases increased.
Several researchers have suggested a reaction mechanism for the decomposition
of cellulose in subcritical and supercritical water as a way of understanding biomass
gasification [3, 14, 22, 36, 37]. For example, Matsumura et al. [14] suggests that
cellulose is initially hydrolysed followed by decomposition to form water soluble
sugars and non-sugars and also gas, oil and char. It was suggested that once the oil is
formed it is difficult to gasify, instead it reacts to form char at higher temperatures.
Gas is produced directly from the decomposition of the water soluble products.
Figure 11.2 shows a schematic diagram of the proposed overall mechanism.
Resende et al. [38] examined the reactions of lignin in a supercritical water
quartz reactor in relation to reaction temperature, residence time and lignin feed
loading. The experiments were carried out using capillary quartz tubes used as mini-
batch reactors which were placed in a pre-heated sand bath or tube furnace. They
examined process parameters of 350–725 ı C reaction temperature, lignin loadings
of 5.0, 9.0 and 33.3 wt% and water densities of 0.05, 0.08 and 0.18 g cm1 . They
reported that the non-catalysed products were mainly methane and carbon dioxide
with lower concentrations of hydrogen and carbon monoxide. Maximum hydrogen
yield of 7.1 mmol g1 was reported with higher temperatures maximising hydrogen
yield and minimising carbon monoxide yield via the water-gas shift reaction. Yong
and Matsumura [21] investigated the supercritical water gasification of lignin in
a continuous flow reactor at temperatures between 390–450 ı C and 25 MPa at
very short residence times (0.5–10 s). They undertook a detailed analysis of the
products including gases and phenolic and aromatic compounds. They concluded
that under the heating rate, temperature and short residence times of their system,
lignin is rapidly decomposed. Supercritical conditions produced a high yield of
solid and char was suggested to be formed through cross linking reactions between
reactive degradation fragments and residual lignin to produce high molecular weight
material. Char was formed at both short and long residence times. The gas produced
was composed of mainly hydrogen and carbon dioxide with ethane and ethene
detected at low concentration and no carbon monoxide was detected, this being
converted to hydrogen and carbon dioxide in the water-gas shift reaction. It was
suggested that gas formation mainly arose from lignin during the early period of
lignin depolymerisation. Pinkowska et al. [39] examined the decomposition of alkali
lignin in a batch reactor at subcritical and supercritical water conditions in relation to
temperature and residence time. The main soluble decomposition products derived
from the lignin were phenolic compounds such as guaiacol and catechol, as well
as phenol and cresol isomers. A solid char residue was also formed which differed
significantly in composition compared to the original lignin depending on reaction
temperature and reaction time. They suggested that the solid residue consisted
mainly of phenolic char. The data from subcritical reaction conditions suggested
that the solid residue was largely composed of undissolved alkali lignin. However,
under supercritical water reaction conditions dissolution of the lignin was promoted
due to the decrease in dielectric constant of the water with increasing temperature.
The soluble lignin fragments reacting with the lower molecular weight soluble
compounds to form polymers and cross-linked phenolic biochar. Increasing the
temperature increased polymerisation and carbonisation via radical coupling and
recondensation reactions respectively.
There have been several suggested reaction mechanisms for the subcritical and
supercritical water gasification of lignin [20, 40, 41]. For example, Fang et al. [41]
propose a reaction scheme involving reactions in the oil, aqueous and gas phases in
addition to the solid residue. Decomposition and hydrolysis of the lignin producing
dissolved lignin, oligomers and monomers which further reacted to the four phases.
Biomass includes feedstocks other than ligno-cellulosic material, such as food
wastes and algae, which are largely composed of carbohydrates, proteins and
lipids. These components have been examined as model compounds of this type of
biomass. For example, research has found that carbohydrate-type biomass produces
more hydrogen than other types such as lipids and proteins [29, 42, 43]. It has also
been reported that proteins in biomass suppress the formation of hydrogen [42, 43].
In addition, the metals present in biomass such as the alkali metals (i.e. sodium and
potassium) and alkaline earth metals (i.e. calcium) may catalyse the production of
hydrogen [3, 29].
11.3 Influence of Process Parameters on Hydrogen

Production from the Supercritical Water Gasification
of Biomass
The supercritical water gasification of biomass to produce hydrogen involves several

process parameters that influence the hydrogen yield, including temperature, heating
rate, pressure, residence time, feedstock concentration and feedstock pre-treatment.
There has been extensive research into the influence of the various process
parameters on the yield of hydrogen, including several reviews, for example,
Basu and Mettanant [13], Kruse [3], Matsumura et al. [14]. Lu et al. [44]
undertook an experimental design analysis of the various parameters to maximise
the production of hydrogen from the supercritical water gasification of biomass.
Their analysis involved an orthogonal experimental design method coupled with
an experimental programme using a tubular reactor system for the supercritical
gasification of biomass in the form of corn cob. The corn cob was introduced into
the reactor with sodium carboxymethylcellulose and water to form a gel for ease
of feeding. Their results suggested that the order of reaction parameters with the
most influence on hydrogen production was, temperature > pressure > feedstock
concentration > residence time. They reported that the maximum hydrogen yield
of 15.23 mol kg1 was obtained at a temperature of 650 ı C, pressure of 25 MPa,
40 s residence time and feedstock loading of 2 wt% corn cob with 1 wt% sodium
carboxymethylcellulose [44].
11.3.1 Temperature
The influence of temperature on the production of hydrogen from the hydrothermal

gasification of biomass has been investigated over a range of temperatures from
300 to 800 ı C [12, 23, 27–30]. The subcritical water gasification of biomass
produces a product gas high in methane and low in hydrogen, but at higher operating
temperatures in the supercritical water gasification region, hydrogen gas is favoured
with less methane [7]. Williams and Onwudili [23] investigated the hydrothermal
gasification of glucose over the temperature range of 330–380 ı C in a closed
batch reactor. They reported that the product yield and composition, including
hydrogen yield, at the subcritical water temperatures did not significantly change
as the temperature (and pressure) increased into the supercritical water region.
However, because of the particular properties of a single fluid phase produced
under supercritical water conditions, including the gas-like diffusivity, but liquid-
like density, the reactions are defined by free-radical reactions which are promoted
by high temperatures [45]. Such parameters favour biomass gasification reactions,
leading to increased hydrogen production [46–48].
In general, the biomass gasification efficiency increases as the temperature
increases, leading to increased hydrogen yield [13]. Kruse [3, 49] has reviewed
the supercritical water gasification of biomass and reported that based on thermo-
dynamic analysis, high supercritical water gasification temperatures of 600 ı C
are preferred for hydrogen production and at lower temperatures methane is
preferentially formed. Carbon monoxide content of the product syngas is low
under such conditions. In addition, with increasing process pressure or biomass
concentration, the yield of methane increases and hydrogen decreases.
Lee et al. [50] investigated the influence of temperature on the supercritical water
gasification of glucose over the temperature range of 480–750 ı C at a pressure
of 28 MPa. They used a continuous flow reactor with a reactor residence time
of 10–50 s. They reported that the yield of hydrogen increased markedly with
increasing temperature over 660 ı C, while the carbon monoxide yield decreased
with temperature. They suggested that the water-gas shift reaction resulting in
reduced carbon monoxide and increased hydrogen and carbon dioxide occurred
significantly at temperatures over 660 ı C. Methane was found to be a stable
compound in supercritical water at temperatures as high as 700 ı C. Guo et al.
[51] also reported a marked increase in hydrogen and carbon dioxide yield for the
supercritical water gasification of glucose with increasing temperature from 500 to
650 ı C which was attributed to the increased water-gas shift reaction. They also
reported that for the supercritical water gasification of lignin high temperatures
in the region of 700 ı C are required to achieve above 90 % conversion. Hao
et al. [22] reported that the gasification efficiency, hydrogen yield potential, and
carbon efficiency increased as the temperature was increased up to 650 ı C for the
supercritical water gasification of glucose.
Real biomass samples have been gasified in supercritical water and have
shown similar increase in gasification efficiency and hydrogen yield with increased
temperature [13, 15, 45]. For example Basu and Mettanant [13] gasified rice husks
in supercritical water and showed a more than 50 % increase in hydrogen yield as
the temperature was raised from 650 to 700 ı C at 32 MPa pressure and 60 min
residence time. Lu et al. [45] investigated the supercritical water gasification of
wood sawdust (in the presence of 2–3 wt% of sodium carboxymethylcellulose to
aid feeding) in a continuous flow reactor. They reported higher hydrogen gas yields
and gasification efficiency when the temperature was 650 ı C compared to 600 ı C.
Williams and Onwudili [15] for Cassava biomass similarly reported that higher
hydrogen yields were obtained at higher reaction temperature. Venkitasamy et al.
[52] also used a closed batch reactor and investigated the influence of temperature
for the supercritical water gasification of sawdust and rice straw. Raising the
temperature from 500 to 750 ı C resulted in an increase in gas and hydrogen yields.
11.3.2 Pressure
The reaction equilibria for the subcritical and supercritical water gasification of
biomass suggest that higher pressures would result in a decrease in hydrogen
production (e.g. Eq. 11.1) and increased methane [3]. However, the reactions in
supercritical water are more complex, since the water acts as both a reactant and
a solvent [3]. Higher pressure favours the direction of equilibrium reactions with
less number of moles of gaseous species as indicated by the equations shown
before (e.g. Eqs. 11.1, 11.2, 11.3, 11.4, 11.5, and 11.6). However, as the pressure
increases the physico-chemical properties of water change, for example the fluid
density, dielectric constant and ionic product of water are increased [45]. As such,
Lu et al. [45] reported that hydrogen yield increases and methane and carbon
monoxide slightly decrease with increased pressure from 17 to 30 MPa for the
hydrothermal processing of wood sawdust, suggesting promotion of the water-gas

shift reaction. They suggest that the changing physico-chemical properties of water
influence the gasification reactions. Basu and Mettanant [13] also reported on the
influence of pressure on hydrogen yield and gasification efficiency with increasing
pressure from 23 to 33 MPa at a range of different temperatures for rice husks. At
700 ı C, there was an increase in gasification efficiency and hydrogen yield as the
pressure was increased from 23 to 33 MPa. However, at lower temperatures (i.e.
500 ı C), the influence of pressure was less significant. Basu and Mettanant [13]
suggest that since the fluid density, dielectric constant and ionic product increase
with pressure, increasing ionic reactions and suppressing free radical reactions.
High ionic product implies high [HC ] and [OH ] concentration which allows
supercritical water to act like an acid or base catalyst [13]. Demirbas [10] also
reported an increase in hydrogen yield with increased pressure, which he attributed
to increased diffusion and mass transfer rates of supercritical water leading to
improved biomass gasification efficiency. Madenoglu et al. [53] investigated the
subcritical and supercritical water gasification of glucose in a batch reactor at
pressures between 20 and 42.5 MPa and at fixed temperatures of 400, 500 and
600 ı C. They reported that hydrogen and carbon dioxide yields decrease with
increasing pressure at constant temperature while methane yield increased. This
they suggested was due to the reaction of hydrogen with carbon dioxide to produce
methane through the methanation reaction, thereby producing a reduction in the
yield of hydrogen.
Others have reported little effect of pressure on product yield and composition,
for example, Hao et al. [22] investigated the supercritical water gasification of
glucose and reported that increasing the pressure from 25 to 30 MPa, had no
great effect on the glucose gasification efficiency and the fraction of gas product.
Matsumura et al. [14] have also reported that pressure had hardly any influence
on the product gas composition or the gasification efficiency over a wide range
of pressures from subcritical to supercritical water conditions (5–45 MPa) for the
hydrothermal gasification of glucose and glycerol. Several researchers [15, 52] have
used closed batch reactors where supercritical water conditions are produced by
raising the temperature, resulting in a consequent increase in pressure. In such
systems, it is difficult to separate the conclusions regarding the influence of pressure
and temperature.
11.3.3 Residence Time
The residence time of reactants and intermediates spent under reaction conditions
within the supercritical water reactor can greatly influence the yields and composi-
tion of gas products. Additionally, over long residence times the products of the
supercritical water gasification, such as the hydrogen could undergo subsequent
reactions which may reduce their eventual yields. For example, hydrogen can be
consumed via methanation reactions. Model biomass compounds such as glucose
and glycerol which are soluble in water, can be gasified in short residence times,
of the order of seconds, whereas biomass with its complex composition would
be expected to require longer reaction times since initial hydrolysis reactions to
produce intermediates would be required. Basu and Mettanant [13] have reviewed
the influence of residence time of reactants at the reaction conditions on biomass
gasification efficiency. They conclude that several researchers report that there is
an initial increase in gasification efficiency at increasing residence time, but after a
certain time period, that influence diminishes and there is no further advantage in
increasing residence time. The time required for maximum biomass conversion was
variable and dependent on reaction temperature, biomass type and reactor type. For
the supercritical water gasification of rice husks at 650 ı C and 30 MPa, it has been
reported that hydrogen yield doubled from 7 to 14 mol kg1 when the residence time
was increased from 10 to 40 min, but there was no significant increase when the time
was further increased to 60 min [13]. Williams and Onwudili [23] investigated the
supercritical water gasification of glucose in a batch reactor in relation to residence
time of the reactants from 0 to 120 min. They reported only a slight increase in
gasification efficiency and hydrogen yield with longer residence times. However,
that work, undertaken in a batch reactor, included a reactor heat-up time of >30 min.
Similar results were also reported by the same authors for the influence of residence
time on cellulose, starch and biomass in the form of Cassava [15]. Lee et al. [50]
gasified glucose in a continuous flow reactor, the continuous flow system producing
short residence times of between 10 and 50 s. At 600 ı C and 28 MPa increasing the
residence time from 10 to 50 s produced a significant increase in hydrogen yield,
from 0.5 to 2.5 mol H2 mol1 glucose. However at higher temperature (700 ı C
and 28 MPa pressure), after an initial increase in hydrogen yield from 10 to 16 s
residence time (from 5 to 6.5 mol H2 mol1 glucose), there was very little increase
in hydrogen yield. Susanti et al. [54] used a continuous updraft supercritical water
gasification reactor to investigate the gasification of glucose at temperatures of 600–
767 ı C and residence times of 15–60 s. Total gas yields and hydrogen gas yield
increased with temperature (10.5–11.2 mol mol1 glucose at 740 ı C), however,
hydrogen gas yields did not vary significantly with different residence times.
11.3.4 Feedstock Concentration
The concentration of the feedstock in supercritical water gasification influences the

yield and composition of the products. For example, at low feedstock concentrations
high gasification efficiencies are usually reported due to the minimization of
competing reaction pathways. However, at high feedstock concentrations there is
a tendency to form tar and char via polymerisation reactions. A concentration
of feedstock of 5 wt% is typical [14, 15, 51]. For example, Matsumura et al.
[14] report that for the supercritical water gasification of both glucose and glyc-
erol, that the feedstock concentration has a significant effect on hydrogen yield.
Concentrations of feedstock higher than 5–10 wt% led to a significant decrease
in hydrogen yield and carbon gasification efficiency. Williams and Onwudili [23]
investigated the subcritical and supercritical water gasification of glucose in relation
to feedstock concentration from 2.5 to 20 wt%. The overall gasification of the
glucose decreased from 100 % to about 59 wt% as the concentration of glucose was
increased. The yield of hydrogen was maximised at 5 wt% glucose concentration
and decreased significantly at high concentrations, in addition, significant increases
in char and oil production were found. Similar effects were reported by Hao et al.
[22] and Yu et al. [55], who found that an increase in biomass concentration led to a
decrease in total gasification but an increase in CO yield. The high concentration
of feedstock in such experiments leads to polymerisation of the decomposition
products instead of gasification. Lu et al. [45] investigated the influence of feedstock
concentration for glucose, cellulose and biomass in a continuous flow reactor.
They reported that increasing the biomass (wood sawdust) concentration from 1.95
to 4.12 wt% resulted in a 10 % decrease in gasification efficiency and a 30 %
decrease in hydrogen yield with a consequent increase in carbon dioxide yield.
They also reported that at higher biomass concentrations (6.19 wt% biomass), they
observed high char formation and reactor plugging. Guo et al. [56] investigated the
supercritical water gasification of glucose for hydrogen production by supercritical
water gasification using a continuous flow reactor at 445–600 ı C and 25 MPa with
a short residence time of 3.9–9.0 s. They reported that at temperatures above 487 ı C
the gasification efficiency increased markedly with increasing temperature. With
the increase of glycerol concentration from 10 to 50 wt%, the gasification efficiency
decreased from 88 to 71 % at 567 ı C.
In addition to the hydrogen and gas yield from the supercritical water gasification
of biomass, there is also the potential to produce solid carbonaceous char and
high molecular weight tar. Supercritical water gasification of biomass is reported to
produce lower levels of product char and tar compared to conventional gasification
[14]. Williams and Onwudili [15] analysed the tar produced from the supercritical
water gasification of glucose, cellulose, starch and biomass using Fourier transform
infrared (FT-IR) spectrometry. They reported that the tar was highly oxygenated
containing carboxylic acids, ketones, aldehydes, primary, secondary, and tertiary
alcohols and phenols, and substituted phenols. The composition of the tars from
the model biomass compounds and biomass were similar, reflecting the similar
chemical composition of the feedstocks, based around the glucose monomer leading
to the formation of similar compounds and functional groups in the derived thermal
degradation products. The tars produced from supercritical water gasification of
biomass contained similar compounds to those found in conventional biomass
gasification tars [15].
11.4 Catalytic Supercritical Water Gasification of Biomass
The role of catalysts in supercritical water gasification of biomass has been reviewed
by several authors, for example, Elliott [7], Azadi and Farnood [11], Basu and
Mettanant [13] and Guo et al. [57]. The use of different types of catalyst for
hydrogen production from the supercritical water gasification of biomass involves

several possible options. For example; conversion of biomass into intermediates that
can be more readily gasified; lowering the activation energy required to achieve
complete gasification; offering selectivity of the reaction routes which favour
hydrogen production. The commonly investigated catalysts are either homogeneous
water-soluble alkaline catalysts, or heterogeneous solid, metal-based catalysts.
11.4.1 Homogeneous Alkali Catalysts
A range of alkaline homogeneous catalysts/additives, such as KOH, NaOH, K2 CO3 ,

and Na2 CO3 have been investigated for subcritical and supercritical water gasifica-
tion [56, 58, 59]. Such alkali materials have the advantage that they can dissolve
in the hydrothermal medium as well as increase gasification efficiency [29, 60–
62]. For the alkaline hydroxide catalysts in particular, hydrogen yield is enhanced
by accelerating the water-gas shift reaction through removal of carbon dioxide as
carbonate. Sinag et al. [60, 61] investigated the supercritical water gasification of
glucose with the addition of K2 CO3 and reported an increase in hydrogen and carbon
dioxide gas production. Sinag et al. [60] explained the increase in the hydrogen yield
as well as in the carbon dioxide yield was due to the enhancement of the water–
gas shift reaction by intermediate formation of formates. Schmieder et al. [63]
added KOH or K2 CO3 to glucose and glycine under supercritical water gasification
conditions of 600 ı C and 25 MPa and reported the production of a hydrogen-rich
gas product. Muangrat et al. [59] investigated the subcritical water gasification of
glucose at 330 ı C temperature and 13.5 MPa pressure in the presence of various
alkali catalysts; NaOH, KOH, Ca(OH)2 , Na2 CO3 , K2 CO3 and NaHCO3 . The results
showed that hydrogen yield was increased which was attributed to promotion
of water-gas shift reaction and that the effectiveness of the alkaline additives
was in the order, NaOH > KOH > Ca(OH)2 > K2 CO3 > Na2 CO3 > NaHCO3 . The
hydrogen yield indicates that alkali hydroxides were more effective than carbonates
and bicarbonates. Catalysts of the hydroxide ion group, NaOH, KOH and Ca(OH)2 ,
were also reacted with biomass in the form of molasses and rice bran and enhanced
hydrogen production was confirmed and also char and tar formation was suppressed
[59]. Guo et al. [56] conducted supercritical water gasification of glucose in
a continuous flow tubular reactor in the presence of several alkali catalysts at
445–600 ı C and 25 MPa. The alkali catalysts greatly enhanced the water-gas
shift reaction and the hydrogen yield in relation to catalysts was in the order,
NaOH > Na2 CO3 > KOH > K2 CO3 . In addition, they also reported that no char
or tar was observed. Madenoglu et al. [64] used a continuous flow, supercritical
water gasification reactor to study the influence of alkali catalysts (K2 CO3 , Na2 CO3
and NaHCO3 .H2 O) on the gasification of five biomass samples. They reported
that carbon gasification efficiencies and hydrogen yields were improved by the
addition of the catalysts. The gasification efficiencies were different for different
types of biomass, attributed to the different ratios of cellulose, hemicelluloses
and lignin in the five biomass types. They later undertook an examination of the
supercritical water gasification of glucose with K2 CO3 as the catalyst in relation
to temperature and pressure [53]. They showed that carbon gasification efficiencies
were increased with the addition of K2 CO3 , reaching a maximum of 94 % at 600 ı C
and 20 MPa. They also reported that the yield of hydrogen was increased with
increasing temperature and decreasing pressure.
The route to increased hydrogen production in the presence of alkali catalysts
involves reactions to produce simpler intermediates which are more readily con-
verted to hydrogen, and also the absorption of product carbon dioxide which drives
the water-gas shift reaction forward producing enhanced hydrogen. Onwudili and
Williams [62] investigated the role of sodium hydroxide in the subcritical and
supercritical water gasification of glucose. In the absence of alkali, and at lower
reaction conditions, they reported that glucose decomposed to produce mainly
carbon dioxide, water, char and tar, and that furfural and 5-hydroxymethylfurfural
were found in the water-soluble products. However, in the presence of the sodium
hydroxide, furfural and 5-hydroxymethylfurfural were not detected, rather ketones,
aldehydes, carboxylic acids and their alkylated and hydroxylated derivatives were
formed. No tar or char was formed, suggesting that the suppression of furfural and
5-hydroxymethylfurfural indicated that these compounds were precursors to char
and tar formation in the absence of alkali. At reaction conditions of 450 ı C and
34 MPa pressure, more than 70 vol% of the gaseous product was hydrogen gas,
while the balance was mostly methane. Onwudili and Williams [62] proposed a
reaction scheme for the supercritical water gasification of glucose into hydrogen via
the formation of sodium formate intermediate;
C6 H12 O6 C 6NaOH C 3H2 O ! C6 H6 O6 :Na6 :9H2 O (11.11)
C6 H6 O6 :Na6 :9H2 O ! 6NaCOOH C 6H2 C 3H2 O (11.12)
6NaCOOH ! 6CO C 6NaOH (11.13)
6CO C 6H2 O ! CO2 C 6H2 (11.14)
6CO2 C 6NaOH ! 6NaHCO3 (11.15)
The overall equation,
C6 H12 O6 C 6NaOH C 6H2 O ! 6NaHCO3 C 12H2 (11.16)
The formation of methane [62] suggests that glucose may react with alkali to
produce methane through the formation of sodium acetate;
C6 H12 O6 C 3NaOH ! 3CH3 COONa C 3H2 O (11.17)

The overall hydrothermal reaction of glucose in the presence of sodium hydrox-

ide;
C6 H12 O6 C 4H2 O C 5NaOH ! 5NaHCO3 C CH4 C 8H2 (11.18)
Similar reaction schemes were proposed by Sinag et al. [60, 61] for the
supercritical water gasification of biomass with potassium carbonate involving the
formation of potassium formate as an intermediate for the enhanced formation of
hydrogen. Akgul and Kruse, [65] also investigated the role of NaHCO3 and KHCO3
in the water-gas shift reaction at 10 and 23 MPa and 230–300 ı C in a vertical
continuous flow reactor. They reported that the presence of NaHCO3 and KHCO3
significantly promotes the water-gas shift reaction resulting in the formation of
sodium and potassium formate salts.
Further research by Onwudili and Williams [28] has suggested that the role
of sodium hydroxide involves the formation of formate and acetate intermediates
which preferentially produce hydrogen or methane depending on process condi-
tions; lower temperatures (400 ı C) producing hydrogen from formate intermediates
and higher temperatures (500 ı C) favouring methane from acetate intermediates
(Eqs. 11.19, 11.20, and 11.21).
2HCOONa C H2 O ! 2H2 C CO2 C Na2 CO3 (11.19)
2CH3 COONa C H2 O ! 2CH4 C CO2 C Na2 CO3 (11.20)
The formation of methane is a pre-requisite for its reforming to hydrogen;
CH4 C 2H2 O ! CO2 C 4H2 (11.21)
11.4.2 Metal-Based Catalysts
The subcritical and supercritical water gasification of biomass has been investigated
using heterogeneous solid catalysts with a range of different metals or metal
oxides for example, nickel, cobalt, ruthenium, molybdenum, copper, tungsten, zinc,
rhenium, tin, lead, and chromium [7]. One of the difficulties of trying to promote
hydrogen production via the use of catalysts in supercritical water gasification
is that the most effective metals such as nickel, ruthenium and platinum can
catalyse both steam reforming reactions of hydrocarbons to hydrogen as well as
methanation reactions, which consumes hydrogen [66–70]. Nickel and ruthenium
catalysts are amongst the most commonly used catalysts for hydrogen production
from supercritical water gasification of biomass [7, 30, 68, 70–73]. Although metals
and metal oxides have been used, the reduced metals are significantly more effective
than their corresponding metal oxides for gas production [7, 57, 67].
Azadi et al. [70] investigated the production of hydrogen from the supercritical
water gasification of cellulose, lignin, bark and model carbohydrates with nickel
and ruthenium catalysts in a batch reactor. They reported that the Ni/Al2 O3
and Ni/hydrotalcite catalysts exhibited higher activities and superior hydrogen
selectivity compared to the other catalysts used. Li et al. [74] used a batch reactor
to investigate the supercritical water gasification of biomass in the form of glucose
model compound in the presence of nickel alumina catalysts with the addition of
copper, cobalt and tin to produce bimetallic catalysts. They reported that of the
added metals examined, the order of hydrogen yield was highest for the nickel-
copper alumina catalyst, followed by Ni-Co/”Al2 O3 and then Ni-Sn/”Al2 O3 . The
presence of copper led to an increased catalytic activity of the nickel catalyst which
increased hydrogen yield due to the reforming of methane. In addition, cobalt was
found to be an excellent promoter of the nickel-based alumina catalyst in relation
to hydrogen selectivity. Zhang et al. [75] used a continuous down-flow supercritical
water gasification reactor to compare hydrogen generation using a nickel-alumina
catalyst and a ruthenium-nickel-alumina catalyst. The ruthenium modified catalyst
was more stable than the nickel alumina catalyst in terms of hydrogen yield over a
4-day catalyst testing period.
Lu et al. [76] studied the influence of cerium as an additive to nickel based
catalysts for the supercritical water gasification of glucose at 400 ı C and 24.5 MPa
in a batch reactor. They reported that hydrogen yield and hydrogen selectivity
increased markedly with addition of the Ni/CeO2 /Al2 O3 catalyst. The addition
of cerium was important in inhibiting carbon deposition on the catalyst and
consequently deactivating the catalyst. The role of the cerium was via increasing
the dispersion of the nickel in the catalyst and reducing carbon by reaction of the
deposited carbon with the cerium. Xu and Donald [77] investigated the supercritical
water gasification of peat in relation to reaction temperature between 350 and 440 ı C
in a batch reactor. Alkali and metal based catalysts were used and it was found that
ruthenium oxide produced the highest yield of hydrogen-rich gases. Chakinala et al.
[78] reported on the catalytic gasification of the water-soluble fraction of biomass
pyrolysis oil in a batch reactor. The catalysts investigated were platinum, palladium,
ruthenium, rhodium and nickel supported on alumina. The gasification efficiency
of the catalysts was in the order: Ru > Pt > Rh > Pd > Ni, where the gasification
efficiency was defined as the degree of conversion of the carbon in the feed to
carbon-containing gases in the product gases. However, the hydrogen selectivity
for the catalysts was different, being in the order Pd > Ru > Rh > Pt > Ni. Van
Bennekom et al. [79] undertook a screening study of several catalysts; Pt/CeZrO2 ,
Ni/ZrO2 , Ni/CaO–6Al2 O3 , NiCu/CeZrO2 , and a CuZn alloy to investigate the
influence on the carbon-to-gas efficiency and gas composition in the reforming of
glycerol in supercritical water using a continuous reactor in relation to temperature
over the range of 375–700 ı C. In the absence of catalyst, 40 % conversion of the
glycerol was obtained at 674 ı C, but increased to almost 100 % in the presence
of the catalysts. All the catalysts promoted the water-gas shift reaction, however,
the nickel-based catalysts also promoted methanation but the rate of methanation
was lower than the water-gas shift reaction. Osada et al. [80] investigated the
supercritical water gasification of sugarcane bagasse, cellulose and lignin with a

titania-supported ruthenium (Ru/TiO2 ) catalyst in a batch reactor. They reported that
the biomass bagasse sample was completely gasified to methane, carbon dioxide,
and hydrogen at 400 ı C. The ruthenium catalyst was reported to enhance methane
and carbon dioxide production rather than hydrogen.
Azadi et al. [70] investigated nickel-based catalysts in relation to a wide range
of catalyst support materials for the supercritical water gasification of glucose in
a batch reactor. They demonstrated that the support material is very important in
determining catalyst activity in terms of carbon conversion and hydrogen selectivity;
with results indicating inactivity through to highly active depending on the physical
and chemical properties of the support material. For example, at 5 % nickel
loading, catalysts supports exhibiting high carbon conversions were, ’-Al2 O3 ,
carbon nanotubes, and MgO, whereas only modest catalytic activity was shown
for SiO2 , Y2 O3 , hydrotalcite, yttria-stabilized zirconia, and TiO2 supports. Ni/’-
Al2 O3 catalyst showed high hydrogen yield and carbon conversion as well as
superior catalyst stability in the supercritical water gasification experiments. They
also showed that the addition of alkali promoters (K, Na, Cs) to the supercritical
water gasification of glucose in the presence of the Ni/’-Al2 O3 catalyst improved
the hydrogen selectivity and the carbon conversion.
11.5 Challenges
The supercritical water gasification of biomass, particularly very wet biomass

is an attractive solution to produce syngas and hydrogen from wastes that are
difficult to process using conventional technologies. There are several high water
content biomass wastes that are produced in large quantities, for example, food
industry wastes, glycerol produced as waste from the biodiesel industry, spent
grain from bioethanol production and sewage sludge. Moisture contents can range
from 40 wt% for waste glycerol to 70 wt% for waste distillers grain and to
95 wt% for sewage sludge. Such wet biomass wastes are difficult to process in
terms of energy recovery and fuels production. Conventional processing, such as
combustion, gasification and pyrolysis require a major input of energy to dry the
biomass prior to processing. However, the presence of high water contents in such
types of biomass is advantageous for energy and fuels production by treatment in
supercritical water. The biomass does not have to be dried prior to processing and
the water in the biomass can contribute to hydrogen for syngas production via the
water gas shift reaction. In addition, the product syngas and hydrogen is produced at
high pressure, consequently the further compression of the gas requires less energy
input. This chapter has reviewed the wide range of research into syngas and in
particular hydrogen production from biomass under sub-critical and supercritical
water gasification conditions. However, there are several challenges involved in
scaling-up and commercialising the process. In this section some of the challenges
are discussed.
There have been several reviews of supercritical water gasification of biomass

which include discussion of the technological issues related to process operation
and scale-up [3, 8, 14]. By definition, supercritical water gasification involves
high pressures and moderate temperatures, around the critical point of water, i.e.
22.1 MPa and 374.8 ı C. These conditions require specialist reactor design, non-
corrosive materials and health and safety measures which involve high capital
and operational costs; particularly for the high throughputs of wet biomass that
would be required for the high outputs of hydrogen needed for economic viability.
The optimum feedstock solid loading has been estimated to be above about 15–
20 wt% [8], since lower feedstock loadings involve high water volumes which
require heating and larger volume throughputs which inevitable result in higher
costs. The challenges of raising large volumes of wet biomass with high water
contents to supercritical water conditions should also not be underestimated in terms
of technology requirements and cost. The delivery of the feedstock to the high
pressure reactor also poses problems, particularly at high solids content and where
a continuous process is desired. Pre-treatment of the feedstock to create a slurry, for
example, by suspending the biomass in a starch gel or using a hydrothermal pre-
treatment step have been used successfully [14]. Much of the research discussed
in this chapter has involved the use of model biomass compounds rather than
‘real-world’ biomass which is often heterogeneous, contains inorganic materials
and may be contaminated with impurities. The impurities may block pipework,
valves, pumps etc., and any inorganic salts may precipitate under supercritical
water conditions, again causing blockage problems [81]. Gasification of biomass
produces tar and char which can also block the reactor system [45]. However,
the tar and char from supercritical water gasification is less than that produced
by conventional biomass gasification and in addition the gas produced is cleaner
since any tar or char formation remains in the water phase when the gas and liquid
are separated during the down-stream depressurising stage [14]. Matsumura et al.
[14] have highlighted the importance of energy recovery and the energy balance
of the reactor system and in particular, heat exchange between the output stream
and the input feed stream to enhance the efficiency of the process. The product
syngas will have a high concentration of hydrogen, but will also contain carbon
monoxide, carbon dioxide, methane etc. To produce a pure hydrogen product stream
will require further processing such as membrane separation and pressure swing
absorption (PSA) technologies.
The use of homogeneous and heterogeneous catalysts to enhance the production
of hydrogen from biomass has been discussed in this chapter, but their use in
scaled-up systems may also create challenges. For example, homogeneous alkali
catalysts KOH, NaOH, K2 CO3 , and Na2 CO3 may cause corrosion of the reactor
material and associated equipment, since the required concentrations of the alkali
are often very high [45, 62]. In addition, the solubility of some salts in supercritical
water is low, causing precipitation and potential blockage issues [3]. Heterogeneous
catalysts such as those based on nickel, ruthenium and palladium can be expensive.
In addition, they may degrade or oxidise under supercritical water conditions, can
become coked and deactivated after long periods of use and the catalyst pores
Flue
gas
Biomass
... .. . .
Water Excess gas
Pre- H2, CH4
flared
heater Reactor storage
.. Hot
. Heat
Feed
. .. exchanger
gas
tank .
.. Flue gas
recirculation CO2
Scrubber
Cutting HP Brine
Milling pump Water
removal
. .. . . . .
Cooler Lean
gas
Gas/Liquid
separator
Effluent tank
Fig. 11.3 Schematic diagram of the VERENA supercritical water pilot plant at Forschungszen-
trum, Karlsruhe, Germany (Adapted with permission from Kruse [49]. Copyright © 2009, Elsevier)
may become blocked by the impurities and inorganic material found in biomass
[8, 81]. Moreover, the catalyst supports may hydrolyse, solubilize or undergo phase
transformation under subcritical and supercritical water conditions [7].
Whilst scale-up may seem a daunting task, there have been some successes, for
example, the VERENA supercritical water biomass gasification pilot plant at the
Forschungszentrum in Kalsruhe [82–85]. Figure 11.3 shows a schematic diagram
of the pilot plant [49, 82, 84]. The pilot plant has been used for the supercritical
water gasification of a range of different biomass wastes, for example, glycerol
as waste from biodiesel production, industrial organic wastes and food industry
wastes [82–84]. The maximum operating parameters of the plant are 100 kg h1
throughput, and temperature of 700 ı C and pressure of 35 MPa [82]. The biomass
feed process enables the particle size of the biomass and the water content to be
adjusted to obtain a homogeneous feed suitable for pumping to the reactor using a
mass flow controlled high pressure pump (35 MPa, 100 kg h1 ).
The feed is heated by a heat exchanger using heat supplied from the effluent fluid
from the main reactor which is typically at a temperature of 600 ı C. The feed is
further pre-heated prior to transfer to the main reactor using external heating from
hot flue gases which heats the input fluid to 700 ı C.
The main supercritical water reactor is a down-flow reactor operating typically
at 50–100 kg h1 throughput, and reaction conditions of 660 ı C and pressure
of 28 MPa [82]. The effluent product fluid passes through the initial feed heat
exchanger which cools the effluent fluids and then further cooled via water cooled
cooling unit. The gas-liquid separators separate the gases and water, the second of
which includes a carbon dioxide water scrubber which reduces the CO2 content of
the gases. The water from the scrubbers passes to an effluent tank where further
separation of water and gases and any solids takes place. The product syngas rich in
hydrogen and methane can be stored in high pressure gas bottles and any excess gas
may be flared [82, 84].
Gasafi et al. [85] have undertaken an economic analysis of the VERENA plant for
the supercritical water gasification of sewage sludge, with a throughput of 5 t h1 ,
in terms of the costs of hydrogen production off-set against the disposal costs of the
sewage sludge waste. Their economic model suggested that hydrogen production
costs were in the range of the costs associated with hydrogen production from the
reforming of natural gas. When income from sewage sludge waste disposal costs
was taken into account, the economic benefits of the process were further increased.
The main costs were associated with the carrying charges of the feedstock. Overall,
the economic analysis suggested that supercritical water gasification of sewage
sludge may be competitive due mainly to the revenues from sewage sludge waste
disposal.
Fiori et al. [86] have used process simulation software to produce a conceptual
design for hydrogen production from the supercritical water gasification of biomass
with a high throughput of 1,000 kg h1 . They simulated several different biomass
feedstocks, for example, glycerol, microalgae, sewage sludge and a range of process
conditions i.e. temperatures of 500, 700 and 900 ı C, pressures of 25, 35 and
35 MPa and biomass feed concentrations from 5 to 35 wt%. Their results showed
that supercritical water gasification of biomass produced a maximum hydrogen
production of 8.5 kg/1,000 kg of biomass. The conceptual use of the product
hydrogen in a fuel cell and expanding the high pressure syngas in turbines would
yield a net power output of 150 kWe /1,000 kg of biomass feed per hour [86].
11.6 Conclusions
The supercritical water gasification of biomass has generated considerable research

investigations of the last two decades. The conversion of wet biomass wastes
into hydrogen is of particular interest since such wastes are difficult to handle
and process in terms of energy recovery. In addition, compared to conventional
gasification, the process temperatures of supercritical water processing are sig-
nificantly lower. Research has concentrated on higher temperature non-catalytic
treatment or lower temperature catalysed processing. Much research has been
carried out in batch reactors whereas for the development of a commercial large
scale process, continuous operation is more desirable. There has been work on
continuous supercritical water gasification systems at the bench scale and at the
pilot scale at the VERENA, plant in Germany. The selection of the biomass type
such that it can be easily delivered into the reactor on a continuous basis is optimal
and therefore there is interest in the supercritical water gasification of wet biomass
wastes and wastewaters, for example, sewage sludge, glycerol, food waste etc. The
use of catalysts is mostly through either homogeneous alkali catalysts or solid
heterogeneous metal-based catalysts. The challenges lie in the recovery of the
alkali catalysts and for the heterogeneous catalysts, through catalyst stability and
effectiveness under supercritical water gasification conditions.
References
1. Ramage PR, Agrawal R. The hydrogen economy: opportunities, costs, barriers and R&D needs.
Washington, DC: National Academies Press; 2004.
2. Yanik J, Ebale S, Kruse A, Saglam M, Yuksel M. Biomass gasification in supercritical water:
Part 1. Effect of the nature of biomass. Fuel. 2007;86:2410–5.
3. Kruse A. Supercritical water gasification. Biofuel Bioprod Biorefin. 2008;2:415–37.
biomass utilization in Japan. Energy Convers Manag. 2002;43:1301–10.
5. Sricharoenchaikul V. Assessment of black liquor gasification in supercritical water. Bioresour
Technol. 2009;100:638–43.
6. Kimura H. Oxidation assisted new reaction of glycerol. Polym Adv Technol. 2001;12:697–710.
2008;2:254–65.
8. Peterson AA, Vogel F, Lachance RP, Fröling M, Antal J, Tester W. Thermochemical biofuel
production in hydrothermal media: a review of sub- and supercritical water technologies.
Energy Environ Sci. 2008;1:32–65.
9. Onwudili JA, Williams PT. Hydrothermal gasification and oxidation as effective flameless
conversion technologies for organic wastes. J Energy Inst. 2008;81:102–9.
10. Demirbas A. Hydrogen-rich gas from fruit shells via supercritical water extraction.
Int J Hydrog Energy. 2004;29:1237–43.
11. Azadi P, Farnood R. Review of heterogeneous catalysts for sub- and supercritical water
gasification of biomass and wastes. Int J Hydrog Energy. 2011;36:9529–41.
12. Lu Y, Guo L, Zhang X, Yan Q. Thermodynamic modelling and analysis of biomass gasification
for hydrogen production in supercritical water. Chem Eng J. 2007;131:233–44.
13. Basu P, Mettanant V. Biomass gasification in supercritical water – a review. Int J Chem React
Eng. 2009;7(R3):1–61.
14. Matsumura Y, Minowa T, Potic B, Kersten SRA, Prins W, van Swaaji WPM, van de Beld B,
Elliott C, Neuenschwander GG, Kruse A, Antal MJ. Biomass gasification in near and super-
critical water: status and prospects. Biomass Bioenergy. 2005;29:269–92.
15. Williams PT, Onwudili JA. Subcritical and supercritical water gasification of cellulose, starch,
glucose and biomass waste. Energy Fuel. 2006;20:1259–65.
16. Resende FLP, Neff ME, Savage PE. Noncatalytic gasification of cellulose in supercritical water.
Energy Fuel. 2007;21:3637–43.
17. Kumar S, Gupta RB. Hydrolysis of microcrystalline cellulose in subcritical and supercritical
water in a continuous flow reactor. Ind Eng Chem Res. 2008;47:9321–9.
18. Yoshida T, Matsumura Y. Gasification of cellulose, xylan, and lignin mixtures in supercritical
Appl Catal B Environ. 2012;117:330–8.
20. Okuda K, Ohara S, Umetsu M, Takami S, Adschiri T. Disassembly of lignin and chemical
recovery in supercritical water and p-cresol mixture: studies on lignin model compounds.
Bioresour Technol. 2008;99:1846–52.
21. Yong TLK, Matsumura Y. Reaction kinetics of the lignin conversion in supercritical water.
Ind Eng Chem Res. 2012;51:11975–88.
as a model compound of biomass gasified in supercritical water. Int J Hydrog Energy.
2003;28:55–64.
23. Williams PT, Onwudili J. Composition of products from the supercritical water gasification of
glucose: a model biomass compound. Ind Eng Chem Res. 2005;44:8739–49.
process: a study of model compounds. Ind Eng Chem Res. 2010;49:4055–62.
25. Wahyudiono SM, Goto M. Conversion of biomass model compound under hydrothermal
conditions using batch reactor. Fuel. 2009;88:1656–64.
26. Wahyudiono SM, Goto M. Thermal decomposition of guaiacol in sub- and supercritical water
and its kinetic analysis. J Mater Cycle Waste Manag. 2011;13:68–79.
27. Kruse A, Gawlik A. Biomass conversion in water at 330–410 ı C and 30–50 MPa. Identification
of key compounds for indicating different chemical reaction pathways. Ind Eng Chem Res.
2003;42:267–79.
28. Onwudili JA, Williams PT. Hydrothermal reactions of sodium formate and sodium acetate as
model intermediate products of the sodium hydroxide-promoted hydrothermal gasification of
biomass. Green Chem. 2010;12:2214–22.
29. Onwudili JA, Williams PT. Reactions of different carbonaceous materials in alkaline media for
hydrogen gas production. Green Chem. 2011;13:2837–43.
30. Onwudili JA, Williams PT. Hydrogen and methane selectivity during alkaline supercritical
water gasification of biomass with ruthenium-alumina catalyst. Appl Catal B Environ.
2013;132:70–9.
31. Jin F, Zhou Z, Kishita A, Enomoto H, Kishida H, Moriya T. A new hydrothermal process for
producing acetic acid from biomass waste. Chem Eng Res Des. 2007;85(A2):201–6.
32. Jin F, Yun J, Li G, Kishita A, Tohji K, Enomoto H. Hydrothermal conversion of carbohydrate
biomass into formic acid at mild temperatures. Green Chem. 2008;10:612–5.
33. Jin F, Zhou Z, Moriya T, Kishida A, Higashijima H, Enomoto H. Controlling hydrothermal
reaction pathways to improve acetic acid production from carbohydrate biomass. Environ Sci
Technol. 2009;39:1893–902.
34. Hsieh Y, Du Y, Jin F, Zhou Z, Enomoto H. Alkaline pre-treatment of rice hulls for hydrothermal
production of acetic acid. Chem Eng Res Des. 2009;87:13–8.
35. Minowa T, Inoue S. Hydrogen production from biomass by catalytic gasification in hot
compressed water. Renew Energy. 1999;16:1114–7.
36. Minowa T, Fang Z, Ogi T, Varhegyi G. Decomposition of cellulose and glucose in hot
compressed water under catalysts free conditions. J Chem Eng Jpn. 1998;31:131–6.
37. Sasaki M, Goto K, Tajima K, Adschiri T, Arai K. Rapid and selective retro-aldol condensation
of glucose to glycolaldehyde in supercritical water. Green Chem. 2002;4:285–90.
38. Resende FLP, Fraley SA, Berger MJ, Savage PE. Noncatalytic gasification of lignin in
supercritical water. Energy Fuel. 2008;22:1328–34.
39. Pinkowska H, Wolak P, Zlocinska A. Hydrothermal decomposition of alkali lignin in sub- and
supercritical water. Chem Eng J. 2012;187:410–4.
40. Saisu M, Sato T, Watanabe M, Adschiri T, Arai K. Conversion of lignin with supercritical
water–phenol mixtures. Energy Fuel. 2003;17:922–8.
41. Fang Z, Sato T, Smith RL, Inomata H, Arai K, Kozinski JA. Reaction chemistry and
phase behavior of lignin in high-temperature and supercritical water. Bioresour Technol.
2008;99:3424–30.
42. Kruse A, Maniam P, Spieler F. Influence of proteins on the hydrothermal gasification and
liquefaction of biomass. 2. Model compounds. Ind Eng Chem Res. 2007;46:87–96.
43. Muangrat R, Onwudili JA, Williams PT. Influence of NaOH, Ni/Al2 O3 and Ni/SiO2 catalysts
on hydrogen production from the subcritical water gasification of model food waste com-
pounds. Appl Catal B Environ. 2010;100:143–56.
44. Lu YJ, Guo LJ, Zhang XM, Ji CM. Hydrogen production by supercritical water gasification of
biomass: explore the way to maximum hydrogen yield and high carbon gasification efficiency.
Int J Hydrog Energy. 2012;37:3177–85.
45. Lu YJ, Guo LJ, Ji CM, Zhang XM, Hao XH, Yan QH. Hydrogen production by biomass
gasification in supercritical water: a parametric study. Int J Hydrog Energy. 2006;31:822–31.
46. Sealock LJ, Elliott DC, Baker EG, Butner RS. Chemical-processing in high-pressure aqueous
environments. 1: Historical-perspective and continuing developments. Ind Eng Chem Res.
1993;32:1535–41.
47. Broll D, Kaul C, Kramer A, Krammer P, Richter T, Jung M, Vogel H, Zehner P. Chemistry in
supercritical water. Angew Chem Int Ed. 1999;38:2998–3014.
48. Kumar A, Jones DD, Hanna MA. Thermochemical biomass gasification: a review of the current
status of the technology. Energies. 2009;2:556–81.
49. Kruse A. Hydrothermal biomass gasification. J Supercrit Fluid. 2009;47:391–9.
50. Lee IG, Kim MS, Ihm SK. Gasification of glucose in supercritical water. Ind Eng Chem Res.
2002;41:1182–8.
51. Guo LJ, Lu YJ, Zhang XM, Ji CM, Guan Y, Pei AX. Hydrogen production by biomass
gasification in supercritical water: a systematic experimental and analytical study. Catal Today.
2007;129:275–86.
52. Venkitasamy C, Hendry D, Wilkinson N, Fernando L, Jacoby WA. Investigation of ther-
mochemical conversion of biomass in supercritical water using a batch reactor. Fuel.
2011;90:2662–70.
53. Madenoglu TG, Saglam M, Yuksel M, Ballice L. Simultaneous effect of temperature and
pressure on catalytic hydrothermal gasification of glucose. J Supercrit Fluid. 2013;73:151–60.
54. Susanti RF, Dianningrum LW, Yum T, Kim Y, Lee BG, Kim J. High-yield hydrogen production
from glucose by supercritical water gasification without added catalyst. Int J Hydrog Energy.
2012;37:11677–90.
55. Yu D, Aihara M, Antal MJ. Hydrogen production by steam reforming glucose in supercritical
water. Energy Fuel. 1993;7:574–80.
56. Guo S, Guo L, Cao C, Yin J, Lu Y, Zhang X. Hydrogen production from glycerol by
supercritical water gasification in a continuous flow tubular reactor. Int J Hydrog Energy.
2012;37:5559–68.
57. Guo Y, Wang SZ, Xu DH, Gong YM, Ma HH, Tang XY. Review of catalytic supercritical water
gasification for hydrogen production from Biomass. Renew Sustain Energy Rev. 2010;14:
334–43.
58. Xu DH, Wang SZ, Hu X, Chen CM, Zhang QM, Gong YM. Catalytic gasification of glycine
and glycerol in supercritical water. Int J Hydrog Energy. 2009;34:5357–64.
59. Muangrat R, Onwudili JA, Williams PT. Influence of alkali catalysts on the production of
hydrogen-rich gas from the hydrothermal gasification of food processing waste. Appl Catal B
Environ. 2010;100:440–9.
60. S{nag A, Kruse A, Schwarzkopf V. Key compounds of the hydropyrolysis of glucose in
supercritical water in the presence of K2 CO3 . Ind Eng Chem Res. 2003;42:3516–21.
61. Sinag A, Kruse A, Rathert J. Influence of the heating rate and the type of catalyst on the
formation of key intermediates and on the generation of gases during hydropyrolysis of glucose
in supercritical water in a batch reactor. Ind Eng Chem Res. 2004;43:502–8.
62. Onwudili JA, Williams PT. Role of sodium hydroxide in the production of hydrogen gas from
the hydrothermal gasification of biomass. Int J Hydrog Energy. 2009;34:5645–56.
63. Schmieder H, Abeln J, Boukis N, Dinjus E, Kruse A, Kluth M, Petrich G, Sadri E, Schacht M.
Hydrothermal gasification of biomass and organic wastes. J Supercrit Fluid. 2000;17:145–53.
64. Madenoglu TG, Boukis N, Saglam M, Yuksel M. Supercritical water gasification of real
biomass feedstocks in continuous flow system. Int J Hydrog Energy. 2011;36:14408–15.
65. Akgul G, Kruse A. Influence of salts on the subcritical water-gas shift reaction. J Supercrit
Fluid. 2012;66:207–14.
66. Hao X, Guo L, Zhang X, Guan Y. Hydrogen production from catalytic gasification of cellulose
in supercritical water. Chem Eng J. 2005;110:57–65.
67. Byrd AJ, Pant KK, Gupta RB. Hydrogen production from glycerol by reforming in supercritical
water over Ru/Al2 O3 catalyst. Fuel. 2008;87:2956–60.
68. Azadi P, Khodadadi AA, Mortazavi Y, Farnood R. Hydrothermal gasification of glucose
using Raney nickel and homogeneous organometallic catalysts. Fuel Process Technol.
2009;90:145–51.
69. Zhang L, Champagne P, Xu C. Supercritical water gasification of a aqueous by-product from
biomass hydrothermal liquefaction with novel Ru modified Ni catalysts. Bioresour Technol.
2011;102:8279–87.
70. Azadi P, Afif E, Azadi F, Farnood R. Screening of nickel catalysts for selective hydrogen
production using supercritical water gasification of glucose. Green Chem. 2012;14:1766–77.
71. Osada M, Sato O, Arai K, Shirai M. Stability of supported ruthenium catalysts for lignin
gasification in supercritical water. Energy Fuel. 2006;20:2337–43.
72. Lee IG, Ihm SK. Catalytic gasification of glucose over Ni/activated charcoal in supercritical
73. Yamaguchi A, Hiyoshi N, Sato O, Bando KK, Osada M, Shirai M. Hydrogen production
from woody biomass over supported metal catalysts in supercritical water. Catal Today.
2009;146:192–5.
74. Li S, Lu Y, Guo L, Zhang X. Hydrogen production by biomass gasification in supercritical
water with bimetallic Ni-M/”Al2 O3 catalysts (M D Cu, Co, Sn). Int J Hydrog Energy.
2011;36:14391–400.
75. Zhang L, Xu C, Champagne P. Activity and stability of a novel Ru modified Ni catalyst for
hydrogen generation by supercritical water gasification of glucose. Fuel. 2012;96:541–5.
76. Lu Y, Li S, Guo L. Hydrogen production by supercritical water gasification of glucose with
Ni/CeO2 /Al2 O3 : effect of Ce loading. Fuel. 2013;103:193–9.
77. Xu C, Donald J. Upgrading peat to gas and liquid fuels in supercritical water with catalysts.
Fuel. 2012;102:16–25.
78. Chakinala AG, Chinthaginjala JK, Seshan K, van Swaaij WPM, Kersten SRA, Brilman DWF.
Catalyst screening for the hydrothermal gasification of aqueous phase of bio-oil. Catal Today.
2012;195:83–92.
79. Van Bennekom JG, Kirillov VA, Amosov YI, Krieger T, Vendenbosch RH, Assinka D,
Lemmens KPJ, Heeres HJ. Explorative catalyst screening studies on reforming of glycerol
in supercritical water. J Supercrit Fluid. 2012;70:171–81.
80. Osada M, Yamaguchi A, Hiyoshi N, Sato O, Shirai M. Gasification of sugarcane Bagasse over
supported ruthenium catalysts in supercritical water. Energy Fuel. 2012;26:3179–86.
81. Xu DH, Wang SZ, Gong YM, Guo Y, Tang XY, Ma HH. A novel concept reactor design for
preventing salt deposition in supercritical water. Chem Eng Res Des. 2010;88:1515–22.
82. Mobius A, Boukis N, Galla U, Dinjus E. Gasification of pyroligneous acid in supercritical
water. Fuel. 2012;94:395–400.
83. Boukis N, Galla U, Muller H, Dinjus E. Hydrothermal gasification of glycerol on the pilot plant
scale. 16th European biomass conference & exhibition, 2–6 June 2008, Valencia, Spain; 2008.
84. Hammerschmidt A, Boukis N, Hauer E, Galla U, Dinjus E, Hitzmann B, Larsen T, Nygaard
SD. Catalytic conversion of waste biomass by hydrothermal treatment. Fuel. 2011;90:555–62.
85. Gasafi E, Reinecke MY, Kruse A, Schebek L. Economic analysis of sewage sludge gasification
in supercritical water for hydrogen production. Biomass Bioenergy. 2008;32:1085–96.
86. Fiori L, Valbusa M, Castello D. Supercritical water gasification of biomass for H2 production:
process design. Bioresour Technol. 2012;121:139–47.
Chapter 12
Production of CH4 from Biomass
via Supercritical Water Gasification
Izad Behnia, Zhongshun Yuan, Paul Charpentier, and Chunbao (Charles) Xu
Abstract A common route to generate methane-rich gas from renewable biomass

feedstocks (such as agricultural/forestry residues and municipal/industrial wastew-
ater sludges) is anaerobic digestion. However, biological processes are limited in
many cases by low reaction rates (days to weeks reaction) and low tolerance to
bacteria, low tolerance to temperature and a high footprint (large reactor and land
requirement). Supercritical water gasification (SCWG) of biomass/waste can be a
promising alternative technology for renewable CH4 production as the process has a
much smaller footprint and higher efficiency. Many early studies on biomass SCWG
proved that methane-rich gas can be achieved at lower temperatures (<500 ı C),
compared with a high yield of hydrogen at above 600 ı C. Production of methane-
rich gas or synthetic natural gas (SNG) via SCWG can be carried out at low
temperatures, since the methanation reaction is a highly exothermic reaction which
requires less external heat to maintain the desired reaction temperature. Processing
at low temperatures can improve cold gas efficiency of the process and utilize
waste heat from some high-temperature processes (i.e. iron/steel manufacturing).
Moreover, methane-rich syngas has a higher Btu value per unit volume than
hydrogen, so it is a promising substitute for natural gas for heat or electricity
production, or it can be injected into existing natural gas grids. However, a low
temperature process generally requires the presence of a catalyst to maximize its
reaction rate. For CH4 production via low-temperature SCWG of biomass, the
presence of an active catalyst is critically important for reducing the tar/char yields,
lowering the reactor volume and increasing the carbon conversion rate. This chapter
I. Behnia • Z. Yuan • P. Charpentier () • C. Xu ()

and Biochemical Engineering, Western University, London, ON N6A 5B9, Canada
324 I. Behnia et al.
overviews the state-of-the-art in production of CH4 from biomass via SCWG. The
focus of this chapter is placed on effects of operating conditions (temperature,
residence time, pressure, etc.) and types of catalysts on the gasification efficiency
and CH4 yield.
Keywords Supercritical water gasification • Methane • Synthetic natural gas •

Biomass • Catalysts
12.1 Introduction
Depletion of fossil fuels and environmental issues such as CO2 emissions and
global warming have intensified worldwide efforts to secure alternative sources
of energy. Methane is considered to be a more eco friendly fuel than other
hydrocarbon fossil fuels (coals and petroleum), as it generates less CO2 emissions
per joule of heat. Methane, being the simplest hydrocarbon, has 802.6 kJ/mol net
enthalpy of combustion with the ratio of heat of combustion to its molecular mass
releasing more heat per mass unit (50.2 kJ/g) than other complex hydrocarbons
[1]. Methane can be used as a fuel in turbines, steam boilers, heat for households,
or in car engines. Natural gas with 95 % methane content is a rich source of
methane. This natural gas is piped for domestic heating and cooking purposes
throughout the world. In a recent speech by US president Barack Obama on June
25, 2013 at Georgetown University in Washington DC, it was reaffirmed that the
US government will commit to burning cleaner natural gas instead of dirtier fuel
sources (e.g., coal) for power generation [2].
Synthetic natural gas (SNG) from biomass is very promising as biomass is
renewable, abundant, clean, and carbon neutral. The main applications and benefits
of SNG include: (1) using SNG can meet emission requirement easily; (2) SNG
can be distributed for transport, heat and electricity uses; (3) it can be an excellent
back-up for natural gas [3].
As is well known, a common route to generate methane-rich bio-gas is anaerobic
digestion. However, biological processes face many challenges such as a large
footprint (large reactor and land requirement), low reaction rates (days to weeks
reaction) and low tolerance of bacteria to temperature changes and different types of
feedstock. In contrast, the low-temperature supercritical water gasification (SCWG)
process can be a promising alternative technology for renewable CH4 production
[4, 5].
Supercritical water (SCW) is defined as water above its critical point (374.0 ı C,
22.1 MPa), which has properties different from those of steam and liquid water, such
as tunable densities and dielectric constant. In particular, water loses its hydrogen
bonding capacity in the supercritical state, giving SCW low polarity. This lack
of polarity makes it an ideal solvent for non-polar compounds and gases, hence
eliminating mass transfer limitations between different phases in a reaction system.
SCW can be employed as a suitable solvent or reactant for reactions, e.g., SCWG
12 Production of CH4 from Biomass via Supercritical Water Gasification 325
of biomass for renewable hydrogen generation, leading to shorter residence times

and a smaller reactor volume and process footprint. In SCWG of biomass, the main
possible reactions are provided as follows using glucose as the model compound for
biomass with the major gas products of H2 , CO2 as well as CO and CH4 , depending
on reaction conditions [4, 6]:
Steam-reforming:
C6 H12 O6 .l/ C H2 O .g/ ! 6CO C 6H2 C H2 O .g/; H0 298 D 607:8 kJ=mol
(12.1)
Methanation:
CO .g/ C 3H2 .g/ ! CH4 .g/ C H2 O .g/; H0 298 D 206:2 kJ=mol
(12.2)
Water-gas shift:
CO .g/ C H2 O .g/ CO2 .g/ C H2 .g/; H0 298 D 41:2 kJ=mol

(12.3)
At relatively lower temperatures (in the vicinity of the critical point) and with
the presence of a suitable catalyst, the methanation reaction leads to significant
formation of methane-rich gas products, while at higher temperatures, H2 is the
dominant gas product from SCWG of biomass [4], due to the water-gas shift
reaction.
The water-gas shift reaction is a weakly exothermic reaction (Hı 298 D
41.2 kJ/mol) compared with the methanation reaction (Hı 298 D 206.2 kJ/mol),
so it is more thermodynamically favorable at temperatures above water’s critical
temperature. In comparison, low temperatures favor the methanation reaction,
leading to generation of methane-rich syngas [7, 8]. An equilibrium model
calculation shows that the maximum CH4 production is achievable when the
gasification reaction operates at a lower temperature but at a higher pressure [5,
9], while Antal et al. [10] showed pressure does not have a significant effect on the
gas distribution at high temperatures (>550 ı C). Substrate concentration also plays
an important role in the product gas composition. Antal et al. [11] showed the total
carbon efficiency and H2 yield dropped when increasing the substrate concentration,
whereas CH4 and CO2 yields increased. Although a higher concentration of organic
feedstock is more favorable for CH4 production, it also results in higher tar and char
yields. Therefore, the use of a suitable catalyst is critical in the process to speed up
the steam-reforming reaction, while also minimizing tar and char formation.
Production of methane-rich gas or synthetic natural gas (SNG) via SCWG may
be economically preferred to hydrogen generation due to the lower required temper-
atures. As shown above, the methanation reaction is a highly exothermic reaction
which requires less external heat to maintain the desired reaction temperature.
In addition, operation at low temperatures can improve cold gas efficiency of

the process and utilize waste heat from high-temperature processes (i.e. iron/steel
manufacturing) [12, 13]. Moreover, methane-rich syngas has a higher heating value
per unit volume than hydrogen, so it is a promising substitute for natural gas for heat
or electricity production, or it can be integrated into the existing natural gas grids.
However, a low temperature process generally requires the presence of a catalyst
to enhance the reaction rate. For CH4 production via low-temperature SCWG of
biomass, the presence of an active catalyst is critically important for reducing the
tar/char yields and increasing the carbon conversion rate.
Previous studies on biomass SCWG showed that methane-rich gas can be
achieved at lower temperatures (<500 ı C), compared with a high yield of hydrogen
at above 600 ı C [6, 14–16]. For example, it was previously shown that gasification
of a 5 wt.% aqueous glucose solution at 600 ı C demonstrated that hydrogen-
rich gas with a low concentration of methane was produced, where it was also
found that a reduced nickel catalyst favored the water-gas shift reaction, while
ruthenium suppressed tar and char production [6]. SCWG of lignin and cellulose
in the presence of Ru/TiO2 at 400 ı C in a bath reactor led to 31 and 74 %
carbon conversion, respectively, and gas products consisting of CH4 as the main
component (41 45 vol.%) [8]. Park and Tomiyasu [17] almost completely gasified
cellulose at 450 ı C for 120 min in a bath reactor using RuO2 catalyst, producing
a gas product containing CH4 (34 vol.%), H2 (14.6 vol.%) and CO2 (50.9 vol.%).
Elliot et al. achieved complete gasification of dairy manure at 350 ı C over Ru/C
catalyst [18], generating a produced gas comprising almost 50 vol.% CH4 . However,
interestingly Cortright et al. [19] also demonstrated that hydrogen can be produced
from sugars and alcohols at temperatures near 500 K (approx. 225 ı C) and 2.9
and 5.6 MPa pressure in a single-reactor aqueous-phase reforming (APR) process
using a platinum-based catalyst. They found that the selectivity for hydrogen
production increased when feedstocks were used that are more reduced than sugars,
e.g., using ethylene glycol and methanol which were almost completely converted
into hydrogen and carbon dioxide using a low weight hourly space velocity
(WHSV).
The above results from aqueous phase reforming of biomass suggest that a
catalyst plays a critically important role in the gasification of biomass in sub-
critical/supercritical water. In a recent study by the authors, SCWG of 5 wt.%
glucose-water solution was operated at 500 ı C in a continuous flow reactor with
various catalysts. Among all the catalysts tested, Ru2% Ni20% /”-Al2 O3 (containing
2 wt.% Ru and 20 wt.% Ni) as an inexpensive catalyst achieved nearly complete
carbon conversion into gas products, generating 40 vol.% methane and 20 vol.%
hydrogen, balanced by CO2 and other minor components. No reactor-plugging was
observed in more than 8 h on-stream.
This chapter overviews the state-of-the-art in production of CH4 from biomass
via SCWG. The focus of this chapter is on how operating conditions (temperature,
residence time, pressure, etc.) and the type of catalyst affect the gasification
efficiency and CH4 yield.
12.2 Supercritical Water – A Unique Medium

for Biomass Gasification
When water is above its critical point, its diffusivity increases, viscosity decreases,
as well as it loses H-bonding ability, attaining superior solubility for organic
compounds, while drastically decreasing the solubility for inorganics [20]. When
gasifying biomass in SCW, i.e., hydrothermal gasification or SCWG, SCW acts
as both a medium and a reactant for the reforming/gasification reactions [21].
Hydrothermal gasification, especially SCWG, can be used for gasifying organic
biomasses or for the treatment of hazardous waste materials. Being superior to other
thermo-chemical processes (combustion, air-blown gasification, pyrolysis, etc.),
the process of biomass SCWG can eliminate the need for drying of wet biomass
even if its water content is more than 90 %. Catalytic gasification of biomass
in hot-compressed water environment or SCW efficiently produces hydrogen and
methane, and the product gas composition depends on the type of feedstock, reactor
design, temperature/pressure of the operation, and application of proper catalyst.
The gas product formed contains mostly methane, hydrogen, carbon dioxide,
carbon monoxide, water vapor, and a low concentration of C2 –C3 . As shown
in Fig. 12.1, the thermodynamic equilibrium calculation of SCWG of biomass
(using Lactide, C6 H8 O4 , as a model compound) reveals that H2 is the main gas
product formed at high temperatures, but temperature does not influence the CO2
concentration significantly. In contrast, at low temperatures, CH4 dominates the gas
products formed although lower temperatures favor tar/char formation and reduced
gasification efficiency [20, 22].
Fig. 12.1 Equilibrium gas composition of biomass model compound (Lactide, C6 H8 O4 ) under
subcritical and supercritical water (C6 /H2 O D 1/10 mol/mol) at 30 MPa
SCWG operation can be classified into three main regions based on the operating
temperatures, i.e., sub-critical, low temperature supercritical and high temperature
supercritical operation [23].
• Region I (500–800 ı C) – the high temperature supercritical region: Gasification
reactions of biomass are greatly promoted in this region, but a catalyst is still
desirable to enhance the hydrogen yield and carbon conversion while avoiding
char formation. In this region, catalysts such as alkali compounds promote the
water-gas shift reaction towards production of hydrogen-rich gas. The authors’
group [21] utilized novel Ru2% Ni10% /”-Al2 O3 catalyst for gasification of glucose,
simulated aqueous biomass (glucose, acetic acid, Guaiacol) and a aqueous waste
stream from wastewater sludge hydrothermal liquefaction process. For all these
biomass feedstocks, approx. 100 % carbon conversion into gaseous products
consisting of mainly H2 , CO2 and CH4 was achieved at temperatures >650 ı C.
• Region II (374–500 ı C) – the low temperature supercritical region: A catalyst
is required to facilitate hydrolytic degradation of biomass and the subsequent
reforming/gasification reactions. A high yield of tarry materials is expected in
this region due to the favorable conditions for re-polymerization of the hydrolytic
products. The presence of suitable catalysts such as novel metal catalysts can
effectively reduce tar formation and promote the gasification reactions [24].
• Region III (below 374 ı C) – subcritical region: in this region, the biomass feed-
stock is more likely to be liquefied to obtain bio-crude oil products via hydrother-
mal liquefaction/gasification, while aqueous phase reforming of biomass is
possible but the process is slow as expected and requires highly active metal
catalysts [4].
The chemistry involved in hydrothermal gasification is relatively complex, which
makes it difficult to uncover the mechanism of methane formation under the harsh
SCWG conditions. Elliott and Sealock [25] suggested the overall stoichiometric
reaction for gasification of cellulose as a carbon source feedstock at 350 ı C and
21 MPa follows: C6 H12 O6 (l) C H2 O (g) ! 3CH4 (g) C 3 CO2 (g). Some possible
reaction pathways were proposed to describe the catalytic reactions of biomass
(phenol) gasification in SCW using Ni/C catalyst [26], which will be discussed in
the following sections.
12.3 Effects of Operating Conditions on Biomass SCWG

for Methane Production
12.3.1 Effects of Temperature
Although operation at lower temperatures is more economical in terms of operating

cost than at higher temperatures, it is more challenging to achieve suitable biomass
conversions and product yields. It is also still a challenge to operate a flow-type
Fig. 12.2 Effects of temperature on yields of gas products from gasification of 5 wt.% glucose-
water solution at 27.2 MPa, WHSV D 3 h1 , and in presence of Ru2% Ni20% /”-Al2 O3
reactor at low temperatures due to the formation of tar and char, which will cause
reactor plugging issues. On the other hand, severe operational conditions with
high temperature supercritical water also poses a major challenge on equipment
specifications, requiring special alloys for reactor construction due to the higher
operating pressure and increased corrosion problems.
As discussed previously in this chapter, there are many studies on the hydrother-
mal conversion of wet biomass. Elliott et al. [27] achieved high methane production
from four different feedstocks: p-cresol 2 %, cheese whey, lactose 10 %, and lactose
all at 5 wt.% in a continuous-flow tubular reactor at 360 ı C, and 20.4 MPa pressure
with different commercial Ni catalysts. Their research reported 94.3, 94.3 87.5, and
94.8 % conversion of carbon in the above 4 feedstocks into gas products of high
CH4 concentration (53, 48, 47, and 40 vol.%), respectively. Aquatic and food waste
materials with 1–20 wt.% carbon composition was treated using near-/supercritical
water at 350–450 ı C to obtain methane-rich gaseous products [28]. Experiments by
various research groups on low-temperature supercritical water gasification have
shown high methane gas production under these conditions, which supports the
thermodynamic predictions [29, 30]. However, the lower the reaction temperature,
the more tar and char formation is expected in non-catalytic SCWG. To overcome
these undesirable effects, different catalysts are commonly used. The authors of
this chapter were successful in the production of methane from glucose via SCWG
at temperatures ranging from 400 to 600 ı C using a nickel-ruthenium catalyst of
Ru2% Ni20% /”-Al2 O3 . Some typical results are illustrated in Fig. 12.2 [31].
As shown in Fig. 12.2, increasing the reaction temperature from 400 to 500 ı C
produced a drastic increase in total carbon conversion and gas total yield. SCWG
of glucose at 500 ı C and above achieved almost complete carbon conversion into
carbon-containing gases, rich in methane and hydrogen. The yield of CH4 increased
first with increasing temperature from 400 to 500 ı C but dropped after 500 ı C,
attributed to the competing effects of kinetics and thermodynamics. The peak CH4
yield was 0.5 mol/mol-C in the feedstock at 500 ı C. As expected, the H2 yield
increases with increasing temperature, attaining 0.8 mol/mol-C in the feedstock at
600 ı C [31].
The effects of various catalysts on biomass low-temperature SCWG will be
discussed in the subsequent section.
12.3.2 Effects of Catalyst
Metallic catalysts (Ni, Ru, etc.) are commonly used for the production of methane
with SCWG of biomass. An efficient catalyst is one with high stability and high
activity during the long reaction times, good selectivity towards the desirable prod-
ucts, high carbon conversions, and using relatively inexpensive metals. However, a
major challenge for using metallic catalysts for SCWG reactions is that the harsh
conditions of SCWG cause oxidation of the catalyst. Both catalyst and support
should remain stable in the hot compressed water environment. Common problems
responsible for deteriorated catalyst stability and activity in SCWG are sintering and
phase transformation, and changes in the surface properties of the catalysts. Sharma
et al. [26] tested a novel carbon-supported Ni catalyst to treat wastewater samples
from a coal dewatering process and the electronic industry. This catalyst was highly
active and stable even after 200 h of testing, without any observable sintering
problems. In the authors’ recent study, a highly active catalyst, Ru2% Ni20% /”-Al2 O3 ,
also retained its activity during SCWG of 5 wt.% glucose for 20 h on stream at
500 ı C and 27.2 MPa. Among all catalyst support materials, carbon, zirconia and
alumina were demonstrated to be the most stable supports in the SCW environment
[20, 32, 33]. Azadi et al. [34] studied the catalytic hydrothermal gasification of
biomass (glucose, cellulose, fructose, xylan, pulp, lignin and bark) in the presence
of Ni/’-Al2 O3 , Ni/hydrotalcite, Raney nickel, Ru/C, and Ru/’-Al2 O3 catalysts at
350 ı C in a batch reactor. It was observed that with ruthenium based catalysts,
high methane production was achievable. Among the catalysts studied, Ni and
Ru were found to be the most efficient methanation catalysts. These were also
found to be the most effective catalysts for reducing tar and char formation in low
temperature SCWG [23]. Ruthenium has also been used as a co-catalyst to promote
the performance of Ni catalyst by decreasing char formation [6].
As was mentioned previously, an important factor in choosing catalysts for
SCWG is the catalyst’s stability. Xu et al. [35] reported a significant loss of catalyst
activity during SCWG operation. For example, many metallic catalysts supported
on coconut shell activated carbon lost activity after 4 h on stream during the SCWG
of glucose at 600 ı C. Nickel and copper catalysts showed excellent stability during
the process. However, copper is not normally as active as ruthenium and nickel
[20, 25]. Elliott et al. [25] investigated the life-time of a wide range of catalysts
and stability of supports in hot compressed water environment at 350 ı C, and their
Table 12.1 Low-temperature gasification of high-moisture biomass feedstocks with Ni-catalysts

at 34 MPa, 400 ı C, and 15 min reaction time
Gas, % of
Feedstock carbon fed CH4 scf/lb H2 CH4 CO2 CO C2 C
Cellulose 97.6 5.5 3.9 43.2 47.8 0 3.7
Sorghum 94.6 5.5 9.5 38.4 50.5 0 1.6
Sunflower 87.8 5.1 5.0 45.7 45.5 0 2.4
Napler grass 100.8 5.2 6.9 40.6 51.1 0 1.4
Corn stover 72.9 1.5 20.1 16.3 62.0 0 1.3
Water hyacinth 73.1 4.0 11.3 35.9 49.6 0.8 0.9
Kelp 78.8 2.9 7.1 41.9 48.4 0 1.5
Douglas fir 49.8 2.6 21.6 29.0 47.6 0 1.8
Grape pomace 44.5 3.1 9.6 40.2 47.7 0 1.0
Spent grain 55.7 3.9 9.6 43.1 44.2 0 1.4
Potato wasteC 46.4 2.0 27.6 20.4 50.2 0 1.8
Reprinted (adapted) from Elliott et al. [28]. Copyright © 1988, Springer
results showed that Ru and Rh are very stable with satisfactory lifetimes under
hydrothermal gasification conditions. These noble metal catalysts have been widely
applied in methanation reactions [23, 36–38]. However, the limitation of these noble
metals as catalysts is their high cost. In this regard, inexpensive metallic catalysts
such as nickel are advantageous. In the following section, the performance of Ni-
based catalysts in the SCWG of biomass is discussed in detail.
12.3.2.1 Nickel Catalysts
In a study by Sealock Jr. et al. [28], nickel catalysts were utilized for the
SCWG of various high moisture-containing biomass feedstocks, and the results are
summarized in Table 12.1. The nickel catalyst shows great potential for catalyzing
the methanation reaction at 400 ı C, leading to a very high carbon conversion.
However, the catalyst in this study exhibited poor stability. It was observed in
the same study that addition of alkali to the catalytic system led to an increase in
hydrogen production, accompanied by a reduced methane yield.
Lee and coworkers [33] demonstrated high stability and activity of a Ni16% /AC
catalyst for gasification of concentrated 0.6 M glucose at 575–725 ı C in a packed
bed reactor. Based on a patent by Sealock Jr. et al. [28], there was a direct
relation between methane formation and concentration of reduced nickel catalyst:
in low-temperature SCWG, higher nickel concentrations resulted in higher methane
formation. Sato et al. [39] observed that increasing nickel loading on a MgO support
produced more methane in the SCWG of lignin. Again, as a common observation in
many studies, an increase of nickel metal loading enhances gasification efficiency
and increases the yield of various gas products, in particular methane formation.
In a recent study by the authors of this chapter [31], ”-Al2 O3 -supported NiRu
catalysts containing 10–30 wt.% Ni with 2 wt.% Ru as promoter showed high
activity for gasification of 5 wt.% water-glucose solution at 500 ı C in a flow reactor.

Increasing Ni loading promoted both CH4 and H2 formation, while such an effect
leveled off as the metal loading increased above 20 wt.%, likely due to the poor
metal dispersion at high metal loading. Ru2% Ni20% /”-Al2 O3 showed the highest
activity in the SCWG of the 5 wt.% glucose-water solution into gas products rich in
methane.
Several possible reaction pathways were proposed by Sharma et al. [26] to
describe the catalytic reactions of biomass (phenol) gasification in SCW in the
presence of a Ni catalyst supported on carbon (Ni/C), as follows:
I. Decomposition of large molecules into smaller molecules at an elevated
temperature;
II. Steam gasification of the small molecules into CO and H2 (steam reforming
reaction);
III. Methanation and water-gas shift reaction of CO to produce CH4 and CO2 over
metallic catalysts.
1. Reduction of water by Ni: Ni C H2 O ! NiO C H2
2. Reduction of organic compounds by Ni: C6 H5 OH C n Ni ! C6 Hx Nin C
(2 x) H2 C H2 O
3. Reduction of NiO: C6 Hx Nin C 6NiO ! 6CO C (x/2) H2 C (6 C n) Ni
4. Methanation reaction: 3CO C 3H2 $ CH4 C H2 O
5. Water-gas shift reaction: CO C H2 O H2 C CO2
The non-stoichiometric overall reaction: C6 H5 OH C H2 O ! CH4 , C2 H6 ,
H2 , CO2
In step 1, nickel was oxidized by water to form nickel oxide. Ni metal can also
reduce organic compounds to form organic Ni compounds as a reducing agent (step
2) and can reduce NiO back to Ni metal (step 3). Meanwhile, the methanation and
water-gas shift reactions also take place in the reactor system. Overall, the formation
of CO, H2 , CO2 , and CH4 over Ni catalyst surface can be realized.
12.3.2.2 Ruthenium Catalysts
As introduced briefly previously, Ru is a highly active metallic catalyst or co-catalyst

for the SCWG of aqueous biomass [31–33, 37, 38, 40]. Results of the gasification
of p-cresol at 350 ı C and 20 MPa over different ruthenium catalysts in comparison
with other noble metal catalysts in a batch reactor are summarized in Table 12.2
[28]. As clearly shown in this Table, Ru has a much higher activity than other noble
metal catalysts (Rh, Pt, Pd) for SCWG.
As reported in a patent report by Elliott and Sealock [38], addition of ruthenium
as a co-catalyst to nickel could retard the sintering and crystallization of nickel
particles, which might account for its positive effect on enhancing the carbon
conversion in the SCWG of biomass. However, the effect on gas distribution was
found to be minimal. In a recent study by the authors of this chapter, it was also
Table 12.2 Performance of different Ruthenium catalysts in SCWG of p-cresol at 350 ı C and
20 MPa
Product gas, vol.%
Catalyst % carbon conv. CH4 CO2 H2 Time (min)
Ru5% /”-Al2 O3 89.1 58.9 38.8 1.0 90
Ru5% /•-Al2 O3 43.5 55.2 35.9 7.7 120
Ru5% /ZrO2 (reduceda ) 28.5 49.0 35.6 13.4 90
Ru3% /”-Al2 O3 18.8 40.9 37.3 20.6 120
Ru5% /carbon 21.4 24.1 58.6 10.2 110
Ru1% /carbon (reduced) 17.3 40.4 42.3 8.4 125
Ru5% /’-Al2 O3 (reduced) 0.01 0.5 0.6 98.5 120
Ru5% /carbon (reduced) 26.6 57.5 36.3 2.3 125
Rh1% /”-Al2 O3 (reduced) 7.06 42.1 39.1 14.4 110
Rh1% /’-Al2 O3 1.54 12.7 32.9 52.5 105
Pt5% /”-Al2 O3 1.29 26.9 63.1 3.9 90
Pd5% /”-Al2 O3 0.42 11.0 29.3 59.8 95
Pd2% /carbon (reduced) 1.20 7.2 56.3 35.2 90
Reprinted (adapted) from Elliott et al. [28]. Copyright © 1988, Springer
a
Catalyst reduction was performed in-situ in the presence of 1 MPa hydrogen at 400 ı C overnight
observed that addition of Ru to a Ni catalyst promoted carbon conversion efficiency

in the SCWG of glucose.
At high temperatures and low substrate concentrations, Ru proved to be a very
active catalyst for improving hydrogen generation and reducing methane formation
in SCWG [41]. Thus, ruthenium can either reduce or increase methane formation
in the SCWG of biomass, depending on the operational conditions. Byrd et al. [41]
reported that ruthenium suppressed methanation reaction at 700 ı C and 25 MPa for
SCWG of 1 wt.% glucose.
In summary, ruthenium or nickel are the most efficient catalysts for low
temperature SCWG, with Ru being an effective promoter for Ni catalysts.
12.3.2.3 Carbon Catalysts
Although Ni and Ru are the most common catalysts utilized for the SCWG of
various feedstocks, a few researchers have also investigated carbon catalysts. Xu
et al. [35] reported the catalytic activity of carbon substrates with a high methane
gas yield and carbon gasification efficiency. Table 12.3 presents the results obtained
with glucose as a model compound in a bench-scale flow type reactor at 600 ı C and
35 MPa with 10 h1 WHSV.
Comparison of spruce wood charcoal and coconut activated carbon shows that
these carbon catalyst surface areas do not influence the catalytic activity of the
catalyst significantly. Although these carbon-based catalysts demonstrated high
activity, they began deactivating after only 4 h. As reported by other researchers,
Table 12.3 Effect of various carbon catalysts on gasification of 1.2 M glucose in supercritical
water at 600 ı C, 35 MPa
Coal activated Coconut shell Macadamia Spruce wood
Catalyst type carbon activated carbon shell charcoal charcoal
WHSV, h1 19.9 22.2 25.7 12.6
Gas products Gas yield (mol of gas/mol of feed)
H2 1.48 2.24 2.71 3.86
CO 2.34 0.97 0.54 0.34
CO2 1.45 3.09 1.09 3.72
CH4 1.04 1.23 3.18 1.36
C2 H4 0.002 0.0 0.002 0.0
C2 H6 0.39 0.35 0.27 0.23
C3 H6 0.002 0.0 0.003 0.0
C3 H8 0.13 0.13 0.11 0.01
Carbon gasification 97 103 95 99
efficiency, %
pH of liquid effluent 4 5 4 4
Reprinted (adapted) with permission from [35]. Copyright © 1996, American Chemical Society
carbon based catalysts significantly lose their activity after only a few hours [4, 18,
35], which might be due to high carbon deposition on their highly porous surface.
12.3.3 Effects of Pressure
Total pressure of the reaction system does not appear to have a significant effect
on the SCWG of biomass. Xu et al. [35] studied the pressure effect on SCWG of
glucose to find the correlation of this effect on gas distribution. Although higher
pressures did not significantly affect carbon conversion in gasification of glucose
at 600 ı C, pressure did slightly influence the gas distribution: SCWG at higher
pressure promoted H2 and CH4 formation. The positive effect of pressure on
methane formation in high temperature SCWG is also evidenced by the research
of Kruse et al. [42], where pressure was observed to promote methane formation in
the SCWG of pyrocatechol at 700 ı C in a tubular reactor.
Although higher pressures promote the methanation reaction, it is however not
economically wise to increase the reactor pressure due to the increased capital costs
for higher pressure rated equipment.
12.3.4 Effects of Substrate Concentration
Apart from temperature, catalysts and pressure, substrate concentration is another

important factor affecting gasification efficiency. Kinetically, a higher substrate
Fig. 12.3 Effect of substrate

concentration on gas yields in
SCWG of glucose solution in
the presence of ruthenium
catalyst at 600 ı C, WHSV of
10 h1 and 30 MPa
(Reprinted (adapted) with
permission from Kersten
et al. [37]. Copyright © 2006,
concentration enhances the reaction rate, but it will also lead to a reduced conversion
for the thermodynamic equilibrium equations due to the reduced water fraction
in the reactor system. In a study by Kersten et al. [37] on the SCWG of glucose
over ruthenium catalyst at 600 ı C and 30 MPa, some typical results are shown
in Fig. 12.3. Generally, the yields of H2 and CO2 decrease while the substrate
concentration increases, while the CH4 yield climbs with increasing substrate
concentration, which is likely resulted from suppressed steam reforming reaction
of methane due to less water at an increased substrate concentration. In other
words, higher yields of H2 and CO2 are generally realized at a lower substrate
concentration. This can be easily explained by the fact that a higher water content
drives the equilibrium of the steam-reforming reaction towards H2 and CO2
production [20].
Thermodynamic equilibrium calculations can help predict the increase of
methane production at a higher substrate concentration and at a lower temperature
[43]. However, this operational condition, although favoring increased methane
production, would also lead to a higher tar and char formation.
12.4 Tar and Char Reduction
Formation of tar and char in a biomass SCWG process will result in a considerable
loss of gasification efficiency [43]. Normally char refers to the unconverted biomass,
while tar is the undesired reaction products [24]. The formation of tar/char is
intensified at a lower temperature or a higher feed substrate concentration, or
a low heating rate [4, 33]. Formation of char and tar during gasification limits
the efficient production of methane and hydrogen and biomass conversion. Tar
formation is more favored at low temperatures and longer residence times [35]
Fig. 12.4 Proposed formation pathways for char/coke in SCWG of glucose (Reprinted (adapted)
with permission from Chuntanapum and Matsumura [47]. Copyright © 2010, American Chemical
Society)
while at high temperatures above 700 ı C, most of the tars convert into gas [4].
Corella and his coworkers [44] successfully converted almost all the tar into gas
by increasing temperature to 700 ı C, whereas char remained unconverted, and
it is well-recognized that deposition of char/coke over a catalysts’ surface will
cause loss to the catalyst’s activity. During hydrothermal gasification of cellulose
at 200–350 ı C by Minowa et al. in a batch reactor [45], it was evident that alkali
compounds such as Na2 CO3 could inhibit tar and char formation. The presence
of alkali compounds also reduces the degradation temperature of cellulose, and
promotes hydrogen production.
On the other hand, in the presence of an alkali catalyst, char formation might
be very low in the reactor, but can still cause problems like plugging due to the
formation of coke, promoted at low temperatures and heating rates. When the
biomass water mixture is heated for a long time at subcritical temperatures, furfurals
and other unsaturated compounds are formed significantly that may polymerize
when free radicals are formed above the critical temperature [4]. The main by-
products in SCWG of glucose are fructose, dihydroxyacetone, glyceraldehydes,
erythrose, glycolaldehyde, pyruvaldehyde, 1,6 anhydrogluse, acetic and formic acid,
and 5-hydroxymethylfurfural (5-HMF) [46]. Then, through repolymerization of
these ring compounds, solid particles, referred to as char or coke form. It is also
believed that 5-HMF is a key intermediate that causes repolymerization reactions
resulting in coke formation via a pathway as illustrated in Fig. 12.4 [47], where k is
the rate constant of each step reaction.
12.5 Comparison with Biological Processes

for CH4 Generation
Currently bio-methane is commonly produced via biological processes, i.e., anaer-

obic digestion: organic matter is converted in the absence of air (anaerobic) into
biogas containing about 60 vol.% methane balanced by CO2 . Biomethanation
naturally takes place in environments such as landfills where there is a lack of
light and air. Digested materials from microorganisms could be used further as
a fertilizer – a potential replacement for synthetic fertilizers [48]. High methane
content and cheap production processes has made this process very common for
municipal wastewater treatment and bio-energy generation in rural areas. Despite
the multiple advantages of this method, it suffers from problematic issues such
as odor and the need for a large footprint or land requirement. Technically, there
are many parameters that need to be continuously monitored/controlled during
digestion such as temperature and pH [49], which means any disturbance of these
conditions can change the gas production efficiency. Moreover, limited by the
process temperature and the activity of microorganisms, the substrate in a bio-
reactor requires a long time for digestion. The hydraulic retention time (HRT) is
an expression of this reaction time. Technically, this parameter can be 30–100 days.
The larger the HRT, the bigger the digester should be, which means very high capital
costs [50].
Matsumura [51] conducted a study on the evaluation of supercritical water
gasification in comparison with a biomethanation process. In both cases, methane
production was the main goal. Energy, environmental, and economic aspects were
considered in this evaluation. 1 dry-t/day of hyacinth-water solution was taken as
the model compound. The evaluation was made based on energy efficiency, CO2
payback time, and the price of the gas products. Supercritical water gasification
was found to be more desirable in terms of its small footprint and high processing
efficiency, but the gas production cost was higher than the anaerobic process.
However, by further improvements in the SCWG process such as implementing
better heat exchangers, optimizing process conditions, designing more active and
stable catalysts, and reducing tar/coke formation, the gas production cost can be
significantly decreased, making this a competitive process.
12.6 Conclusions
Supercritical water gasification is a viable option for converting high water content
biomass into valuable gases such as methane and hydrogen. As methane forma-
tion is thermodynamically favorable at lower temperatures, converting aqueous
biomass/waste to methane is more economically viable than hydrogen.
A high amount of methane gas is expected at low SCW temperatures based on
thermodynamic calculations, but the gasification efficiency is generally low without
catalyst due to the kinetic barrier at low temperatures. Utilizing a proper catalyst
is essential to enhance the methanation reaction. Nickel and ruthenium supported
on alumina and carbon are highly active catalysts for the methanation reaction. On
the other hand, ruthenium can also be used as a promoter to enhance the catalytic
activity of nickel catalysts. A high amount of nickel loading in a supported catalyst
generally favors methane formation, although there is an optimal nickel loading for
methane production in the SCWG of biomass.
Comparing with anaerobic digestion, SCWG of biomass/waste can be a promis-
ing alternative technology for renewable CH4 production as the process has a
smaller footprint and higher efficiency. However, there are many challenges for
industrial realization of this new technology, e.g., the gas production cost is much
higher than the anaerobic process. The SCWG process can be further improved by
various measures such as enhancing heat exchange efficiency to decrease the energy
costs, and optimizing the process conditions (e.g., via designing more active and
stable catalysts) to improve gasification efficiency and reduce tar/coke formation at
lower temperatures.
Acknowledgements The authors are grateful for the financial support from NSERC/FPInnovations
Industrial Research Chair Program in Forest Biorefinery and the Ontario Research Fund-Research
Excellence (ORF-RE) from Ministry of Economic Development and Innovation. Support from the
industrial partners including FPInnovations, Arclin Canada, BioIndustrial Innovation Centre is also
acknowledged. The authors are also in debt to Prof. Yasuo Ohtsuka at Tohoku University for his
assistance in the calculation of equilibrium gas composition of biomass using model compounds
under subcritical and supercritical water conditions.
References
1. Green DW, Perry RH. Perry’s chemical engineers’ handbook. 8th ed. New York: McGraw-Hill;
2008.
2. The Realities of Climate Change. Presidential rhetoric by President Barack Obama. 2013.
http://www.presidentialrhetoric.com. Accessed June 2013.
3. The applications and benefits of SNG. 2007. http://www.biosng.com. Accessed June 2013.
4. Matsumura Y, Minowa T, Potic B, Kersten RAS, Prins W, Van Swaaij PMW, Ven de Beld B,
Elliotte DC, Neuenschwander GG, Kruse A, Antal Jr MJ. Biomass gasification in near- and
super-critical water status and prospects. Biomass Bioenergy. 2005;29:269–92.
5. Mozaffarian M, Deurwaarder EP, Kersten SRA. Green gas (SNG) production by supercritical
gasification of biomass. ECN-C–04-081. 2004. http://www.ecn.nl/docs/library/report/2004/
c04081.pdf. Accessed June 2013.
6. Zhang L, Champagne P, Xu C. Screening of supported transition metal catalysts for hydrogen
production from glucose via catalytic supercritical water gasification. Int J Hydrog Energy.
2011;36:9591–601.
7. Elliott DC, Sealock JLJ. Low temperature gasification of biomass under pressure. In: Funda-
mentals of thermochemical biomass conversion. Dordrecht: Springer; 1985. p. 937–50.
8. Osada M, Sato T, Watanabe M, Adschiri T, Arai K. Low-temperature catalytic gasifica-
tion of lignin and cellulose with a ruthenium catalyst in supercritical water. Energy Fuel.
2004;18:327–33.
9. Voll FAP, Rossi CCRS, Silva C, Guirardello R, Souza ROMA, Cabral VF, Cardozo-Filho L.
Thermodynamic analysis of supercritical water gasification of methanol, ethanol, glycerol,
glucose and cellulose. Int J Hydrog Energy. 2009;34(24):9737–44.
10. Antal MJ, Manarungson S, Mok WS. Hydrogen production by steam reforming glucose
in supercritical water. In: Bridgewater AV, editor. Advances in thermochemical biomass
conversion. Dordrecht: Springer; 1993. p. 1367–77.
water. Energy Fuel. 1993;7:574–7.
12. Lee I, Kim M-S, Ihm S-K. Gasification of glucose in supercritical water. Ind Eng Chem Res.
2002;41(5):1182–8.
13. Lee I, Jae-Sung GL, Kim M-S. Hydrogen production by the gasification of biomass in
supercritical water. Paper presented at 5th Korea-Japan joint symposium ‘99 on hydrogen
energy, Yusong, Taejon, Korea; 1999.
II. Effect of catalyst. Int J Hydrog Energy. 2008;33(17):4520–6.
15. Guo Y, Wang SZ, Xu DH, Gong YM, Ma HH, Tang XY. Review of catalytic supercritical
water gasification for hydrogen production from biomass. Renew Sustain Energy Rev.
2010;14(1):334–43.
16. Chowdhury MBI, Hossain MM, Charpentier P. Effect of supercritical water gasification
treatment on Ni/La2 O3 -Al2 O3 -based catalysts. Appl Catal A Gen. 2011;405:84–92.
17. Park KC, Tomiyasu H. Gasification reaction of organic compounds catalyzed by RuO2 in
supercritical water. Chem Commun. 2003;6:694–5.
18. Elliott DC, Neuenschwander GG, Hart TR, Butner RS, Zacher AH, Engelhard MH, Young
JS, McCready DE. Chemical processing in high-pressure aqueous environments: 7 Process
development for catalytic gasification of wet biomass feedstocks. Ind Eng Chem Res.
2004;43(9):1999–2004.
19. Cortright RD, Davda RR, Dumesic JA. Hydrogen from catalytic reforming of biomass-derived
hydrocarbons in liquid water. Nature. 2002;418(6901):964–7.
2008;2(3):254–65.
21. Zhang L, Champagne P, Xu C. Supercritical water gasification of an aqueous by-product from
biomass hydrothermal liquefaction with novel Ru modified Ni catalysts. Bioresour Technol.
2011;102(17):8279–87.
22. Kee RJ, Miller JA, Coltrin ME, Grcar JF, Meeks E, Moffat HK, Lutz AE, Dixon G, Lewis MDS,
Warnatz J, Evans GH, Larson RS, Mitchell RE, Petzold LR, Reynolds WC, Caracotsios WES,
Glarborg MP, Wang C, Adigun O. CHEMKIN collection, Release 3.6, in a software package
for the analysis of gas-phase chemical and plasma kinetics. San Diego: Reaction Design Inc.;
2001.
23. Osada M, Sato T, Watanabe M, Shirai M, Arai K. Catalytic gasification of wood biomass in
subcritical and supercritical water. Combust Sci Technol. 2006;178(1–3):537–52.
24. Calzavara Y, Joussot-Dubien C, Boissonnet G, Sarrade S. Evaluation of biomass gasi-
fication in supercritical water process for hydrogen production. Energy Convers Manag.
2005;46(4):615–31.
25. Elliott DC, Sealock Jr LJ. Chemical processing in high-pressure aqueous environments: low-
temperature catalytic gasification. Chem Eng Res Des. 1996;74(5):563–6.
26. Sharma A, Nakagawa H, Miura K. A novel nickel/carbon catalyst for CH4 and H2 production
from organic compounds dissolved in wastewater by catalytic hydrothermal gasification. Fuel.
2006;85(2):179–84.
27. Elliott DC, Neuenschwander GG, Baker EG, Butner RS, Sealock LJ. Bench-scale reactor
tests of low-temperature, catalytic gasification of wet industrial wastes. Paper presented at
the Energy Conversion Engineering Conference, 1990. IECEC-90. Proceedings of the 25th
Intersociety. Reno, Nevada; 1990.
28. Elliott DC, Butner RS, Sealock Jr LJ. Low-temperature gasification of high-moisture biomass.
In: Bridgwater AV, Kuester JL, editors. Research in thermochemical biomass conversion.
Dordrecht: Springer; 1988. p. 696–710.
29. Youssef EA, Nakhla G, Charpentier P. Oleic acid gasification over supported metal catalysts
in supercritical water: hydrogen production and product distribution. Int J Hydrog Energy.
2011;36(8):4830–42.
30. Youssef EA, Chowdhury MBI, Nakhla G, Charpentier P. Effect of nickel loading on hydrogen
production and chemical oxygen demand (COD) destruction from glucose oxidation and
gasification in supercritical water. Int J Hydrog Energy. 2010;35(10):5034–42.
31. Behnia I. Treatment of aqueous biomass and waste via supercritical water gasification for the
production of CH4 and H2 . M.E.Sc. thesis. The University of Western Ontario; 2013.
32. Elliott DC, Sealock Jr LJ, Baker EG. Chemical processing in high-pressure aqueous environ-
ments. 2 Development of catalysts for gasification. Ind Eng Chem Res. 1993;32(8):1542–8.
33. Lee IG, Ihm SK. Catalytic gasification of glucose over Ni/activated charcoal in supercritical
Appl Catal B Environ. 2012;117–118:330–8.
35. Xu X, Matsumura Y, Stenberg J, Antal Jr MJ. Carbon-catalyzed gasification of organic
feedstocks in supercritical water. Ind Eng Chem Res. 1996;35(8):2522–30.
36. Elliott DC, Sealock LJ, Phelps MR, Neuenschwander GG, Hart TR. Development of a catalytic
system for gasification of wet biomass. Conference: biomass conference of the Americas:
energy, environment, agriculture, and industry; Burlington, VT; 1993.
37. Kersten SR, Potic B, Prins W, Van Swaaij WP. Gasification of model compounds and wood in
hot compressed water. Ind Eng Chem Res. 2006;45(12):4169–77.
38. Elliott DC, Sealock JL. US patent no 5,814,112, US Patent and Trademark Office; 1998.
39. Sato T, Furusaw T, Ishiyama Y, Sugito H, Miura Y, Sato M, Itoh N. Effect of water density
on the gasification of lignin with magnesium oxide supported nickel catalysts in supercritical
40. Elliott DC, Sealock LJ Jr, Baker EG. US patent no 5,616,154, US Patent and Trademark Office;
1997.
41. Byrd AJ, Pant KK, Gupta RB. Hydrogen production from ethanol by reforming in supercritical
water using Ru/Al2 O3 catalyst. Energy Fuel. 2007;21(6):3541–7.
42. Kruse A, Meier D, Rimbrech P, Schacht M. Gasification of pyrocatechol in supercritical water
in the presence of potassium hydroxide. Ind Eng Chem Res. 2000;39(12):4842–8.
43. Mozaffarian M, Deurwaarder EP, Kersten SRA. “Green Gas” (SNG) production by supercrit-
ical gasification of biomass. ECN-C-04-081. 2004. http://www.biosng.com/fileadmin/biosng/
user/documents/reports/c04081.pdf
44. Delgado J, Aznar MP, Corella J. Biomass gasification with steam in fluidized bed: effec-
tiveness of CaO, MgO, and CaO-MgO for hot raw gas cleaning. Ind Eng Chem Res.
1997;36(5):1535–43.
nickel catalyst. J Supercrit Fluid. 1998;3(1):253–9.
46. Kabyemela BM, Adschiri T, Malaluan RM, Arai K. Glucose and fructose decomposition in
subcritical and supercritical water: detailed reaction pathway, mechanisms, and kinetics. Ind
Eng Chem Res. 1999;38(8):2888–95.
process: a study of model compounds. Ind Eng Chem Res. 2010;49(9):4055–62.
48. Angelidaki I, Karakashev D, Batstone DJ, Plugge CM, Stams AJ. Biomethanation and its
potential. Method Enzymol. 2011;494:327–51.
49. Chen Y, Cheng JJ, Creamer KS. Inhibition of anaerobic digestion process: a review. Bioresour
Technol. 2008;99(10):4044–64.
50. Sreekrishnan TR, Kohli S, Rana V. Enhancement of biogas production from solid substrates
using different techniques-a review. Bioresour Technol. 2004;95(1):1–10.
biomass utilization in Japan. Energy Convers Manag. 2002;43(9):1301–10.
Chapter 13
Catalysis in Supercritical Water Gasification
of Biomass: Status and Prospects
Youjun Lu, Sha Li, and Liejin Guo
Abstract High pressure and temperature demand containment structure of costly

alloys for the supercritical water gasification (SCWG) apparatus. The catalyst is of
great significance for realizing high gasification efficiency of biomass under milder
circumstances in supercritical water (SCW). This chapter gives a comprehensive
review on the original work in the area of catalysis in SCWG of biomass. The
development status and role of different catalysts including homogeneous and
heterogeneous catalysts used in SCWG were discussed. The alkali catalyst is a
homogeneous catalyst, which has a key role in the SCWG of model compounds,
biomass, organic wastes and coal. As a heterogeneous catalyst, Ni has been widely
used in sub or supercritical water gasification due to its low cost and high activity.
The crystallite sintering, supports’ breakdown and carbon deposition under SCW
conditions will cause deactivation of catalysts. Therefore, the work of the Ni
catalysts modification mentioned in this chapter is necessary and of great value.
Keywords Catalyst • Supercritical water gasification • Biomass • Ni
13.1 Introduction
Biomass is a renewable energy resource, but due to its low energy density, direct
use of biomass is not convenient. Thus it is necessary to convert biomass to gas
such as hydrogen or liquid fuel, in which can be used cleanly and high-efficiently
in fuel cells [1]. Hydrogen production from renewable biomass is a sustainable way
because the utilization of biomass is carbon neutral [2].
Y. Lu () • S. Li • L. Guo
State Key Laboratory of Multiphase Flow in Power Engineering (SKLMFPE), Xi’an Jiaotong
University, Xi’an 710049, Shaanxi, China
344 Y. Lu et al.
One of the methods for producing hydrogen from biomass is supercritical water
gasification (SCWG) that can avoid high drying costs in conventional thermo-
chemical gasification process, especially for wet biomass [3]. The reaction tem-
perature required for non-catalytic complete gasification of biomass in supercritical
water (SCW) is about 873 K that is much lower than the reaction temperature of the
conventional thermo-chemical gasification for hydrogen production [4]. However,
the operation pressure of SCWG is generally higher than the critical pressure of
water (22.1 MPa). High pressure and temperature demand containment structure
of costly alloys for the gasification apparatus [5]. Therefore, reducing the reaction
temperature is crucial for reducing the hydrogen production cost of SCWG at
high operation pressure. The catalysts enable SCWG of biomass at mild reaction
temperature. Osada et al. identified three temperature regions for hydrothermal
gasification as follows [6]: (1) Region I (773–973 K SCW) the biomass decomposes
to gas without presence of catalyst, and activated carbon catalyst is used to
avoid char formation or alkali catalyst facilitates the water-gas shift reaction. (2)
Region II (647–773 K, SCW) biomass hydrolyzes and metal catalysts facilitate
gasification. (3) Region III (below 647 K, subcritical water) biomass hydrolysis
is slow and catalysts are required for gas formation. Therefore, the catalysts are
of great significance for realizing high gasification efficiency of biomass under
milder circumstances in SCW [7]. In Chaps. 11 and 12, catalytic SCWG is briefly
introduced. This chapter gives a comprehensive review on catalysis in SCWG of
biomass.
13.2 Homogeneous Catalysts
13.2.1 Alkaline Catalysts
In the homogenous catalysts, the alkali catalysts such as NaOH, KOH, Na2 CO3
K2 CO3 etc. are important in the process of SCWG, have been widely used. Catalytic
gasification of model compound, biomass, organic wastes and coal in SCW were
widely investigated by some researchers [8–26] due to being easy to transport and
enhance the gasification in the system of continuous operation.
In model compound gasification, Kruse et al. [8] also studied the gasification
of pyrocatechol in the presence of KOH, and found that the smallest yield of CO
can be obtained and the production of H2 and CO2 increases with increasing KOH
content from 0 to 5 wt%. Signa et al. [9, 10] studied the effect of K2 CO3 on glucose
gasification in SCW with batch reactor and they found that the addition of K2 CO3
enhanced the gasification of glucose. Gadhe and Gupta [11] studied the effect of
KOH and K2 CO3 on methanol gasification in SCW. They found that the addition
of catalysts enhanced the formation of H2 and the maximum molar fraction of H2
reached 267 mol/s when the concentration of KOH was 0.68 wt%. Schmieder et al.
[12] studied the effect of KOH and K2 CO3 on the glucose and vanillin gasification
13 Catalysis in Supercritical Water Gasification of Biomass: Status and Prospects 345
in SCW. They found that the addition of alkali catalyst enhanced the gasification
and a complete gasification of glucose could be achieved when the temperature
was higher than 823 K. In SKLMFPE, Hao et al. [13] investigated the effect of
KOH and Na2 CO3 on glucose gasification in SCW with continuous reactor and they
found that the addition of some alkali into the feedstock solution could decrease
the CO yield and increase the CO2 yield. But it wasn’t found apparent effect on
the gasification efficiency. The addition of Na2 CO3 could accelerate the gasification
reaction and result in a gasification efficiency reach to 100 % in less resident time.
Guo et al. [14] investigated the hydrogen production from glycerol by SCWG in a
continuous flow tubular reactor. It is found that the addition of KOH, NaOH, K2 CO3
significantly increased the H2 and CO2 yields and decreased CO yield, and the
gasification, carbon gasification and hydrogen gasification efficiency with increase
of catalyst amount. The maximum of hydrogen yield was 4.93 mol/mol and the
hydrogen gasification efficiency was up to 128 % when the amount of NaOH was
0.1 wt%, Guo et al. [15] studied the influence of NaOH on gas yields for SCWG
of HAc (0.5 wt%) and phenol (1.0 wt%) mixture. They found that the changing
tendency of total gas yield nearly kept, and both the yields of H2 and CO2 were the
highest at 0.2 wt% of NaOH addition.
In real biomass gasification, Yanik et al. [16] investigated the effect of K2 CO3
catalyst on the gasification of corncob gasification and sunflower stalk in SCW.
They found that the yield of H2 increased significantly in the presence of catalyst
and the K2 CO3 showed the best catalytic activity. Kruse et al. [17] studied the phyto
mass and the zoo mass gasification in the presence of K2 CO3 and they found that
the addition of K2 CO3 increased the gasification efficiency. Schmieder et al. [12]
investigated the effect of K2 CO3 on straw and wood gasification in SCW and found
that the addition of K2 CO3 enhanced the gasification. Watanabe et al. [18] studied
the catalytic effect of NaOH for partial oxidation of n-hexadecane and organosolv-
lignin by use of a batch type reactor in SCW. The n-hexadecane experiments were
carried out at 673 K, 40 MPa and they found that the addition of catalyst did not
increase the conversion of n-C16 and promoted the formation of H2 . The lignin
experiments were performed at 673 K and 30 MPa and they found that the H2
yield with NaOH was almost 4 times higher than that without catalyst (with and
without O2 ).
In organic wastes gasification, the effect of KOH catalyst on the gasification of
oleaginous wastewater and alcohol distillery wastewater in SCW were investigated
by Jarana et al. [19]. They found that the addition of KOH catalyst decreased the
COD and enhanced the formation of H2 and CH4 at the same reaction condition.
Nakhla et al. [20] studied the effect of NaOH catalyst on the gasification of hog
manure. They found that the NaOH catalyst could enhance the reduction of COD
better than other catalysts, and the COD removal efficiency achieved 81 %. Yan
et al. [21] studied the effects of KOH catalyst on SCWG of polyvinyl alcohol-
contaminated wastewater in continuous reactor. They found that the molar fraction
of CO decreased to zero with the concentration of KOH increases from 0–100 mg/L
at 813 K, 25 MPa and 40 s. The molar fraction of CH4 increased from 26.57 to
31.9 %, however, the molar fraction of H2 was increased (40.63–44.31 %) when
346 Y. Lu et al.
the KOH concentration increased. They suggested that the methanation reaction
may be dominative at 813 K for 40 s and the H2 formed water-gas shift reaction
was converted to CH4 which results in the increase of CH4 fraction. Yanik et al.
[16] investigated the effect of K2 CO3 catalyst on the gasification of leather waste
in SCW. They found that the yield of H2 increased significantly in the presence
of catalyst. The K2 CO3 catalyst was the best catalyst on the gasification of leather
waste. Schmieder et al. [12] studied the effect of K2 CO3 catalyst on the gasification
of sewage sludge at 723 K in batch reactor and found that the TOC destruction
efficiency could reach 85.3 % whit the addition of K2 CO3 catalyst. Zhai et al. [22]
investigated the digested sewage sludge gasification in SCW in the presence of
K2 CO3 . They found that the carbon gasification efficiency was almost four times
higher than the efficiencies without catalyst. In SKLMFPE, Chen et al. [23] studied
the effect of NaOH, KOH, Na2 CO3 and K2 CO3 on the gasification of sewage sludge
in fluidized bed reactor. They found that the addition of alkali catalysts enhanced the
gasification of sewage sludge and the formation of H2 in SCW with fluidized bed
reactor. The maximum yield of H2 reached to 15.49 mol/kg in the presence of KOH.
The maximum value of GE and CE reached to 53.22 % and 46.39 % respectively
in the presence of K2 CO3 . The effect of KOH on the gasification of black liquor
in batch reactor was also studied in SKLMFPE. The yield of H2 increased in the
presence of KOH. The maximum yield of H2 reached to 11.15 mol/kg and the COD
destruction efficiency reached to 95 %.
In coal gasification, Xu et al. [24] investigated the effect of KOH on the product
distributions for the treatment of the peat in SCW for water/peat D 10/1 at 683 K
for 60 min. They found that the use of the K2 CO3 increased the yields of water
soluble oil and gas product, while it slightly suppressed the formation of heavy
oil. In SKMFLPE, Li et al. [25] investigated the hydrogen production from coal
gasification in SCW in a continuous flowing system. They found that the addition
of K2 CO3 enhanced the gasification of coal and the maximum molar fraction of
hydrogen reached 68.9 %. Jin et al. [26] studied the hydrogen production from coal
gasification in SCW in a fluidized bed reactor, and found that a hydrogen yield
of 32.26 mol/kg was obtained and the hydrogen fraction reached 69.78 % in the
presence of K2 CO3 .
Generally, the alkali catalysts have a key role in the SCWG of model compounds,
biomass, organic wastes and coal. They enhance the gasification efficiency and
increase the hydrogen yield effectively. Because of the dissolution of alkali catalysts
in water, the wide application of alkali catalysts will provide great benefits in SCW.
13.2.2 Metal Catalysts
The homogeneous metal catalysts can be used in hydrothermal of biomass in

addition to alkaline catalysts, but there are not many examples reported in the
literature. Azadi et al. used three organometallic salts as potential homogeneous cat-
alysts for hydrothermal gasification of glucose solutions (between 0.06 to 0.65 M)
[27]. The hydrothermal gasification of glucose in the presence of transition metal

chelates consisting of nickel(II) acetylacetonate (Ni(acac)2 ), cobalt(II) acetylacet-
onate (Co(acac)2), iron(III) acetylacetonate (Fe(acac)3 ) and Raney-nickel particles
was carried out. They found that the presence of the organometallic compounds
can slightly facilitate the rate of decomposition of glucose. This enhancement in
reaction rate was more pronounced at higher concentrations of the feed (0.65 M)
compared with lower concentrations (0.06 M). Raney-nickel was a more effective
catalyst for glucose gasification compared to homogeneous Ni(acac)2 , Co(acac)2 ,
and Fe(acac)3 catalysts at a same catalyst weight.
13.3 Heterogeneous Catalysts
13.3.1 Activated Carbon
Activated carbons produced from different sources are found to be catalytically

active for high temperature SCWG reactions. Lee et al. carried out the gasifi-
cation of glucose over the 16 wt% Ni/activated charcoal in SCW at 28 MPa.
Effects of temperature (848–1,048 K), feed concentration (0.3–0.9 M), and LHSV
(6–24 h1 ) were investigated on the product distribution. For comparison, the
gasification experiments were performed with activated charcoal. However, the
activated charcoal itself led to significantly low catalytic activity for hydrogen
production compared with the Ni/AC catalyst [28]. Xu et al. used carbon-based
catalysts, such as coal activated carbon, coconut shell activated carbon, macadamia
shell charcoal and spruce wood charcoal, for organic compounds gasification
in SCW. The feedstocks included glycerol, glucose, cellobiose, whole biomass
feedstocks and sewage sludge [29, 30]. Complete conversions of these feedstocks
were achieved at 873 K and 34.5 MPa. Antal et al. gasified completely the high
concentration biomass feedstocks above 923 K with carbon-based catalyst in SCW.
The gas formed was mainly composed of hydrogen and carbon dioxide, and the
maximum hydrogen yield could be more than 100 g per kg dry biomass [31].
13.3.2 Metal Oxides
Although metal-oxide is not usually used as a catalyst for biomass gasification, few
oxides such as CaO [32], ZrO2 [33], CeO2 [34] and RuO2 [35, 36] have been also
employed for catalyzing the SCWG. Zhang et al. [32] gasified coal in SCW with
CaO. In the presence of calcium oxide, gas product is abundant in hydrogen with
no CO2 emission. It was found that CaO could increases the hydrocarbon reforming
and the water- gas shift reaction. Watanabe et al. found that the hydrogen yield
of glucose and cellulose gasification in SCW with zirconia was almost twice as
348 Y. Lu et al.
much as that without catalyst [33]. Hao et al. carried out the cellulose gasification
in SCW with different noble metal and metal-oxides, and showed that the metal-
oxide including CeO2 , nano-CeO2 and nano-(ZrCe)xO2 shows a little activity for
cellulose gasification in SCW [34]. Park et al. achieved nearly complete gasification
of aromatic compounds in SCW with RuO2 [35]. Yamamura et al. studied SCWG
of model biomass samples (glucose, cellulose, and heterocyclic compounds), and
low-purity biomass samples obtained from a paper-recycling facility (paper sludge)
and from a sewage treatment plant (sewage sludge) in the presence of RuO2
[36]. In clear contrast to another catalysts, the RuO2 catalyst led to completely
gasification of cellulose to produce mainly hydrogen, methane, and carbon dioxide
under various conditions (e.g., 673 K at 30 MPa and 773 K at 50 MPa). The
catalytic gasification of natural biomasses (lignocellulosic and proteinous materials)
in SCW was investigated by Yanik et al. [16]. Trona (a natural mineral) and red mud
were used as catalysts besides K2 CO3 and Raney-Ni, which are commonly used
catalysts in SCWG. Red mud contains mainly Fe2 O3 (37.7 %), Al2 O3 (17.3 %),
SiO2 (17.1 %), TiO2 (4.8 %), Na2 O (7.1 %) and CaO (4.5 %). Although, the yield
of hydrogen in the presence of red mud was lower than that obtained with alkali
catalysts for all kinds of biomass tested. In addition, iron based catalysts also showed
catalytic activity for the production of hydrogen from biomass [16].
13.3.3 Noble Metal Catalysts
The noble metal catalysts including Pt, Pd, Rh, Ru have been widely used in
SCWG because of their high activity (Table 13.1). Onwudili and Williams used
Ruthenium supported on alpha-alumina spheres as a catalyst for the gasification of
glucose and other biomass-related samples in SCW at 823 K, 36 MPa. In general,
carbon gasification efficiencies (CGE) over 96 % were achieved in the presence of
Ru/Al2 O3 , while hydrogen gasification efficiencies (HGE) based on result-derived
reaction stoichiometries reached 87 % for glucose [37]. Chakinala et al. gasified
the water soluble fraction of bio-oil with heterogeneous metal catalysts Pt, Pd, Ru,
Rh, and Ni supported on alumina. The GE for the metals decreases in the order
Ru > Pt > Rh > Pd > Ni. For optimum H2 selectivity the order of the catalysts is
Pd > Ru > Rh > Pt > Ni. Ru was found to be the most active catalyst in terms of
gasification efficiency as well as alkane selectivity [38]. Stucki et al. conducted a
complete gasification of microalgae (Spirulina platensis) to a methane-rich gas in
SCW using ruthenium catalysts [39]. Guan et al. determined the effects of different
process variables on the gasification of Nannochloropsis sp., a marine microalga,
in water with a Ru/C catalyst at 683 K. About 45 % gasification efficiency was
achieved at 75 min with a catalyst loading of 1 g/g(mass of Ru/C catalyst/mass
of dry algal biomass), a water density of 0.096 g/cm3 , and a 4.3 wt% loading
of algae [40]. Youssef et al. used Oleic acid as a model compound for lipids,
which was gasified in SCW using a batch reactor from 673 to 773 K at 28 MPa.
The influences of operating temperature and several commercial catalysts on the
Table 13.1 Selected results of Noble metal catalysts used for SCWG in literatures
T/P(or water
density) Reactor Materials Catalyst Support Results References
823 K, 36 MPa Batch Glucose, cellulose, Ru ’Al2 O3 Without alkaline additive, Ru/Al2 O3 almost [37]
xylan, sawdust, completely converted glucose into a gas
HCOONa, product comprising of carbon dioxide and
CH3COONa relatively high yields of hydrogen and methane.
The selectivity of Ru/Al2 O3 catalyst towards
hydrogen over methane was enhanced in the
presence of NaOH and Ca(OH)2
723–853 K, Batch Bio-oil Pt, Pd, Rh, ”Al2 O3, TiO2 , The GE for the metals is decreasing in the order [38]
28–30 MPa Ru, Ni ZrO2 , SiO2 , Ru > Pt > Rh Pd > Ni. For optimum H2
Ce–ZrO2 , AC selectivity the order of the catalysts is
Pd > Ru Rh > Pt > Ni. Ru was found to be
the most active catalyst in terms of gasification
efficiency as well as alkane selectivity. Stable
support materials identified for the
hydrothermal gasification of biomass include
ZrO2 , Ce–ZrO2 and TiO2
672–682 K, Batch S. platensis Ru AC, ZrO2 Complete gasification of microalgae (Spirulina [39]
30.8– platensis) to a methane-rich gas is now possible
34.5 MPa in SCW using ruthenium catalysts. 60–70 % of
the heating value contained in the algal
biomass would be recovered as methane
683 K, Batch Microalgae paste Ru AC Complete gasification of the microalga was [40]
0.096 g. cm3 achieved with a catalyst loading of 2 g/g. The
presence in algae of sulfur and perhaps other
13 Catalysis in Supercritical Water Gasification of Biomass: Status and Prospects
elements, such as Cl, that are not as prevalent

in terrestrial biomass indicates that efficient
and effective gasification of microalgae could
present new challenges in engineering and
catalyst design
349
(continued)
350
Table 13.1 (continued)

T/P(or water
473–773 K, Batch Oleic acid Ru, Pd, Pt, Al2 O3 , ™Al2 O3 , The H2 yield was 15 mol/mol oleic acid [41]
28 MPa Ni AC converted using both the pelletized
Ru/Al2 O3 and powder Ni/Silica- alumina
catalysts which gave four times higher than
the equilibrium yield
588 K, 0.07– Batch Microgranular Pt Al2 O3 In the presence of Pt/Al2 O3 , the effect of [42]
0.51 g. cm3 cellulose headspace fraction became more
pronounced, with gas increasing by
approximately a factor of forty from the
lowest to highest headspace fraction
783–923 K, Tubular Glycerol Ru ZrO2 Complete glycerol conversion was achieved at [43]
35 MPa fixed-bed a residence time of 8.5 s at 783 K, and at
reactor around 5 s at 823 K with the
1 wt%Ru/ZrO2 catalyst
373–673 K, A continuous Glucose Pt Al2 O3 Quantitative analysis of the total organic [44]
10 MPa flow carbon in the liquid residue indicated 67 %
reactor carbon gasification efficiency at 603 K
673 K, Batch Lignin Pt, Pd, Rh, AC The supported metal salt catalysts showed [45]
0.5 g. cm3 Ru, Ni, activities for the lignin gasification,
metal salt especially the catalysts without chloride
anion showed higher activities for the
gasification. The order of activity for the
gasification among the metal species was
following: ruthenium > rhodium >
platinum > palladium nickel
673 K, 0– Batch Sugarcane, Ru TiO2 , AC Sugarcane bagasse was completely gasified to [46]
0.5 g. cm3 bagasse, methane, carbon dioxide, and hydrogen
cellulose, lignin over Ru/C and Ru/TiO2 catalysts at 673 K
Y. Lu et al.
473–673 K, Batch Bean cured refuse Ru, Rh, Pt, AC The activity of catalyst was in the order of [47]
0–0.5 g. cm3 Pd, AC Ru/C> > Rh/C > Pt/C > Pd/C
673 K, Batch Lignin Ru salt, TiO2 , AC The Ru(NO)(NO3 )3 /C and [48]
0.5 g. cm3 Charcoal Ru(NO)(NO3 )3 /TiO2 catalysts showed lignin
gasification activities as high as a charcoal
supported ruthenium catalyst (Ru/C)
673 K, Batch Lignin Ru salt, TiO2 , AC Ruthenium (III) nitrosyl nitrate on charcoal [49]
0.5 g. cm3 Charcoal (Ru(NO)(NO3 )3 /C) was more active than
ruthenium (III) chloride on charcoal
(RuCl3 /C) for the Lignin gasification
reaction. Size of ruthenium metal in
Ru(NO)(NO3 )3 /C was smaller than that in
RuCl3 /C
673 K, Batch Lignin, Ru, S-Ru TiO2 In the absence of sulfur, lignin was completely [50]
0.5 g. cm3 4-propylphenol, gasified to methane, carbon dioxide, and
formaldehyde hydrogen over the titania-supported
ruthenium catalyst in SCW. In the presence of
sulfur, the overall gas yield decreased and
THF-insoluble products (namely, char) were
formed. Also, the content of hydrogen
increased in the resulting gas composition
673 K, Batch Lignin Ru, Pt, Pd TiO2 , Al2 O3 , In the presence of sulfur, the gas yield decreased [51]
0.5 g. cm3 AC with the amount of sulfur added. The carbon
dioxide composition in the presence of sulfur
was larger than that in the absence of sulfur,
and the methane composition was lower
(continued)
351
352

T/P(or water
673 K, Batch Lignin Ru TiO2 , Al2 O3 , The initial activity of the catalysts was in the [52]
0.5 g. cm3 AC order of Ru/TiO2 > Ru/”-A2 O3 > Ru/C.
The Ru/TiO2 catalyst maintained high
gasification activities for three subsequent
uses in SCW. The Ru/C and Ru/”-A2 O3
catalysts showed the high gasification
activity at the initial stage
673 K, 0– Batch Lignin Pt, Pd, Rh, TiO2 , Al2 O3 , The TON values were the following: [53]
0.5 g. cm3 Ru, Ni AC Ru/TiO2 > Ru/Al2 O3 > Ru/C,
Pt/C > Pt/Al2 O3 ,
Rh/C > Pd/C > Pd/Al2 O3 > Ni/Al2 O3 ,
indicating that supported ruthenium
catalysts were very effective for gasification
in water
673 K, Batch Lignin, cellulose, Ru, Ni, TiO2 , Al2 O3 In the presence of the ruthenium catalyst, a [54]
0.33 g. cm3 formaldehyde NaOH high CH4 yield was obtained and no solid
product was formed for lignin and cellulose
gasification, and formaldehyde was rapidly
decomposed to gases such as CH4 , CO2 ,
and H2
Y. Lu et al.
gasification efficiency, hydrogen yield, and residual liquid product quality were
examined and discussed [41]. Dolan et al. studied on the effect of headspace fraction
and alkalinity on the hydrothermal gasification of cellulose has been studied at
588 K in the presence of Pt/Al2 O3 [42]. May et al. converted glycerol in SCW at
783–823 K and a pressure of 350 bars using both a bed of inert and non-porous ZrO2
particles and a bed of a 1 % Ru/ZrO2 catalyst, and complete glycerol conversion
was achieved [43]. Hashaikeh et al. studied on the degradation and gasification of
cellulose-based biomass in compressed water in the 373–673 K temperature range.
Catalytic effects of Pt/Al2 O3 on the gasification temperature were determined [44].
Shirai’s group did a lot of experiment to study on different feedstock gasification in
SCW with Ru, Rh, Pt, Pd noble metal catalysts [45–54].
In all noble metal catalysts, Ruthenium is found to be very active for reactions
involved in SCWG. The most frequently used supported ruthenium catalysts for
SCWG are Ru/C, Ru/TiO2 and Ru/Al2 O3 . However, presence of sulfur containing
compounds even at very low concentrations dramatically deactivates the catalyst by
successive adsorption and/or solid state reaction on the metal surface [55].
13.3.4 Ni Catalysts
Due to the comparable activity to that of noble metal catalysts (e.g. Ru, Rh) but
relatively low-cost, nickel-based catalysts were extensively studied in the sub and
SCWG. Modell and his coworkers studied the gasification of cellulose over a
nickel catalyst with a batch type reactor at 647 K and 22 MPa, and the gaseous
products mainly consisted of CO, CO2 and H2 [56]. Elliott’s group evaluated several
different forms of nickel catalysts in a batch reactor (623 K, 17–23 MPa) filled
with a mixture of p-cresol and water, and CH4 -rich gaseous products were obtained.
Although ’-alumina and zirconia revealed high hydrothermal stability, ’-alumina
and zirconia supported nickel catalysts showed low catalytic activity in this process
[57]. Minowa’s and the coworkers studied the activities of reduced nickel catalysts
supported by alumina, silica-alumina, aluminum silicate, silica, kieselguhr and mag-
nesia on cellulose decomposition in hot-compressed water (473–673 K, 8–22 MPa).
Magnesia supported catalyst showed the highest catalytic activity, and the supports
showed a strong effect on the gas yield but no effect on the hydrogen selectivity
[58]. Yoshida and Matsumura’s group also gasified the mixtures of cellulose and
lignin over a nickel catalyst (Ni-5312P, Engealhard) in SCW (673 K, 25 MPa), and it
indicated that tar product could be suppressed by nickel catalysts, while the catalysts
could be deactivated by tar product [59]. Sato’s et al. conducted the gasification
of lignin in SCW (523–673 K) catalyzed by magnesium oxide supported nickel,
and MgO only facilitated the reaction of lignin decomposition but did not enhance
the gas yield. Increasing nickel loading could enhance the catalytic activity, but
the stability of magnesium supported nickel catalyst needs to be improved [60].
The stability of Ni/MgO for SCWG process was evaluated by Furusawa et al.
[61]. Char-like carbonaceous products and the formation of Mg(OH)2 led to the
354 Y. Lu et al.
deactivation of Ni/MgO. Zhang et al. screened 17 heterogeneous catalysts including

nickel supported on ”Al2 O3 , ZrO2 and activated carbon for hydrogen production
from SCWG (873 K, 24 MPa) of glucose in a bench-scale continuous-flow tubular
reactor. Metallic nickel catalysts supported on ”-Al2 O3 investigated for a time-on-
stream of 5–10 h exhibited very high activity and H2 selectivity. It is also concluded
that Mg and Ru are effective promoters for Ni/”-Al2 O3 catalysts [62]. Azadi and
the coworkers tested Ni/’-Al2 O3 , Ni/hydrotalcite, Raney nickel and Ru/”-Al2 O3
catalysts for hydrogen production from lignocelluloses biomass (glucose, fructose,
xylan, pulp, lignin and bark) in SCW (653 K). Ni/’-Al2 O3 and Ni/hydrotalcite
were found to be active not only for the gasification of carbohydrate, but also for
a higher hydrogen selectivity compared with Raney nickel, Ru/C and Ru/”-Al2 O3
[63]. DiLeo et al. found that nickel wire could greatly increase the gasification rate
of methanol, phenol, guaiacol and glycine in SCW [64–66]. Youssef et al. conducted
the gasification and partial oxidation of glucose in SCW over alumina supported
nickel catalysts [67]. Taylor and coworkers synthesized carbon nanotube/fiber and
aluminosilicate supported Ni catalysts which were confirmed to be active in the
gasification of biomass in SCW [68]. Yan Bo et al. investigated Ni/ZrO2 , Co/ZrO2
and W/ZrO2 catalysts in SCWG (663 K, 24 MPa) of polyethylene glycol, and
Ni/ZrO2 showed the highest catalytic activity [69].
Table 13.2 shows the selected results of Ni catalysts used for SCWG in
literatures. Generally, these nickels can promote the gasification efficiency by a large
amount versus to the non-catalytic tests. Although these screening tests were carried
out in different types of reacting systems under different operating parameters
(e.g. heating rate, temperatures, feedstock and its ratio to catalyst loading, water
density and reaction time,) that also impact gaseous production besides the catalysts,
many promising supports have been found. These studied nickel catalysts are found
to be very active in hydrogen generation reactions, water-gas shift reaction and
methanation reaction, showing the comparable catalytic performance to that of
noble metal catalysts.
13.4 Ni Catalysts: Deactivation and Modification
Although the nickel catalysts have high catalytic activity in SCWG for hydrogen
production, crystallite sintering, supports’ breakdown and carbon deposition under
SCW conditions cause deactivation of catalysts. Supercritical fluids are often used
as media for the synthesis of many kinds of materials. Thus for nickel and the
supporting metals, SCW is also a kind of hydrothermal media that the crystalline
structures of these material can be changed during the gasification reactions. The
growth of nickel crystals is a kind of consequence of exploring them under the
SCW conditions even though these materials have been previously treated at high
temperatures. The growth of nickel obviously could cause the deactivation of the
catalyst by reducing the active sites, and the crystalline change is irreversible.
The supports also undergo the critical hydrothermal conditions and then, phase
Table 13.2 Selected results of Ni catalysts used for SCWG in literatures
T/P(or water
density) Reactor Materials Catalyst Results References
673 K, SUS 316 tube Organosolv-lignin, Commercial In the presence of a nickel catalyst, the H2 [54]
0.33 g. cm3 bomb reactors cellulose, Ni/Al2 O3 yield of lignin and cellulose
formaldehyde gasification was greater than that
without a catalyst, and solid product
was also formed. Formaldehyde was
rapidly decomposed to gases such as
CH4 , CO2 , and H2 , whereas without a
catalyst, formaldehyde was converted
to methanol and CO2
623 K, Batch stirred p-cresol, phenol, Commercial nickel Many commercial catalysts provide [57, 70–72]
20–21 MPa reactor, wastewater catalysts, short-term activity, but these catalysts
fixed-bed Raney nickel lose activity readily in almost all cases
catalytic because of crystallite growth and
tubular reactor resulting loss of active surface area,
although the loss is not significant in a
few cases. The careful monitoring and
control of feedstock trace components
(e.g., calcium, sulfate, and
chloride)are critical for maintaining
long-term catalyst activity
623 K, Autoclave reactor Cellulose Ni 3288 The nickel catalyst could catalyze the [58, 73, 74]
0.234 g. cm3 (Engelhard), Ni steam reforming reaction of aqueous
supported on products as intermediates and the
different methanation reaction. Supports
supports showed the strong effect on the gas

yield, and the supports show no effects
on methanation. Magnesia showed the
highest catalytic activity
(continued)
355
356

T/P(or water
673 K, SS316 Cellulose, xylan, and Ni-5132P Sufficient amount of catalyst achieves [59, 75]
0.1663 g. cm3 stainless-steel lignin mixtures, (Engelhard) high gasification efficiency even for
tube bomb Sawdust, rice the mixtures of cellulose and softwood
reactors straw lignin
523–673 K, Stainless steel tube Lignin Ni/MgO The highest total gas yield in a carbon [60]
0 0.5 g. cm3 bomb reactors basis was 78 % with 20 % Ni/MgO
catalyst at 673 K and 0.3 g. cm3
water density
673 K, Stainless steel tube Lignin Ni/MgO There is an optimal Ni particle size for [61]
0.3 g. cm3 bomb reactor lignin gasification in SCW
673–998 K, Quartz tube batch Cellulose, lignin Nickel wire, Raney SCWG at 773 K generated 16 mmol.g1 [76]
0.05– reactor Ni catalysts of H2 from cellulose with nickel wires
0.18 g. cm3 in the reactor
613–653 K, 316 stainless steel Glucose Raney Ni catalysts Raney-nickel and ruthenium were a more [63, 77]
0.214 g. cm3 tube effective catalyst for glucose
gasification compared to
Raney-copper and Raney-cobalt
catalysts. At the same catalyst weight,
Raney-nickel powder has more active
metal, and therefore has likely more
number of active sites compared to the
supported ruthenium catalysts
773–823 K, Quartz tube batch Methanol Nickel wire In the absence of nickel, conversions up to [64]
0.079 g. cm3 reactors 20 % were reached after 2 h
Y. Lu et al.
873 K, 0.064– Quartz tube batch Phenol, glycine Nickel wire Of the gases that were formed, CO was [65]
0.159 g. cm3 reactors most abundant from the homogeneous
reactions, while hydrogen was the
most abundant in the presence of a
nickel catalyst. The Ni catalyst
assisted in glycine gasification, as less
carbon was found in the aqueous
phase, and gas yields were increased
673–973 K, Quartz tube batch Guaiacol, phenol Nickel wire Nickel does not affect the conversion of [66]
0.079 g. cm3 reactors guaiacol to phenol and o-cresol, but it
significantly changes the gas product
compositions in the presence of Ni
wire
773 K, Batch C276 reactor Glucose Synthesized Ni/™- The hydrogen yield was relatively [67]
0.117 g. cm3 Al2 O3 , insensitive on the pellet catalyst at
commercial nickel loading above 11 % as well as
nickel on silica the different catalyst supports. The
alumina commercial Ni/silica–alumina catalyst
surpassed the synthesized Ni/alumina
in terms of H2 yield by 0.3 mol.mol1
glucose due to the higher active metal
surface area the commercial
Ni/silica–alumina catalyst posses
773 K, Quartz tube batch Cellulose Ni/CF, Ni/CNT, These catalysts were effective for the [68]
0.08 g. cm3 reactors Ni/SiAl, production of hydrogen from cellulose
Ni/SiAl(Sigma), in SCW. The synthesized catalysts
Ni/SiAl produced similar hydrogen yields
(BASF) compared to Sigma and BASF
catalysts commercially available silica–alumina

supported nickel catalysts and
demonstrated a higher (H2 /CH4 )
product ratio
357
(continued)
358
T/P(or water
663 K, 24 MPa Continuous flow Polyethylene glycol Synthesized The catalyst 15%Ni/ZrO2 gave the highest [69]
Inconel 625 Ni/ZrO2 activity and H2 yields, which were 6.1 and
reactor 2.4 times higher than those without catalyst
and with only ZrO2
848–998 K, Continuous flow Glucose Ni/activated The Ni/activated charcoal catalyst showed a [28]
28 MPa C276 reactor charcoal good yield of hydrogen, it was deactivated
due to coke deposition especially at low
temperatures below 923 K and also due to
sintering of nickel particles
473 623 K, Autoclave reactor Cellulose Ni-3288, At 623 K, only 4 % aqueous yield, and no [78]
0.234 g. cm3 Engelhard residue or oil products were achieved with
Ni catalysts
673–873 K, Quartz tube batch Algae NiMo/Al2 O3 , Ni The activity of catalysts with respect to the [79]
24 MPa reactor wire gasification efficiency of algae in SCW at
873 K decreases in the order of
Inconel Ni > Ru > PtPd > CoMo > NiMo,
and the activity of catalysts for high H2
yields is in the order of
Ru > NiMo > Inconel > Ni > PtPd > CoMo
773 K, Inconel 625-lined Sunflower stalk, Raney Ni catalyst Hydrogen yield for sunflower stalk gasification [16]
0.14 g. cm3 tumbling batch corncob increases from 3.65 mol.kg1 to
autoclave 7.99 mol.kg1 with Raney Ni catalyst
addition
773 K, 30 MPa Inconel 625-lined, Glucose Raney Ni catalyst More CH4 is formed in the presence of nickel, [9]
tumbling batch which is a hydrogenation catalyst. Raney
autoclave nickel as well as high heating rate leads to a
reduced yield of unwanted furfurals and
increased yield of the wanted burnable
gases, mainly H2
Y. Lu et al.
Fig. 13.1 A simplified reaction scheme for SCWG of cellulose (Reprinted with permission from
Minowa and Inoue [74]. Copyright © 1999, Elsevier)
changing and aggregation may happen. They cause the deactivation of nickel or
lose the activities when they are reused. It has been found that Al2 O3 except ’-
Al2 O3 suffered phase changing to boehmite (AlO(OH)) under SCW and lost lots of
surface area. Ni/MgO was found quite active for SCWG of biomass for hydrogen
production. Also, also, MgO could not avoid the phase change to Mg(OH)2
under hydrothermal conditions. Phase structures of materials like Rutile TiO2 ,
monoclinic ZrO2 and ’-Al2 O3 were found stable under the SCW. However, some
of these materials still undergo crystalline growth upon the exposure to the SCW
environment.
For the nickel catalyzed SCWG, coking over the catalyst using organic com-
pounds as reaction materials can’t be avoid as well. Minowa’s et al. used cellulose
as the biomass model compound, and proposed a simplified reaction scheme for
SCWG of cellulose after testing all products (gases, oil, char and water-soluble
products) of the hydrothermal reactions under different temperatures. A simplified
reaction scheme for SCWG of cellulose was proposed as Fig. 13.1. They concluded
that cellulose decomposed between 533 and 593 K in which water-soluble products
and gases with minor amount of char-like residue were produced. The water-soluble
products were maximum at 573 K, then they were gasified with the increase of
temperature or with the aid of the catalysts. If the catalytic activity of nickel was
not high enough, the water-soluble intermediates could be polymerized into oil/char
byproducts, and as a consequence, coke deposition over the catalysts accumulated
and deactivation of the catalyst happened [74].
As we just reviewed, several research groups studied nickel catalysts for SCWG
process, while most of them were focused on the screening of proper materials
as the supports or the optimizing of nickel loadings on the catalyst by comparing
the gaseous yields. Few works supplied specific information about the coking of
the used catalysts, and limited analysis about the relationship between the catalyst
properties and the covered carbon over them during the deactivation process was
reported. According to the references [80, 81], the deposited carbon over the
used catalysts could be simply described as two kinds: the layered amorphous
carbon and the crystallized graphite carbon. These two kinds of carbon are often
distinguished and measured by the thermo-gravimetric analyses corresponding to
360 Y. Lu et al.
Ni/γAl2O3
Ni/CeO2-γAl2O3
Deriv.weight (%/K)
T=859K
T=786K
400 500 600 700 800 900 1000

Temperature (K)
Fig. 13.2 Thermogravimetry analysis (TGA) corresponding to the temperature programmed

oxidation of Ni catalysts used in SCWG of glucose (Reprinted with permission from Lu et al.
[82]. Copyright © 2010, Elsevier)
the temperature programmed oxidation [82]. As shown in Fig. 13.2, oxidation peaks
at low temperature can be ascribed to carbon on nickel surfaces [80] whereas the
peaks at temperatures higher than 823 K are ascribed to oxidation of coke deposits
with different degree of graphitization [81].
There are many factors of SCWG that could affect the properties of the deposited
carbon, such as heating rate of the reactors, reaction time, temperatures, and
concentrations of the feedstock, etc. It was found that slow heating rate of the
catalytic reaction system could allow more polymerized production and thus more
amount of depositing carbon over the used catalyst [9]. For a longer reaction, the
amorphous carbon may be crystallized into graphite from.
Properties of the supports, of course, may also affect the anti-carbon performance
of nickel catalyst. For example, Al2 O3 has the acid sites favoring the dispersion of
nickel during the preparation of the catalysts and dehydration during the reaction.
However, these acid sites were proved to be kind of active in adsorbing the
intermediate products and transformation them into carbon as the reaction progress
and thus, sever deactivation of nickel by coking was observed [83, 84]. CeO2 , on the
other hand, could enhance anti-carbon ability of nickel due to its oxidation-reduction
properties when it was used as promoters to supported nickel catalysts.
For the future practical use, the properties of the real feedstock (origin of
biomass, excrement of the animals, wastewaters and industrial organic wastes)
should be fully considered during the study. They could have both the negative and
positive effect on the catalytic performance of nickel catalyst. For example, the real
biomass actually contains a certain amount of sulfur that could cause the poison of
nickel. Chlorine can also deactivate the active site of nickel. The alkali salts in the
original biomass, on the other hand have been proved to be active for its gasification
in SCW. However, their effect on the catalytic performance of nickel has not been
clearly determined yet.
Although the reduced nickel catalysts have high catalytic activity in SCWG for
hydrogen production, as it just discussed, crystallite sintering, coking and supports’
breakdown under SCW conditions can cause deactivation of catalysts. Most of the
previous works were screening of suitable materials for the SCWG. The study of the
modified nickel catalysts tailored for the SCWG seems to be quite limited. Adding
promoters to nickel catalysts was found to be an effective way. These additives
can form bimetallic nickel-based catalysts or compound metal oxides supports,
promoting the properties (nickel dispersion, anti-carbon activity, and phase stability)
of nickel catalysts for SCWG.
Bimetallic Ni-Sn, Ni-Co and Ni-Cu catalysts were studied by many researches.
Sn-promoted Raney-Ni catalysts for aqueous-phase reforming were conducted by
Dumesic’s group [85]. It was found that the addition of tin to nickel decreased the
rate of methane formation from C–O bond cleavage while maintaining the high rate
of methane formation from C–O bond cleavage required for hydrogen formation.
They ascribed the effect of Sn on the selectivity of H2 in the presence of Sn at Ni-
defect sites and the formation of Ni-Sn alloy surfaces like Ni3 Sn. Cu was found to
be a strong inhibitor of coke formation and active in the water-gas shift reaction to
produce hydrogen. Bimetallic Ni-Co also showed higher activity and stability than
nickel catalysts.
In our previous work, Ni-Cu, Ni-Co and Ni-Sn catalysts for SCWG were also
studied [86]. In this work, the bimetallic Ni-M (M D Cu, Co and Sn)catalysts were
prepared by a co-impregnation method. By the XRD test, as shown in Fig. 13.3, the
additives of Cu, Co and Sn could enhance the dispersion of nickel over the alumina
support. The catalytic SCWG results (Fig. 13.4) showed that Cu could improve
the catalytic activity of Ni catalyst in reforming reaction of methane to produce
more hydrogen. Co was found to be an excellent promoter for nickel in relation
to hydrogen selectivity. But bimetal alloys Ni3 Sn formed in Ni-Sn/”Al2 O3 catalyst
resulted in a reduction of catalytic activity.
CeO2 , La2 O3 , ZrO2 and MgO have usually been used as the promoter of
carbon removal from metallic surfaces. In our previous work, Ni catalysts
with the supports (CeO2 /Al2 O3 , La2 O3 /Al2 O3 , ZrO2 /Al2 O3 , MgO/Al2 O3 )
were prepared by two-step impregnation method. The results showed that
hydrogen yield for different supports decreased in order: CeO2 /Al2 O3 > La2 O3 /
Al2 O3 > MgO/Al2 O3 > Al2 O3 > ZrO2 /Al2 O3 , and hydrogen selectivity decreased
in order: CeO2 /Al2 O3 > La2 O3 /Al2 O3 > ZrO2 /Al2 O3 > Al2 O3 > MgO/Al2 O3. CeO2
was thus the best promoter of carbon removal from catalyst surfaces. The effects
of Ce loading in catalysts on glucose gasification were studied [82, 87]. The
results showed that hydrogen yield and hydrogen selectivity increased sharply with
addition of Ni/CeO2 /Al2 O3 catalysts. When the Ce loading content was 8.46 wt%,
the maximum H2 yield and H2 selectivity were obtained (Fig. 13.5). The carbon
deposition and coking will lead to the deactivation of the catalysts. Based on the
thermo-gravimetric analyses, the oxidant kinetic data of carbon deposited on the
used catalysts with air was obtained.
362 Y. Lu et al.
Ni-Sn/γAl2O3
♦ ♦ ♦ ♦
♦
Ni-Co/γAl2O3
Intensity(A.U.)
Ni-Cu/γAl2O3
Ni/γAl2O3
20 30 40 50 60 70 80
2θ (degree)
Fig. 13.3 X-ray diffraction (XRD) patterns of fresh Ni-Cu, Ni-Co and Ni-Sn catalysts (• ”-Al2 O3 ,
Ni, ♦ Ni3 Sn) (Reprinted with permission from Li et al. [86]. Copyright © 2011, Elsevier)
14
H2 CH4 CO2 H2 selectivity 70
12
60
Hydrogen Selectivity, %
10
Gas Yield, mol/kg
50
8
40
6
30
4 20
2 10
0 0
No Catalyst Ni Ni-Cu Ni-Co Ni-Sn
Fig. 13.4 Gaseous yields and hydrogen selectivity of nickel catalyzed SCWG of glucose.
Materials: 11 g glucose solution (9.09 wt%) C 0.2 g catalyst; Temperature: 673 ˙ 3.0 K; Pressure:
24.5 ˙ 0.5 MPa; Reaction time: 20 min (Reprinted with permission from Li et al. [86]. Copyright
© 2011, Elsevier)
a 16
H2
CH4
12 CO2
Gas Yield (mol/kg)
0
No catalyst 1.22 3.66 6.07 8.46 10.83
x (%)
b 60
50
H2 selectivity (%)
40
30
20
10
0
No Catalyst 1.22 3.66 6.07 8.46 10.83
x (%)
Fig. 13.5 Effects of Ce loading on gas yield (a), H2 selectivity (b) of glucose gasification in
SCW: X D content of Ce in catalyst (wt%). Materials: 11 g glucose solution (9.09 wt%) C 0.2 g
catalyst; Temperature: 673 ˙ 3.0 K; Pressure: 24.5 ˙ 0.5 MPa; Reaction time: 20 min (Reprinted
with permission from Lu et al. [87]. Copyright © 2013, Elsevier)
During the SCWG of glucose, the carbon could be formed by two pathways.
One is by decomposition of intermediate liquid products and the other is by product
gas. Sinag et al. proposed the mechanism of solid carbon formation in SCWG of
glucose [9]. Glucose decomposed into furfurals and organic acids firstly and then the
furfurals could decompose in SCW with the formation of solid carbon, phenols and
364 Y. Lu et al.
organic acids. The phenolic compounds could convert to further into solid carbon
and organic acids. The product gas consisting of H2 , CH4 , CO, and CO2 is formed by
organic acids decomposition. Ni catalysts can promote the decomposition of glucose
into organic acid, but inhibit the formation of the phenolic compounds from the
furfurals. Therefore, addition of Ni catalysts will reduce the amount of solid carbon.
The carbon formation could also occur due to the side reactions of gas produced.
Laosiripojana proposed the most probable reactions that could lead to the carbon
formation, in steam reforming of CH4 [88]. The reactions are as follows,
CH4 $ 2H2 C C (13.1)
2CO $ CO2 C C (13.2)
CO C H2 $ H2 O C C (13.3)
CO2 C 2H2 $ 2H2 O C C (13.4)
In SCWG of glucose, the carbon is most likely formed by reaction (13.3) and
(13.4) but not by reaction (13.1) and (13.2), because at low reaction temperature
(673 K), Eqs. (13.3) and (13.4) are favorable, while Eqs. (13.1) and (13.2) are
thermodynamically unfavorable [88]. At the same time, the equilibrium of water-
gas shift reaction moves forward and produces more CO2 rather than CO with the
increase of water to glucose ratio. Therefore, high water feed in SCWG can also
inhibit carbon deposition via Eq. (13.2). In comparison, at low temperature solid
carbon is more likely formed by the first pathway rather than the second.
High catalytic activity and resistance toward carbon formation of Ni/CeO2 /Al2 O3
could be mainly due to the redox property of ceria [88]. CeO2 -based materials have
high oxygen storage capacity and oxygen mobility. These characteristics are related
to their rapid reduction/oxidation capability by releasing and uptaking oxygen
owing to the reversible reaction of [89],
CeO2 $ CeO2x C Ox (13.5)
Where Ox is the lattice oxygen at CeO2 surface. Carbon monoxide can adsorb and
react with the lattice oxygen on the surface of ceria to form carbon dioxide,
CO C Ox ! CO2 C Ox1 (13.6)
Where Ox1 is the reduced site of ceria. More CO2 are produced rather than CO,
which can inhibit carbon deposition via the Eqs. (13.2) and (13.3).
As a lattice oxygen provider, CeO2 may oxidize the solid carbon in the following
reaction [89],
C .s/ C Ox ! Ox1 C CO (13.7)

Similarly, in presence of lattice oxygen, Ni oxidation may also take place

according to the following reaction,
Ni C Ox ! Ox1 C NiO (13.8)
It is unlikely that CeO2 oxidizes Ni to NiO, which has less catalyst activity
than Ni species, whereas NiO possibly reacts with reduced site of Ce, Ox1 , and
is reduced to Ni, maintaining Ni activity for hydrogen production [89].
Also, H2 O can react with the reduced site of ceria, Ox1 . The steady state reform-
ing rate is mainly due to the continuous supply of the oxygen source by H2 O [89].
Ox1 C H2 O ! H2 C Ox (13.9)
Because Ce as promoter can help to remove the deposited carbon, Ni/CeO2

catalysts, having more active sites, will inhibit the carbon formation further. At the
same time, Ni interacts with CeO2 strongly over the catalysts prepared by the co-
impregnation method and this interaction can lead to high catalytic performance
[90], which also benefits the inhibition of carbon formation. On the other hand,
introducing alkali element to alumina supported nickel catalysts has been suggested
to neutralize the acid sites of alumina and reduce carbon formation by suppressing
cracking and polymerization reactions.
In work of Li et al., alumina supported nickel modified with various Mg loading
was prepared by a co-precipitation method for SCWG [91]. In their work, a series
of Ni-Mg-Al catalysts with different Mg/Al molar ratios were synthesized. Physical
and chemical structures of the materials were fully characterized. It was found the
high dispersion of nickel catalysts was crucial for efficient gasification of glucose in
SCW. The co-precipitated nickels with aluminum and magnesium had a crystalline
size of less than 10 nm (as shown Fig. 13.6) and they showed highly catalytic
activities. The results indicated that there was an optimum amount of magnesium
in this nickel-aluminum system. The optimum magnesium additives could take up
the vacancies around aluminum and form spinel MgAl2 O4 (as shown in Fig. 13.7).
These vacancies around aluminum would be filled with nickel ions with the absence
of promoters and then the acid sites over the alumina support could cause severe
coking of nickel metals. Their experiments showed that the addition of the optimum
magnesium could effectively reduce the graphite carbon deposition of alumina
support nickel catalyst by 41 wt%. Moreover, they tested that the co-precipitated
MgAl2 O4 supported nickel catalyst had a better resistance of being dissolved
under the critical hydrothermal conditions. However, further increase of Mg content
resulted in the formation of MgNiO2 during the co-precipitation process. The Ni2C
MgNiO2 is found to be difficult to be activated, and the catalytic activity is reduced
as a consequence.
To summarize, besides the high catalytic activity (hydrogen yield, carbon
gasification efficiency and hydrogen selectivity), the hydrothermal stability related
to the deactivation (sintering and carbon deposition et al.) of nickel in SCW should
also be the key challenges to make SCWG process technically and economically
viable for hydrogen production from biomass.
366 Y. Lu et al.
Fig. 13.6 Transmission Electron Microscopy (TEM) images of NiMg0.6 Al1.9 catalysts (a, b: fresh
NiMg0.6 Al1.9 ; c, d: used NiMg0.6 Al1.9 ) (Reprinted with permission from Li et al. [91]. Copyright
© 2013, Elsevier)
13.5 Conclusion and Future Perspectives
High pressure and temperature makes the gasification apparatus more demand for
containment structure and use of costly alloys, which increases the investment costs
of this hydrogen production technology. The catalyst is of great significance for real-
izing high gasification efficiency of biomass under milder circumstances in SCW.
Catalysis in SCWG could be simply classified into two kinds: Homogeneous and
Heterogeneous catalysts. The alkali catalyst is a homogeneous catalyst, which has a
key role in the SCWG of model compounds, biomass, organic wastes and coal. They
promoted the gasification reaction and increased the yield of hydrogen effectively.
Because of the dissolution of alkali catalysts in water, the wide application of alkali
catalysts will bring great benefits in SCWG. However, the recycle of alkali catalysts
is difficult. Heterogeneous catalysts as noble metal have usually been used in SCWG
of biomass because of its high activity. In all noble metal catalysts, Ruthenium is
found to be very active for reactions involved in SCWG. However, high cost and
limited availability prevent their further development. Nickel catalysts have been
widely used in sub or supercritical water gasification, due to its low cost and high
♦ Ni γ-Al2O3 MgAl2O4 MgNiO2

♦
♦ ♦
e
d
Intensity(a.u.)
JCPDS No. 01-084-0377 (MgAl2O4)
♦
a ♦ ♦
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
2θ
Fig. 13.7 XRD patterns of reduced Ni-Al and Ni-Mg-Al catalysts (a: NiAl3.1 ; b: NiMg0.6 Al1.9 ;
c: NiMg0.7 Al1.5 ; d: NiMg1.0 Al0.9 ; e: NiMg1.0 Al0.5 ) (Reprinted with permission from Li et al. [91].
activity, but crystallite sintering, supports’ breakdown and carbon deposition under
SCW conditions cause deactivation of catalysts. Therefore, it is necessary to study
on the Ni catalysts modification, such as the work mentioned in this chapter, in the
future.
Acknowledgements We greatly acknowledge the financial supports from the National Natural
Science Foundation of China (No. 5132206) and the National Key Project for Basic Research
of China (No. 2012CB215303). We will also thank Drs. Yunan Chen and Liya Zhu for their
contributions to this chapter.
References
1. Hemmati S, Saboohi Y, Hashemi N. Thermodynamic modeling for hydrogen production

from biomass and evaluation of biomass energy technologies. In: Proceedings of the 3rd
international congress on biotechniques for air pollution control, Delft, The Netherlands; 2009.
2. Dincer I. Technical, environmental and exergetic aspects of hydrogen energy systems. Int J
Hydrog Energy. 2002;27(3):265–85.
3. Kruse A, Gawlik A. Biomass conversion in water at 330–410 ı C and 30–50 MPa: identification
of key compounds for indication different chemical reaction pathways. Ind Eng Chem Res.
2003;42(2):267–79.
368 Y. Lu et al.
4. Lu YJ, Zhao L, Guo LJ. Technical and economic evaluation of solar hydrogen production by
supercritical water gasification of biomass in China. Int J Hydrog Energy. 2011;36:14349–59.
2008;2:254–65.
6. Osada M, Sato T, Watanabe M, Shirai M, Arai K. Catalytic gasification of wood biomass in
subcritical and supercritical water. Combust Sci Technol. 2006;178(1–3):537–52.
7. Elliott DC, Hart TR, Neuenschwander GG. Chemical processing in high-pressure aqueous
environments. 8. Improved catalysts for hydrothermal gasification. Ind Eng Chem Res.
2006;45(11):3776–81.
8. Kruse A, Meier D, Rimbrecht P, Schacht M. Gasification of pyrocatechol in supercritical water
in the presence of potassium hydroxide. Ind Eng Chem Res. 2000;39(12):4842–8.
9. Sinag A, Kruse A, Rathert J. Influence of the heating rate and the type of catalyst on the
formation of key intermediates and on the generation of gases during hydropyrolysis of glucose
in supercritical water in a batch reactor. Ind Eng Chem Res. 2004;43(2):502–8.
10. Kruse A, Sinag A, Schwarzkopf V. Key compounds of the hydropyrolysis of glucose in
supercritical water in the presence of K2 CO3 . Ind Eng Chem Res. 2003;42(15):3516–21.
11. Gadhe JB, Gupta RB. Hydrogen production by methanol reforming in supercritical water:
suppression of methane formation. Ind Eng Chem Res. 2005;44(13):4577–85.
12. Schmieder H, Abeln J, Boukis N, Dinjus E, Kruse A, Kluth M, Petrich G, Sadri E,
Schacht M. Hydrothermal gasification of biomass and organic wastes. J Supercrit Fluid.
2000;17(2):145–53.
as a model compound of biomass gasified in supercritical water. Int J Hydrog Energy.
2003;28(1):55–64.
14. Guo S, Guo L, Cao C, Yin J, Lu Y, Zhang X. Hydrogen production from glycerol by
supercritical water gasification in a continuous flow tubular reactor. Int J Hydrog Energy.
2012;37(7):5559–68.
15. Guo Y, Wang S, Wang Y, Zhang J, Xu D, Gong Y. Gasification of two and three-components
mixture in supercritical water: influence of NaOH and initial reactants of acetic acid and
phenol. Int J Hydrog Energy. 2012;37(3):2278–86.
II. Effect of catalyst. Int J Hydrog Energy. 2008;33(17):4520–6.
17. Kruse A, Krupka A, Schwarzkopf V, Gamard C, Henningsen T. Influence of proteins on the
hydrothermal gasification and liquefaction of biomass. 1. Comparison of different feedstocks.
Ind Eng Chem Res. 2005;44(9):3013–20.
18. Watanabe M, Inomata H, Osada M, Sato T, Adschiri T, Arai K. Catalytic effects of NaOH and
ZrO2 for partial oxidative gasification of n-hexadecane and lignin in supercritical water. Fuel.
2003;82(5):545–52.
19. Jarana MBG, Saanchez-Oneto J, Portela JR, Sanz EN, de la Ossa EJM. Supercritical water
gasification of industrial organic wastes. J Supercrit Fluid. 2008;46(3):329–34.
20. Nakhla G, Youssef EA, Elbeshbishy E, Hafez H, Charpentier P. Sequential supercritical water
gasification and partial oxidation of hog manure. Int J Hydrog Energy. 2010;35(21):11756–67.
21. Yan B, Wei CH, Hu CS, Xie C, Wu JZ. Hydrogen generation from polyvinyl alcohol-
contaminated wastewater by a process of supercritical water gasification. J Environ Sci China.
2007;19(12):1424–9.
22. Zhai Y, Wang C, Chen H, Li C, Zeng G, Pang D, Lu P. Digested sewage sludge gasification in
supercritical water. Waste Manag Res. 2013;31(4):393–400.
23. Chen Y, Guo L, Cao W, Jin H, Guo S, Zhang X. Hydrogen production by sewage
sludge gasification in supercritical water with a fluidized bed reactor. Int J Hydrog Energy.
2013;38(29):12991–9.
24. Xu C, Donald J. Upgrading peat to gas and liquid fuels in supercritical water with catalysts.
Fuel. 2012;102:16–25.
25. Guo LJ, Li YL, Zhang XM, Jin H, Lu YJ. Hydrogen production from coal gasification in super-
critical water with a continuous flowing system. Int J Hydrog Energy. 2010;35(7):3036–45.
26. Jin H, Lu YJ, Liao B, Guo LJ, Zhang XM. Hydrogen production by coal gasification
in supercritical water with a fluidized bed reactor. Int J Hydrog Energy. 2010;35(13):
7151–60.
27. Azadi P, Khodadadi AA, Mortazavi Y, Farnood R. Hydrothermal gasification of glucose
using Raney nickel and homogeneous organometallic catalysts. Fuel Process Technol.
2009;90:145–51.
28. Lee IG, Ihm SK. Catalytic gasification of glucose over Ni/Activated Charcoal in supercritical
water. Ind Eng Chem. 2009;48:1435–42.
29. Xu X, Matsumura Y, Stenberg J, Antal MJ. Carbon-catalyzed gasification of organic feedstocks
in supercritical water. Ind Eng Chem. 1996;35:2522–30.
30. Xu X, Antal MJ. Gasification of sewage sludge and other biomass for hydrogen production in
supercritical water. Environ Prog. 1998;17:215–20.
31. Antal JM, Allen SG, Schulman D, Xu X, Divilio RJ. Biomass gasification in supercritical
32. Zhang R, Jiang W, Cheng L, Sun B, Sun D, Bi J. Hydrogen production from lignite via
supercritical water in flow-type reactor. Int J Hydrog Energy. 2010;35:11810–5.
33. Watanabe M, Inomata H, Arai K. Catalytic hydrogen generation from biomass (glucose and
cellulose) with ZrO2 in supercritical water. Biomass Bioenergy. 2002;22:405–10.
34. Hao XH, Guo LJ, Zhang XM, Guan Y. Hydrogen production from catalytic gasification of
cellulose in supercritical water. Chem Eng J. 2005;110:57–65.
35. Park KC, Tomiyasu H. Gasification reaction of organic compounds catalyzed by RuO2 in
supercritical water. Chem Commun. 2003;6:694–5.
36. Tomoo Y, Tomonori M, Ki CP, Yasuhiko F, Hiroshi T. Ruthenium(IV) dioxide-catalyzed
reductive gasification of intractable biomass including cellulose, heterocyclic compounds, and
sludge in supercritical water. J Supercrit Fluid. 2009;51:43–9.
37. Onwudili JA. Williams PT (2013) Hydrogen and methane selectivity during alkaline super-
critical water gasification of biomass with ruthenium-alumina catalyst. Appl Catal B Environ.
2013;132–133:70–9.
38. Chakinala AG, Chinthaginjala JK, Seshan K, van Swaaij WPM, Kersten SRA, Brilman DWF.
Catalyst screening for the hydrothermal gasification of aqueous phase of bio-oil. Catal Today.
2012;195:83–92.
39. Stucki S, Vogel F, Ludwig C, Haiduc AG, Brandenberger M. Catalytic gasification of
algae in supercritical water for biofuel production and carbon capture. Energy Environ Sci.
2009;2:535–41.
40. Guan QQ, Wei CH, Savage PE. Hydrothermal gasification of Nannochloropsis sp. with Ru/C.
Energy Fuels. 2012;26:4575–82.
41. Youssef EA, Nakhla G, Charpentier PA. Oleic acid gasification over supported metal catalysts
in supercritical water: hydrogen production and product distribution. Int J Hydrog Energy.
2011;36:4830–42.
42. Dolan R, Yin SD, Tan ZC. Effects of headspace fraction and aqueous alkalinity on subcritical
hydrothermal gasification of cellulose. Int J Hydrog Energy. 2010;35:6600–10.
43. May A, Salvadó J, Torras C, Montané D. Catalytic gasification of glycerol in supercritical
water. Chem Eng J. 2010;160:751–9.
44. Hashaikeh R, Fang Z, Butler IS, Kozinski JA. Sequential hydrothermal gasification of biomass
to hydrogen. Proc Combust Inst. 2005;30:2231–7.
45. Yamaguchi A, Hiyoshi N, Sato O. Gasification of organosolv-lignin over charcoal supported
noble metal salt catalysts in supercritical water. Top Catal. 2012;55(11–13):889–96.
46. Osada M, Yamaguchi A, Hiyoshi N. Gasification of sugarcane bagasse over supported
ruthenium catalysts in supercritical water. Energy Fuels. 2012;26(6):3179–86.
47. Sato T, Inda K, Itoh N. Gasification of bean curd refuse with carbon supported noble metal
catalysts in supercritical water. Biomass Bioenergy. 2011;35(3):1245–51.
48. Yamaguchi A, Hiyoshi N, Sato O. Lignin gasification over supported ruthenium trivalent salts
in supercritical water. Energy Fuels. 2008;22(3):1485–92.
370 Y. Lu et al.
49. Yamaguchi A, Hiyoshi N, Sato O. EXAFS study on structural change of charcoal-supported

ruthenium catalysts during lignin gasification in supercritical water. Catal Lett. 2008;
122(1–2):188–95.
50. Osada M, Hiyoshi N, Sato O. Reaction pathway for catalytic gasification of lignin in presence
of sulfur in supercritical water. Energy Fuels. 2007;21(4):1854–8.
51. Osada M, Hiyoshi N, Sato O. Effect of sulfur on catalytic gasification of lignin in supercritical
water. Energy Fuels. 2007;21(3):1400–5.
52. Osada M, Sato O, Arai K. Stability of supported ruthenium catalysts for lignin gasification in
supercritical water. Energy Fuels. 2006;20(6):2337–43.
53. Osada M, Sato O, Watanabe M. Water density effect on lignin gasification over supported noble
metal catalysts in supercritical water. Energy Fuels. 2006;20(3):930–5.
54. Osada M, Sato T, Watanabe M, Adschiri T, Arai K. Low-temperature catalytic gasification
of lignin and cellulose with a ruthenium catalyst in supercritical water. Energy Fuels.
2004;18(2):327–33.
gasification of biomass and wastes. Int J Hydrog Energy. 2011;36:9529–41.
56. Modell M. Gasification and liquefaction of forest products in supercritical water. In: Overend
RP, Milne TA, Mudge LK, editors. Fundamentals of thermochemical biomass conversion.
London: Elsevier Applied Science Publishers; 1985. p. 937–50.
57. Elliott DC, Sealock LJ, Baker EG. Chemical-processing in high-pressure aqueous environ-
ments.2. Development of catalysts for gasification. Ind Eng Chem Res. 1993;32(8):1542–8.
58. Minowa T, Ogi T. Hydrogen production from cellulose using a reduced nickel catalyst. Catal
Today. 1998;45(4):411–6.
59. Yoshida T, Oshima Y, Matsumura Y. Gasification of biomass model compounds and real
biomass in supercritical water. Biomass Bioenergy. 2003;26(1):71–8.
60. Sato T, Furusawa T, Ishiyama Y, Sugito H, Miura Y, Sato M, Suzuki N, Itoh N. Effect of
water density on the gasification of lignin with magnesium oxide supported nickel catalysts in
supercritical water. Ind Eng Chem Res. 2006;45(2):615–22.
61. Furusawa T, Sato T, Sugito H, Miura Y, Ishiyama Y, Sato M, Itoh N, Suzuki N. Hydrogen
production from the gasification of lignin with nickel catalysts in supercritical water. Int J
Hydrog Energy. 2007;32(6):699–704.
production from glucose via catalytic supercritical water gasification. Int J Hydrog Energy.
2011;36(16):9591–601.
bark and model carbohydrates in supercritical water using nickel and ruthenium catalysts. Appl
Catal B Environ. 2012;117–118:330–8.
64. Dileo GJ, Neff ME, Kim S, Savage PE. Supercritical water gasification of phenol and glycine
as models for plant and protein biomass. Energy Fuels. 2008;22(2):871–7.
65. Dileo GJ, Neff ME, Savage PE. Gasification of guaiacol and phenol in supercritical water.
Energy Fuels. 2007;21(4):2340–5.
66. DiLeo GJ, Savage PE. Catalysis during methanol gasification in supercritical water. J Supercrit
Fluid. 2006;39(2):228–32.
67. Youssef EA, Chowdhury MBI, Nakhla G, Charpentier P. Effect of nickel loading on hydrogen
production and chemical oxygen demand (COD) destruction from glucose oxidation and
gasification in supercritical water. Int J Hydrog Energy. 2010;35(10):5034–42.
68. Taylor AD, Dileo GJ, Sun K. Hydrogen production and performance of nickel based catalysts
synthesized using supercritical fluids for the gasification of biomass. Appl Catal B-Environ.
2009;93(1–2):126–33.
69. Yan B, Wu J, Xie C, He F, Wei C. Supercritical water gasification with Ni/ZrO2 catalyst
for hydrogen production from model wastewater of polyethylene glycol. J Supercrit Fluids.
2009;50(2):155–61.
70. Elliott DC, Sealock LJ, Baker EG. Chemical processing in high-pressure aqueous environ-
ments. 3. Batch reactor process development experiments for organics destruction. Ind Eng
Chem Res. 1994;33:558–65.
71. Elliott DC, Phelps MR, Sealock Jr LJ, Baker EG. Chemical processing in high-pressure
aqueous environments. 4. Continuous-flow reactor process development experiments for
organic destruction. Ind Eng Chem Res. 1994;33:566–74.
72. Elliott DC, Neuenschwander GG, Phelps MR, Hart TR, Zacher AH, Silva LJ. Chemical pro-
cessing in high-pressure environments. 6. Demonstration of catalytic gasification for chemical
manufacturing wastewater cleanup in industrial plants. Ind Eng Chem Res. 1999;38:879–83.
nickel catalyst. J Supercrit Fluid. 1998;13(2):253–9.
74. Minowa T, Inoue S. Hydrogen production from biomass by catalytic gasification in hot
compressed water. Renew Energy. 1999;16:1114–7.
75. Yoshida T, Matsumura Y. Gasification of cellulose, xylan, and lignin mixtures in supercritical
76. Resende FLP, Savage PE. Effect of metals on supercritical water gasification of cellulose and
lignin. Ind Eng Chem Res. 2010;49(6):2694–700.
77. Azadi P, Syed KM, Farnood R. Catalytic gasification of biomass model compound in near-
critical water. Appl Catal A Gen. 2009;358:65–72.
78. Fang Z, Minowa T, Fang C, Smith Jr RL, Inomata H, Kozinski JA. Catalytic hydrothermal
gasification of cellulose and glucose. Int J Hydrog Energy. 2008;33:981–90.
79. Chakinala AG, Brilman DWF, van Swaaij WPM, Kersten SRA. Catalytic and non-
catalytic supercritical water gasification of microalgae and glycerol. Ind Eng Chem Res.
2010;49(3):1113–22.
80. Wang S, Lu GQM. CO2 reforming of methane on Ni catalysts: effects of the support phase and
preparation technique. Appl Catal B Environ. 1998;16(3):267–77.
81. Natesakhawat S, Watson RB, Wang X, Ozkan US. Deactivation characteristics of lan-
thanide promoted sol-gel Ni/Al2 O3 catalysts in propane steam reforming. J Catal.
2005;234(2):496–508.
82. Lu YJ, Li S, Guo LJ, Zhang XM. Hydrogen production by biomass gasification in supercritical
water over Ni/”Al2 O3 and Ni/CeO2 -”Al2 O3 . Int J Hydrog Energy. 2010;35:7161–8.
83. Vizcaino A, Arena P, Baronetti G, Carrero A, Calles J, Laborde M, Amadeo N. Ethanol
steam reforming on Ni/Al2 O3 catalysts: effect of mg addition. Int J Hydrog Energy.
2008;33(13):3489–92.
84. Gac W. Acid–base properties of Ni–MgO–Al2 O3 materials. Appl Surf Sci.
2011;257(7):2875–80.
85. Huber GW, Shabaker JW, Dumesic JA. Raney Ni-Sn catalyst for H2 production from biomass-
derived hydrocarbons. Science. 2003;300(5628):2075–7.
86. Li S, Lu YJ, Guo LJ, Zhang XM. Hydrogen production from biomass gasification in
supercritical water with bimetallic Ni-M/Al2 O3 catalysts (M D Cu, Co and Sn). Int J Hydrog
Energy. 2011;36:14391–400.
87. Lu YJ, Li S, Guo LJ. Hydrogen production by supercritical water gasification of glucose with
Ni/CeO2 /Al2 O3 : effect of Ce loading. Fuel. 2013;103:193–9.
88. Laosiripojana N, Sangtongkitcharoen W, Assabumrungrat S. Catalytic steam reforming of
ethane and propane over CeO2 -doped Ni/Al2 O3 at SOFC temperature: improvement of resis-
tance toward carbon formation by the redox property of doping CeO2 . Fuel. 2006;85:323–32.
89. Kumar P, Sun Y, Idem RO. Comparative study of Ni-based mixed oxide catalyst for carbon
dioxide reforming of methane. Energy Fuels. 2008;22:3575–82.
90. Tomishige K, Kimura T, Nishikawa J, Miyazawa T, Kunimori K. Promoting effect of
the interaction between Ni and CeO2 on steam gasification of biomass. Catal Commun.
2007;8:1074–9.
91. Li S, Guo LJ, Zhu C, Lu YJ. Co-precipitated Ni-Mg-Al catalysts for hydrogen production by
supercritical water gasification of glucose. Int J Hydrog Energy. 2013;38(23):9688–700.
Chapter 14
Hydrothermal Conversion in Near-Critical
Water – A Sustainable Way of Producing
Renewable Fuels
Jessica Hoffmann, Thomas H. Pedersen, and Lasse A. Rosendahl
Abstract Liquid fuels from biomass will form an essential part of meeting
the grand challenges within energy. The need for renewable and sustainable
energy sources is triggered by a number of factors; like increase in global energy
demand, depletion of conventional resources, climate issues and the desire for
national/regional energy independence. Especially in marine, aviation and heavy
land transport suitable carbon neutral drop-in fuels from biomass are needed,
since electrification of those is rather unlikely. Hydrothermal conversion (HTC)
of biomass offers a solution and is a sustainable way of converting biomass
feedstocks to valuable bio-crude. HTC is a high pressure and medium temperature
thermochemical biomass conversion process and converts aqueous biomasses under
sub- or super-critical conditions to a bio-crude similar to fossil crude oil.
This chapter deals with the chemical reaction pathways during hydrothermal
conversion of lignocellulosic biomass and upgrading pathways of bio-crude com-
ponents with focus on hydrodeoxygenation reactions.
Keywords Hydrothermal liquefaction • Lignocellulosic biomass • Upgrading

• Hydrodeoxygenation • Supercritical water • Bio-crude
14.1 Introduction
Biomass to biofuels is a major ongoing research focus worldwide due to the renew-
able aspects and abundance of biomass. Biomass is an energy carrier containing
carbon and hydrogen; the backbone constituents of current fuels. Breaking down
and chemically modifying biomasses provides numerous pathways to synthetic
J. Hoffmann () • T.H. Pedersen • L.A. Rosendahl

Department of Energy Technology, Aalborg University, Pontoppidanstræde 101, DK-9220
Aalborg Ø, Denmark
374 J. Hoffmann et al.
Fig. 14.1 Carbon number of common hydrocarbons
renewable biofuels that can offer an alternative to the current fossil based hydro-
carbon infrastructure.
Producing marketable renewable fuels from biomass is, however, confronting a
multitude of challenges. Fuels produced from a fossil origin are still economically
favorable over sustainable biofuels, despite the fact that fossil crude prices have
increased significantly and most future projections to 2035–2040 forecast still
increasing or steady costs depending on political actions [1, 2]. More important
is that any final biofuels should be what is known as an advanced biofuel, and need
to conform to the following:
• Be sustainably produced
• Be based on an abundant non-food feedstock or feedstock mixture to have a high
impact factor, locally and globally
• Exhibit drop-in properties in the sense that hydrocarbon infrastructure and
downstream processing remain unaltered
• Obey all current regulations on fuel specifications
Hence, a major goal of producing biofuels to obtain products which perform like
existing conventional hydrocarbon fuels mainly consist of aliphatic, alicyclic and
aromatic hydrocarbons in a variety of different compounds.
Often, crude oil is characterized by the boiling points of the fractions of the
specific products (diesel, kerosene, naphtha, etc.) obtained from the oil, which to an
extent can be translated to carbon numbers as shown in Fig. 14.1.
Unfortunately, the natural composition of biomass is different from the composi-
tion of fossil crude, and advanced conversion methods of biomass are needed to
obtain the fuel requirements mentioned previously. Like hydrocarbon fuels, raw
biomass consists of hydrogen and carbon but also of heteroatoms like oxygen,
nitrogen and sulphur. A broad excerpt of different biomasses was examined by [3]
for which average biomass figures are presented in Table 14.1.
The challenges in using biomass as a sustainable feedstock for fuel production
are those heteroatoms. Currently biomass conversion technologies produce biofuels
with high oxygen contents depending on the feedstock used.
14 Hydrothermal Conversion in Near-Critical Water – A Sustainable Way. . . 375
Table 14.1 Ultimate analysis and high heating values (HHV, MJ/kg) of different biomasses
Biomass HHV C H N S Cl O Ash H/C O/C
Energy grass, 19.14 48.30 5.50 0.60 0.10 0.20 41.50 3.80 1.37 0.64
miscanthus
Energy grass, 18.04 45.00 5.30 2.10 0.20 0.50 37.60 9.30 1.41 0.63
other
Wood material 19.58 49.00 5.70 0.40 0.10 0.10 41.90 2.90 1.40 0.64
Cereals 18.61 46.50 6.10 1.20 0.10 0.20 42.00 3.90 1.57 0.68
Millet 18.17 45.90 5.30 0.90 0.10 0.30 41.10 6.50 1.39 0.67
Sunflower 20.26 50.50 5.90 1.30 0.10 0.40 34.90 6.90 1.40 0.52
Hemp 18.04 45.70 6.30 0.60 0.00 0.10 44.10 3.20 1.65 0.72
Data reprinted with permission from Friedl et al. [3]. Copyright © 2005, Elsevier
Hence, a target in any advanced biomass-to-liquid (BtL) process is to reduce the

great amount of especially oxygen in biomass to approach the composition of fossil
products as illustrated by Eq. (14.1).
CH2 O ! CH2 (14.1)
Hydrothermal liquefaction (HtL) is approaching this by using high pressures

(10–30 MPa) and medium temperatures (250–400 ı C) and the advantageous
characteristics of water in sub- or supercritical conditions to convert biomass to a
bio-crude relatively low in oxygen [4].
On a daily basis some 90 million barrels of oil are consumed worldwide [2], of
which approximately two third is consumed by the transportation sector, why any
marketable product should be able to maintain, in volume as in energy, a significant
market share of any targeted market segment.
Therefore, when considering the diversity of biomass composition, BtL-
processes need to be flexible in terms of biomass input. Achieving this,
competitiveness with conventional fuels can be realized. Unlike biochemical
processes, thermochemical processes like HtL, pyrolysis and gasification have
displayed such ability. Additionally, since HtL utilizes an aqueous medium, it
benefits from the absent need of feedstock drying prior processing and can process
wet materials. Figure 14.2 schematically displays different BtL routes based on a
lignocellulosic feedstock platform. It clearly appears that a wide range of liquid
and gaseous fuels and fuel precursors can readily be produced directly from HtL;
some even with chemical resemblance to fossil fuels. Literature studies reveal that
from a lignocellulosic biomass origin, hydrocarbon products in the carbon range of
C1–C20 are achievable [5, 6].
The first part of this chapter reviews general hydrothermal reactions of fun-
damental biomass constituents in a lignocellulosic context. Hence, conversion of
cellulose, hemicellulose and lignin and in hot-compressed water and obtainable
products is being discussed.
Fig. 14.2 Flow chart of lignocellulosic conversion route using an aqueous medium (Adapted with
permission from Huber and Dumesic [7]. Copyright © 2006, Elsevier)
The second part discusses the properties of bio-crude obtained from HtL. In the
last part upgrading pathways from bio-crude to finished transportation fuel are being
discussed.
14.2 Lignocellulosic Biomass – Valorization of the Complete

Feedstock Potential
Lignocellulose is a biomass subgroup consisting primarily of three principal

subcomponents: Cellulose, hemicellulose and lignin. The relative amount of the
three different subcomponents is highly biomass dependent, but generally the mass
distribution is approximately 35–50 % cellulose, 20–40 % hemicellulose and 10–
35 % lignin [8]. A range of examples are shown in Table 14.2 together with their
HHV measured by [9].
Physically, the structure of lignocellulose is highly complex and consists of a
rigid crystalline cellulose fraction embedded in a matrix of hemicellulose and lignin
defining the woody architecture [10, 11], Fig. 14.3.
Table 14.2 Cellulose, hemicellulose and lignin distribution of several lignocellulosic materials
including measured HHV
Species Cellulose mass % Hemicellulose mass % Lignin mass % HHV MJ/kg
Tobacco leaf 43.45 41.54 15.01 17.7
Corncob 52.49 32.32 15.19 17.48
Corn straw 51.53 30.88 17.59 18.27
Wheat straw 33.82 45.2 20.98 18.55
Beech wood 46.27 31.86 21.87 19.23
Hardwood 45.85 32.26 21.89 18.97
Ailanthus wood 47.52 26.73 25.75 19.37
Tobacco stalk 44.32 28.89 26.79 18.43
Softwood 42.68 24.82 32.5 19.55
Spruce wood 47.11 21.31 31.58 19.77
Hazelnut shell 26.7 30.29 43.01 20.05
Wood bark 25.59 30.28 44.13 20.74
Olive cake 23.08 21.63 55.29 21.53
Reprinted with permission from Demirbaş [9], copyright © 2001, Elsevier
Values are given on an extractive-free and ash-free basis
Fig. 14.3 Structure of lignocellulosic biomass (Reprinted with permission from Chatel and
Rogers [11]. Copyright © 2013, American Chemical Society)
Pretreatment of the complex structure towards accessibility of all macro-

molecules has been recognized for a long time. The majority of literature on
this topic mostly serves the purpose of efficient delignification and removal of
hemicellulose to recover cellulose fibers or glucose units by hydrolysis. The reason
is that cellulose itself almost exclusively has been recognized as the lignocellulosic
subcomponent holding the greatest valorization potential. However, to realize full
BtL valorization potential, utilization of complete biomasses subcomponents is a
necessity. In HtL the purpose is to break down all principal subcomponents into
fragments, rearrange and combine fragments selectively in order to maximize the
biomass valorization by subcomponent synergetic interactions; synergetic reactions
which not only aim for high yields but also for high quality hydrocarbon products.
Another important aspect to be considered when elucidating valorization poten-

tials is the energy distribution between principal subcomponents. Assuming that the
HHV of lignocellulosic matter is given by a weighted sum of the individual heating
values of cellulose, hemicellulose and lignin, Eq. (14.2) can be derived from the
data given in Table 14.2. It appears that the lignin fraction is a key subcomponent
with respect to the inherent energy distribution.
HHV D 17:17 %Cellulose C 15:76 %Hemicellulose C 25:687 %Lignin

(14.2)
Despite this fact, today nearly all lignin residues are burned in order to
produce process heat, and have so far not been considered as feasible liquid fuel
precursors. More than 50 % of the total energy may be inherent within the lignin
fraction and discarding this fraction will significantly penalize the BtL energy
efficiency.
14.2.1 Hydrothermal Treatment of Cellulose
As mentioned, cellulose is one of three major subcomponents of lignocellulosic

biomass. In order to understand the reaction pathways during HtL of lignocellulosic
biomass it is crucial to look at cellulose decomposition during hydrothermal
treatment separately.
Cellulose is a well-defined polymer of anhydro-D-glucopyranose units linked by
glycosidic “(1 ! 4) linkages in a crystalline structure. When processing cellulose
hydrothermally, the first degradation mechanism of cellulose polymers is scission
of the glycosidic bonds for depolymerization. Two scission pathways have been
suggested; thermal cleavage via dehydration at reducing-end units and hydrolysis of
the glycosidic bonds by which glucose units are obtained. Thermal cleavage takes
place at intermediate temperatures and low pressure, while hydrolysis becomes
predominant in the near and supercritical region.
Similarly, the degradation of cellobiose, the dimeric model of cellulose, occurs
along two pathways; thermally induced Retro-Aldol condensation and hydrolysis.
In the subcritical region the rate constants of the two pathways are almost invariant
with pressure, but show great variation in the supercritical region. This is partly
explained by the variance in density in the two regions. As for cellulose degradation,
the hydrolysis rate is enhanced by increasing pressure [12]. Hence, reaction
pathways are tunable by manipulating reaction conditions.
The degradation rates are naturally also affected by catalysts present. The
depolymerization of cellulose is both acidic and alkaline aided, where the latter
demonstrates the highest rates. However, acidic conditions exhibit the highest rate
for further degradation reactions of fragmented cellulose [13].
Fig. 14.4 Degradation pathways and products of glucose (Adapted with permission from Watan-
abe et al. [14]. Copyright © 2005, Elsevier)
14.2.1.1 Degradation Pathways of Glucose During Hydrothermal

Treatment
As for cellulose and cellobiose, the degradation of glucose units obtained from
hydrolysis also follows two pathways; Retro-Aldol condensation and dehydration,
Fig. 14.4.
At low temperatures in the subcritical region glucose conversion is mainly dom-
inated by dehydration reactions. Shifting to higher temperatures in the supercritical
region, Retro-Aldol condensation reactions are predominant. As for cellobiose, the
conversion rates of glucose degradation pathways show hardly any response to
pressure in the subcritical region [15]. However, when entering the supercritical
region, pressure effects can be quite significant; hence selectivity in the supercritical
region can be controlled by changing reaction conditions [12]. At high temperatures,
glucose tends to epimerize into fructose.
The rate of glucose degradation displays an Arrhenius relation through both
the subcritical and supercritical region. At subcritical conditions the degradation
rate of glucose is much faster than its formation from cellulose scission. However,
around the critical point of water the degradation rate of cellulose discontinuously
increases by approximately an order, thereby the glucose formation rate exceeds
its degradation rate. Therefore, by applying short reaction times high selectivity
towards glucose units can be obtained in the supercritical region [16].
As for hydrolysis scission, the selectivity of the different glucose degradation
components is highly affected by whether an acidic or alkaline catalyst is present.
At alkaline conditions 5-hydroxymethylfurfural (5-HMF) is not stable why mainly
lactic acid or other carboxylic acids are formed. In acidic solutions high concentra-
tions of 5-HMF and furfural are found [14].
5-HMF is a major dehydration product not only from glucose but from various
hydrothermally treated hexoses [17]. It is a versatile intermediate including an inter-
esting fuel precursor, since it can be derived from a multitude of feedstocks [18].
One of the research fuel compounds derived from 5-HMF is dimethylfuran (DMF),
a C6 member, which is obtained by selective hydrogenolysis and hydrogenation of
5-HMF to remove oxygen. DMF yields a high research octane number (RON), low
miscibility in water and a lower heating value (LHV) approximately 25 % higher
than that of ethanol. Furthermore, DMF has shown similar properties to gasoline
with regards to combustion and might well yield drop-in properties [19].
14.2.2 Hydrothermal Treatment of Hemicellulose
In contrast to cellulose, hemicellulose is a heteropolymer consisting of both pentoses

and hexoses branched together into short polymers with an amorphous structure.
The hemicellulose sugar distribution is greatly biomass dependent [13].
It is generally known, that due to the branched structure of hemicellulose, which
inhibit inter- and intramolecular hydrogen bonds, hemicellulose is less resistant
to hydrolysis than cellulose [13]. Therefore, hemicellulose is also an interesting
precursor for bio-fuels, since less effort is needed for its conversion.
Lü and Saka investigated the degradation of the several monosaccharides present
in hemicellulose for which the distributions of degradation products were found
dissimilar. For pentoses, furfural is the major degradation product, whereas 5-HMF
is a major product of hexoses. The amount of 5-HMF and furfural increases with
increased reaction time and temperature [17].
14.2.2.1 Degradation Pathways of Xylose/Xylan During Hydrothermal

Treatment
Next to arabinan, galactan and mannan, xylan is commonly the major component
of hemicellulose. This also explains why xylan or more frequently xylose, the
monomeric model of xylan, is used as a representative model compound for
hemicellulose [20–22].
Like glucose, xylose degrades through the two main pathways: The Retro-Aldol
condensation and dehydration [21]. In the low temperature range (160–250 ı C) the
main decomposition products of D-xylose are furfural and lactic acid [17, 22]. For
xylose hydrolysis the activation energy is not changed by increasing the acidity of
the aqueous solution, where the activation energy ranges from 119 to 130 KJ/mol.
On the other hand, an alkaline solution significantly reduces the activation energy to
only 63 KJ/mol, which also changes the reaction pathway from furfural to increased
acids formation [23, 24]. In the temperature range of 160–250 ı C the furfural
formation from xylose increases with increased reaction time and temperature.
As water temperature increases to near critical values the dehydration reactions
selectively decrease, hence the conversion of furfurals also decreases [25, 26].
Not surprisingly furfural has, like 5-HMF, also been pinpointed an interesting
platform chemical. Similarly to the production route of the DMF from 5-HMF,
Fig. 14.5 Different conversion route of furfural, (a) conversion route of MF from furfural, (b) C10
compound, diketodiacid, from furfural and levulinic acid, (c) C8-C13 members from furfural
furfural conversion into 2-methylfuran (MF), a C5 compound, shows almost the

same advantages, see Fig. 14.5a. In comparison to DMF, MF yields almost the same
RON and a slightly lower heating value, but still superior to for instance ethanol.
Due to the similarity of the just mentioned reactions, a co-process of 5-HMF
and furfural, based on a sugar platform for C5 and C6 gasoline substituents, is an
interesting concept. A second remarkable reaction of furfural involving levulinic
acid, another common glucose derivative is shown in Fig. 14.5b. The product is
a C10 member, diketodiacid, which could turn out a valuable diesel compound.
Compounds with even higher carbon numbers are potentially obtained of furfural
as shown in Fig. 14.5c which manifest the versatility of products obtainable from
hydrothermal treatment of sugars [27].
14.2.3 Hydrothermal Treatment of Lignin
Lignin is a natural amorphous phenolic polymer of high molecular weight. The

backbone units are coniferyl alcohol, sinapyl alcohol and p-coumaryl alcohol.
Fig. 14.6 Backbone

polymeric units of lignin
Table 14.3 Common lignin Percentage of total

linkages and distribution linkages
in softwood and hardwood
Linkage type Softwood Hardwood
“-O-4 45–50 60
5-5 19–27 3–9
“-5 9–12 3–11
Spirodienone 2 3–5
4-O-5 4–7 7–9
“-1 1–9 1–2
Dibenzodioxocin 5–7 0–2
“-“ 2–6 3–12
’-O-4 2–8 7
Reprinted with permission from Zakzeski
[28]. Copyright © 2010, American Chemical
Society
The schematic appearances of the three units are shown in Fig. 14.6. The distribution
of the individual units is biomass dependent but normally softwood comprises
mainly of coniferyl alcohol units, hardwood of coniferyl alcohol and sinapyl alcohol
units and grasses of p-coumaryl alcohol units.
Though lignin is a random polymer in the sense that backbone unit distribution
and unit linkages are randomized, general trends are still of high importance since
the probabilistic product scattering naturally depends hereof (Table 14.3).
Studies on lignin degradation in an aqueous medium suggest two degradation
pathways: An ionic pathway through hydrolysis of ether bonds leading to the forma-
tion of phenols, and a radical pathway, along which ether bonds are thermolytically
cleaved. In non-catalyzed water the general trend is that the most abundant “-O-
4 linkage readily cleaves whereas the 4-O-5 bonds together with C–C bonds are
stable under common HtL conditions. As an example, diphenyl ether (DPE) has
been studied as a 4-O-5 ether bond lignin model compound in supercritical water.
Penninger et al. investigated the degradation of DPE in supercritical water with
and without an acid catalyst, NaCl. It was found that the acidic solution promotes
degradation, but at a reaction temperature of 430 ı C and after 5 h the conversion of
DPE was only around 3 % [29, 30]. However, Roberts et al. found that the 4-O-5
bond is far less stable when adding an alkaline catalyst (Li2 CO3 , Na2 CO3 , K2 CO3 ).
At 400 ı C in the presence of a K2 CO3 catalyst conversion of more than 50 % was
obtained after 1 h [31]. It was further found that the DPE conversion decreased
at increased densities but that the hydrolysis reaction pathway was found favored
at high water densities. It is noteworthy that the hydrolysis selectivity was found
100 % for all catalysts at low temperature (320–370 ı C) but decreased at 400 ı C for
all but K2 CO3 . The observation was explained by the decreased catalyst solubility
at supercritical conditions. The observed hydrolysis pathway selectivity may also
explain why oil formation and char reduction is enhanced in liquefaction by adding
an alkaline catalyst.
Radicals formed in the radical pathway are responsible for the formation of high
molecular weight compounds by re-polymerization, oligomerization, condensation
etc. and undesirable char formation comprises mainly of condensed C-C bonds.
Hence, the radical pathway is an undesirable reaction pathway of lignin if radicals
cannot be stabilized. A second negative side effect of the radical reaction is observed
when processing whole lignocellulosic biomasses. In [32] it was visually observed
during hydrothermal treatment of willow, that lignin and hemicellulose dissolute at
around 200 ı C. As temperature increased to 250 ı C lignin precipitated and formed
capping fragments preventing the cellulose dissolution. As a consequence, when the
temperature was further increased to 350 ı C cellulose underwent pyrolysis instead
of hydrolysis. In the pursuance to prevent char formation, radical scavengers like
phenols and alcohols have successfully been used to cap reactive fragments.
Base catalyzed alcoholysis of Kraft and Organosolv derived lignin and lignin
model compounds in methanol and ethanol at 290 ı C was studied by Miller
et al. [33]. Diethyl ether insoluble fraction levels as low as 5 % were achievable
with an ethanol solvent. From the model compound study it was clear that at
these conditions alkylation and dealkylation of benzene rings occurred due to the
interaction with ethanol. In addition it was found that C–C bonds were not cleaved
at these conditions, hence compounds like biphenyl, diphenyl methane and bibenzyl
were unreactive. A similar fact has been confirmed in a hydrous environment. Here
it was concluded that aryl-aryl and methylene bonds can only be cleaved above
400 ı C with the addition of a proper base catalyst.
The alcoholysis is somewhat different from hydrolysis. In hydrothermal medium
phenols are stable compounds, whereas in alcoholic solvents they undergo alky-
lation. The difference between degradation of alkali lignin in a pure ethanol,
pure water or a co-ethanol-water solution was investigated by Cheng et al. [34].
The following sequence was concluded for the yield of degraded lignin; co-
solution > water > ethanol. The higher yield of degraded lignin was accompanied
by the lowest yield of solid residues. Solid production could almost be neglected in
a 1:1 co-solution.
Although the depolymerization/repolymerization mechanisms have been widely
studied, previous work dedicated for turning isolated lignin into a bio-oil is scarce.
Roberts et al. [31] reported the yield of a dark-brown product, formed by the
utilization of base catalyzed depolymerization and boric acid as a repolymerization
inhibition agent. A maximum yield of 52 % was obtained at a NaOH/boric acid
weight ratio of 0.75. It was found that a multitude of process parameters influenced
the yield like temperature, pressure, residence time and weight ratios between lignin,
catalyst and boric acid.
Valuable aromatic fuel components can be derived from lignin monomers, if

bonds are properly broken down and fragments stabilized to prevent repolymer-
ization. Different lignin derived model compounds were investigated by Zhao et al.
[35] for hydrotreating in an aqueous phase using a Pd/C and H3 PO4 catalyst at
250 ı C. High selectivity of saturated compounds was obtained of which the carbon
numbers were mainly preserved. In this particular study C6-C9 components were
obtained, however if the breakdown of lignin polymers can be controlled in terms
of molecular weight, the carbon length of products may also be controlled. In the
next sections upgrading and hydrotreating of bio-crude and bio-crude compounds
are discussed in detail.
14.3 Properties and Upgrading of Bio-crude

from Hydrothermal Conversion by Hydroprocessing
This part of the chapter will discuss the properties of bio-crude from hydrothermal
conversion, upgrading possibilities and reaction pathways during upgrading. Overall
the focus will be on the applied side of bio-crudes from hydrothermal conversion.
Bio-crude from hydrothermal conversion do have a high potential for future
replacement of conventional crude oils. When comparing bio-crudes from
hydrothermal conversion (HTC) with bio-oils from pyrolysis, it is evident that
HTC crudes properties are superior. Table 14.4 shows the elemental composition
of HTC bio-crude compared to bio-oil derived from pyrolysis. Pyrolysis oils have
a significantly higher amount of oxygen compounds and water in the oil. Higher
oxygen content of the oil leads to a lower stability and heating value and a higher
viscosity of the oil [36]. Water in the oil has a complex effect on heating value,
viscosity, pH, homogeneity and other characteristics [36]. Those unwanted contents
can lead to handling and pumping, storage, and corrosion problems and therefore
need to be improved.
Bio-crude from HTC and bio-oil properties are shown in Table 14.4, in compar-
ison to conventional crude oil and gasoline.
Table 14.4 Property ranges of renewables oil compared to conventional hydrocarbons

HTC bio-crude [36] Pyrolysis oil [36] Conventional crude Gasoline
wood wood [37] (Isooctane)
C (wt.-%) 64.5 56.4 83–86 87
H (wt.-%) 7.7 6.2 11–14 13
N (wt.-%) <0.1 <0.1 <1 –
S (wt.-%) <0.1 <0.1 <4 –
O (wt.-%) 22.5 37.3 <1 –
Water (wt.-%) 6–25 25–42 –
Fig. 14.7 Van-Krevelen diagram (Reprinted with permission from Kersten et al. [38]. Copyright
© 2007, Wiley-VCH)
If considering use of bio-crude as a transportation fuel, upgrading is needed.

Upgrading of bio-crude from hydrothermal conversion demands removal of
heteroatoms like oxygen, nitrogen and sulphur in the oil. The amount of hetero
atoms in the oil highly depends on the feedstock being used. Lignocellulosic
biomasses (e.g. like straw shown in the table above) with a low sulphur and nitrogen
content will produce bio-crudes low in nitrogen and sulphur. Sulphur rich feedstock,
like e.g. black liquor a leftover product from the paper industry, will produce a bio-
crude with higher amounts of sulphur containing compounds.
An often discussed characteristic of renewable oils or crudes, be it pyrolytic or
hydrothermally produced is the H/C and O/C ratio in the liquid. The higher the H/C
ratio and the lower the O/C ratio in the oil, respectively, the higher the heating value
and the quality of the oil will be. A comprehensive illustration of H/C and O/C ratio
of conventional and renewable fuels has been published by Kersten et al. [38] shown
in Fig. 14.7. The red arrows shows the intended change during upgrading, oxygen
has to be removed and hydrogen as to be added to the crude.
Table 14.5 Qualitative characterization of bio-crude from hydrothermal conversion (volatile

organic fraction) by GC-MS reported in [39, 40]
Biomass
Compounds Softwood Xu, 2008 [33] Straw Yuan, 2009 [34]
Hydrocarbon
Alcohols
Phenols
Guaiacols & Syringols
Acids
Aldehydes & Ketones
Ethers & Esters
Furans
Carbohydrates
N-compounds
When studying the detail of upgrading possibilities of bio-crudes, analysis of the

crude are necessary. HTC bio-crude is a complex mixture of components. It makes
detailed characterization and analyses of the oil difficult. Most characteristics on
bio-crudes reported in literature are reported in terms of chemical compounds based
on GC/MS analyses results. These results represent only the volatile organic fraction
of bio-crude. Besides this, the bio-crude consists of non-volatile organics, high
molecular weight material and water. Two lignocellulosic feedstock examples have
been chosen from literature and results of qualitative GC/MS measurement of these
examples is summarized in Table 14.5. Xu and Lad studied direct liquefaction of
jack pine sawdust (softwood) in sub/near-critical water at temperatures of 280–
380 ı C without and with catalysts (alkaline earth and iron ions). The heavy oil
product has been analyzed using GC/MS. The GC/MS measurements for the heavy
oil products revealed that the HOs consist mainly of phenolic compounds and
derivatives, long-chain carboxylic acids/esters and hydrocarbons, which are the
decomposition products from lignin and cellulose. The greatly reduced oxygen
content; 40.4 wt.-% in the crude wood sample versus 15.17–23.2 wt.-% in the heavy
oil product, may partially be due to the dehydration reactions (forming H2 O) and de-
carbonylation and de-carboxylation reactions (forming CO/CO2 ) in the liquefaction
process [39]. Yuan et al. studied the bio-oil structure of thermochemical liquefaction
of straw by hot compressed water. The temperature range for conversion was 200–
310 ı C. The obtained bio-crude was analyzed by GC/MS. The results suggest that
the main component of the biomass feedstock including cellulose and hemicellulose
of straw began to decompose at 200 ı C, but the lignin decomposed at 250–300 ı C.
The main compounds of bio-oil were butylated hydroxytoluene and dibutyl phtha-
late; some high molecular compounds were produced by further repolymerization
as the temperature increased. A comparison of the bio-crude characteristics of both
studies is given in Table 14.5.
14.4 Hydroprocessing of Bio-crudes
Like mentioned above upgrading of bio-crudes is needed. To use bio-crudes as

transportation fuels the characteristics of the oils have to be improved. Hydropro-
cessing is one way of upgrading conventional crude oils and an adaption for
bio-crudes or co-processing of bio-crudes in existing refinery infrastructures seems
attractive.
Hydroprocessing is a generic term that describes the use of hydrogen and an
appropriate catalyst to remove undesired components form refinery streams. In
refinery hydroprocessing it is used to remove reactive compounds like olefins,
sulphur, nitrogen, but also oxygen from crudes. Mentioned heteroatoms are removed
as hydrogensulfide, ammonia and water, respectively. Lower temperature pro-
cesses are referred to as hydrotreatment. At higher temperatures and pressures,
aromatic rings can be saturated and all sulphur and nitrogen is removed from
the feed. Hydroprocessing at higher temperatures and pressures is referred to as
hydrocracking. Large molecules are broken down and saturated with hydrogen.
In standard refineries this process is used for cracking of vacuum gas oil (VGO).
Hydrocracking in standard refineries yields a high percentage of products in the
diesel and kerosene boiling range. If upgrading of bio-crude under standard refinery
conditions can be achieved, existing infrastructure and available know-how can be
used. An objective in bio-crude and biofuel production should be to investigate
direct adaption possibilities of conventional crude oil upgrading procedures. This
would include the development of standards for the characterisation of bio-crude
comparable and according to ASTM standards.
During hydrotreatment the following hydrogenolysis reactions are involved:
Hydrodeoxygenation (HDO)
.CH2 O/ CH2 ! .CH2 / CH2 O (14.3)
Hydrodesulfurization (HDS)
.CH2 S/ CH2 ! .CH2 / CH2 S (14.4)
Hydrodenitrogenation (HDN)
.CH2 N/ C1:5H2 ! .CH2 / CNH3 (14.5)
Furthermore, isomerisation, cyclization and hydrolysis reactions occur.

14.5 Oxygen Removal from HTC Bio-crudes:

Hydrodeoxygenation
Knowing the bio-oil composition and possible hydrotreating reaction products, the
upgraded liquid composition can be estimated. A wide range of upgrading process
parameters (temperature: 250–400 ı C, pressure: 10–12 MPa) and catalyst are used
in practice due to oxygenated compounds diversity in biofeeds and their functional
group reactivity. Ordering the chemical compounds based on their Hydrodeoxy-
genation (HDO) reactivity is difficult. A classification based on activation energies
will be specific for the catalyst used in the upgrading process. A review on methods
used to define reactivity trends is presented by Furimsky [41] and a tentative order
for the HDO reactivity of O-containing group is established as:
alcohol > alkylether > carboxylic acid m-; p-phenol naphtol > phenol >
diarylether o-phenol alkylfuran > benzofuran > dibenzofuran:
Hydrogen consumption is also an important parameter indicating the increase

in H/C ratio of the bio-oil and consequently an improvement of the HHV value.
It is also an indicator for the extend of hydrogenation reactions taking place.
In Fig. 14.8, oxygenated compounds are ordered based on their HDO reactivity
and H2 consumption per group or molecule based on Mortensen et al. [42]. The
reactivity trend is defined as the overall conversion of an O-compound into a
hydrocarbon rather than into an oxygen containing intermediate [41]. The hydrogen
consumption of a functional group or molecule also refers to complete conversion
into hydrocarbon.
14.6 HDO of Model Compounds
Literature studies on conversion of model compounds found in bio-crude have been

done. Model compound reactions are summarized in Fig. 14.9.
14.6.1 Phenols
Substituted phenols are predominant phenolic compounds in bio-oils. They also

play an important role in overall HDO mechanism as intermediates. For alkyl
substituted phenols, -OH elimination is more difficult compared to simple phenol.
Among substituted phenols, o-alkyl phenols have the most adverse effect on
HDO [43].
HDO of alkyl phenols may follow two paths: (1) direct HDO (DDO) resulting
in alkyl benzenes formation or (2) hydrogenation (HYD) of benzene ring followed
High Low
Alcohol
R OH
Alcohol R OH
1 H2
O
Ketone
O
C
R1 R2 Ketone C
2 H2 R1 R2
Hydrogen consumption for the deoxygenation

HDO reactivity of O-containing compounds
Ether
O
R1 R2 O
OH Carboxylic
OH
O
acid C
Carboxylic
acid 3 H2 R OH
m-,p-phenol C R
Naphthol R OH OH OH
m-,p- Phenols
OH 4 H2 R
Phenol o-,m-,p-
OH
R O
Diaryl ether
O OH
O-phenol
Furan
Ar Ar Methoxy
phenol
OCH3
6 H2
BF
O
O O
DBF
DBF
8 H2
Low High
Fig. 14.8 HDO reactivity of different bio-crude compounds
by water elimination resulting alkyl cyclohexane or alkyl cyclohexene. DDO route

dominate the overall HDO when moderate H2 pressures are employed [43].
Odebunmi and Ollis [44] investigated cresols HDO under 225–400 ı C and
6.8 MPa, using CoO-MoO3 /Al2 O3 supported catalyst. The major products iden-
tified were toluene and methylcyclohexane. On this catalyst, the reactivity of
methyl substituted phenols was found: meta > para > ortho. At lower temperatures,
methylcyclohexane was obtained trough consecutive reactions: cresol–> toluene–
> methylcyclohexane. At high temperatures 350–400 ı C, cresols were directly
converted to methylcyclohexane. Other authors suggest an alternative pathway
where cresol is first hydrogenated and then hydroxyl group is eliminated as H2 O.
But ring saturation prior to hydrodeoxygenation is not a necessary step [44].
14.6.2 Guaiacols
Guaiacols are compounds of great interest due to their abundance in ligno-cellulosic

bio-oils and their resistance to HDO treatment. Several studies were carried out on
guaiacol hydrotreatment.
Nimmanwudipong et al. [45] studied the catalytic conversion of guaiacols on
Pt/Al2 O3 catalyst. Their experiments were conducted under mild conditions, 300 ı C
and near atmospheric H2 pressure. Under these conditions three main classes of
reactions leads to phenol, catechol (benzene-1,2-diol) and 3-methylcatechol as
predominant products. The reactions involved were: transalkylation, hydrodeoxy-
genation and hydrogenation. A detailed reaction scheme was proposed and it was
noticed that during HDO reactions, oxygen is removed as H2 O or CH3 OH. Products
are formed as a result of three different C-O bonds cleavage. CAL-O bond from
methoxy group is the weakest, thus bi-phenols (e.g. catechol) and methyl-catechol
are major intermediates. Phenol intermediates may also be formed by CAR-OCH3
bond cleavage which has a lover selectivity comparing with CAL-O bond. Anisol
was found in small amounts, thus -OH is difficult to remove under these conditions,
breaking CAR-OH bond.
Further conversion into hydrocarbons involves dehydration of phenols (com-
pletely or partially) or hydrogenation. As indicated in other studies, this require
more severe conditions (300 ı C & 8 MPa, 300 ı C & 5 MPa, 320 ı C & 17 MPa).
Final products are aromatic or cyclic hydrocarbons like cyclohexane, cyclohex-
anone, benzene or toluene. In these experiments various catalysts and supports were
tested: CoMo/Al2 O3 , ZrO2 /Pd, ZrO2 /Pt, ZrO2 /Rh, ZrO2 /PtRh, Ni/ZrO2 Ni5 Cu/ZrO2
etc. [46, 47].
14.6.3 Ethers
Ethers are found in small amounts in bio-oils composition comparing with other
oxygenated compounds, as is shown in Sect. 14.1. They usually contain benzene
rings. Aryl ethers are of higher interest because are more difficult to convert into
hydrocarbons.
A review on diphenylether HDO is given by Furimsky [41]. It is mentioned that
under 400 ı C and 6.9 MPa H2 pressure, diphenylether is converted first into phenol
and benzene. When temperature is increased, phenol dehydration and hydrogenation
of benzene occurs. Both benzene and cyclohexane are final products along with
small amounts of cyclohexane, resulted from cyclohexane isomerisation.
To show the higher resistance of CAR-O-CAR bond comparing with CAL-O-
CAL bond, Shukla’s [48] review on dibenzylether states that hydrocarbons are
obtained (toluene) at much lower temperature, 200 ı C.
14.6.4 Furans
The furanic structures constitute an important fraction of O-containing compounds

in bio-oils [49]. They are one of the most difficult oxygenated compounds to convert
into hydrocarbons [42].
Furan degradation was studied by Furimsky [49] under 400 ı C and near
atmospheric H2 pressure and reduced or sulphided catalyst. C2, C3 and C4 saturated
and unsaturated hydrocarbons formation was reported. Two reactions paths are
proposed: (1) butadiene formation (cleavage of both C-O ring) followed by a rapid
hydrogenation and (2) partial hydrogenation of furan (one of the C-O bond cleavage)
followed by hydrocracking of C-C bond attached to O heteroatom. Sulphided
catalyst increases furans HDO conversion [43].
Benzofuran (BF) is one of the major heterocyclic compounds found in coal
liquids. Benzofuran reactions occur in two steps: first, the oxygen containing ring
is hydrogenated yielding phenol and second, the oxygen is removed at higher
temperatures as H2 O. Hydrogenation occurs at temperatures above 200 ı C. At low
temperatures the hydrogenation takes place only at the double bond of 5 carbon (5C)
ring. Ring opening (5C ring) and o-ethylphenol formation occurs when temperature
increases. Phenol dehydration starts at 310 ı C and the major products identified are
ethylbenzene, ethyl-cyclohexene and ethylcyclohexane. Ethyl group expulsion was
scarcely observed [50].
Dibenzofuran (DBF) experiments were conducted by LaVopa and Satterfield
[51] at 350–390 ı C and 7 MPa total pressure, with sulphided catalyst. DBF
HDO mechanism is different than furan or benzofuran. The products isolated
were mainly hydrocarbons. Also no linear alkanes were found. Around 75 % of
hydrocarbons were single ring, predominating cyclohexane. Other compounds were
methylcyclopentane, cyclopentane, benzene, methylcyclohexane and cyclohexene.
The presence of H2 S has a positive effect on single ring products formation.
14.6.5 Furfural
Furfural is a highly reactive compound found in bio-oils and obtained by dehy-

dration of pentoses. The furfural behavior is different on various metal catalysts.
For example, on Cu/SiO2 furfuryl alcohol is obtained (by hydrogenation) while
on Pd/SiO2 the product is furan (by decarbonylation). In Ni/SiO2 presence, ring
opening and hydrogenation occur and lead to butane formation [52].
14.6.6 Carboxylic Acids and Esters
Carboxylic acids and esters are representative compounds found in all bio-oils
studied in Sect. 14.1. GS-MS analyses results are in accordance with Furimsky [43]
which states that small molecules are found in cellulosic oil (C1 to C8) while in
vegetable oil and algae bio-oil, long chain of acids (fatty acids) and esters are present
(C18).
Hydrogenolysis of esters bound results in formation of carboxylic acid and
alcohol. De-esterification (DES) of two ester group is usually achieved in two steps
[43].
There are two possible routes of carboxylic acids conversion to hydrocarbons:
(1) decarboxylation (-CO2 ) and (2) hydrogenation of carboxylic group followed by
water elimination. The second path is more selective for high enough H2 pressure
and it was found to predominate at 12.5 MPa and 400 ı C. CO formation was also
noticed thus decarbonylations reactions coexist with decarboxylation [37].
14.6.7 Ketones and Aldehydes
Cyclohexanone HDO was studied by Kong et al. [53] and the HDO mechanism
was similar to other linear or cyclic aliphatic ketones. Their experiments show that
aliphatic ketones are converted into the corresponding alkanes over alkali treated
Ni/HZSM-5 catalyst, at 160 ı C and 2 MPa H2 pressure. A two step mechanism is
proposed: first, hydrogenation of CDO bond and second, dehydration of alcohol.
The same authors studied previously the reactions over Ni/Al2 O3 at mild conditions
(160 ı C) and obtained as major compounds the ketones corresponding phenols.
Thus, the catalyst plays an important role in activation energy reduction.
Aldehydes are important intermediates in esters HDO mechanism. There are
two main pathways which may lead to hydrocarbons formation: HDO and decar-
bonylation (DCO). At 250 ı C and 1.5 MPa it was noticed that hydrogenation
of CDO bound is more favored than decarbonylation pathway. Thus, similar to
ketones, HDO route implies aldehydes conversion into corresponding alcohols and
afterwards O removal as H2 O. The alkenes obtained are rapidly hydrogenated to
alkanes. DCO mechanism leads to one carbon less alkanes by CO elimination from
molecule [54].
14.6.8 Alcohols
Alcohol dehydration leads to alkenes formation. Usually, the reaction is acid

catalyzed and the temperature required vary with alcohol type: tertiary alcohols
50–80 ı C, secondary alcohols 100–140 ı C, primary alcohols 170–180 ı C. If
the temperatures are not high enough, primary alcohols give symmetrical ethers.
Secondary and tertiary alcohols, generally give only elimination products. Tertiary
alcohols easily eliminate water even in small amounts of acids [55].
Knozinger et al. [56] studied alcohols dehydration over Al2 O3 catalyst. Treated
on this catalyst, secondary alcohols lead to alkenes at temperature range 157–180 ı C
while primary alcohols were dehydrated at temperatures between 188 and 351 ı C
with temperature increasing for the most branched alcohols.
14.6.9 Fatty Acids
Fatty acids were found especially in algae derived bio-oils and their presence
is related to biomass lipids content. However, previous studies also showed that
fatty acids can also be formed from extractives of the lignocellulosic biomass
under hydrothermal conditions. Because their structure is alike carboxylic acids
it is expected that oxygen removal to follow the same routes. Snåre et al. [57]
studied several fatty acids degradation over Pd/C catalyst under 15–27 bar and
300–360 ı C. Long chain hydrocarbons were obtained by hydrogenation of double
bonds first, followed by decarboxylation. De-carboxylation and de-carbonylation
reactions occur especially in inert atmosphere. Under high H2 pressure, HYD and
HDO dominate the overall mechanism [43]. Through any reaction path presented,
fatty acids may be converted into long chain hydrocarbons.
14.7 Kinetics of HDO
For the HDO reactions described in Sect. 14.6, Arrhenius parameters and some
kinetic constant (k) values reported in several studies are summarized in Table 14.6.
Pre-exponential factor A was calculated from linearised Arrhenius equation, using
reported values of k or ln k and T or 1/T. Often k and Ea are reported as apparent
values.
The data presented in Table 14.6 are based on kinetic studies on mixtures
of reactant, solvent (e.g. n-hexadecane, n-tetradecane, n-heptadecane) and small
amounts of other compounds used for example to prevent catalyst ageing. Differ-
ences between values reported by different authors indicate that a special attention
need to be paid for the experimental conditions employed. Major influence in kinetic
results has the initial concentration of reactant (C0M ), the catalyst (form, type),
experimental parameters (temperature, pressure), reactor type, the solvent etc. Most
of these are included in Table 14.6 for the corresponding Arrhenius parameters.
Reaction rate constant k is catalyst weight based. Most of the reactions are
pseudo-first order with unit [l/h gcat. ], excepting for guaiacol and ketones which
are first order reactions.
In order to compare the compounds reactivity, k values at 300 ı C were calculated
with Arrhenius parameters listed in Table 14.6. In agreement with HDO reactivity
order proposed by Furimsky [35] and presented in Sect. 14.6, Fig. 14.10, ketones
are the most reactive compounds while furans reactions are much slower.
394
Table 14.6 Arrhenius parameters of common bio-oil compounds HDO reactions

Arrhenius parameters Experimental conditions
ı
Model Temperature k300 C
HDO
a 0 Œmol= l ı
compound ! Product Ea [kJ/mol] lnA A units CM pH2 [MPa] range [ C] Catalyst [l/h gcat. ] Ref.
Phenol ! Benzene 126 (apparent) 22.99 [l/h gcat. ] 7.2 102 225–275 NiMo 3.12 102 [58]
o-cresol ! Toluene 96.23 (apparent) 11.1 (10.1?) [l/h gcat. ] 0.15 6.8 350–400 CoMo 1.1 104 [44]
m-cresol ! Toluene 121.3 (apparent) 22.7 [l/h gcat. ] 0.15 6.9 250–350 CoMo 6 102 [59]
112.96 (apparent) 15 [l/h gcat. ] 0.15 6.8 350–400 CoMo 1.62 104 [59]
p-cresol ! Toluene 154.80 (apparent) 21.9 [l/h gcat. ] 0.15 6.8 350–400 CoMo 2.46 105 [44]
Guaiacol ! Catech. C ph. 112 (apparent) 22.74 [1/h gcat. ] 0.232 7 260–300 CoMo 4.62 101 [60]
111 (apparent) 22.82 [1/h gcat. ] 0.232 7 260–300 NiMo 6.90 101 [60]
Diphenyl ! Hydroc. 148 – – – – – CoMo – [41]
Ether
! Ph C C6H6 86.60 16.94 [l/h gcat. ] 5.8 102 6.89 220–300 CoMo 2.27 101 [48]
Dibenzyl ! Toluene 65.53 13.10 [l/h gcat. ] 0.25 6.89 110–200 CoMo 4.5 101 [48]
Ether
BF ! Hydroc. 106 13.7 [l/h gcat. ] 0.6 ptot D 3.5 300–400 NiMo 1.92 104 [61]
138 (apparent) 21.46 [l/h gcat. ] 0.15 6.9 310–345 CoMo 5.35 104 [50]
DBF ! Hydroc. 67 (apparent) 8.86 [l/h gcat. ] 0.245 ptot D 7 350–390 NiMo 3.69 102 [51]
Ketones ! Hydroc 50 – [1l/h gcat. ] – – 2.6– [62]
3.24(250 ı C)
a
lnA calculated values from linearized Arrhenius equation
J. Hoffmann et al.
Reactant Intermediates Possible products Experimental

Name Structure (Mild Conditions) (Severe Conditions) Conditions
OH OH OH2
HYD HDO HYD

[37]
Phenol Cyclohexane
OH
Phenols R
DDO(HDO) HYD
R
CH3
p=6.8 MPa
HY
D
t=225-275°C
Cresols
[35,38]
OH OH
p=6.8 MPa
D Methyl-
HY t=350-400°C
R HYD HDO cyclohexane
R R
OH
OH
DDO HYD [37,35,39]
Cyclohexane
OH O P=140
HD
E
CH3
M
O
kPa
D
O
OH
HYD CH3
t=300°C
DMO (HDO)
Guaiacol Methyl-
cyclohexanone p=5 MPa
OH CH3
CH3
t=300°C
HD
O
CH3
E
p=8 MPa
DM
OCH3 Xylene t=300°C

OCH3
HYD
p=17 MPa
DM
t=320°C
Tran
O(
HD
O)
salkyl
Methoxy
ation
OH cyclohexan
OH OH
HDO
CH3 CH3
Ethers
[42,35]
O CH3
HDO
dibenzylether Toluene
p=6.9-7 MPa
O
OH t=300-400°C
HYD HDO HYD
Izo.
diphenylether
CH3
Methyl-
cyclopentane
Fig. 14.9 Reaction schemes for removal of oxygen during hydrodeoxygenation

Reactant Intermediates Possible products Experimental

Name Structure (Mild Conditions) (Severe Conditions) Conditions
O
HDO [43,35]
H 2C CH 2
Ethene
Furan
H3 C CH3 p=1 atm
Propane
t=400°C
Furans H3 C CH2
Propene
OH CH2 CH3
O O HYD CH2 CH3
HYD
HDO Ethyl-benzene p=6.9 MPa
CH2 CH3 [44,35]
Benzofuran t>310°C
Ethyl cyclohexene
CH2 CH3
Ethyl cyclohexane
O OH CH3
HYD HDO
Dibenzofuran
HY
CH3 [45,35]
D(
hy
dro
P=7 MPa
gen
OH t=350-390°C
oly
sis
HDO
)
Dicyclohexyl
O CPMCH
OH OH
O O
HYD HYD O
Furfural
p=1 atm
Furfuryl alcohol t=230-290°C [46]
DCO O HYD HDO
OH
Butane
O OH
HYD HDO HYD
R1 C R2 R1 CH R2 R1 CH R2 R1 CH 2 R2 p=2 MPa [47]
Ketone t=160°C
&
Aldehyde O OH
HYD HDO
R1 C H R1 CH H R1 CH2 HYD R1 CH 3 P=1.5 MPa [48]
t=250°C
O O
DES
R1 C O R2 R1 C OH +R 2 OH
[37,52]
O p=6-9 MPa
DCO CH 3 CH 2
CH3 CH2 C CH3 t=200-250°C
x x-1
OH HDO
CH3 CH2 CH3
x p=7.5 MPa [53]
t=250°C
Esters O
HYD
CH3 CH2 CH CH3 CH2 CH2 OH
& x x p=12.5 MPa
D
C
-C
Carboxylic acids
O
O
t=400°C
CH3 CH 2 CH3
x-1
Fig. 14.10 Reaction schemes for removal of oxygen from bio-crude compounds (HYD hydro-
genation, HDO hydrodeoxygenation, DCO decarboxylation)
14.8 Conclusions
The chapter discussed conversion of ligno-cellulosic biomass through hydrothermal

treatment and the upgrading possibilities and pathways to obtain a valuable
transportation fuel. The intention of the first section is to give an idea of how the
ligno-cellulosic biomass composition (share of lignin, cellulose and hemicellulose)
influences the product distribution in the hydrothermal conversion product and
opens the discussion if intelligent mixing of biomass feedstock can maximize the
bio-crude quality and yield.
The second section presents characteristics of bio-crude from HTC found in
literature and upgrading strategies and pathways, based on compounds present
in bio-crudes. Hydrodeoxygenation is discussed in detail and kinetic data found
in literature are presented.
References
1. International Energy Agency. World energy outlook 2011; 2011 IEA. doi:10.1787/weo-2013-
en.
2. U.S. Energy Information Administration. Annual energy outlook 2013. EIA. DOE/EIA-
0484(2013); 2013.
3. Friedl A, Padouvas E, Rotter H, Varmuza K. Prediction of heating values of biomass fuel from
elemental composition. Anal Chim Acta. 2005;544:191–8.
4. Huber GW, Iborra S, Corma A. Synthesis of transportation fuels from biomass: chemistry,
catalysts, and engineering. Chem Rev. 2006;106:4044–98.
5. Huber GW, Chheda JN, Barrett CJ, Dumesic JA. Production of liquid alkanes by aqueous-phase
processing of biomass-derived carbohydrates. Science. 2005;308:1446–50.
6. Kleinert M, Barth T. Towards a lignincellulosic biorefinery: direct one-step conversion of lignin
to hydrogen-enriched biofuel. Energy Fuel. 2008;22:1371–9.
7. Huber GW, Dumesic JA. An overview of aqueous-phase catalytic processes for production of
hydrogen and alkanes in a biorefinery. Catal Today. 2006;111:119–32.
8. Lee S, Shah YT, editors. Biofuels and bioenergy processes and technologies. Boca Raton: CRC
Press; 2013.
9. Demirbaş A. Relationships between lignin contents and heating values of biomass. Energy
Convers Manag. 2001;42:183–8.
10. Saha BC. Lignocellulose biodegradation and applications in biotechnology. ACS. 2004;889:
2–34.
11. Chatel G, Rogers RD. Review: oxidation of lignin using ionic liquids – and innovative
strategy to produce renewable chemicals. ACS Sustain Chem Eng. 2013;2(3):322–39.
doi:10.1021/sc4004086.
12. Sasaki M, Furukawa M, Minami K, Adschiri T, Arai K. Kinetics and mechanism of cellobiose
hydrolysis and retro-aldol condensation in subcritical and supercritical water. Ind Eng Chem
Res. 2002;41(26):6642–9.
13. Bobleter O. Hydrothermal degradation of polymers derived from plants. Polym Sci.
1994;19:797–841.
14. Watanabe M, Aizawa Y, Toru I, Nishimura R, Inomata H. Catalytic glucose and fructose
conversions the TiO2 and ZrO2 in water at 473 K: relationship between reactivity and acid-
base property determined by TPD measurement. Appl Catal A Gen. 2005;295:150–6.
15. Kabyemela BM, Adschiri T, Malaluan RM, Arai K. Kinetics of glucose epimerization and
decomposition in subcritical and supercritical water. Ind Eng Chem Res. 1997;36(5):1552–8.
hydrolysis in subcritical and supercritical water. J Supercrit Fluid. 1998;13:261–8.
17. Lü X, Saka S. New insights on monosaccharides’ isomerization, dehydration and fragmenta-
tion in hot-compressed water. J Supercrit Fluid. 2012;61:146–56.
18. Rosatella AA, Simeonov SP, Frade RFM, Afonso CAM. 5-hydroxymethylfurfural (HMF) as
a building block platform: biological properties, synthesis and synthetic applications. Green
Chem. 2011;13:754–93.
19. Guohong T, Ritchie D, Xu H. DMF – a new fuel candidate. In: Bernardes MADS, editor.
Biofuel production – recent developments and prospects. Rijeka, Croatia: InTech; 2011.
p. 487–520.
20. Pińkowska H, Wolak P, Złocińska A. Hydrothermal decomposition of xylan as a model
substance for plant biomass waste – hydrothermolysis in subcritical water. Biomass Bioenergy.
2011;35(9):3902–12.
21. Aida TM, Shiraishi N, Kubo M, Watanabe M, Smith Jr RL. Reaction kinetics of d-xylose in
sub- and supercritical water. J Supercrit Fluid. 2010;55(1):208–16.
22. Jing Q, Lü X. Kinetics of non-catalyzed decomposition of D-xylose in high temperature liquid
water. Chin J Chem Eng. 2007;15(5):666–9.
23. Oefner PJ, Lanziner AH, Bonn G, Bobleter O. Quantitative studies on furfural and organic acid
formation during hydrothermal, acidic and alkaline degradation of D-xylose. Monatsh Chem.
1992;123(6–7):547–56.
24. Antal Jr MJ, Leesomboon T, Mok WS, Richards GN. Mechanism of formation of
2-furaldehyde from d-xylose. Carbohydr Res. 1991;217:71–85.
25. Gao Y, Chen-Han-Ping, Wang-Jun, Shi-Tao, Yang-Hai-Ping, Wang-Xian-Hua. Characteriza-
tion of products from hydrothermal liquefaction and carbonation of biomass model compounds
and real biomass. J Fuel Chem Technol. 2011;39(12):893–900.
26. Sasaki M, Hayakawa T, Arai K, Adichiri T. Measurement of the rate of retro-aldol condensation
of D- xylose in subcritical and supercritical water. Presented at the proceeding of the 7th
international symposium hydrothermal reactions; 2003. p. 169–76.
27. Lange J, van der Heide E, van Buijtenen J, Price R. Furfural – a promising platform for
lignocellulosic biofuels. ChemSusChem. 2012;5:150–66.
28. Zakzeski J, Bruijnincx PCA, Jongerius AL, Weckhuysen BM. The catalytic valorization of
lignin for the production of renewable chemicals. Chem Rev. 2010;110:3552–99.
29. Penninger JML, Kersten RJA, Baur HCL. Hydrolysis of diphenylether in supercritical water:
effects of dissolved NaCl. J Supercrit Fluid. 2000;17:215–26.
30. Penninger JML, Kersten RJA, Baur HCL. Reactions of diphenylether in supercritical water –
mechanism and kinetics. J Supercrit Fluid. 1999;16:119–32.
31. Roberts VM, Knapp RT, Li X, Lercher JA. Selective hydrolysis of diphenyl ether in supercrit-
ical water catalyzed by alkaline carbonates. ChemCatChem. 2010;11:1407–10.
32. Hashaikeh R, Fang Z, Butler IS, Hawari J, Kozinski JA. Hydrothermal dissolution of willow in
hot compressed water as a model for biomass conversion. Fuel. 2007;86:1614–22.
33. Miller JE, Evans L, Littlewolf A, Trudell DE. Batch microreactor studies of lignin and lignin
model compound depolymerization by bases in alcohol solvents. Fuel. 1999;78:1363–6.
34. Cheng S, D’cruz I, Wang M, Leitch M, Xu C. Highly efficient liquefaction of woody biomass
in hot-compressed alcohol-water co-solvents. Energy Fuel. 2010;24:4659–67.
35. Zhao C, He J, Lemonidou AA, Li X, Lercher JA. Aqueous-phase hydrodeoxygenation of bio-
derived phenols to cycloalkanes. J Catal. 2011;280:8–16.
36. Furimsky E. Hydroprocessing challenges in biofuels production. Catal Today. 2012;217:13–56.
37. Mortensen PM, Grundwaldt JD, Jensen PA, Knudsen KG, Jensen DA. A review of catalytic
upgrading of bio-oil to engine fuels. Appl Catal Gen. 2011;407(1–2):1–19.
38. Kersten SRA, van Swaaij WPM, Lefferts L, Seshan K. Options for catalysis in the thermo-
chemical conversion of biomass into fuels. In: Centi G, van Santen RA, editors. Catalysis for
renewables: from feedstock to energy production. Chichester: Wiley-VCH; 2007.
39. Xu C, Lad N. Production of heavy oils with high caloric values by direct liquefaction of woody
biomass in sub/near-critical water. Energy Fuel. 2008;22:635–42.
40. Yuan XZ, Tong JY, Zeng GM, Li H, Xie W. Comparative studies of products obtained
at different temperatures during straw liquefaction by hot compressed water. Energy Fuel.
2009;23:3262–7.
41. Furimsky E. Catalytic hydrodeoxygenation. Appl Catal A Gen. 2000;199:144–90.
42. Mortensen PM, Grundwaldt JD, Jensen PA, Knudsen KG, Jensen AD. A review of catalytic
upgrading of bio-oil to engine fuels. Appl Catal A Gen. 2011;407:1–19.
43. Furimsky E. Hydroprocessing challenges in biofuels production. Catal Today. 2013;217:13–56.
44. Odebunmi E, Ollis D. Catalytic hydrodeoxygenation. I. Conversions of o-, p-, and m-cresols.
J Catal. 1983;80:56–64.
45. Nimmanwudipong T, Runnebaum RC, Block EC, Gates BC. Catalytic reactions of guaiacol:
reaction network and evidence of oxygen removal in reactions with hydrogen. Catal Lett.
2011;141:779–83.
46. Gutierrez A, Kaila RK, Honkela ML, Slioor R, Krause AOI. Hydrodeoxygenation of guaiacol
on noble metal catalysts. Catal Today. 2009;147:239–46.
47. Zhang X, Wang T, Ma L, Zhang Q, Yu Y, Liu Q. Characterization and catalytic properties of Ni
and NiCu catalysts supported on ZrO2 –SiO2 for guaiacol hydrodeoxygenation. Catal Commun.
2013;33:15–9.
48. Shukla YV. Catalytic hydrodeoxygenation studies of aromatic ethers, phenols and oxygen-rich
synfuels. PhD thesis. Salt Lake City: The University of Utah, Department of Fuels Engineering;
1985.
49. Furimsky E. The mechanism of catalytic hydrodeoxygenation of furan. Appl Catal.
1983;6:159–64.
50. Lee C-L, Ollis DF. Catalytic hydrodeoxygenation of benzofuran and o-ethylphenol. J Catal.
1984;87:325–31.
51. LaVopa V, Satterfield CN. Catalytic hydrodeoxygenation of dibenzofuran. Energy Fuel.
1987;1(4):323–31.
52. Sitthisa S, Resasco DE. Hydrodeoxygenation of furfural over supported metal catalysts: a
comparative study of Cu, Pd and Ni. Catal Lett. 2011;141:784–91.
53. Kong X, Lai W, Tian J, Li Y, Yan X, Chen L. Efficient hydrodeoxygenation of aliphatic ketones
over an alkali-treated Ni/HZSM-5 catalyst. ChemCatChem. 2013;5:2009–14.
54. Ruinart de Brimont MR, Dupont C, Daudin A, Geantet C, Raybaud P. Deoxygenation
mechanisms on Ni-promoted MoS2 bulk catalysts: a combined experimental and theoretical
study. J Catal. 2012;286:153–64.
55. Streitwieser A, Heathcock CH, Kosower EM. Introduction to organic chemistry. New York:
Macmillan; 1992.
56. Knozinger H, Buhl H, Kochloefl K. The dehydration of alcohols on alumina XIV. Reactivity
and mechanism. J Catal. 1972;24:57–68.
57. Snåre M, Kubičková I, Mäki-Arvela P, Chichova D, Eränen K, Murzin DY. Catalytic
deoxygenation of unsaturated renewable feedstocks for production of diesel fuel hydrocarbons.
Fuel. 2008;87:933–45.
58. Chon S, Allen DT. Catalytic hydroprocessing of chlorophenols. AIChE. 1991;37:1730–2.
59. Odebunmi E, Ollis D. Interactions between catalytic hydrodeoxygenation of m-cresol and
hydrodenitrogenation of indole. J Catal. 1983;80:76–89.
60. Laurent E, Delmon B. Study of the hydrodeoxygenation of carbonyl, carboxylic and guaiacyl
groups over sulfide CoMo/Al2O3 and NiMo/Al2O3 catalysts. I. Catalytic reaction schemes.
Appl Catal. 1994;109:77–96.
61. Edelman MC, Maholland MK, Baldwin RM, Cowley SW. Vapor-phase catalytic hydrodeoxy-
genation of benzofuran. J Catal. 1988;111:243–53.
62. Grange P, Laurent E, Maggi R, Centeno A, Delmon B. Hydrotreatment of pyrolysis oils from
biomass: reactivity of the various categories of oxygenated compounds and preliminary techno-
economical study. Catal Today. 1996;29:297–301.
63. Şenol OI, Ryymin E–M, Viljava T-R, Krause A. Reactions of methyl heptanoate hydrodeoxy-
genation on sulphided catalysts. J Mol Catal A Chem. 2007;268:1–8.
64. Ryymin E-M, Honkela ML, Viljava T-R, Outi I, Krause A. Competitive reactions and
mechanisms in the simultaneous HDO of phenol and methyl heptanoate over sulphided
NiMo/G-Al2 O3 . Appl Catal A Gen. 2010;389:114–21.
Chapter 15
Supercritical Water Oxidation (SCWO) of Solid,
Liquid and Gaseous Fuels for Energy
Generation
M. Dolores Bermejo, Ángel Martín, Joao Paulo Silva Queiroz, Pablo Cabeza,
Fidel Mato, and M. José Cocero
Abstract The SCWO (Supercritical Water Oxidation) process is well known for
being able to destroy any kind of compound without producing prejudicial byprod-
ucts. This fact together with the high potential for energy production (because
the high pressure high temperature effluent generated in the process) makes it
a good candidate for generating energy from bio-fuels, especially those which
valorization by conventional combustion can be problematic. In this work, different
literature energetic studies of the SCWO process are analyzed. When comparing
the heat produced by direct expansion of the effluent and by indirect heating steam
generation it is observed that when direct expansion is used, the energetic efficiency
is much higher than when the effluent is used to heat an auxiliary fluid of a Rankine
or Brayton cycle. Nevertheless, the production of energy by direct expansion of
the SCWO is not technically available in the short term. In any case, obtaining a
high temperature effluent it is a key point for optimizing energy utilization. To do
so, reactors in which the effluent is not diluted or reactor working at hydrothermal
flame regime are desirable. Also the lay-out of the plant is important for energy
utilization and factors as preheating scheme must be thoroughly studied. In addition
to all of this, SCWO process has the additional advantage of the possibility of CO2
sequestration.
Keywords Supercritical water oxidation • Energy integration • Power generation

• Biofuel • CO2 sequestration
M.D. Bermejo () • Á. Martín • J.P.S. Queiroz • P. Cabeza • F. Mato • M.J. Cocero
High Pressure Process Group, Department of Chemical Engineering and Environmental
Technology, University of Valladolid, C/Doctor Mergelina s/n, 47011 Valladolid, Spain
402 M.D. Bermejo et al.
15.1 Introduction
Supercritical Water Oxidation (SCWO) is a technique for the complete oxidation of

organic wastes, based on the physical properties of water in supercritical conditions:
a complete miscibility with organics and with oxygen, due to the reduced dielectric
constant of supercritical water, equivalent to those of organic solvents. This
technique stands out by its capacity to achieve very high efficiencies of elimination
of wastes in very short residence times, of the order of minutes or less. Toxic or
xenobiotic wastes that cannot be treated with conventional biological processes can
be eliminated with the SCWO technique. Moreover, this is accomplished in a single-
step process that does not produce toxic effluents, and in particular do not release
gaseous contaminants such as nitrogen oxides or dioxins. Due to these advantages,
considerable efforts have been devoted to the development of SCWO processes, and
this technique has already been tested and operated at industrial scale [1, 2].
Due to these intense development efforts, the limitations and challenges in
the implementation of SCWO processes are well known, including the control
of corrosion and salt precipitation processes [3]. Corrosion and scale formation
are responsible for the need of complex equipment built with specialty corrosion
resistant materials, which imply that the investment costs of SCWO plants are
relatively high [4]. Another well known challenge of SCWO are the energy
requirements, which can be high, particularly if simple plug-flow tubular reactors
are used, since these designs require preheating of the influent to temperatures above
the critical point of water.
However, under appropriate operating conditions, the SCWO process can be
energetically self-sufficient or even produce a net excess of energy [5, 6]. This
excess energy is produced when the oxidation of concentrated wastes with high
energy content is produced. In this way the operation is thermally self-sufficient,
obtaining an energy excess for preheating or even for generating power. To do so, the
correct use of the energy produced by the oxidation is a crucial step in order to make
SCWO processes economically viable. Thus, the use of different energy recovery
schemes (for influent preheating, etc.) is common in full-scale SCWO systems.
The possibilities for the use of energy produced by SCWO processes are
not limited to these simple energy recovery schemes. On the contrary, energy
production processes based on SCWO show considerable potential due to the
peculiar characteristics of this process, including:
SCWO processes produce high-temperature (and thus, high-quality) thermal
energy. Indeed, the oxidation is normally performed at temperatures above 400 ı C.
Due to its high temperature, the thermodynamic efficiency of the conversion of this
thermal energy into electricity is high.
SCWO processes can achieve a complete combustion of a wide range of fuels
without producing gaseous contaminants. This includes low-quality fuels such
as heavy oils or sulphur-containing coal, biomass, biological residues, etc. Thus,
SCWO offers possibilities for an energetic valorization of these low-quality fuels
and residues that would be difficult to achieve using conventional combustion
techniques.
15 Supercritical Water Oxidation (SCWO) of Solid, Liquid and Gaseous. . . 403
The realization in practice of this big potential is far from straightforward.

Several aspects must be considered, including the design of reactors and plants
against possible corrosion and plugging problems, operability and safety issues,
and economic aspects. This chapter presents a review of recent developments
and pending challenges in the development of SCWO systems for the production
of energy. The viability of oxidizing different substances considered biofuels
is discussed in Sect. 15.2. In Sect. 15.3, different energy recovery and energy
production systems theoretically applied to SCWO process are described and the
different variables of the process such as oxidant, preheating scheme etc. are
discussed. In Sect. 15.4 the CO2 capture in energy production processes by SCWO
is discussed. In Sect. 15.5 different technologies patented or already used in industry
in the SCWO for energy recovery are used. It will be seen that the net energy
produced is very variable, depending on the waste, the reaction temperature and
the heat recovery scheme.
15.2 SCWO of Solid, Liquid and Gaseous Bio-fuels
It is well known that SCWO, with the appropriate temperature and residence time, is
able to completely mineralize (totally oxidize to carbon dioxide, water and inorganic
salts and oxides) almost every kind of compound, without generating by-products
typical from conventional combustion [1, 2]. Bio-fuels are not an exception to this
behavior. A bio-fuel can be any type of fuel which energy is derived from biological
carbon fixation. Thus, bio-fuels can include a variety of substances such as pure
processed fuels derived from biomass conversion, such as bio-ethanol or biogas that
are essentially ethanol and methane respectively. Other complex substances such as
various solid biomass, sludge, algae or algae waste, oils, different waste and many
others can be also considered a bio-fuel. It is not necessary to insist on the fact
that the combustion of this last kind of bio-fuel is far much more complicated than
that of the first kind both using conventional combustions or SCWO process. In any
case the peculiarities of each bio-fuel must be considered when selecting the power
generation technology.
In principle, the treatment of gaseous, liquid or solid bio-fuels by SCWO is
technically feasible. However, the treatment of gaseous compounds is hampered
by the need of compressing these gases to the pressures of operation of SCWO
reactors, which implies considerable compression and equipment purchase costs.
Due to this, the treatment of gases by SCWO has been seldom considered. When
dealing with gaseous fuels such as methane, the conventional combustion systems
are easy to implement, and the combustion is cleaner than when dealing with liquid
or specially solid fuels and also the energy recovery system is currently much more
developed, existing combined cycle technologies able to generate power with an
efficiency higher of 60 % [7].
When dealing with processed liquid fuels such as bio-ethanol, methanol, or bio-
fuel, the disadvantage of an expensive compression system is reduced. It is the
simplest situation from a technical point of view, since liquid fuels can be easily
dissolved or emulsified in water, and pressurized together with it using a pump for
liquids. Nevertheless, conventional combustion, despite its disadvantages, can be
still an attractive option.
But when dealing with complex semisolid mixtures, in most cases with high
water content such as biomass, algae or sludge is when SCWO is presented as
a very attractive option over combustion system. From an energy point of view,
the treatment of such solid suspensions by SCWO can be very advantageous if
the initial solid feed contains a certain amount of water. Treatment of wet wastes
by incineration or other alternative techniques normally requires a previous drying
step with a high consumption of energy, while for SCWO processes some water
content in the fuel of course is not a problem. In SCWO the necessity of drying the
biomass or sludge disappears, saving a high amount of money and energy as well
as operational complications. In addition, lower oxidant excess is necessary due to
the good solvating properties of supercritical water and its total miscibility with
oxygen that also avoids the formation of hot spots, achieving that the CO and NOx
production are not a problem. Nevertheless, the treatment of solids adds technical
complications to the SCWO process due to the need of using high pressure pumps
capable of dealing with concentrated solid suspensions.
Abundant experimental results are available discussing the SCWO of different
substances that can be considered a bio-fuel. Even if these studies were not specif-
ically focused on the production of energy, they can provide valuable information
for the design and development of energy production systems based on SCWO.
In Table 15.1, a number of examples of the supercritical water oxidation of
potential bio-fuels are summarized together with the oxidation conditions and the
Total Organic Carbon (TOC) Elimination. These substances include olive oil waste
waters [8–10], waste of distillery [11], municipal sludge [11, 12], food waste [13]
or oily sludge [14]. In Table 15.1 only temperature conditions have been listed.
Pressure conditions are usually between 23 and 25 MPa, because normally no
additional waste elimination is achieved by increasing pressure beyond that point.
The oxidant amount of results summarized in Table 15.1 is also not mentioned as
most of these data were obtained in laboratory scale systems where this parameter
is not optimized. Nevertheless, the amount of oxidant is one of the most important
factors to take into account for energy production and also from the economical
point of view. Thus, it is convenient to optimize this parameter to keep as low
as possible for obtaining total oxidation of the waste/bio-fuel in order to obtain
energetic and economical efficiency.
Sometimes to simplify the study the SCWO of complicated wastes, these are
assimilated to model compounds. For example cellulose can be assimilated by
glucose; lignin by phenol; proteins can be assimilated to ammonia etc. In literature
there are abundant kinetic studies of this kind of compounds that can be used to
make an initial design of an SCWO plant for the oxidation of bio-fuels, i.e. methanol
[15]; phenol [16] or ammonia [17].
Table 15.1 Summary of SCWO conditions of potential bio-fuels
Waste Residence
Biofuel Reference removal (%) Temp. (ı C) time (tR ) (s) TOC
Olive oil wastewater Rivas et al. [8] 99.9 380–500 60 1,200–944 ppm
Olive oil wastewater Chkoundali et al. [9] 200–400
Olive oil wastewater Erkonak et al. [10] 99.96 400–650 5.0–30.0 23.64–291.86 mmol/L
Waste of distillery Goto et al. [11] 99.48 400–500
Municipal sewage sludge Cabeza et al. [12] 99.9 512 21 73,180 ppm
Food wastes (carrots and beef suet) Fang-Ming Jin et al. [13] 97.5 400–450 10–600
Municipal sewage sludge Goto et al. [11] 99.8 400–500
Oily sludge Cui et al. [14] 390–450 60–600
15 Supercritical Water Oxidation (SCWO) of Solid, Liquid and Gaseous. . .
405
15.3 Energy Production in SCWO Processes
It is known that, with appropriate processing conditions, the energy produced by

the combustion reactions in SCWO can provide all the requirements of the process.
Basic theoretical calculations indicate that feeds with an energy content of 930 kJ/kg
(roughly equivalent to an aqueous solution with 2 wt.% of hexane) can supply
enough energy to preheat the feed from room temperature to 400 ı C, and to generate
electric power equivalent to that consumed by the high-pressure pump and the
air compressor [5]. In practice, higher energy contents are required to achieve an
energetically self-sustaining process, due to inefficiencies in the conversion and use
of energy.
The first key parameter that must be considered for the design of an SCWO
system for energy production is the choice of the oxidant. For the reaction point of
view it has been proved that using air or oxygen is not influencing the conversion of
the feed treated [18]. Thus, the election of the oxidant is a purely an economic issue,
depending on most cases of the size of the facility, the energy source availability or
the working time per year of the facility.
Air is of course the cheapest material, but it contains a large proportion of
nitrogen that has to be pressurized using energy-consuming compressors, and that
acts as a diluent that reduces the temperature of effluents, and therefore its thermal
quality for the production of energy.
On the other side of the spectra, cryogenic liquid oxygen carries no diluents; high
electrical consumption of air compressor could be replaced by low consumption
cryogenic pumps. Furthermore, pure oxygen does not need to be preheated up to
feed injection temperature. However, the cost and energy consumption of producing
pure oxygen could affect the viability of the process, i.e. if oxygen is produced in
situ by the consumption in the process is between 0.99 and 0.40 kWh/kg-O2 [19].
An intermediate option is the use of oxygen-enriched air. It is commercially
available at different proportions with oxygen content as high as 95 %. Using rich-
air with 40 % of oxygen, the higher cost of equipment, including compressors and
selective membranes for air enrichment, are compensated by moderate electrical
cost, since the air flow is nearly half of the corresponding flow of atmospheric air.
Table 15.2 summarizes the main advantages and disadvantages of these oxidants.
It is clear that oxidant is the major cost in an SCWO process; nevertheless, it
is difficult to make a general economical estimation of the cost of the oxidant in
the SCWO. The cost of air compression is influenced by the cost, size and cost of
the equipment and the cost of the energy used to operate the compressor: electrical,
Table 15.2 Qualitative Air Rich-air O2 (cryo.) O2 (gas)

comparison among oxidant
alternatives Equipment cost C C C
Operational cost C C
Equipment size C CC CC
Operability C C
Table 15.3 Cost of the oxidant for a facility with a treatment capacity of 200 kg waste/h to be
seated in according to prices level in year 2011
Cost (A
C/year)
O2 O2 95 % Aira
8,000 h/year 60,281 89,670 26,589
2,000 h/year 18,370 78,235 17,342
a
Air compressor is redeemed in 10 years
fuel-oil or even the air can be compressed by a turbo-compressor connected to a

turbine. This energy cost is strongly dependant of the cost of the energy source that it
is highly variable. Oxygen cost depends also of the cost of the pumping/compressing
equipments, but sometimes compressed gas companies can lease the equipment for a
certain period of time. Also the cost of the oxygen itself it is an important parameter
to take into account. It depends mainly of the oxygen consume per year, that fix
the prize that the supplier charges. For large consumes it is even possible to rent a
cryogenic plant for oxygen or enriched air at 95 % oxygen to feed its facility.
As an example in Table 15.3 the cost of oxidant (including equipment cost,
oxygen purchase and electricity cost) for a facility for oxidizing 200 kg/h of
concentrated waste is shown, when the plant is intended to be operated 8,000 or
2,000 h/year.
In Table 15.3 it is observed that in the case of a medium size, while the oxidant
consumption is small air and pure oxygen are similar options while for larger
oxidants consumes air becomes a better alternative.
In literature different arrangements for the production of electrical energy from
SCWO has been proposed. Here they are analyzed as a function of the global energy
production efficiency defined as the electrical energy produced minus the energy
consumed divided by the heat content of the waste, as shown in Eq. 15.1.
Work_produced Work_consumed
Global_Efficiency D (15.1)
Fuel_calorific_value
Lavric et al. [6] studied theoretically the process of SCWO using a tubular
reactor for diluted organic waste. Their objective was to obtain an energetically
self sufficient process for a mobile SCWO plant. To do so they compared several
technologies for power generation from the heat of reaction using the same tubular
SCWO reactor:
1. Closed Brayton Cycle. A Brayton cycle is a thermodynamic cycle using a gas as
a working fluid that consists of compression, an heating in a heat exchanger and
expansion. Lavric et al. [6] considered helium or carbon dioxide as working fluids
and they found that it was not even possible to achieve a sufficient production of
energy to cover energy consumption in the process, unless unrealistic efficiencies
were assumed for compressors and turbines.
2. Supercritical water expansion in a turbine. Using a small supercritical turbine, a
part of the effluent at 650 ı C is used to preheat the effluent, expanding only a
Fig. 15.1 Proposed plant schematic diagram for the energy production using a supercritical
turbine (Reprinted with permission from Lavric et al. [6]. Copyright © 2005, Elsevier)
27.5 % of the effluent to produce electricity, as shown in Fig. 15.1. They report a
production of 154 kW, enough to cover the consumptions of pump (28 kW) and
compressor (45 kW). Having into account that the heat released in the reactor is
673.4 kW, the efficiency in producing electricity is 12 %.
3. Organic Rankine Cycle. Rankine cycle is a thermodynamic cycle using a con-
densable fluid as working fluid that consists of pressurization of the fluid as a
liquid, evaporation by supplying external heating (normally a boiler), expansion
in a turbine and condensation and cooling. It is the thermodynamic cycle
that described the work of a steam turbine. Nevertheless as steam turbines
normally works at temperatures higher than 300 ı C. Organic Rankine Cycles
use organic substances with lower boiling temperatures than water as working
fluids developed in order to use low temperature residual heat.
In the application of an Organic Rankine Cycle [6] the authors proposed to use
the effluent of the reactor at 650 ı C to preheat the feed, and use the cooled effluent
at 250 ı C as a heat source of the cycle as shown in Fig. 15.2. They found several
organic solvents that can be used in the Organic Rankine Cycle covering the power
need of the plant. The best efficiencies were achieved using R123 as working fluid,
and ammonia and isopentane also were suitable choices. The global efficiencies
obtained were between 0.15 and 2.6 % considering the heat released in the reactor.
The results obtained showed that a small scale SCWO plant can be energetically
self-sufficient using either a small supercritical turbine, or an Organic Rankine Cycle
when considering realistic efficiencies for equipment.
Fig. 15.2 Proposed plant schematic diagram for the energy production using an Organic Rankine
Cycle (Reprinted with permission from Lavric et al. [6]. Copyright © 2005, Elsevier)
In 2004, our research group performed a theoretical study of power generation

plant from oxidation of coal by SCWO using a transpiring wall reactor [20]. Two
versions of SCWO power plant were proposed. In the first one coal is grinded in a
wet way to obtain a coal suspension that is introduced into the reactor and mixed
with the air stream and a recirculated water stream, both at supercritical conditions.
These two streams were preheated in a natural gas boiler. The effluent leaves the
reactor at 650 ı C and 30 MPa and then expanded in a steam turbine down to
atmospheric pressure. The second alternative of SCWO power plant considers an
intermediate reheating for partially expanded steam to get a higher efficiency. That
is the partially expanded stream is reheated again to be completely expanded and
obtain a high energy production. The schematic diagrams of the process can be
found in Fig. 15.3a, b respectively. The energetic efficiencies were of 37 % in case
of SCWO power plant and of 40 % in case of SCWO power plant with a reheating
cycle. These efficiencies were found to be more than 4 and 8 % higher than those of
conventional coal power plants, respectively, at the same steam conditions.
It is observed that much higher amount of energy is obtained in this last work
than in the work performed by Lavric et al. [6]. In one hand, the objectives of both
works are different: in the first one it is important to make self sufficient a small
facility [6] and in the other the viability of a full scale power plant it is investigated
[20]. Thus, the solutions adopted in any case are different. In addition, in the second
work [20] an additional fuel is used to make the preheating of the fuel. Even when
the calorific power of the additional fuel was taken into account for calculating the
global energy efficiency, it is much favorable from the energetic point of view to
Fig. 15.3 Schematic diagram of a power plant for electricity production by supercritical water
oxidation of coal. (a) Single expansion; (b) double expansion with intermediate reheating
(Reprinted with permission from Bermejo et al. [20]. Copyright © 2004, Elsevier)
use this fuel to preheat the feed than using the effluent to preheat it. This can be
explained because the effluent has a certain enthalpy due to the high pressure that it
is lost if the heat content of this stream is used by heat transmission instead of using
by expanding in a turbine.
A more thorough analysis was made by Cabeza et al. [12]. In this work, a
theoretical analysis of the feasibility of producing energy from the SCWO of sludge
was performed. According to the research performed, optimal conditions for the
reactions were temperatures of 600 ı C and pressure of 23 MPa. Thus, these ones
Table 15.4 Qualitative Direct expansion Rankine cycle

comparison of alternatives for
energy recovery Equipment cost
Electric produc. CC C
Operability C C
Efficiency 9% 17 %
Efficiencies are calculated as the electrical output (pro-
duced power minus consumed power) divided by the
thermal energy (caloric power of fuel). Air is used as
oxidant. Efficiency calculated without feed preheating.
Rankine cycle uses steam at 4.6 MPa and 400 ı C
were the conditions fixed for the effluent. In this work the energy production was
considered taking into account different options:
• Oxidant: cryogenic oxygen or air.
• Technologies for energy production: direct expansion of the effluent in a
supercritical turbine or Rankine steam working cycle at 4.6 MPa and 400 ı C.
The advantages and disadvantages of both technologies are summarized in
Table 15.4, where results are calculated considering feed at room temperature.
Preheating the feed with an external heat source improves the process efficiency.
Efficiencies shown in Table 15.4 are calculated as the net electrical production
(i.e. discounting the requirements of pumps and compressor) divided by the caloric
power of the fuel. Negative efficiency means that the production of electricity is not
enough for covering plant requirements, and additional energy is needed. As it may
be expected a direct expansion in a turbine allows for a higher theoretical energy.
However, this alternative is hampered by higher equipment and operation costs due
to harsh operational conditions in the turbine. To make this alternative technically
feasible, the SCWO effluent should be oxygen and particle free, as otherwise the
operational life of the turbine would be very short. High water content in this
effluent is another problem of direct expansion processes, since condensation of
water during the expansion should be avoided as water droplets can cause important
mechanical damage to the blades.
• Injection temperature of the feed.
This last point was not studied in the previous works. Nevertheless, since influent
preheating is a major contribution to the energy consumption of the process, another
important parameter is the minimum inlet temperature required for the operation of
the reactor. This minimum temperature depends on the design of the reactor, as
plug-flow tubular reactors typically require preheating to supercritical temperatures
for a successful operation, while some vessel reactors may operate with lower
influent temperatures. If influent temperature is decreased, the reduction in the
thermal energy of the influent must be compensated increasing the concentration
of combustible waste. This effect is illustrated in Fig. 15.4, that shows the minimum
concentration of sludge required to achieve a reaction temperature of 600 ı C as a
Fig. 15.4 Minimum sludge

concentration required to
achieve a reaction
temperature of 600 ı C as a
function of feed temperature
(a) using air as the oxidant;
(b) using pure oxygen as the
oxidant (Reprinted with
permission from Cabeza et al.
[12]. Copyright © 2013,
Elsevier)
function of the feed temperature results in the part (a) of this figure were calculated
air as the oxidant and in the part (b) considering oxygen as the oxidant, and a typical
sludge caloric power of 18,780 kJ/kg pure in dry basis [21]. Figure 15.4 shows that
in order to reach 600 ı C at the reaction chamber, feed must have 20.6 % solids
at 25 ı C. If inlet temperature increases, concentration decreases down to 3.7 % at
425 ı C. In the case of using oxygen the values are 17.4 % and 3.5 % respectively.
Together with efficiency analysis it must be taken into account that lower injection
temperatures allow to avoid corrosion and salt deposition issues. On the other hand,
when the feed temperature increased, sludge is easily pumpable (very concentrated
sludge may be impossible to pump!!).
• Preheating of the feed using the effluent of the reactor or using a external heat
source.
In direct gas expansion systems, another design consideration that must be taken
into account is whether reactor feed preheating is carried out with the effluent, or
if an external heating system is used for this purpose. Figure 15.5 shows the global
efficiency in energy generation by direct expansion as a function of feed injection
temperature, considering preheating the feed with the effluent (HEAT RECOVERY)
or using an external heat source to preheat the feed and all the effluent is expanded in
Fig. 15.5 Efficiency in

energy generation by direct
expansion as a function of
feed injection temperature,
considering preheating the
feed with the effluent (HEAT
RECOVERY) or using an
external heat source to
preheat the feed and all the
effluent is expanded in the
turbine (COMPLETE
EXPANSION) (a) using air
as the oxidant; (b) using pure
oxygen as the oxidant
from Cabeza et al. [12].
the turbine (COMPLETE EXPANSION). This can be explained because the effluent
has a certain enthalpy due to the high pressure that it is lost if the heat content of
this stream is used by heat transmission instead of using by expanding in a turbine.
In part (a) air is used as the oxidant; while in part (b) pure oxygen is used as the
oxidant As shown in Fig. 15.5a, when air is used as the oxidant, it is clear that more
electrical energy is produced with a complete expansion of the effluent than when
the effluent is used for preheating. In part (b), nevertheless it is observed that when
oxygen is used as a fuel the efficiency results are comparable both using the effluent
for preheating or using a supplementary heating source unless that we are working
at feed injection temperatures higher than the critical point of water. This is due
to the high work consumed in air compression compared to that of pumping liquid
oxygen. In addition this work is higher when the injection temperature is lower
because higher concentration of fuel (and higher amount of oxidant) is needed.
In Fig. 15.5a, b it is observed that a sharp change in the curves is produced around
the critical point of water. Even when the change from liquid to supercritical water
it is not “phase transition” with a proper latent heat, in this area a sharp change of
the properties (including specific enthalpy) is produced in the vicinity of the critical
point. In this area, heat capacity of water presents a sharp maximum with values
several times higher than the heat capacity of liquid water. Thus the change from
liquid to supercritical water can be considered a kind of “phase change” with a
Fig. 15.6 Temperature – enthalpy profiles (composite curves) of the partially expanded effluent
of an SCWO reactor and the feed and air streams
pseudo latent heat, making that the calorific energy to increase the temperature in
this area is non linear and very high, producing this irregularities in the curves.
In addition to all of this, when preheating the feed with the effluent it is necessary
to be extremely careful with the enthalpy content at each temperature level of the
streams considered. It can happen that even when the effluent has much higher
energy content than the feed, this enthalpy is at a lower temperature, and thus, cannot
be transferred to the feed.
This is made evident by representing the composite curves (enthalpy versus
temperature) of the streams to be cooled and the streams to be heated, as illustrated
in the following examples.
For example, if the system consists of a direct effluent expansion with heat
recovery by preheating of feed with the effluent, it is important to take into account
that it is normally not possible to implement a layout with an expansion of reactor
effluent in a turbine followed by heat recovery via a heat exchanger between reactor
feed and turbine effluent. This is due to the evolution of temperature-enthalpy
profiles of streams at different pressures that are sketched in Fig. 15.6.
Even if at supercritical conditions a vapor-liquid phase transition with the
associated latent heat of vaporization is not observed, in the vicinity of the critical
point a “pseudo-latent” heat is evolved. Moreover, this heat is released at lower
temperatures as pressure is decreased. Due to this phenomenon, the preheating of a
feed stream at high pressure using a turbine effluent partially expanded to a lower
pressure normally is not possible, even if stream supply and target temperatures
and heat loads may suggest otherwise. As shown in Fig. 15.6, due to the shift
of the “pseudo-latent” region to lower temperatures as pressure is decreased, this
layout would result in a temperature cross in the recovery heat exchanger. Feasible
solutions in this situation are:
Fig. 15.7 Composite curves

for feed and air preheating.
Red line corresponds to
heating each stream before
mixing them. Blue line
corresponds to mixed feed
and air in one stream and
heat them together
1. Implementation of the recovery heat exchanger before the gas turbine. This
layout can result in important losses in the energy efficiency of the process due
to large temperature differences in the recovery heat exchanger.
2. Division of the reactor effluent into two branches: one used for feed preheating,
and a second one expanded in the gas turbine. This layout increases process
complexity, but allows for higher energy efficiency.
It is necessary to point that the shape of the curve of preheating will be different
whether feed and air are heated together or separated. Figure 15.7 shows the
composite curves for both alternatives, supposing 400 ı C as final temperature. It
is observed that if aqueous feed and air are mixed, most of the heat (1,850 kW) is
absorbed below 300 ı C, and the final heating up to 400 ı C demands a small portion
of enthalpy (320 kW). On the other side, if the two streams are heated before mixing,
the initial heating up to 350 ı C demands 1,500 kW; and the last 50 ı C up to the final
temperature need 710 kW. Preheating feed and air together is more energetically
efficient, however the reaction could initiate during the preheating process, what
must be considered for safety reasons.
Another example is the use of an SCWO reactor as heat source in a conventional
Rankine cycle. Considering the oxidation products at 600 ı C and 23 MPa, one
stream of 5,552 kg/h at these conditions could be cooled down to 45 ı C, releasing
2.7 MW of enthalpy. A simple energy balance points that this amount of enthalpy
could be used for heating 3,145 kg/h of water at 4.6 MPa from 35 up to
400 ı C (steam condition for Rankine cycle). However, this process is not feasible.
Figure 15.8 shows the composite curves: red line corresponds to the hot stream
(reactor products); dotted blue line corresponds to the cold stream (water). Heat
transfer from hot to cold stream is limited by the pinch point, where the temperature
difference is minimal. In practice, the amount of steam that could be produced in
this case is 2,218 kg/h, represented by continuous blue line in Fig. 15.8.
Thus, these composite curves must be taken into account before designing a
preheating system in an SCWO plant.
In summary, different options for energy production in SCWO plants have
been analyzed. The selection of one or other option will depend on every specific
Fig. 15.8 Composite curves

of the effluent of an SCWO
reactor and water stream used
in a Rankine cycle. Dotted
blue line has the same amount
of enthalpy as red line.
Continuous blue line has the
“feasible” amount of steam,
observing limitations from
pinch analysis
situation. The size of the plant must be taken into account. Currently, the largest
SCWO reactors are between 3,000 and 5,000 kg Feed per hour, while energy
production system normally works at much larger scales.
Thus, for a small scale plants, technologies such as an organic Rankine cycles
can be a more realistic option than supercritical turbine or a Steam Rankine cycle,
even when It is less efficient.
When comparing the energy produced by direct expansion of the effluent and by
indirect heating steam generation it is observed that when direct expansion is used,
the energetic efficiency is much higher than when the effluent is used to heat an
auxiliary fluid of a Rankine or Brayton cycle.
Even when the option of direct expansion of the effluent is, by far, the most
energetically efficient, it will be not applicable in the short term. This is mainly
due to the fact that the composition of the effluent (50–80 % mol of water, carbon
dioxide and nitrogen if air is used as oxidant) makes it not suitable for expansion in
a conventional turbine. This composition makes the effluent intermediate between
the pure water used in steam turbine and the flue gases, products of combustion used
in gas turbines. The starting conditions of this mixture, around 600 ı C and 23 MPa,
determine the near-isoentropic path needed for an efficient expansion and route it
down this path to an early condensation in terms of a full harnessing of the mixture
enthalpy content; depending of course on the exact composition of the mixture.
Thus, technical issues concerning the expansion of two-phased streams prevent the
effective implementation of direct expansion in the short term. Furthermore, the
detailed design of a dedicated, effective turbine would be costly and would take a
long time to be carried out.
15.4 CO2 Sequestration
In recent years, enormous investments have been done for the development and
implementation of CO2 sequestration processes in power plants, particularly those
involving the combustion of coal, biomass or other low-quality fuels [22]. Therefore,
Fig. 15.9 Schematic diagram of an SCWO power plant with CO2 recovery (Reprinted with
permission from Donatini et al. [25]. Copyright © 2009, Elsevier)
combustion technologies that offer advantages for the sequestration or revalorization

of CO2 are of great interest.
The combustion by SCWO has the additional advantage of making easier CO2
sequestration. The effluent of this process consists of a mixture of an aqueous stream
and pressurized CO2 , and N2 if air is used as the oxidant. What is more, if oxygen is
used as the oxidant, a pressurized stream of nearly pure CO2 can be obtained as gas
effluent, that can constitute a valuable by-product rather than a waste. Some frequent
uses of the CO2 are as a cryogenic agent, inert gas or protective atmosphere in
food packaging, beverages carbonization, fire suffocation agent, or pH control agent.
Sometimes it is simply captured for storing to avoid release it to the atmosphere.
Several commercial SCWO plants had already considered the capture and storage
of CO2 in their designs.
In 2002, Griffith and Raymond [23] recovered a part of the carbon dioxide of the
effluent and used it in for neutralization of Basic stream generated in other part of
the processes. More recently, in 2005, SuperWater [24] incorporates a recovery of
the produced CO2 via gas-liquids separators at different operation pressures.
Donatini et al. [25] proposed and simulated the application of SCWO of coal for
power generation, including also a method for capturing and storing CO2 as liquid.
The schematic diagram proposed for the process is shown in Fig. 15.9.
At first, aqueous coal slurry is pressurized and preheated. Then, the coal slurry is
mixed with pure O2 coming from a multistage compressor, forming a three -phase
system that is preheated in a counter-current heat exchanger by a regenerative closed
loop. The regenerative loop consist of heating steam at 1 bar as thermal fluid with
the effluent of the reactor, and using this reheated steam to preheat the three phase
mixture of oxygen and aqueous coal slurry before entering to the reactor. It should
be noted that the shape of the composite curves of the heating process change if the
oxidant and the aqueous feed are preheated mixed together or separately, because
of the thermodynamic properties of the mixture are not the addition of the thermal
properties of the separated substances as shown in Sect. 15.3.
Combustion takes place in a tubular reactor constituted of two parts. The first
one is adiabatic and the combustion temperature rises. In the latter part, thermal
power developed during the combustion process is absorbed by pressurized water

circulating in a countercurrent tube bundle exchanger, which is located inside the
internal surface of the reactor. This steam is introduced in a typical reheat Rankine
cycle operating with maximum steam conditions of 700 ı C and 350 bar for the first
turbine and 565 ı C and 190 bar after reheating [25].
The effluent of the reactor is used first to heat the steam at 1 bar acting as a
thermal fluid of the regenerative loop used for preheating the feed, and later, then to
a pre-economized for the steam of the Rankine cycle and finally to preheat the coal
slurry. Then the effluent is sent to a gas-liquid separator.
The gaseous mixture exiting the gas-liquid separator expands in a biphase turbine
(rotary separator turbine), where the separation between CO2 in liquid phase and
CO2 in gas phase is directly executed. Indeed, the inlet thermodynamic conditions
of the CO2 assure that it is mainly in liquid phase downstream the expansion.
The liquid phase is first purified from ash in ash–liquid separator and then
feeds a multistage degasification process in order to extract the CO2 -gas dissolved
in liquid water producing at the same time power. The degasification process is
executed expanding the liquid in three successive biphasic turbines. Reducing the
pressure, the CO2 -gas concentration in liquid phase evidently decreases and the CO2
is released in gas phase. CO2 -gas obtained in this process is then compressed and
liquefied ready for storage.
After optimization the global net efficiency having into account the consumption
of the Air Separation Unit for O2 production is 27.9 % with the reactor working at
250 bar and 560 ı C in the inlet of the reactor. The pure oxygen production process
results the main energy penalty (reducing the efficiency from 38 to 28 %). Thus, it is
demonstrated that a relative high efficiency in energy production can be realistically
obtained despite of CO2 capture.
Other option different from recovering liquid CO2 at high pressure can be
recovering it as a solid in the form of carbonate. If a basic stream such as Ca(OH)2
is added to the effluent, the CO2 can be absorbed by the effluent as carbonate, and
eventually precipitated as calcium carbonate [26].
15.5 Proposed Reactor Designs and Plant Layouts
15.5.1 Patented Designs
Due to the importance of energy recovery and production schematic diagrams for
the economic feasibility of SCWO facilities, different forms of energy integration
are considered in most commercial SCWO plants and in patented designs [27]. In
particular, most patented reactor designs consider different ways of using the heat
in reactor effluent to preheat reagents, for example using an external heat exchanger
[28], by thermal contact through the walls of a cooled wall reactor [29] or by mixing
the hot effluent with reagents [30, 31], which if performed in a mixing chamber
directly before the reaction chamber also has the advantage of providing the mixture
with radicals that facilitate the initiation of the combustion, reducing the temperature
required to ignite it [32]. As previously described, in SCWO processes aimed to the
production of energy, all these types of preheating may be the cause of large drops
in the global efficiency of the process due to the use of large temperature differences
in pre-heaters, and therefore they must be very carefully designed.
Besides these basic forms of energy integration, some patents describe SCWO
layouts specifically designed to produce electrical energy. The use of SCWO as
an alternative combustion technology for the production of energy was already
described in early patents of this technology [33]. Molnar patented a power cycle
using an SCWO reactor to generate steam, expanded in a turbine to produce
electricity [34]. Other patents deal with the treatment of specific residues, that due
to the volume of residue produced or to the peculiarities of their treatment can
be particularly interesting for the production of energy. Some relevant examples
include the treatment of common household garbage patented by Hayakawa, based
on the advantages of SCWO over conventional incineration because it does not
produce gaseous contaminants such as dioxins, and allows a valorization of the
residue by producing electricity from the heat of combustion, and also by recovering
valuable compounds from the solid combustion residue such as metal oxides
[35]. Another interesting application is the treatment of undigested wastewater
sludge from biological water treatment facilities patented by Griffith et al. from
Hydroprocessing LLC. This application is of particular interest due to the large
volumes of such residues produced worldwide, the increasing restrictions to the
disposal of these wastes in dumping sites, and the capacity of SCWO for the
treatment and energy valorization of these residues without a previous drying
step [36].
Regarding the different reactor designs proposed for SCWO, it must be taken into
account that most designs are aimed at solving the problems of corrosion and scale
formation [37], proposing solutions that can be disadvantageous from the energy
production point of view. In case of vessel reactors, most designs rely on the use
of cooling streams to protect reactor walls from the high temperature, corrosive
conditions in the reaction chamber. Some examples include the injection of a curtain
of cold water near reactor walls in MODAR reactors [37], the refrigeration of
the reaction chamber in cooled-shell reactors [29], or the injection of a coolant
stream into the reaction chamber through the pores of a transpiring wall reactor
[38]. All these arrangements imply a dilution of the effluent and a reduction of its
temperature, which is disadvantageous for the efficiency of the production of energy.
Specific reactor designs for the production of energy should consider the possibility
of retrieving at least a fraction of the effluent as a non-diluted, high-temperature,
low-salt concentration stream suitable for the production of electricity either by
direct expansion or by indirect methods [37] as described in the Chap. 7 of this
book.
Similarly, due to safety and materials resistance issues, the operation of SCWO
tubular reactors often is limited to lower temperatures than vessel reactors. Indeed,
several patents describe methods for controlling and restricting the temperature
increase in tubular reactors, using methods such as multiple injection schematic

diagrams of oxidant or fuel [39, 40], or the continuous injection of a coolant through
a porous wall [41]. Again, all these methods dilute the effluent and reduce its
thermal quality. Some alternatives have been proposed in order to mitigate this
problem, such as the direct injection of cold oxidant without water, which allows
carefully controlling the evolution of the reaction without diluting the effluent with
cold water [42], or the injection of cold feed in several points of the reactor in
order to decrease the reaction temperature. This last solution has the additional
advantage of avoiding preheating of a fraction of the feed making easier efficient
energy integration. In all these cases, the maximum temperature limitations imposed
by safety and materials resistance considerations must be carefully balanced taking
into account the negative effect of a dilution or a reduction of temperature on the
production of energy from the effluent.
15.5.2 Hydrothermal Flames
SCWO operating at hydrothermal flame regime was described in Chap. 7. It is

also mentioned here because it presents evident advantages for energy integration
layouts.
Hydrothermal flames are combustion flames produced in aqueous environments
at supercritical conditions [43]. Such flames are formed when fuel and oxidant
streams are mixed at conditions that enable autoignition, producing an oxidation
process at sufficient temperature and rate to produce a luminous flame. In general,
flames ignite spontaneously beyond a temperature in the range between 400 and
500 ı C. The autoignition temperature is higher when the concentration of fuel is
lower, and there is a minimum temperature and fuel concentration under which
a flame is not formed. As fuel concentration is reduced, flames present a lower
temperature and luminosity, but even when luminosity is not observed, the flame
structure is maintained. In general, the conditions of flame ignition depend on
the fuel, the oxidant, the fuel/oxidant ratio and the geometry of the injection
system. Interestingly, in general the autoignition temperatures under pressure are
significantly lower that under ambient conditions and decrease with pressure
[44–46].
A first attractive characteristic of hydrothermal flames is that they enable an
intensification of the SCWO process, reducing the time scales of the oxidation
process from minutes to 10–100 ms, with waste combustion efficiencies above 99 %
[43]. This intensification of the oxidation reaction allows for a significant reduction
of the reactor volume and its associated cost. For example, a continuous reactor of
1 L treating a flow of 104 kg/h working with a residence time of 1 min in a flameless
regime in principle could be substituted by a reactor of 14 mL and a residence
time of 0.5 s working with a hydrothermal flame, without detrimental effects on
combustion efficiencies.
Other advantages of hydrothermal flames concerning energy integration/power

generation are the possibility to achieve very high reaction temperatures, and the
viability of the operation without feed preheating.
Compared to typical reaction temperatures of 450–600 ı C in conventional
SCWO reactors, hydrothermal flames can reach temperatures above 1,000 ı C,
even when normally temperatures between 600 and 700 ı C are normally used.
Higher temperatures enable a higher thermodynamic efficiency in the production
of electricity, but extraction of hot fluid directly from the hydrothermal flame pose
significant technical challenges, including important corrosion crevice problems.
Most reactor designs working in hydrothermal flame regime rely on the injection
of coolant streams in order to confine the hydrothermal flame within the interior of
the vessel, protecting reactor walls from damage caused by the oxidation conditions
at such high temperatures [47–49]. Such coolant streams dilute the reactor effluent
and reduce its thermal quality.
The possibility of operating reactors in hydrothermal flame regime reducing
or even eliminating feed preheating is well documented [48–50]. This is made
possible by the capacity of the flame to generate sufficient heat to immediately
preheat reagents when they are directly injected into the flame. This is an important
advantage for the production of energy, because as described in Sect. 15.3, feed
preheating requirements limit and complicate the recovery of energy from SCWO
processes. Moreover, by eliminating feed preheating systems, the frequently severe
problems of corrosion and plugging in these equipments are also eliminated, when
these equipments are not necessary for stationary operation.
The formation of a stable flame that can be maintained even when cold reagents
are injected into it requires a flow pattern with a certain degree of back-mixing
and low flow velocities that it is normally not possible in plug-flow tubular
reactors [47, 51]. In such reactors, each fluid element must ignite by itself, which
requires introducing it into the tubular reactor at conditions above its auto-ignition
temperature [51]. This is well illustrated in the video found in the web page of the
High Pressure Process Group of the University of Valladolid [52] Besides, operation
of a tubular reactor with a flame is not desirable due to safety concerns related to the
formation of hot spots that can change their position along the reactor in response
to small fluctuations in feed temperature, composition or flow rate. In contrast, in
vessel reactors it has been demonstrated that a stable hydrothermal flame can be
maintained, enabling elimination efficiencies of organic compounds in feed up to
99.95 % with Total Organic Carbon (TOC) concentrations in the effluent below
20 ppm, even when feeds were directly injected into the flame at temperatures as
low as 50 ı C [50].
Recently, our research group patented a new cooled wall reactor working with a
hydrothermal flame as a heat source that presents the additional advantage that part
of the products can be extracted of the reactor by its upper part, without mixing with
the cold water, allowing better energy integration because an effluent at temperatures
between 600 and 700 ı C can be obtained [53]. A schematic diagram of the reactor
working with that lay out is shown in Fig. 15.10.
Fig. 15.10 Schematic diagram of the new cooling wall reactor working with an outlet for hot
products designed by the High Pressure Process Group of the University of Valladolid
15.5.3 Energy Integration and Power Generation

in Commercial SCWO Processes
Currently, there are a number of SCWO commercial plants in operation, and many
others have operated in the past for several years and they stopped operation for
different reasons. These subjects has been recently revised by Marrone [54], and
it will not be considered here, focusing exclusively in the heat/energy recovery
systems of some of these facilities.
It has been extensively discussed so far that energy recovery of the energy
produced by SCWO has a tremendous impact on the economy of the process. There
are a number of examples of this in the industry.
Hydroprocessing LLC reported that the profitability of their sludge treatment
plant by SCWO of Harlingen (Texas, USA) strongly depended on income obtained
by selling excess thermal energy as hot water and CO2 as a product for the
neutralization of industrial effluents [23]. More recently in 2005, the SuperWater

process [24], used for the construction of a sludge SCWO treatment plant in
Orlando, used the energy generated in the process to produce steam and hot water
and later to produce electricity. They claim that the efficiency of conversion of the
energy to power in SuperWater Process is higher than that of a coal plant with the
equivalent feedstock. The reason is that in a normal coal fired plant a great amount
of heat energy is lost in the stack as hot exhaust while in an SCWO plant there is
no stack effluent and all the heat of condensation of the steam, lost in a coal plant,
can be used here for steam generation. With their technology, they claim that a
plant processing 10 dry tons per day of sludge will generate approximately 0.6 MW
of electrical energy. Considering a typical heat of combustion of 18,780 kJ/kg in
dry basis [21], this would mean a global efficiency of 27 % in power generation.
Nevertheless, no information is provided whether the energy used for oxygen
purification is considered.
To achieve this value they propose a heat integration consisting in a regenerative
heat loop consisting of an auxiliary fluid that is heated with the effluent of the reactor
and this heat used to preheat their feed up to 275 ı C. The rest of the heat contained
in the effluent is recovered in the steam generators. Information about the conditions
of the steam produced is not provided.
In the Aquacritox® process commercialized by SCFI (Supercritical Fluids
International) the used of the heat produced in the effluent is also considered. They
propose to preheat the feed with the effluent of the reactor to above the supercritical
point in an economizer. The waste heat from the economizer/reactor set can be used
to generate steam at temperatures up to 500 ı C or hot water and in larger plants for
power generation. In this case, they propose that a heat exchanger takes the place of
a boiler in a standard steam loop (i.e. a Rankine cycle) [55].
15.6 Conclusions and Future Perspectives
The capacity of SCWO to destroy almost every kind of compound without

producing pollutant byproduct together with the potential of this technology for
thermal and electrical energy production makes it a promising candidate for the
combustion of bio-fuels, especially if these bio-fuels present difficulties in a
conventional combustion process such as necessity of drying or generation of
dangerous byproducts.
The possibility of heat integration and the production of energy to cover the
necessities obtaining a small energy surplus is currently a realistic option for
small/medium size SCWO facilities. Nevertheless, the large scale energy production
by SCWO is still far to be implemented. In first place the size of the larger facilities
existing right now is orders of magnitude lower than that necessary to be considered
a full scale power plant. In addition, the technology necessary to build a turbine to
perform direct expansion of the effluents is far to be available.
Currently, after 30 years of its invention, SCWO of different wastes is producing

again a lot of interest. In the near future this will easily conduct to the development
of medium size plants, of several tons per/hour of treatment, in which special
attention will be paid to energy integration/production systems as well as CO2
capturing. In our opinion the construction of really large SCWO plants which main
purpose is energy integration is still very far to be a reality.
References
1. Bermejo MD, Cocero MJ. Supercritical water oxidation: a technical review. AIChE J.
2006;52:3933–51.
2009;47:382–90.
supercritical water oxidation – Part B: Commercial/full-scale applications. J Supercrit Fluids.
2004;29:289–312.
5. Cocero MJ, Alonso E, Sanz MT, Fdz-Polanco F. Supercritical water oxidation process under
energetically self-sufficient operation. J Supercrit Fluids. 2001;24:37–46.
6. Lavric ED, Weyten H, De Ruyck J, Plesu V, Lavric V. Delocalized organic pollutant destruction
through a self-sustaining supercritical water oxidation process. Energy Convers Manag.
2005;46:1345–64.
7. Martin S. Fred up about efficiency. 2013. http://www.siemens.com/innovation/apps/pof_
microsite/_pof-spring-2013/_html_en/combined-cycle-power-plants.html. Last accessed 14
Aug 2013.
8. Rivas FJ, Gimeno O, de la Portela JRM, Ossa E. Supercritical water oxidation of olive oil mill
wastewater. Ind Eng Chem Res. 2001;40:3670–4.
9. Chkoundali S, Alaya S, Launay JC, Cansell F. Kinetic parameters of hydrothermal oxidation
process on olive mill wastewater. Environ Eng Sci. 2008;25:131–7.
10. Erkonak H, Sögüt OO, Akgün M. Treatment of olive mill wastewater by supercritical water
11. Goto M, Nada T, Kodama A, Hirose T. Kinetic analysis for destruction of municipal sewage
sludge and alcohol distillery wastewater by supercritical water oxidation. Ind Eng Chem Res.
1999;38:1863–5.
12. Cabeza P, Queiroz JPS, Arca S, Jiménez C, Gutiérrez A, Bermejo MD, Cocero MJ. Sludge
destruction by means of a hydrothermal flame. Optimization of ammonia destruction condi-
tions. Chem Eng J. 2013;232:1–9.
13. Jin F-M, Kishita A, Takehiko Moriya HE. Kinetics of oxidation of food wastes with H2 O2 in
14. Cui B, Cui F, Jing G, Xu S, Huo W, Liu S. Oxidation of oily sludge in supercritical water.
J Hazard Mater. 2009;165:511–7.
15. Vogel F, DiNaro Blanchard JL, Marrone PA, Rice SF, Webley PA, Peters WA, Smith KA,
Tester JW. Critical review of kinetic data for the oxidation of methanol in supercritical water.
16. Oshima Y, Hori K, Toda M, Chommanad T, Koda S. Phenol oxidation kinetics in supercritical
17. Segond N, Matsumura Y, Yamamoto K. Determination of ammonia oxidation rate in sub- and
supercritical water. Ind Eng Chem Res. 2002;41:6020–7.
18. Phenix BD, DiNaro JL, Tester JW, Howard JB, Smith KA. The effects of mixing and oxidant
choice on laboratory-scale measurements of supercritical water oxidation kinetics. Ind Eng
Chem Res. 2002;41:624–31.
19. Kansha Y, Kishimoto A, Nakagawa T, Tsutsumi A. A novel cryogenic air separation process
based on self-heat recuperation. Sep Purif Technol. 2011;77:389–96.
20. Bermejo MD, Cocero MJ, Fernández-Polanco F. A process for generating power from the
oxidation of coal in supercritical water. Fuel. 2004;83:195–204.
21. Channiwala SA, Parikh PP. A unified correlation for estimating HHV of solid, liquid and
gaseous fuels. Fuel. 2002;81:1051–63.
22. Abu-Khader M. Recent progress in CO2 capture/sequestration: a review. Energy Source, Pt A
Recovery Util Environ Eff. 2006;28:1261–79.
23. Griffith JW, Raymond DH. The first commercial supercritical water oxidation sludge process-
ing plant. Waste Manag. 2002;22:453–9.
24. SuperWater Solutions. 2008. http://www.superwatersolutions.com/technology.html. Last
accessed 14 Aug 2013.
25. Donatini F, Gigliucci G, Riccardi J, Schiavetti M, Gabbrielli R, Briola S. Supercritical water
oxidation of coal in power plants with low CO2 emissions. Energy. 2009;34:2144–50.
26. Domingo C, García-Carmona J, Loste E, Fanovich A, Fraile J, Gómez-Morales J. Control
of calcium carbonate morphology by precipitation in compressed and supercritical carbon
dioxide media. J Cryst Growth. 2004;271:268–73.
27. Martín A, Bermejo MD, Cocero MJ. Recent developments of supercritical water oxidation: a
patents review. Recent Pat Chem Eng. 2011;4:219–30.
28. Li L, Gloyna EF. High temperature wet oxidation using sintered separators. Patent WO
9302969; 1993.
29. McBrayer RN, Eller JM, Deaton JE. Turbulent flow cold-wall reactor. Patent WO 9602471;
1996.
30. Eller JM, McBrayer RN, Peacock RD, Barber JS, Stanton WH, Applegath F, Lovett GH.
Heating and reaction system and method using recycle reactor. Patent US 6001243; 1999.
31. Wofford WT, Griffith JW, Humphries RW, Lawrence JW. Hydrothermal treatment system and
method. Patent US 6709601; 2004.
32. Bourhis AL, Hong GT, Killilea WR. Method for hydrothermal oxidation. Patent WO 9705069;
1997.
33. Modell M. Processing methods for the oxidation of organics in supercritical water. Patent WO
8103169; 1981.
34. Molnar C. System and method for generating electricity from supercritical water oxidation
process. Patent WO 2008118348; 2008.
35. Hayakawa K. Refuse/waste treatment system and power generation system using supercritical
water. Patent JP 2003164831; 2003.
36. Griffith W, Wofford WT, Griffith JR. Apparatus and method for oxidizing undigested
wastewater sludges. Patent WO 9847822; 1998.
37. Huang CY, Barner HE, Albano JV, Killilea WR, Hong GT. Method for supercritical water
oxidation. Patent WO 9221621; 1992.
38. Mueggenburg HH, Rousar DC, Young MF. Supercritical water oxidation reactor with wall
conduits for boundary flow control. Patent US 5387398; 1995.
39. McBrayer R, Eller JM, Swan JG, Deaton JE, Gloyna RR. Method and apparatus for treating
waste water streams. WO 9526929; 1995.
40. Eller JM, McBrayer RN, Swan JG. Method for controlling reaction temperature. Patent US
5770174; 1998.
41. McGuiness TG. Supercritical oxidation reactor apparatus and method. Patent WO 9418128;
1994.
42. Cansell F. Method for treating waste by hydrothermal oxidation. Patent US 6929752; 2005.
43. Agustine C, Tester JW. Hydrothermal flames: from phenomenological experimental
demonstrations to quantitative understanding. J Supercrit Fluids. 2009;47:415–30.
44. Schillonf W, Franck EU. Combustion and diffusion flames at high pressures to 2000 bar. Ver
Bun Phys Chem. 1988;92:631–6.
45. Hirth T, Franck EU. Oxidation and hydrothermolysis of hydrocarbons in supercritical water at
high pressures. Bun Phys Chem. 1993;97:1091–8.
46. Pohsner GM, Franck EU. Spectra and temperatures of diffusion flames at high pressures to
1000 bar. Bun Phys Chem. 1994;98:1082–90.
47. Bermejo MD, Cabeza P, Queiroz JPS, Jiménez C, Cocero MJ. Analysis of the scale up of a
transpiring wall reactor with a hydrothermal flame as a heat source for the supercritical water
48. Wellig B, Weber M, Lieball K, Prikopsky K, Rudolf von Rohr P. Hydrothermal methanol
diffusion flame as internal heat source in a SCWO reactor. J Supercrit Fluids. 2009;49:59–70.
49. Wellig B, Lieball K, Rudolf von Rohr P. Operating characteristics of a transpiring-wall SCWO
reactor with a hydrothermal flame as internal heat source. J Supercrit Fluids. 2005;34:35–50.
50. Bermejo MD, Jiménez C, Cabeza P, Matías Gago A, Cocero MJ. Experimental study of
hydrothermal flames formation using a tubular injector in a refrigerated reaction chamber.
Influence of the operational and geometrical parameters. J Supercrit Fluids. 2011;59:140–8.
51. Bermejo MD, Cabeza P, Bahr M, Fernández R, Ríos V, Jiménez C, Cocero MJ. Experimental
study of hydrothermal flames initiation using different static mixer configurations. J Supercrit
Fluids. 2009;50:240–9.
52. High Pressure Process Group of the University of Valladolid. http://hpp.uva.es/. Last accessed
17 April 2014
53. Bermejo MD, Cabeza P, Cocero MJ, Jiménez C, Queiroz JPS. Device and method for
generating autothermal hydrothermal. Application number PCT/ES2011/070727; 2012.
54. Marrone PA 2013. Supercritical water oxidation—current status of full-scale commercial
activity for waste destruction. J Supercrit Fluids. 79:283–8.
55. Super Critical Fluid International (SCFI). http://www.scfi.eu/. Last accessed 14 Aug 2013.
Chapter 16
Production of Chemicals in Supercritical Water
Yukihiko Matsumura and Tau Len-Kelly Yong
Abstract Supercritical water (SCW) is expected to be a green solvent for

dissolving substances, for chemical synthesis, or for chemical reactions, and thus
various study has been carried out. Production of chemicals has been found possible
in SCW. This chapter reviews the chemical production in terms of feedstocks
and reactions. The feedstocks can include biomass (cellulose, hemicellulose and
lignin), plastics, other wastes (e.g., tire, rubber), inorganics and waste water. The
reactions include SCW gasification (where hydrolysis and pyrolysis take important
roles), SCW oxidation, depolymerization, precipitation, hydrothermal synthesis,
other organic reactions for synthesis of chemicals. In these reactions, roles of
H2 O are: (1) Reactant/product (hydrolysis, hydration, hydrogen source and free-
radical chemistry) or catalyst (Acid/base catalyst precursor and catalyst in the
transition state); (2) Intermolecular interactions in high temperature water: Solvation
effects (effects of preferential solvation, hydrophobicity and solvent dynamics) and
density inhomogeneity effects (ions, organic compounds, noble gases and radicals);
(3) Medium: Energy transfer, diffusion and solvent cages and phase behavior.
However, low selectivity and yield and high production cost are barriers for the
commercialization of production of chemicals in SCW.
Keywords Feedstocks • Biomass • Wastes • Reaction • Hydrolysis • Hydration

• Catalyst
Y. Matsumura ()
Division of Energy and Environmental Engineering, Institute of Engineering,
Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima 739-8527, Japan
T.L.-K. Yong
Department of Mechanical Science and Engineering, Graduate School of Engineering,
Hiroshima University, 1-4-1 Kagamiyama, Higashi-Hiroshima 739-8527, Japan
428 Y. Matsumura and T.L.-K. Yong
16.1 Introduction
Chemical reactions proceed in supercritical water (SCW), and conversion of

feedstock compounds into various products is possible in SCW. This characteristic
has drawn attention of various researchers, some of whom tried to apply this for the
production of chemicals. In these reactions, roles of H2 O are: (1) Reactant/product
(hydrolysis, hydration, hydrogen source and free-radical chemistry) or catalyst
(Acid/base catalyst precursor, catalyst in the transition state); (2) Intermolecular
interactions in high temperature water: Solvation effects (effects of preferential
solvation, hydrophobicity and solvent dynamics) and density inhomogeneity effects
(ions, organic compounds, noble gases and radicals); (3) Medium: Energy transfer,
diffusion and solvent cages and phase behavior.
The target products are often organic compounds, but there are some examples
which aim to get inorganic products, too. The most well-known reaction in SCW
should be SCW oxidation (SCWO), where organic compounds are oxidized into
carbon dioxide and water. It was Michael Modell, professor at Massachusetts Insti-
tute of Technology, who found the principle of this reaction. He found that glucose
was completely decomposed into gas in SCW. Since he thought decomposition
of hazardous organics should be more economically feasible, he developed his
study in the direction of complete decomposition using oxygen. This technology
employed the advantage of SCW which can completely mix with various organic
compounds as well as oxygen, thus quick reaction in a single phase is achieved.
Meanwhile, Douglus Elliott from Pacific Northwest Laboratory and Michael Antal
from University of Hawaii worked on biomass decomposition in SCW, and showed
biomass can be decomposed into combustible gas including hydrogen, methane
and carbon dioxide in SCW [1]. Assuming these are also chemicals, this SCW
gasification is one example of production of chemicals in SCW.
It was Michael Antal who further studied the reactions in SCW, and pointed out
that at lower temperature, ionic reaction is favored, while at higher temperature,
radical reaction if favored. This fact shows one of the main characteristics of
reactions in SCW. By adjusting temperature and pressure, reaction pathway may
be controlled. This leads to the possibility of producing various organic compounds
in SCW. Ionic or radical decomposition of specific feedstock can produce various
organic compounds.
However, soon, researchers were challenged by the severe reaction condition of
the SCW. Since radical reactions were mostly favored under the SCW condition
often studied, reaction products were too various. Usually, controlling radical
reaction is not easy. Radicals are very reactive, and one radical can attack various
position of the same molecule, and then produced much more various radicals.
Unless very simple compounds such as H2 and I2 are employed, the results of
the SCW reaction are a soup of many kinds of organic compounds. Sometimes,
the target product, if any, shows very low yields. Thus, hydrothermal reaction for
chemicals production is studied by many researchers, but employment of SCW is
limited [2].
16 Production of Chemicals in Supercritical Water 429
When reaction temperature is sufficiently high, or reaction time is sufficiently

long, the composition of the products goes to equilibrium. This is observed for
SCW gasification. Unfortunately, the products at this equilibrium are usually very
simple compound such as hydrogen, methane and carbon dioxide for the case of
organic compound. For the case of inorganic compound, it is known that reaction
product can be predicted also by chemical equilibrium in the SCW reactor. There
is a proposal to produce hydroxyapatite using this equilibrium, and the production
of inorganic chemicals can be another possibility. However, to the best of authors’
knowledge, the reports on this inorganic product are not many. Still the production
of chemicals in SCW kept attention of researchers. There are two main reasons
for this.
As a reaction medium, SCW is known as an adjustable fuel whose properties can
be altered and controlled by temperature and pressure in a wide range. For example,
the dielectric constant of water changes from around 80 at room temperature to 4
at 873 K. This fact implies that SCW can behave like organic solvent, and thus can
be employed in place of organic solvents used in the chemical industry. Since water
is harmless and non-toxic, replacement of organic solvents used in conventional
chemical industry by SCW is expected to be environmental friendly. This is one of
the reasons why chemical production in SCW is attracting attention.
Another reason for the production of chemicals in SCW is to make the various
SCW technologies more economically feasible. Various SCW process such as
SCWO and SCW gasification are facing difficulty for commercialization. It is not
economically advantageous to use this kind of technology for hazardous material
decomposition or energy production from renewable feedstocks. If by-products
could be produced in supercritical process as chemical feedstocks, it could be sold
at much higher prices so that the total process gets economically feasible.
From these reasons, the production of chemicals in SCW has been studied by
various researchers. To have a global idea of the activities in this topic, it is better
to discuss the reactions studied from the viewpoint of feedstocks and reactions
since the production of chemicals is conducted by converting some feedstocks using
chemical reactions in SCW. In this chapter, the production of chemicals in SCW is
classified first from the viewpoint of feedstocks and then from the viewpoint of
reactions employed.
16.2 Feedstocks for the Production of Chemicals
Various feedstocks have been studied for the production of chemicals in SCW.
Jin et al. reviewed chemicals and fuels production from carbohydrate, lignin, and
glycerin, and reduction of CO2 with metals, biomass, and organic wastes, under
hydrothermal conditions [3]. Biomass and its components including cellulose,
hemicellulose and lignin are often studied since biomass is expected to be renewable
feedstock. Wastes such as waste plastic and waste rubber are also studied from the
viewpoint of recycle and waste reduction.
16.2.1 Biomass
Biomass can be considered as organic substance produced by the activities of

living creatures that can be used as energy or material. Its utilization is studied
hard because it is renewable and carbon neutral. These characteristics can be well
understood when compared with fossil fuels, which is the main feedstock for the
energy and material production presently. Fossil fuel such as oil and coal is obtained
from the earth as thermally treated remains of ancient creatures. Since it takes long
time for these conversions to take place, we cannot produce fossil fuel by ourselves,
and theoretically, it is depleted when we use up all the fossil fuel which is now
lying underground. On the other hand, biomass such as trees and grasses grows
by photosynthesis, and thus can be regenerated as far as there is land and solar
power. This is the renewable characteristics of biomass. Meanwhile, it is to be noted
that utilization of fossil fuel produces carbon dioxide, and this release of carbon
dioxide into the atmosphere is causing the climate change. Biomass also produces
carbon dioxide when burnt, but it is the same amount which has been absorbed
by the photosynthesis for its growth. Thus, as far as the biomass production is
conducted in reproducible way, biomass utilization does not increase the carbon
dioxide concentration in the atmosphere. This is the carbon neutral characteristics
of biomass. Considering the fact that the two largest problems in terms of energy
and environment in the world are energy depletion and climate change, biomass
utilization should be sought intensely.
From the viewpoint of chemistry, main components of biomass are cellulose,
hemicellulose, lignin, starch, fat, and protein, whose composition differs largely
depending on the kind of biomass. Lignocellulosic biomass, which is mainly
composed of cellulose, hemicellulose, and lignin, is often employed as feedstock
for research recently, mainly due to the fact that it exist in large amount, and
does not compete with food use. Biomass utilization such as ethanol production
conventionally employed starch and sugar as feedstock, but it sometimes resulted
in the increase in food price, and considering the increase in global population, the
chemical production from starch and sugar is more and more regarded as improper
deed.
Cellulose is the polymer of glucose, and thus can produce glucose when
hydrolyzed. The difference between starch (another polymer of glucose) and
cellulose, is the bond between glucose unit. For the case of cellulose, the bond
between glucose unit is “-1,4-glycoside bond, meanwhile for the case of starch,
it is ’-1,4-glycoside bond. Just because of this difference of bonding, cellulose
has the polymer structure of straight line, while starch has the structure of helix.
The linear structure of cellulose polymer results in rigid crystal which is hard to be
hydrolyzed. Human beings have enzyme for starch hydrolysis, but does not have that
for cellulose hydrolysis, which means cellulose cannot be food for human beings.
On the other hand, cellulose is abundant on the earth in the form of wood and grass.
If we hydrolyze the cellulose using acid or base catalyst, we still can get
glucose, which can be further converted into various products. The most well-known
conversion should be ethanol fermentation, where yeast ferment glucose to produce

ethanol and carbon dioxide. This reaction was employed mainly alcoholic beverage
production and bread production for long years. The other product obtained from
glucose is lactic acid. Polymerization of lactic acid produces polylactic acid, a
bioplastic. Acetic acid is also obtained from fermentation of glucose. Various
compounds can be produced based on glucose, and this system of glucose based
refinery is being studied from the viewpoint of biomass utilization.
Cellulose decomposition in SCW was studied by Kabyemela et al. [4], Fang et al.
[5], and Kabyemela further studied the glucose decomposition [6]. Behrendt et al.
[7] also reviews some reaction of cellulose under hydrothermal conditions. As
for decomposition of glucose, which is one of the main product of cellulose
decomposition in SCW, Chunthanapum tried to determine the reaction rate of each
reaction of the reaction network [8–12], and Promdeij showed the temperature
dependence of these reactions [13, 14]. Reaction pathways for cellulose in the
presence of Na2 CO3 catalyst is given by Kruse and Gaulik [15].
One of the characteristics of cellulose behavior in SCW is that it gets completely
dissolved in it. Behavior of cellulose dissolution has been studied by Smith and Fang
[16].
Decomposition of cellulose and glucose is often explained by parallel reaction
of gasification and char production [17–19]. So far, more attention has been paid to
the gasification, and chemicals production has not been considered so much. When
chemicals are to be obtained from supercritical decomposition of cellulose, it is the
intermediate to the gas product or polymerization product which further leads to
char production. Behrendt introduced Hubert et al.’s discussion on the liquefaction
of carbohydrates from biomass to produce liquid alkanes [7].
Hemicellulose is also a polymer of sugar, but not glucose. It is a co-polymer
of various sugars, but main component is xylose. In lignocellulosic materials such
as wood and grass, cellulose is covered by hemicellulose, and thus separation of
hemicellulose is needed for effective treatment of cellulose. On the other hand,
hemicellulose itself can produce sugars including xylose, mannose, arabinose, and
some amount of glucose. When treated in hot compressed water, hemicellulose is
the first to get dissolved leaving part of lignin and most of cellulose. Usually it get
dissolved below 573 K, which is completely in the range of subcritical temperature.
Lignin is another kind of polymer found in lignocellulosic biomass, whose
monomer unit has the phenylpropane structure. For it has the structure of benzene
ring, it sometimes is considered as the feedstock of aromatic compounds. One of
the products from lignin treatment is vanillin. Decomposition pathway of lignin
has been proposed by Fang et al. [20], which includes more than ten groups of
chemicals. Yong and Matsumura also showed reaction network of lignin in SCW,
considering the decomposition of benzene, phenol and guaiacol [21–24].
When lignocellulosic biomass is placed in SCW, hydrolysis takes place so that
ether bonding including glycoside bond in cellulose and hemicellulose and “-O-4
linkage in lignin are cleaved. Saka obtained methanol soluble oily material and
methanol insoluble char by treatment in SCW [25]. Veski et al. [26] conducted
co-liquefaction of oil shale and pine wood, again to obtain oil. When temperature is
sufficiently high or catalyst is employed, gasification takes place, and hydrogen is

obtained. Demirbas calculated the price of thus obtained hydrogen and found it is
more expensive than conventional hydrogen [27].
Studies of biomass conversion in SCW to produce chemicals have been
made by various researchers [28]. Modell reviewed reactions of biomass in
SCW in 1985. Saka reviewed reactions of lignocellulosics in SCW in which he
pointed out cellulose produces oligosaccharides, monosaccharides (D-glucose and
D-fructose), fragmented products, and dehydrated products such as levoglucosan
and 5-hydroxymethylfurfural [29]. Behrendt’s review [7] also discusses cellulose
and lignin decomposition in SCW.
Some studies on treating biomass related compounds in SCW are introduced by
Arai et al. [30], which includes glycolaldehyde production from glucose. However,
again, due to the severe reaction condition encountered in SCW, more processes are
studied in subcritical water rather than SCW [31] but for gaseous product such as
hydrogen and bio-oil products.
When gaseous product is the target, some reviews on SCW gasification are
available [32–35], although some of which is not totally dedicated to SCW. Catalyst
employed can include metal [36–45], alkali [46], and carbonaceous ones [47]. A
detail review is available for the catalyst employed for SCW gasification [48].
However, its stability is still a challenge [49]. It may be to be noted that hydrogen
thus obtained can be employed for synthesis of other chemicals for hydrogenation
or Fischer-Tropsch synthesis, which is indirect production of chemicals with SCW
treatment of biomass.
Bio-oil is not a pure chemical, and thus is not the focus of this manuscript,
but information is available as a part of reviews by Peterson et al. [50]. SCW
condition is actually rather too severe for bio-oil production, leading to low yield,
and temperatures up to 673 K is usually employed.
16.2.2 Waste Plastics and Waste Rubber
Waste plastic and waste rubber such as waste tires can be also important feedstocks
for the chemical production in SCW. Waste management is an important issue for
the world in terms of conservation of resources. If we can recycle the plastic bottle
to original chemical compound for plastic production, we can save the fossil fuel
utilized for production of the same amount of product.
Waste plastic treatment is often conducted to recover the monomers of the
plastic. This results in chemical recycle, where plastic is converted back to its
monomers, and then re-polymerization is conducted to produce the same plastic.
SCW is sometimes employed in this process, but it is usually effective for polymers
produced by condensation by dehydration. For example, polyethylene terephthalate
(PET) can be hydrolyzed into ethylene glycol and terephthalic acid, if properly
hydrolyzed. However, for the polymers produced by addition polymerization,
hydrolysis does not give original monomer. Thus, effectiveness of treatment in SCW
should arise just for thermal decomposition.
The good review in terms of chemical recycling of plastics using sub- and
supercritical fluids has been provided by Goto [51]. The fundamental behavior
and reactions related to plastics has been introduced by Cansell et al. [52]. Some
examples of polymerization in supercritical fluid are also shown, but not in SCW.
Recycling process is mentioned at the end of their review, with only two reference
and three patents. Siskin also showed some polymer-related reactions in superheated
water, but mainly under subcritical condition [2]. Fang treated polystyrene in SCW,
and found that it is dissolved above 769.3 K, and then homogeneous reaction takes
place above 843.5 K to produce styrenelike product [53]. Kronholm also treated
polystyrene, and obtained styrene recovery of 57 wt% with NaOH addition in
20 min at 673 K [54]. Without NaOH, the highest yield was only 13 wt%. For
the case of polyethylene terephthalate, Jankauskaite gives a thorough review of its
recycle, but no mention of SCW is made [55]. This is because decomposition of
ethylene glycol takes place, and its yield is low when SCW is employed. Instead,
supercritical methanol is more favored, which becomes supercritical state at much
lower temperature (513 K), and higher yield of depolymerization product is obtained
[56, 57]. Smith’s review also mentions treatment of polymers in SCW observed
using diamond anvil cell [16]. They found polyethylene terephthalate and nylon
could dissolve completely in SCW, but addition polymers such as polyethylene,
polystyrene, and polyvinylchloride was degraded heterogeneously.
It is to be mentioned that SCW decomposition of fiber reinforced plastic is
sometimes applied for recovery of fiber, although it is not chemical production.
The same idea as SCW treatment of plastic is applicable to waste rubber. Park
et al. employed partial oxidation in the semi-batch reactor and obtained various
product of destruction including benzene, toluene, ethylbenzene, styrene, phenol,
acetophenone, benzaldehyde, and benzoic acid [58]. Partial oxidation is addition
of small amount of oxygen, which is not sufficient for complete oxidation of the
feedstock. The added oxygen does not only decompose the feedstock, but the
product carbon monoxide reacts with water to produce hydrogen via water gas shift
reaction, and thus produced hydrogen hydrogenates the feedstock or the product. It
enhances the reaction and prevents the formation of tarry material or char to some
extent. Onwudili and Williams studied treatment of waste tire to produce hydrogen,
but found it to be poor feedstock for this purpose [59].
16.3 Reactions for the Production of Chemicals
16.3.1 Decomposition
Decomposition of materials especially organic compounds under supercritical

conditions have been investigated and/or reviewed by several researchers including
Balat et al. [33, 60], Szuppa et al. [61], Funke and Ziegler [62], Lü and Saka
[63], Luterbacher et al. [64], Brebu and Vasile [65], Akhtar and Amin [66],
Zhu et al. [67], Moller et al. [68], Pang et al. [69], Adschiri et al. [70], Toor et al.
[71], Zetzl et al. [72], Kruse [73], Onwudili and Williams [59], Robbins et al.
[74], Yong and Matsumura [21], Kiran [75], and Savage et al. [76]. Usually,
it is not only decomposition but also the reactions between the decomposition
products. Reaction takes place in cascading way so that intermediate compounds
form reaction network. The kind of reactions can be various. It can be radical
production by homolysis of bonds in the molecule. It can be hydrolysis caused by
the attack of water molecules. It can be dehydration which is often observed for
sugar and related compounds. Each of these reactions is also shown below.
16.3.2 Pyrolysis
Pyrolysis is the decomposition reaction by heat, and includes hemolytic scission

of bonding, isomerization, and dehydration, which is accompanied by succeeding
various reactions. Thus, pyrolysis products include oil and char products, such
as flash pyrolysis oil. In SCW, water molecules attacks reactant severely, causing
reaction with water to take place easily, but due to the high temperature associated
with the supercritical state, pyrolysis also takes place. Especially, when large solid
particles are treated, inner part of the particle does not have chance to contact
with water, and pyrolysis is favored. As flash pyrolysis oil is a mixture of various
chemical compounds, pyrolysis products are usually mixture of various chemicals.
When production of chemicals is the target, pyrolysis reaction is not desirable
because the yield of the product is limited, and byproducts are various. For the case
of SCW treatment, direct liquefaction produces oil by treatment at low temperature
close to critical point and with short reaction time. Studies on pyrolysis in SCW
included work conducted by Szuppa et al. [61], Pandey and Kim [77], Mohammed
et al. [78], Brebu and Vasile [65], Akhtar and Amin [66], Toor et al. [71], Zhou et al.
[79], Shen et al. [80], Rezende et al. [81], Yong and Matsumura [21], and Kruse and
Ebert [82].
16.3.3 Hydrolysis
Hydrolysis is one of the main reactions utilized under supercritical conditions

especially in the organic wastes destruction. It is also the decomposition of the
feedstock, but it is the scission by the reaction with water of various bonds including
glycosidic bond, amide bond, and ester bond. Since water molecules in SCW are
under high temperature and high pressure, and ion product is rather high near
critical point, hydrolysis reaction is quick, and proceeds even when no catalyst is
obtained. Thus, glucan, protein, fat, and other polymers can be easily hydrolyzed in
SCW. Especially, for the case of cellulose, it is pointed out that hydrolysis reaction
becomes faster than glucose decomposition under supercritical condition, which
may be partly due to the dissolution of cellulose [83]. Hydrolysis reaction were
reported extensively by several studies such as Guo et al. [84], Rinaldi and Schuth
[49], Funke and Ziegler [62], Lu and Saka [63], Luterbacher et al. [64], Ju et al.
[85], Pandey and Kim [77], Akhtar and Amin [66], Rackemann and Doherty [86],
Zhu et al. [67], Pang et al. [69], Adschiri et al. [70], Toor et al. [71], Zetzl et al. [72]
(2011), Kruse [73], Zhou et al. [79], Rezende et al. [81], Guo et al. [87], Yong and
Matsumura [21], Jerome and Parsons [88].
16.3.4 Carbon-Carbon Bond Breakage
There are numerous studies on the formation and breakage of selected carbon-
carbon bonds in SCW. These include studies reported by Hu et al. [89], Funke and
Ziegler [62], Glezakou et al. [90], Rana and Chandra [91], Fedyaeva and Vostrikov
[92]. Unfortunately, C-C bond scission requires rather high temperature, and usually
is radical reaction. Then, various competitive reaction takes place, and it is not
easy to obtain high yield of the target product. Sasaki et al. studied retro-aldol
condensation to produce glycolaldehyde from glucose [93].
16.3.5 Hydration/Dehydration
Hydration and dehydration can also occur in SCW as the medium of reaction. This
has been discussed extensively by several studies such as Funke and Ziegler [62],
Toor et al. [71], Karinen et al. [94], Lesoin et al. [95], Torres-Alacan et al. [96],
Foustoukos and Mysen [97], and Wallen et al. [98]. Hydration is easy to understand
in SCW due to excessive amount of water around the reactant. Dehydration may
look surprising to take place in water phase. However, dehydration of sugars are
quite usual, and resulting products include 5-hydroxymethylfurfural.
16.3.6 Depolymerization
Depolymerization of inorganic and organic compounds into its monomers and

essentially its starting materials in SCW is an attractive method as way to recover
and recycling of the compounds. Several studies discussed the utilization of this
reaction into converting numerous types of polymers. These included studies done
by Lu and Saka [99], Pandey and Kim [77], Brebu and Vasile [65], Akhtar and
Amin [66], Moller et al. [68], Adschiri et al. [70], Toor et al. [71], Rezende et al.
[81], Gosselink et al. [100], Jin et al. [3], Yong and Matsumura [21]. To recover
the monomers, decomposition of the product monomers is to be prevented, which is
actually difficult when SCW is employed.
16.3.7 Gasification
Gasification of chemicals especially organic compounds such as biomass in SCW

has been discussed and studied extensively as a means to reform these organic
compounds into useful gaseous products (H2 , CO2 , CO, and CH4 ) that can be used
as a source of energy. Among these studies are by Balat et al. [33], Guo et al.
[84], Kalinci et al. [101], Marrone and Hong [102], Lu and Saka [63], Pandey
and Kim [77], Mohammed et al. [78], Zhu et al. [67], Toor et al. [71], Azadi and
Farnood [48], Kruse [73], Onwudili and Williams [59], Zhou et al. [79], Robbins
et al. [74], Yong and Matsumura [21], and Yeh et al. [103]. Another possibility of
gasification is to leave inorganics behind, removing all organics decomposed into
combustible gas. For example, recovery of hydroxyapatite from chicken manure
has been proposed based on the knowledge that distribution of inorganic compound
follows the thermodynamic prediction [104, 105].
16.3.8 H-D Exchange
The studies on H-D exchange reactions under supercritical conditions are essentially
to investigate the mechanism of hydrogen atom exchange reactions. This is done by
investigating the reactions in supercritical D2 O with the aim to produce deuterated
organic compounds [82, 97]. It does not directly related to the chemical production,
but gives insight to the reactions related to the chemical production. However, it is
to be noted that H-D exchange takes place rather fast between water molecule and
hydrogen in the polymers including cellulose.
16.3.9 Hydrogenation
Several researchers reported effective hydrogenation via the water-gas shift reaction
under supercritical conditions. Hydrogenation via water-gas shift reaction is possi-
ble by partial oxidation to produce carbon monoxide, which in turn reacts with water
molecule. It is reported that thus produced hydrogen is more reactive than molecular
hydrogen. Related studies include Pandey and Kim [77], Zhou et al. [79], Jerome
and Parsons [88], and Clarke et al. [106]. Hydrodesulfurization using this reaction
was proposed by Adschiri et al. [107].
16.3.10 Oxidation
Oxidation of organic compounds in SCW especially in waste destruction is widely

utilized and as results there are numerous studies reported on it. Oxidation of
various compounds in SCW was reviewed by Brunner [108]. Studies of SCWO

includes those by Marrone and Hong [102], Pandey and Kim [77], Cabeza et al.
[109], Adschiri et al. [70], Toor et al. [71], Tan and Lagerkvist [110], Fedyaeva and
Vostrikov [92], Jerome and Parsons [88], Mishra et al. [111], Savage et al. [76], Ding
et al. [112], Aki and Abraham [113], Cabeza et al. [109], and Hoffmann and Conradi
[114]. When it is related to production of chemicals rather than destruction of
organic waste, it is to be noted that oxidation of organic compound leaves inorganic
compound, which can be employed for purification and recovery of inorganics.
16.3.11 Partial Oxidation
Studies of partial oxidation in SCW was conducted by Marrone and Hong [102],
Azadi and Farnood [48], Yeh et al. [103]. It is to be noted that partial oxidation can
be used for hydrogenation as was discussed above. For the case of SCW gasification,
partial oxidation can be employed as a gasification enhancer, but too much oxygen
addition results in lower energy recovery due to the increase in CO2 yield [115].
16.3.12 Precipitation
Fedyaeva and Vostrikov [92] discussed the disposal of hazardous organic substances
in SCW through precipitation. This is not chemical reaction, but this kind of
approach is effective for purification of the product chemicals. Other utilization of
precipitation includes the production of titanium dioxide powders [116].
16.4 Conclusions
Studies on the production of chemicals in SCW were reviewed from the viewpoint
of feedstocks and reactions. Various kinds of reactions can take place in SCW, but
the yield and selectivity is the problem. SCW is often too severe to allow only one
single reaction to obtain one compound in a significant yield. Thus, purification
of the target compound from very dilute concentration in mixtures of many kinds
of compound is needed after the reaction process. This purification costs expense
and energy, together with the high initial cost of SCW process and large energy
needed when no heat recovery is made, is a barrier for the production of chemicals
in SCW. Only the exception is the case where chemical equilibrium is achieved by
sufficiently high temperature or sufficiently long reaction time, where main product
is gases such as hydrogen, carbon dioxide, and methane for the case of organic
feedstock, or other stable inorganic compounds such as hydroxyapatite.
References
1. Antal Jr MJ, Allen SG, Schulman D, Xu X. Biomass gasification in supercritical water. Ind
Eng Chem Res. 2000;39:4040–53.
2. Siskin M, Katritzky AR. A review of the reactivity of organic compounds with oxygen-
containing functionality in superheated water. J Anal Appl Pyrolysis. 2000;54:193–214.
3. Jin F, Zeng X, Jing Z, Enomoto H. A potentially useful technology by mimicking nature—
rapid conversion of biomass and CO2 into chemicals and fuels under hydrothermal conditions.
Ind Eng Chem Res. 2012;51(30):9921–37.
4. Kabyemela BM, Adschiri T, Malaluan RM, Arai K. Glucose and fructose decomposition
in subcritical and supercritical water: detailed reaction pathway, mechanisms, and kinetics.
Ind Eng Chem Res. 1999;38:2888–95.
5. Fang Z, Minowa T, Smith Jr RL, Ogi T, Kozinski JA. Liquefaction and gasification
of cellulose with Na2 CO3 and Ni in subcritical water at 350 ı C. Ind Eng Chem Res.
2004;43(10):2454–63.
6. Kabyemela BM, Adschiri T, Malaluan RM, Arai K. Kinetics of glucose epimerization and
decomposition in sub critical and supercritical water. Ind Eng Chem Res. 1997;36:1552–8.
7. Behrendt F, Neubauer Y, Oevermann M, Wilmes B, Zobel N. Direct liquefaction of biomass.
Chem Eng Technol. 2008;31(5):667–77.
8. Chuntanapum A, Matsumura Y. Formation of tarry material from 5-HMF in subcritical and
process: a study of model compounds. Ind Eng Chem Res. 2010;49:4055–62.
10. Chuntanapum A, Matsumura Y. Role of 5-HMF in supercritical water gasification of glucose.
J Chem Eng Jpn. 2011;44(2):91–7.
11. Chuntanapum A, Shii T, Matsumura Y. Acid-catalyzed char formation from 5-HMF in
subcritical water. J Chem Eng Jpn. 2011;44(6):431–6.
12. Chuntanapum A, Yong TL-K, Miyake S, Matsumura Y. Behavior of 5-HMF in subcritical and
13. Promdej C, Chuntanapum A, Matsumura Y. Effect of temperature on tarry material production
of glucose in supercritical water gasification. J Jpn Inst Energy. 2010;89(12):1179–84.
14. Promdej C, Matsumura Y. Temperature effect on hydrothermal decomposition of glucose in
sub- and supercritical water. Ind Eng Chem Res. 2011;50(14):8492–7.
15. Kruse A, Gaulik A. Biomass conversion in water at 330–410 ı C and 30–50 MPa. Identifica-
tion of key compounds for indicating different chemical reaction pathways. Ind Eng Chem
Res. 2003;42:267–79.
16. Smith RL, Fang Z. Techniques, applications and future prospects of diamond anvil cells for
studying supercritical water systems. J Supercrit Fluids. 2009;47(3):431–46.
17. Minowa T, Fang Z. Hydrogen production from cellulose in hot compressed water using
reduced nickel catalyst: product distribution at different reaction temperatures. J Chem Eng
Jpn. 1998;31(3):488–91.
nickel catalyst. J Supercrit Fluids. 1998;13:253–9.
19. Minowa T, Fang Z, Ogi T, Varhegyi G. Decomposition of cellulose and glucose in hot-
compressed water under catalyst-free conditions. J Chem Eng Jpn. 1998;31(1):131–4.
20. Fang Z, Sato T, Smith Jr RL, Inomata H, Arai K, Kozinski JA. Reaction chemistry and
phase behavior of lignin in high-temperature and supercritical water. Bioresour Technol.
2008;99(9):3424–30.
21. Yong TL-K, Matsumura Y. Reaction kinetics of the lignin conversion in supercritical water.
Ind Eng Chem Res. 2012;51(37):11975–88.
22. Yong TL-K, Matsumura Y. Kinetic analysis of lignin hydrothermal conversion in sub- and
supercritical water. Ind Eng Chem Res. 2013;52(16):5626–39.
23. Yong TL-K, Matsumura Y. Kinetic analysis of guaiacol conversion in sub- and supercritical
24. Yong TL-K, Matsumura Y. Reaction pathways of phenol and benzene decomposition in
supercritical water gasification. J Jpn Petrol Inst. 2013;56(5):331–43.
25. Saka S, Konishi R. Chemical conversion of biomass resources to useful chemicals and
fuels by supercritical water treatment. In: Bridgwater AV, editor. Progress in thermochemical
biomass conversion. Oxford: Blackwell Science Ltd; 2001. p. 1338–48.
26. Veski R, Palu V, Kruusement K. Co-liquefaction of kukersite oil shale and pine wood in
supercritical water. Oil Shale. 2006;23(3):236–48.
27. Demirbas AH, Demirbas I. Importance of rural bioenergy for developing countries. Energy
Convers Manag. 2007;48(8):2386–98.
28. Savage PE. A perspective on catalysis in sub- and supercritical water. J Supercrit Fluids.
2009;47(3):407–14.
29. Saka S, Ehara K, Minami E. Efficient utilization of woody biomass with supercritical fluid
technologies (in Japanese). Mokuzai Gakkaishi. 2005;51(4):207–17.
30. Arai K, Smith RL, Aida TM. Decentralized chemical processes with supercritical fluid
technology for sustainable society. J Supercrit Fluids. 2009;47(3):628–36.
31. King JW, Srinivas K. Multiple unit processing using sub- and supercritical fluids. J Supercrit
Fluids. 2009;47(3):598–610.
32. Matsumura Y, Minowa T, Potic B, Kersten S, Prins W, Vanswaaij W, Vandebeld B, Elliott D,
Neuenschwander G, Kruse A. Biomass gasification in near- and super-critical water: status
and prospects. Biomass Bioenergy. 2005;29(4):269–92.
33. Balat M, Balat M, Kırtay E, Balat H. Main routes for the thermo-conversion of
biomass into fuels and chemicals. Part 2: Gasification systems. Energy Convers Manag.
2009;50(12):3158–68.
34. Saxena RC, Seal D, Kumar S, Goyal HB. Thermo-chemical routes for hydrogen rich gas from
biomass: a review. Renew Sust Energ Rev. 2008;12(7):1909–27.
35. Sievers C, Valenzuela-Olarte MB, Marzialetti T, Musin I, Agrawal PK, Jones CW. Ionic-
liquid-phase hydrolysis of pine wood. Ind Eng Chem Res. 2009;48:1277–86.
36. Sealock JLJ, Elliott DC, Baker EG, Butner RS. Chemical processing in high-pressure aqueous
environments. 1. Historical perspective and continuing developments. Ind Eng Chem Res.
1993;32:1535–41.
37. Elliott DC, Phelps MR, Sealock Jr LJ, Baker EG. Chemical processing in high-pressure
aqueous environments. 4. Continuous-flow reactor process development experiments for
organics destruction. Ind Eng Chem Res. 1994;33:566–74.
38. Elliott DC, Sealock Jr LJ. Chemical processing in high-pressure aqueous environments: low-
temperature catalytic gasification. Chem Eng Res Des. 1996;74:563–6.
39. Elliott DC, Sealock Jr LJ, Baker EG. Chemical processing in high-pressure aqueous environ-
ments. 2. Development of catalysts for gasification. Ind Eng Chem Res. 1993;32:1542–8.
40. Elliott DC, Sealock Jr LJ, Baker EG. Chemical processing in high-pressure aqueous
environments. 3. Batch reactor process development experiments for organics destruction.
Ind Eng Chem Res. 1994;33:558–65.
41. Sealock LJ, Elliott DC, Baker EG, Fassbender AG, Silva LJ. Chemical processing in
high-pressure aqueous environments. 5. New processing concepts. Ind Eng Chem Res.
1996;35:4111–8.
42. Elliott DC, Newenschwander GG, Phelps MR, Hart TR, Zacher AH, Silva LJ. Chemical
processing in high-pressure aqueous environments. 6. Demonstration of catalytic gasification
for chemical manufacturing wastewater cleanup in industrial plants. Ind Eng Chem Res.
1999;38:879–83.
43. Osada M, Sato T, Watanabe M, Adschiri T, Arai K. Low-temperature catalytic gasification
of lignin and cellulose with a ruthenium catalyst in supercritical water. Energy Fuel.
2004;18:327–33.
44. Yoshida T, Matsumura Y. Gasification of cellulose, xylan and lignin mixtures in supercritical
45. Yoshida T, Oshima Y, Matsumura Y. Gasification of biomass model compounds and real
biomass in supercritical water. Biomass Bioenergy. 2004;26(1):71–8.
46. Kruse A, Meier D, Rimbrecht P, Schacht M. Gasification of pyrocatechol in supercritical
water in the presence of potassium hydroxide. Ind Eng Chem Res. 2000;39:4842–8.
47. Xu X, Matsumura Y, Stenberg J, Antal Jr MJ. Carbon-catalyzed gasification of organic
feedstocks in supercritical water. Ind Eng Chem Res. 1996;35:2522–30.
gasification of biomass and wastes. Int J Hydrog Energy. 2011;36(16):9529–41.
49. Rinaldi R, Schüth F. Design of solid catalysts for the conversion of biomass. Energy Environ
Sci. 2009;2(6):610.
50. Peterson AA, Vogel F, Lachance RP, Fröling M, Antal JMJ, Tester JW. Thermochemical bio-
fuel production in hydrothermal media: a review of sub- and supercritical water technologies.
Energy Environ Sci. 2008;1(1):32–65.
51. Goto M. Chemical recycling of plastics using sub- and supercritical fluids. J Supercrit Fluids.
2009;47(3):500–7.
52. Cansell F, Rey S, Beslin P. Thermodynamic aspects of supercritical fluids processing:
applications to polymers and wastes treatment. Revue de l’Institut Francais du Petrole.
1998;53(1):71–98.
53. Fang Z, Kozinski JA. A comparative study of polystyrene decomposition in supercritical water
and air environments using diamond anvil cell. J Appl Polym Sci. 2001;81(14):3565–77.
54. Kronholm J, Vastamaki P, Rasanen R, Ahonen A, Hartonen K, Riekkola M-L. Thermal
field-flow fractionation and gas chromatography-mass spectrometry in determination of
decomposition products of expandable polystyrene after reactions in pressurized hot water
and supercritical water. Ind Eng Chem Res. 2006;45:3029–35.
55. Jankauskaite V, Macijauskas G, Lygaitis R. Polyethylene terephthalate waste recycling and
application possibilities: a review. Mater Sci Medziagotyra. 2008;14(2):119–27.
56. Goto M, Koyamoto H, Kodama A, Hirose T, Nagaoka S. Depolymerization of polyethylene
terephthalate in supercritical methanol. J Phys Condens Matter. 2002;14(44):11427–30.
57. Genta M, Iwaya T, Sasaki M, Goto M. Supercritical methanol for polyethylene terephthalate
depolymerization: observation using simulator. Waste Manag. 2007;27(9):1167–77.
58. Park Y, Hool JN, Curtis CW, Robers CB. Depolymerization of styrene-butadiene copolymer
in near-critical and supercritical water. Ind Eng Chem Res. 2001;40:756–67.
59. Onwudili JA, Williams PT. Reaction of different carbonaceous materials in alkaline
hydrothermal media for hydrogen gas production. Green Chem. 2011;13(10):2837.
60. Balat M, Balat M, Kırtay E, Balat H. Main routes for the thermo-conversion of
biomass into fuels and chemicals. Part 1: Pyrolysis systems. Energy Convers Manag.
2009;50(12):3147–57.
61. Szuppa T, Stolle A, Ondruschka B. Fate of monoterpenes in near-critical water and
supercritical alcohols assisted by microwave irradiation. Org Biomol Chem. 2010;8(7):
1560–7.
62. Funke A, Ziegler F. Hydrothermal carbonization of biomass: a summary and discussion of
chemical mechanisms for process engineering. Biofuels Bioprod Biorefin. 2010;4:160–77.
63. Lü X, Saka S. Hydrolysis of Japanese beech by batch and semi-flow water under subcritical
temperatures and pressures. Biomass Bioenergy. 2010;34(8):1089–97.
64. Luterbacher JS, Tester JW, Walker LP. High-solids biphasic CO2 -H2 O pretreatment of
lignocellulosic biomass. Biotechnol Bioeng. 2010;107(3):451–60.
65. Brebu M, Vasile C. Thermal degradation of lignin – a review. Cell Chem Technol.
2010;44:353–63.
66. Akhtar J, Amin NAS. A review on process conditions for optimum bio-oil yield in
hydrothermal liquefaction of biomass. Renew Sust Energ Rev. 2011;15(3):1615–24.
67. Zhu G, Zhu X, Fan Q, Wan X. Recovery of biomass wastes by hydrolysis in sub-critical water.
Resour Conserv Recycl. 2011;55(4):409–16.
68. Moller M, Nilges P, Harnisch F, Schroder U. Subcritical water as reaction environment:
fundamentals of hydrothermal biomass transformation. ChemSusChem. 2011;4:566–79.
69. Pang J, Zheng M, Wang A, Zhang T. Catalytic hydrogenation of corn stalk to ethylene glycol
and 1,2-propylene glycol. Ind Eng Chem Res. 2011;50(11):6601–8.
70. Adschiri T, Lee Y-W, Goto M, Takami S. Green materials synthesis with supercritical water.
Green Chem. 2011;13(6):1380.
71. Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass: a review of
subcritical water technologies. Energy. 2011;36(5):2328–42.
72. Zetzl C, Gairola K, Kirsch C, Perez-Cantu L, Smirnova I. High pressure processes in
biorefineries. Chem Ing Tech. 2011;83(7):1016–25.
73. Kruse A. Behandlung von Biomasse mit überkritischem Wasser. Chem Ing Tech.
2011;83:1381–9.
74. Robbins MP, Evans G, Valentine J, Donnison IS, Allison GG. New opportunities for the
exploitation of energy crops by thermochemical conversion in Northern Europe and the UK.
Prog Energy Combust Sci. 2012;38(2):138–55.
75. Kiran E. Supercritical fluid processing in the pulp and paper and the forest products industries.
ACS Symp Ser. 1995;608:380–401.
76. Savage P, Gopalan S, Mizan T, Martino C, Brock E. Reactions at supercritical conditions –
applications and fundamentals. AlChE J. 1995;41:1723–78.
77. Pandey MP, Kim CS. Lignin depolymerization and conversion: a review of thermochemical
methods. Chem Eng Technol. 2011;34(1):29–41.
78. Mohammed MAA, Salmiaton A, Wan Azlina WAKG, Mohammad Amran MS, Fakhru’l-
Razi A, Taufiq-Yap YH. Hydrogen rich gas from oil palm biomass as a potential source of
renewable energy in Malaysia. Renew Sust Energ Rev. 2011;15(2):1258–70.
79. Zhou CH, Xia X, Lin CX, Tong DS, Beltramini J. Catalytic conversion of lignocellulosic
biomass to fine chemicals and fuels. Chem Soc Rev. 2011;40(11):5588–617.
80. Shen D, Xiao R, Gu S, Luo K. The pyrolytic behavior of cellulose in lignocellulosic biomass:
a review. RSC Adv. 2011;1(9):1641–60.
81. Rezende CA, de Lima MA, Maziero P, Deazevedo ER, Garcia W, Polikarpov I. Chemical and
morphological characterization of sugarcane bagasse submitted to a delignification process
for enhanced enzymatic digestibility. Biotechnol Biofuels. 2011;4(1):54.
82. Kruse A, Ebert K. Chemical reactions in supercritical water.1. Pyrolysis of tert-butylbenzene.
Berichte der Bunsen-Gesellschaft. 1996;100(1):80–3.
hydrolysis in subcritical and supercritical water. J Supercrit Fluids. 1998;13:261–8.
84. Guo Y, Wang SZ, Xu DH, Gong YM, Ma HH, Tang XY. Review of catalytic super-
critical water gasification for hydrogen production from biomass. Renew Sust Energ Rev.
2010;14(1):334–43.
85. Ju Y-H, Huynh L-H, Kasim NS, Guo T-J, Wang J-H, Fazary AE. Analysis of soluble and
insoluble fractions of alkali and subcritical water treated sugarcane bagasse. Carbohydr
Polym. 2011;83(2):591–9.
86. Rackemann DW, Doherty WOS. The conversion of lignocellulosics to levulinic acid. Biofuels
Bioprod Biorefin. 2011;5(2):198–214.
87. Guo F, Fang Z, Xu CC, Smith RL. Solid acid mediated hydrolysis of biomass for producing
biofuels. Prog Energy Combust Sci. 2012;38(5):672–90.
88. Jerome KS, Parsons EJ. Metal-catalyzed alkyne cyclotrimerizations in supercritical water.
Organometallics. 1993;12(8):2991–3.
89. Hu B, Wang K, Wu L, Yu SH, Antonietti M, Titirici MM. Engineering carbon materials from
the hydrothermal carbonization process of biomass. Adv Mater. 2010;22(7):813–28.
90. Glezakou VA, Rousseau R, Dang LX, McGrail BP. Structure, dynamics and vibrational
spectrum of supercritical CO2 /H2 O mixtures from ab initio molecular dynamics as a function
of water cluster formation. Phys Chem Chem Phys PCCP. 2010;12(31):8759–71.
91. Rana MK, Chandra A. Solvation structure of nanoscopic hydrophobic solutes in supercritical
water: results for varying thickness of hydrophobic walls, solute–solvent interaction and
solvent density. Chem Phys. 2012;408:28–35.
92. Fedyaeva ON, Vostrikov AA. Disposal of hazardous organic substances in supercritical water.
Russ J Phys Chem B. 2013;6(7):844–60.
93. Sasaki M, Goto K, Tajima K, Adschiri T, Arai K. Rapid and selective retro-aldol condensation
of glucose to glycolaldehyde in supercritical water. Green Chem. 2002;4(3):285–7.
94. Karinen R, Vilonen K, Niemela M. Biorefining: heterogeneously catalyzed reactions of
carbohydrates for the production of furfural and hydroxymethylfurfural. ChemSusChem.
2011;4:1002–16.
95. Lesoin L, Boutin O, Crampon C, Badens E. CO2 /water/surfactant ternary systems and
liposome formation using supercritical CO2 : a review. Colloids Surf A Physicochem Eng
Asp. 2011;377(1–3):1–14.
96. Torres-Alacan J, Kratz S, Vohringer P. Independent pairs and Monte-Carlo simulations of
the geminate recombination of solvated electrons in liquid-to-supercritical water. Phys Chem
Chem Phys PCCP. 2011;13(46):20806–19.
97. Foustoukos DI, Mysen BO. D/H fractionation in the H2 –H2 O system at supercritical
water conditions: compositional and hydrogen bonding effects. Geochim Cosmochim Acta.
2012;86:88–102.
98. Wallen S, Palmer B, Pfund D, Fulton J, Newville M, Ma Y, Stern E. Hydration of bromide
ion in supercritical water: an X-ray absorption fine structure and molecular dynamics study.
J Phys Chem A. 1997;101(50):9632–40.
99. Lü X, Saka S. New insights on monosaccharides’ isomerization, dehydration and fragmenta-
tion in hot-compressed water. J Supercrit Fluids. 2012;61:146–56.
100. Gosselink RJ, Teunissen W, van Dam JE, de Jong E, Gellerstedt G, Scott EL, Sanders
JP. Lignin depolymerisation in supercritical carbon dioxide/acetone/water fluid for the
production of aromatic chemicals. Bioresour Technol. 2012;106:173–7.
101. Kalinci Y, Hepbasli A, Dincer I. Biomass-based hydrogen production: a review and analysis.
Int J Hydrog Energy. 2009;34(21):8799–817.
102. Marrone PA, Hong GT. Corrosion control methods in supercritical water oxidation and
gasification processes. J Supercrit Fluids. 2009;51(2):83–103.
103. Yeh T, Dickinson J, Franck A, Linic S, Thompson L, Savage P. Hydrothermal catalytic
production of fuels and chemicals from aquatic biomass. J Chem Technol Biotechnol.
2013;88:13–24.
104. Yanagida T, Minowa T, Nakamura A, Matsumura Y, Noda Y. Behavior of inorganic
elements in poultry manure during supercritical water gasification. J Jpn Inst Energy.
2008;87(9):731–6.
105. Yanagida T, Minowa T, Shimizu Y, Matsumura Y, Noda Y. Recovery of activated carbon cata-
lyst, calcium, nitrogen and phosphate from effluent following supercritical water gasification
of poultry manure. Bioresour Technol. 2009;100(20):4884–6.
106. Clarke MJ, Harrison KL, Johnston KP, Howdle SM. Water in supercritical carbon dioxide
microemulsions: spectroscopic investigation of a new environment for aqueous inorganic
chemistry. J Am Chem Soc. 1997;119:6399–406.
107. Adschiri T, Shibata R, Sato T, Watanabe M, Arai K. Catalytic hydrodesulfurization of
dibenzothiophene through partial oxidation and a water-gas shift reaction in supercritical
2009;47(3):382–90.
109. Cabeza P, Bermejo MD, Jimenez C, Cocero MJ. Experimental study of the supercritical water
oxidation of recalcitrant compounds under hydrothermal flames using tubular reactors. Water
Res. 2011;45(8):2485–95.
110. Tan Z, Lagerkvist A. Phosphorus recovery from the biomass ash: a review. Renew Sust Energ
Rev. 2011;15(8):3588–602.
111. Mishra VS, Mahajani VV, Joshi JB. Wet air oxidation. Ind Eng Chem Res. 1995;34:2–48.
112. Ding ZY, Frisch MA, Li LX, Gloyna EF. Catalytic oxidation in supercritical water. Ind Eng
Chem Res. 1996;35(10):3257–79.
113. Aki SNVK, Abraham MA. Mass transfer and chemical reaction during catalytic supercritical
water oxidation of pyridine. ACS Symp Ser. 1997;670:232–41.
114. Hoffmann MM, Conradi MS. Nuclear magnetic resonance probe for supercritical water and
aqueous solutions. Rev Sci Instrum. 1997;68(1):159.
115. Matsumura Y, Harada M, Li D, Komiyama H, Yoshida Y, Ishitani H. Biomass gasification in
supercritical water with partial oxidation. J Jpn Inst Energy. 2003;82(12):919–25.
116. Matějová L, Matěj Z, Fajgar R, Cajthaml T, Šolcová O. TiO2 Powders synthesized by
pressurized fluid extraction and supercritical drying: effect of water and methanol on
structural properties and purity. Mater Res Bull. 2012;47(11):3573–9.
Chapter 17
Techno-economic Analysis of Renewable
Hydrogen Production via SCWG of Biomass
Using Glucose as a Model Compound
Dawood Al-Mosuli, Shahzad Barghi, Zhen Fang, and Chunbao (Charles) Xu
Abstract The world’s hydrogen production is around 45 million tons (500 million
m3 ) per year, primarily derived from fossil fuels, of which about 50 % is by
steam methane reforming at 800–900 ı C in the presence of a catalyst (typically
nickel-based). The production of hydrogen from bio-renewable sources and organic
wastes is promising with respect to the environmental problems associated with
fossil fuel use and reduction of dependence on the declining fossil-fuel reserves.
Hydrogen can be produced by air and/or steam gasification of solid biomass
or biomass-derived feedstock products or by-products (e.g. glucose, ethanol and
glycerol) at temperatures above 700–800 ı C with or without catalysts. Compared
to conventional steam gasification or reforming processes, supercritical water
gasification (SCWG) can increase the gasification efficiency, improve the hydrogen
yield and reduce tar and coke formation. In addition, SCWG can utilize the wet
biomass and wastes directly, eliminating the energy and capital-intensive drying
process. Besides the technical challenges for SCWG of biomass in a continuous flow
reactor system, e.g., the difficulty in feeding the slurry feedstock into a high-pressure
system, determining economic profitability for large-scale processes is essentially
needed for future commercialization of the process. This chapter presents the results
D. Al-Mosuli • S. Barghi ()

London, ON, Canada N6A 5B9
Z. Fang
Biomass Group, Key Laboratory of Tropical Plant Resource and Sustainable Use,
Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences,
88 Xuefulu, Kunming, Yunnan Province 650223, China
C. Xu ()
and Biochemical Engineering, Western University, London, ON, Canada N6A 5B9
446 D. Al-Mosuli et al.
of techno-economic analysis for production of hydrogen from glucose (a biomass

model compound) and sewage sludge waste materials via SCWG. Detailed process
simulation is carried out for the SCWG process evaluation with two feedstocks. Cost
of hydrogen production and the revenue obtained from different feedstock are also
evaluated.
Keywords Supercritical water gasification (SCWG) • Hydrogen generation

• Techno-economic analysis • Glucose • Sewage sludge
List of Abbreviations
ANI Annual net income.

ATE Annual total expenses.
AS Annual sales.
ATX Annual taxes.
PBP Payback period.
PRi Production rate of product i.
PSA Pressure swing absorption.
ROI Return on investment.
SCWG Super critical water gasification.
SPi Sale price of product i.
TR Taxation rate factor.
WHSV Weight hourly space velocity (h1 ).
17.1 Introduction
The world’s hydrogen production is around 45 million tons (500 million m3 ) per
year [1], primarily derived from fossil fuels (around 96 %), of which about 50 % is
by steam methane reforming (SMR) at 800–900 ı C in the presence of a catalyst
(typically nickel-based) [2]. Depletion of conventional fossil fuels and growing
energy demand as well as serious environmental concerns reveal the urgent need for
clean fuels from renewable resources with negligible impact on the environment.
Biomass as a renewable resource has been used in different processes such as
anaerobic digestion, pyrolysis, and gasification to produce fuels, and recently
supercritical water gasification (SCWG) gasifying biomass in water above its critical
points (373.95 ı C and 22.06 MPa), has shown potential to produce gaseous fuel
from low quality biomass feedstock such as high water-containing wastes materials
(food processing wastes, wastewater sludge, etc.). SCWG process is at its infancy
period and despite its unique features it requires more in depth analysis before
commercialization. The main advantage of SCWG is the production of high purity
renewable hydrogen for fuel cells.
17 Techno-economic Analysis of Renewable Hydrogen Production via. . . 447
The production of hydrogen from bio-renewable sources and organic wastes is

promising with respect to the environmental problems associated with the fossil fuel
use and reduction of the dependence on the declining fossil-fuel reserves. Hydrogen
can be produced by air and/or steam gasification of solid biomass or biomass-
derived feedstock products or by-products (e.g. glucose, ethanol and glycerol) at
temperatures above 700–800 ı C with or without catalysts [3–6]. Supercritical water
(SCW) demonstrated to be very effective in reforming glucose or ethanol into H2
and CO2 . Glucose at low concentrations (1.8 wt.%) could be completely gasified
in SCW after about 20 s at 600 ı C and 34.5 MPa [7], or at 700 ı C and 24.8 MPa
with Ru/Al2 O3 catalyst, and stoichiometric limit of hydrogen yield (i.e., 12 mol of
H2 /mol of glucose) was achieved [8], following the reaction of C6 H12 O6 (l) C 6H2 O
(l) ! 12H2 (g) C 6 CO2 (g), Hı D 625.0 kJ/mol. Xu’s group [9] investigated in
total 17 heterogeneous catalysts, with combinations of 4 transition metals (Ni, Ru,
Cu and Co) and various promoters (e.g., Na, K, Mg, or Ru) supported on different
support materials (”-Al2 O3 , ZrO2 , and activated carbon) for SCWG of glucose at
600 ı C and 24 M Pa in a bench-scale continuous-flow tubular reactor. Ni (in a
metallic form) and Ru (in both metallic and oxidized forms) supported on ”-Al2 O3
exhibited very high activity and H2 selectivity among all the catalysts investigated
for a time-on-stream of 5–10 h. With Ni2 0/”-Al2 O3 (i.e., ”-Al2 O3 with 20 wt.% Ni),
the H2 yield of 41 mol/kg-glucose was achieved, 20 times higher than that obtained
from the blank test without catalyst (1.7 mol/kg-glucose). Mg and Ru were found
to be effective promoters for the Ni/”-Al2 O3 catalyst to suppress the coke and tar
formation.
Compared to conventional steam gasification or reforming processes, SCWG
can increase the gasification efficiency, improve the hydrogen yield and reduce tar
and coke formation. In addition, SCWG can utilize the wet biomass and wastes
directly, eliminating the energy and capital-intensive drying process. Thus, it is
particularly suitable for gasifying wet biomass and some waste streams such as
aqueous products (carbohydrates, acetic acid) derived from biomass or bio-oils. The
major challenges for SCWG of biomass in a continuous flow reactor system are the
high-pressure requirement for the reactor system (which could be very expensive)
and the difficulty in feeding the slurry feedstock into a high-pressure system and the
costly feeding system.
The SCWG has attracted significant interest since the 1980s especially in Europe
and Japan, but there is no process demonstrated at a commercial scale. To achieve
future commercial production of renewable hydrogen from wet biomass and waste
materials via SCWG, it requires identifying the optimal technical and economic
scale for given input supplies and target markets. Gasafi and co-workers [10]
reported an economic analysis of sewage sludge gasification in SCW for hydrogen
production, and for disposing of sewage sludge as a bio-waste. Based on their
results, it is found that supercritical gasification may well be competitive due to
the revenues associated with the disposal of sewage sludge as a waste product.
However, data are not available on what scale commercial capacity would be best
for renewable hydrogen production from wet biomass to achieve the best economics
of the SCWG process.
Biomass contains a large amount of water, which makes it difficult for pyrolysis
as it needs dry matter. Drying is an expensive energy intensive process. For example,
the water hyacinth produces 150 tons of dry mat/ha.year, but its water content is
more than 95 % and evaporation of such an amount of water requires substantial
amount of energy. On the other hand, wet biomass is the best feedstock for SCWG
as it requires a high content of water as a reactant. Wet biomass cannot be used
directly in a dry gasification (oxygen or air-blown process). In addition, SCWG is
more preferable than biodegradation of biomass because it is not affected by the
lignin content and no biomass residues is produced.
As mentioned earlier, the SCWG process is new and more analysis is required
to investigate its profitability compared to other processes. Such analysis can be
realized using available software packages such as Hysys or UniSim to perform
plant design and its economic and sensitivity analysis in order to achieve process
optimization and benefit maximization. This chapter discusses the economics and
efficiencies of a SCWG process to convert glucose and sewage sludge into H2 /CH4 -
rich gaseous fuels, which can be used for the generation of clean energy.
17.2 Fundamentals of SCWG
SCW is water at temperatures and pressures higher than its critical point (373.95 ı C
and 22.06 MPa). It has the diffusivity of a gas and dissolving ability of a liquid.
Moreover, it is completely miscible with most of the weak-polar organic compounds
and gases, which would greatly facilitate chemical reactions. The reactivity of water
at these conditions will be very high leading to fast hydrolysis of cellulose to sugar
followed by fast decomposition. Due to the fact that half of the hydrogen produced
in the SCWG comes from water then the high activity water at these conditions will
result in a high hydrogen yield. The water in SCWG will act as a solvent and as
a reacting medium. Adschiri et al. [11] mentioned that cellulose decomposes more
quickly in supercritical conditions than in subcritical conditions. This high activity
could be used efficiently to decompose organic raw materials without expensive
pre-treatment processes. In addition, it is found that the yield of CO gas in SCWG
process is much lower, less than 1 % [10], while in traditional dry process, CO is one
of the main products, which needs further treatment. The main advantage of SCWG
is that SCWG can utilize the wet biomass and wastes directly, eliminating the energy
and capital-intensive drying process that is otherwise required for the conventional
air/steam gasification processes. Thus, it is particularly suitable for gasifying wet
biomass and some waste streams such as waste water sludge or aqueous products
(carbohydrates, acetic acid) derived from biomass or bio-oils.
As a feedstock for gasification, biomass has some advantages over commercial
coal gasification. It is more reactive and thus easier to gasify, and most biomass
contains negligible sulfur, so the costly sulfur removal processes are not required.
On the other hand, while coal facilities can benefit from economies of scale in capital
costs, the higher cost of transporting biomass (due to its low volumetric energy
density and the low efficiency of photosynthesis) will constrain biomass facilities to
more modest sizes [12].
17.3 Flowsheet and Simulation of SCWG
Figure 17.1 shows a simplified general flowsheet for the proposed SCWG process
using Hysys software. The main units for the process are shown with the connecting
streams. The feed could be low quality (purity) glucose water solution, or sewage
sludge. The flowrate of the feedstock is set to 30 t/h of wet biomass with 15 wt.%
solids (i.e., 4.5 tons dry mass/h). This corresponds to the disposal capacity of a
city with a population of approximately 1,300,000 inhabitants based on roughly
30 kg dry material per person per annum [13]. Depending on rough stoichiometric
calculations, and assuming that the amount of organic mass is 80 % in sewage
sludge, then this plant will produce approximately 5.5 % of the hydrogen demand
for a 270,000 ton/h ammonia plant. This quantity could be increased or its
concentration could be adjusted by adding chicken manure, bean curd refuse,
Fig. 17.1 Flowchart of the proposed SCWG process for hydrogen production from biomass. (PSA
pressure swing adsorption unit)
sawdust, etc. Such a plant is also capable of processing low quality glucose, and
relatively complete conversion of the biomass can be achieved at the supercritical
conditions (inorganics are assumed to be mostly removed from sewage feedstock
before processing).
Based on the empirical stoichiometric equation [10];
C6 H12 O6 C 5H2 O ! 11=2 CO2 C 1=2CH4 C 10H2 : : : Hr 0 D 501:4 kJ=mol

(17.1)
In biomass SCWG, glucose is taken as a model compound to represent the

gasification reactions. This is because the biomass gasification includes a complex
set of reactions between many reactants. A detailed model, which includes all of
the reaction pathways is very difficult to develop and simulate. Aforementioned
glucose gasification reaction according to the above equation is taken to simplify
the calculations.
If a complete conversion is assumed, 503.558 kg/h of H2 is produced from the
decomposition of 30 t/h organic biomass. Inorganic compounds are assumed to
be additional inert materials entering and leaving the reactor. Some of these inert
materials are precipitated completely and separated by gravity as a bottom stream
after the reactor [14]. It is assumed that the inorganic salts like chlorides in the
supercritical stream are carried with the stream through packed bed and separated
with effluent stream in the CO2 flash separator.
In Fig. 17.1 the conditioned aqueous mixture (organics 15 wt.% solids) is
pressurized using pump A to 28 MPa, then (the pressurized liquid is heated to
435 ı C using the reactor exit stream at 600 ı C in heat exchanger B). The preheated
mixture is heated to its reaction temperature (600 ı C) in a heat exchanger using the
hot flue gas-air mixture from a furnace. The reactor is a packed bed catalytic reactor
with heating coils to provide heat to the endothermic gasification reaction (Eq. 17.1).
The catalyst is Alumina supported RuNi [9]. Heat is supplied to both the reactor and
heat exchanger C by burning natural gas (and any methane produced in this process)
in the furnace. The reaction product mixture is mainly composed of hydrogen,
CO2 , methane, and water at supercritical conditions. This mixture is cooled to
(130–140 ı C) by providing heat to the feed stream in heat exchanger B, followed
by cooling further to 40 ı C in cooler D using cooling water. Stream pressure is
still around 28 MPa. Under these conditions, condensation occurs, and the gas is
separated from the liquid in two phase separator E. CO2 has a high solubility in
water at this high pressure compared to hydrogen and methane. Around 30 % of
the CO2 produced remains dissolved in water and leaves separator E, representing
around 7 wt.% of the bottom stream. The high pressure stream is then passed into a
back-pressure regulating (expansion) valve, where its pressure is reduced to ambient
pressure.
The pressure of the overflow stream of separator E is reduced to 15 MPa with
a back-pressure regulator before entering the absorber (scrubber). This pressure
reduction wouldn’t affect appreciably the solubility of CO2 in water [15], but
reduces the cost of the absorber and the scrubbing water pump. The gaseous mixture
is sent to an absorption tower where water is the solvent. The bottom stream of
the absorber, which contains 5 wt.% of its weight CO2 is depressurized to 1 atm
and mixed with the depressurized bottom stream of separator E. Dissolved CO2 is
released out of the water stream in CO2 phase separator F. Part of the water produced
at 25 ı C is sent to a cooling tower (not shown in the Figure) where its temperature
drops to 15 ı C, after which it is recycled back to the scrubber to minimize water
consumption in the process.
The upper product of the CO2 absorber consists of hydrogen, methane and a
small amount of CO2 . This stream flows to the pressure swing adsorption unit (PSA)
after reducing the pressure to 3 MPa, which is the optimum pressure for the PSA
unit [10]. High pressure of the hydrogen facilitates its transportation and use in other
plants such as ammonia production. In this unit two columns are in use alternatively
to perform the absorbing/desorbing of the methane and CO2 for purification of the
gas stream. Pressure swing units are capable of reducing the impurities in H2 stream
to less than 10 ppm [16], which is important for some applications such as fuel
cells [17]. Hydrogen purity of 99.999 vol.% could be obtained in PSA units [16].
Common adsorbents used in hydrogen PSA systems include alumina, silica gel,
zeolites and activated carbon [18]. In this study, a commonly used adsorbent –
typically zeolite is used, with unit cell formula of Na86 [(AlO2 )86 (SiO2 )106 ] and a
window and effective channel diameter of 0.84 nm [19]. Using a relatively larger
PSA unit to reduce the load of CO2 scrubbing was found to be beneficial to increase
the efficiency and reduce the operating costs [20, 21].
In this chapter, the calculations are carried out for different scenarios and feed
concentrations. In case of SCWG of glucose, the reaction equation will be different
(only traces of Methane will be produced) at the given gasification conditions [9].
C6 H12 O6 .l/ C 6H2 O .l/ ! 12H2 .g/ C 6 CO2 .g/ ; Hı D 625:0 kJ=mol (17.2)
At 15 % concentration of organics, it is assumed that there will be no big

difference between the glucose solution and the sewage sludge solution in physical
properties of the flow streams for equipment sizing.
17.4 Simulation, Mass and Energy Balances
The units were designed to estimate the equipment cost. The sizing is based on mass
and energy balances using Hysys simulation software. SRK fluid package is used for
the simulation of this process except when condensation of water occurs. Starting
from cooler D, NRTL fluid package is used as it fits better the solubility data for
CO2 in water [15].
Mass and energy balance calculations are carried out for the flow scheme in
Fig. 17.1. Calculations are done for 30 metric ton/h of both 15 and 25 wt.% glucose
concentrations and for 15 wt.% sewage sludge. Results of mass and energy balances
Table 17.1 Mass and energy balances for 15 wt.% glucose feedstock concentration
Stream no. 1 3 4 5
Temperature (ı C) 25.00 435.00 600.00 600.00
Pressure (kPa) 101 28,000 28,000 28,000
Molar flow (kgmol/h) 1,440.45 1,440.45 1,440.45 1,715.19
Mass flow (kg/h) 30,000 30,000 30,000 30,000
Heat flow (kJ/h) 4.09E8 3.35E8 3.14E8 3.34E8
Dextrose (kg/h) 4,500.00 4,500.00 4,500.00 0.00
H2 O (kg/h) 25,500.00 25,500.00 25,500.00 22,800.00
CO2 (kg/h) 0.00 0.00 0.00 6,595.00
H2 (kg/h) 0.00 0.00 0.00 604.00
Stream no. 16 31 14 15
Temperature (ı C) 14.26 22.44 22.44 22.44
Pressure (kPa) 3,000 101 101 101
Molar flow (kgmol/h) 286.17 6,979.91 163.95 6,815.96
Mass flow (kg/h) 580.68 129,419.31 6,528.48 122,890.83
Heat flow (kJ/h) 125,271 20.02E8 0.585E8 19.43E8
Dextrose (kg/h) 0.00 0.00 0.00 0.00
H2 O (kg/h) 0.55 122,799.54 79.17 122,720.37
CO2 (kg/h) 3.41 6,592.25 6,421.81 170.44
H2 (kg/h) 576.71 27.51 27.50 1.64E-02
Stream no. Hydrogen Impurities 6C 7
Temperature (ı C) 14.27 – 133.63 40.00
Pressure (kPa) 150 150 2.80EC04 2.8EC04
Molar flow (kgmol/h) 286.06 0.10 1.72EC03 1.7EC03
Mass flow (kg/h) 576.71 3.96 3.00EC04 3.0EC04
Heat flow (kJ/h) 87,207 39,357 4.07EC08 4.2EC08
Dextrose (kg/h) 0.00 0.00 0.00 0.00
H2 O (kg/h) 0.00 0.55 22,800.10 22,800.10
CO2 (kg/h) 0.00 3.41 6,595.67 6,595.67
H2 (kg/h) 576.71 0.00 604.23 604.23
of different streams as well as the required parameters are shown in Tables 17.1,
17.2, 17.3. The following assumptions were made and used in the mass and energy
balance calculations:
(a) SCWG reactions are carried out at 600 ı C and 28 MPa.
(b) 5 % heat loss in heat exchangers, gasifier, furnace or piping system is assumed
(c) No ash content in the glucose feed
(d) Negligible amount of tar is produced in this process
(e) Hydrogen production per unit weight of dry glucose is constant for the two
feedstock concentrations
(f) The pressure drop across the equipment and piping system is assumed to be
negligible compared to high operating pressure
(g) Hydrogen losses in PSA tail gasses and scrubber and separator downstreams
are negligible
(h) The process is continuous for 328 days (90 % of the year time)
Table 17.2 Mass and energy balances for 25 % glucose feedstock concentration
Stream no. 1 3 4 5
Temperature (ı C) 25.00 435.00 600.00 600.00
Pressure (kPa) 101 28,000 28,000 28,000
Mass flow (kg/h) 30,000 30,000 30,000 30,000
Dextrose (kg/h) 7,500.00 7,500.00 7,500.00 0.0
H2 O (kg/h) 22,500.00 22,500.00 22,500.00 18,000.17
CO2 (kg/h) 0.0 0.0 0.0 10,992.77
H2 (kg/h) 0.0 0.0 0.0 1,007.04
Stream no. 16 31 14 15
Temperature (ı C) 14.00 20.00 20.00 20.00
Pressure (kPa) 3,000 101 101 101
Mass flow (kg/h) 984.03 189,015.96 10,865.41 178,150.54
Heat flow (kJ/h) 270,332 29.16E8 0.97E8 28.19E8
Dextrose (kg/h) 0.0 0.0 0.0 0.0
H2 O (kg/h) 0.93 177,999.23 114.75 177,884.48
CO2 (kg/h) 12.39 10,980.38 10,714.33 266.04
H2 (kg/h) 970.70 36.34 36.32 1.94E-02
Temperature (ı C) 14 – 97.70 40.00
Pressure (kPa) 3,000 137 28,000.0 28,000.0
Molar flow (kgmol/h) 481.50 0.33 1.7EC03 1.7EC03
Mass flow (kg/h) 970.70 13.33 3.0EC04 3.0EC04
Heat flow (kJ/h) 1.505E5 1.257E5 3.8EC08 3.8EC08
Dextrose (kg/h) 0.00 0.00 0.0 0.0
H2 O (kg/h) 0.00 0.94 18,000.2 18,000.2
CO2 (kg/h) 0.00 12.39 10,992.8 10,992.8
H2 (kg/h) 970.70 0.00 1,007.0 1,007.0
A summary of the calculated cost values for 15 and 25 wt.% glucose feed
concentrations is shown in Tables 17.7 and 17.8. These concentrations are chosen to
consider the reaction yield at high feed concentrations and the reduction in overall
plant efficiency at lower feed concentrations.
Table 17.4 depicts the overall mass and energy balances for the two glucose
feedstock (15 and 25 wt.%).
Production of CO2 and hydrogen is higher in the case of more concentrated
glucose feed due to higher amount of carbon and hydrogen in the feed. Since there
is more biomass in concentrated feed more methane is required.
Less cooling water is needed in the case of 25 wt.% glucose feed concentration
compared to 15 wt.% glucose. This is due to the fact that the heat duty of cooler
D is lower in case of 25 wt.% glucose feed. The main reasons are the change in
heat capacity of the product streams (more gases are produced in case 25 wt.%
Table 17.3 Mass and energy balances for 15 wt.% sewage feedstock concentration
Stream no. 1 3 4 5
Temperature (ı C) 25 435 600 600
Pressure (kPa) 101 28,000 28,000 28,000
Mass flow (kg/h) 30,000 30,000 30,000 30,000
Dextrose (kg/h) 4,500.00 4,500.00 4,500.00 0.00
H2 O (kg/h) 25,500.00 25,500.00 25,500.00 23,250.08
CO2 (kg/h) 0.00 0.00 0.00 6,046.02
H2 (kg/h) 0.00 0.00 0.00 503.55
Methane (kg/h) 0.00 0.00 0.00 200.32
Stream no. 16 31 14 15
Temperature (ı C) 14 23 23 23
Pressure (kPa) 3,000 101 101 101
Mass flow (kg/h) 673.79 109,326.20 6,013.04 103,313.16
Heat flow (kJ/h) 1E6 16.87E8 0.53E8 16.33E8
Dextrose (kg/h) 0.00 0.00 0.00 0.00
H2 O (kg/h) 0.48 103,249.59 76.67 103,172.92
CO2 (kg/h) 7.05 6,038.97 5,898.75 140.21
H2 (kg/h) 481.46 22.09 22.08 1.20E-02
Methane (kg/h) 184.78 15.53 15.52 1.50E-02
Temperature (ı C) 14 – 138.4 40.0
Pressure (kPa) 3,000 160 28,000.0 28,000.0
Molar flow (kgmol/h) 26.69 11.68 1,690.2 1,690.2
Mass flow (kg/h) 481 192.00 3.0EC04 3.0EC04
Heat flow (kJ/h) 7.6E6 874,224 4.1EC08 4.2EC08
Dextrose (kg/h) 0.00 0.00 0.0 0.0
H2 O (kg/h) 481.00 0.48 23,250.1 23,250.1
CO2 (kg/h) 0.00 7.05 6,046.0 6,046.0
H2 (kg/h) 0.00 0.00 503.6 503.6
Methane (kg/h) 0.00 184.46 200.3 200.3
glucose) and the degree of condensation in heat exchanger C upstream of cooler D

(see columns 6 and 7 in Tables 17.1 and 17.2).
17.5 Equipment Sizing and Cost Estimation
General Atomics [18] provided cost estimation for SCWG of 5,625 kg/h sewage
sludge for the production of hydrogen. The study was made for both 20 and 40 wt.%
of sewage sludge. The pretreatment liquefaction step was put at the beginning of
Table 17.4 Overall mass and energy balances for the two glucose feedstock
Materials and energy 15 wt.% glucose 25 wt.% glucose
H2 O in product (ton/h) 22.8 18
CO2 in product (ton/h) 6.59 10.99
H2 produced (ton/h) 0.604 1.007
Methane needed (ton/h) 1.306 1.524
Electricity needed (kW) 2,346a 3,008b
Cooling water (ton/h) at 15 ı C 46 26
Scrubbing water (ton/h) at 13 ı C 100 160
a
This power composed of 1,541 kW for furnace blower, 289.4 kW for feed pump and 515.8 kW
for scrubbing water pump. Cooling water pump needs only 1.647 kW then it is negligible
b
This power composed of 1,850 kW for furnace blower, 289.4 kW for feed pump and 868.6 kW
for scrubbing water pump. Cooling water pump needs negligible power
their process to precipitate the insoluble inorganic materials and avoid problems
associated with their presence in the feed line. Hydrogen purification is done using
membrane separation and a pressure swing adsorption unit. In 1999 another cost
estimation was done by Hemmes [22], for a feed rate of 7,500 kg/h of 15 wt.%
starch waste. Gas purification was done by membrane unit only.
Another cost estimation was performed in 2002 by a group at the Hiroshima
University using 5 wt.% of water hyacinths. The flowrate was low (42.67 kg/h).
Their estimation focuses on choosing between SCWG and biomethanization for
wet biomass utilization in Japan [23]. In 2008, Gasafi et al. [10] made another cost
estimation for 5,000 kg/h of sewage waste sludge with 20 wt.% solid content. The
process at such a high pressure (280 MPa) is not practically feasible, although it
leads to a very high solubility of CO2 in water and low scrubbing water requirement.
In the present work, a comprehensive cost estimation is carried out based on
equipment design. Design calculations for the case of 15 wt.% glucose feed concen-
tration are provided as an example, and the calculations’ results for 25 wt.% glucose
as well as 15 wt.% sewage sludge are included. Since hydrogen embrittlement
is a common problem in equipment containing hydrogen at high pressures and
due to the very high pressure and temperature in SCWG process, it may result in
serious equipment failure and explosion. Therefore, the Nickel alloy (INCOLOY
925) [24, 25] was chosen as the material of construction for all the equipment and
lines in contact with hydrogen. At low temperatures, rubber lined carbon steel is
generally used to resist the corrosion caused by carbonic acid and acetic acid traces
formed in the process. Costs are computed using November 2011 cost indices.
Table 17.5 provides a list of plant components in the case of 15 wt.% glucose feed
concentration.
Size and cost of the feed pump, heat exchangers, furnace and separator E didn’t
change appreciably with feed concentrations (15–25 %). Safety allowance was
considered in the design, so that the equipment can tolerate fluctuations in physical
properties and operating conditions. On the other hand, the size of the reactor
Table 17.5 List of plant equipment in the case of 15 wt.% glucose feed concentration (Lang
factor D 6 [26])
Equipment Cost ($) Equipment size
Preheater B 1,457,000 600 tubes, 12 m length, Nominal pipe diameter
NPS D 1 schedule XXS is used. Both of
the shell and the tubes are made of
INCOLOY 925 alloy.
Gas heater C 187,000 360 tubes, 2.4 m length, Nominal diameter
pipe NPS D 1 schedule XXS. Tubes are
made of INCOLOY 925 alloy. Shell is from
carbon steel.
Cooler D 60,000 240 tubes, 2.1 m length, Nominal pipe
diameter NPS D 1 schedule XXS is used.
Tubes are made of 304 SS alloy. Shell is
made of carbon steel.
Gasification reactor 565,000 The reactor is a packed-bed tubular reactor
with four equal size catalyst beds. Each
bed has D D 0.6 m and L D 0.9 m. The
total reactor length is 5.9 m. The reactor is
equipped with a finned tube heating coil
that has a Nominal diameter NPS D 1 with
20 fins 1=2 inch height by 0.035 in (20
BWG) width. Total length of coil is 370 ft.
(111 m).
Separator E 185,000 D D 1.287 m, L D 6.435 m. Made from rubber
lined carbon steel.
PSA unit 2,704,000 Estimation is based on manufacturer’s data by
UOP company [27].
Feed pump A 1,306,000 Cast steel positive displacement pump.
Scrubbing water pump 137,500 Cast iron rotary positive displacement pump.
Scrubber 872,000 D D 3.4 m, H D 8.5 m, 12 trays. Made from
lined rubber carbon steel.
Separator F 25,000 D D 2.58 m, L D 7.74 m. Made from rubber
lined carbon steel.
Flue gas heat recovery 57,000 400 tubes, 6 m length. Nominal pipe
diameter D 1, STD schedule.
Burner 2,786,000 Estimation is based on Gasafi et al. [10].
Total purchase price 10,343,000 –
Fixed capital investment 51,715,000 –
Total capital investment 62,058,000 –
will increase because of larger capacity and more heat-transfer area (heating coils)
needed. The size of the CO2 separator would increase as it should handle larger
quantities of scrubbing water. Table 17.6 provides a list of plant components in case
of 25 wt.% glucose feed concentration.
The design procedure and sizing of some of the main plant components are
briefly described as follows.
Table 17.6 List of plant components in the case of 25 wt.% glucose feed concentration
Equipment Cost ($) Equipment size
Preheater B 1,457,000 600 tube, 12 m length, Nominal pipe diameter
NPS D 1 schedule XXS is used.
Both of the shell and the tubes are made of
INCOLOY 925 alloy.
Gas heater C 187,000 360 tube, 2.4 m length, Nominal diameter pipe
NPS D 1 schedule XXS.
Tubes are made of INCOLOY 925 alloy, Shell
is made of carbon steel.
Cooler D 60,000 240 tube, 2.1 m length, Nominal pipe diameter
NPS D 1 schedule XXS is used.
Tubes are made of 304 SS alloy. Shell is made
of carbon steel.
Gasification reactor 700,000 The reactor has D D 0.6 m, L D 7.5 m, 6 beds
of catalyst with length of packing D 0.9 m
per each bed and finned tube heating coils
with a Nominal diameter NPS D 1 with 20
fins 1=2 in. height by 0.035 in (20 BWG)
width. Total length of coil is 370 ft.
(183 m).
Separator E 185,000 D D 1.287 m, L D 6.435 m. Made from rubber
lined carbon steel.
PSA unit 3,831,000 Estimation is based on manufacturer’s data by
UOP company [27].
Feed pump A 1,306,000 Cast steel positive displacement pump.
Scrubbing water pump 288,000 Cast iron rotary positive displacement pump.
Scrubber 1,458,000 D D 5.2 m, H D 8.5 m, 12 trays. Made from
lined rubber carbon steel.
Separator F 35,000 D D 2.9 m, L D 8.69 m. Made of rubber lined
carbon steel.
Flue gas heat recovery 57,000 400 tube, 6 m length. Nominal pipe
diameter D 1 STD schedule.
Burner 2,786,000 Estimation is based on Gasafi et al. [10].
Total purchase price 12,352,000
Fixed capital investment 61,759,000
Total capital investment 74,111,000
17.5.1 Feed Pump (A)
A positive displacement pump is used to pump 30 t/h of glucose water mixture at

28 MPa. The duty of this pump (calculated and supported by Hysys program) is
290 kW at 75 % efficiency and the present price of this pump (cast steel) is about
$1,306,250 [28].
17.5.2 SCWG Heat Recovery Heat Exchanger (B)
It is imperative to recycle the thermal energy, therefore a heat exchanger (horizontal

shell and tube type), was designed to heat the feed stream from 26 to 435 ı C (in
the tube side) using the hot product stream from 600 to 134 ı C in a countercurrent
fashion. The heat transfer coefficients are generally high for both streams (typically
in the range of 6,000–25,000 W/m2 ı C) [29, 30]. The tube side fouling factor is
assumed to be equal to that of the sea water (0.001 W/m2 ı C) because of no available
published data.
The area was found to be 1,261 m2 (14,008 ft2 ). The Nominal pipe diameter
NPS D 1 schedule XXS is used. [31] Number of tubes is 600 (two passes), each
having a length of 12 m. Both the shell and tubes are made of INCOLOY 925 alloy.
The associate cost is $1,457,000. According to [26].
17.5.3 Heat Exchanger (C)
Heat exchanger C is a horizontal shell and tube type used to raise the feed
temperature to 600 ı C (the reaction temperature) using the flue gases from the
furnace at 712 ı C in the shell side in a countercurrent fashion.
Heat transfer area is 155.4 m2 . The heat exchanger used, has two passes of 360
tubes with a Nominal diameter NPS D 1 schedule XXS [31] and a length of 2.4 m.
Tubes are made of INCOLOY 925 alloy. The cost will be $187,000 following [26].
17.5.4 Cooler (D)
A shell and tube horizontal type heat exchanger is used to cool down the products
mixture to 40 ı C and 28 M Pa. Cold water at 15 ı C is used for this purpose. Heat
transfer area is 90 m2 . The heat exchanger used has two passes of 240 tubes and
2.1 m length and a Nominal pipe diameter NPS D 1 schedule XXS [31]. Tubes are
made of 304 SS alloy. The cost is $60,000 according to [26].
17.5.5 Gasification Reactor
The reactor is a packed-bed cylindrical reactor. The reactor contains heating coils.
In these coils, flue gasses from the furnace pass at 712 ı C to supply heat to
the endothermic reaction. The catalyst used in these beds is alumina supported
RuNi with weight hourly space velocity (WHSV) D 13 h1 recommended for
gasification reaction [32]. WHSV (h1 ) is defined as (weight feeding rate of
feedstock, kg /h)/(weight of the catalyst in the packed bed of the tubular reactor, kg).
At WHSV of 3/h, the amount of catalyst is calculated as: (30,000 kg/h 15 wt.%)/
(3/h) D 1,500 kg catalyst. Price of catalyst is in the range of $1,000–$6,000
per ton [32], so $2,500/t is assumed in this study. The catalyst used has an
apparent particle density of 2,000 kg/m3. Thus, the total catalyst volume
is D (1,500 kg/2,500 kg)/m3 D 0.6 m3 . Assuming a void fraction of 40 %. To avoid
uneven distribution of phases over the catalyst, the length to diameter ratio of 1.5
is selected for each bed. Considering the pressure drop across the bed (less than
10 kPa/m) and using Ergun’s equation [33], a diameter of 0.6 m was considered
for the reactor. Therefore it is required to have 4 beds of catalyst inside the reactor.
Considering the space between the beds (0.3 m) and 0.6 m free space at both
entrance and exit ends, the total length of each reactor is L D 5.9 m.
This reactor will need a finned tube heating coil to provide heat for the
endothermic reaction. This coil is made of INCOLOY 925 alloy with area of
67.5 m2 . It has a Nominal diameter NPS D 1 with 20 fins 1=2 in. height by 0.035 in
(20 BWG) width. And total length of coil is 370 ft. (111 m).
Reactor wall, made of INCOLOY 925 alloy to resist high internal pressure at the
reaction temperature. The cost of the reactor is estimated at $565,000 (including the
reactor vessel, heating coils and catalyst).
17.5.6 Gas Liquid Separators (E)
Gas-liquid separator E is designed to separate the two phase gas-liquid stream

leaving cooler D at 280 bar and 40 ı C. It separates water with about 30 % of the CO2
produced in the process as a down flow. CO2 represents about 7 wt.% of the water
stream. Hydrogen, CO2 and impurities flow from the top of the vessel to the next
stage. A horizontal separator is selected because of large liquid volumetric flowrate.
Diameter of the separator is calculated assuming that the liquid height is set at
half the vessel diameter. The upper half is found to be more than enough for the gas
residence time, demister and liquid level controller. The diameter of this separator is
1.287 m, length D 6.435 m (L/D D 5 for pressures >3.5 MPa [33]). Wall thickness
is 14 mm due to the high pressure rating requirement.
The cost of this separator is estimated at $185,000 (rubber lined carbon steel)
[26].
17.5.7 Scrubbing Tower
A tray tower absorber was selected to separate CO2 from the gaseous stream. It uses
water at 15 MPa to absorb CO2 from the gas stream, because of high solubility of
CO2 in water at high pressures compared to other gases in the stream.
This tower works at 15 MPa and 40 ı C (pressure is reduced to reduce the
cost, since there is no big difference on CO2 solubility between 28 and 15 MPa).
This scrubber is 3.4 m diameter and 8.5 m height. It has 9 theoretical stages as
calculated by Hysys software. Plate efficiency is set at 0.8 [33]. It uses 80 t/h of
15 ı C water to absorb CO2 gas. This reduces the concentration of CO2 in hydrogen
gas stream to 1.6 wt.%. This concentration is easy to be treated in the PSA unit.
Total tower absorber cost (rubber lined carbon steel) with its internals (trays
valves etc.) is estimated at $4,054,000.
17.5.8 Scrubbing Water Pump (P)
This pump is a rotary positive displacement pump, with duty of 435 kW. [28]. It
takes water from the cooling tower to the top of the absorber. Its cost (cast iron
rotary positive displacement pump) is estimated at $137,500.
17.5.9 Pressure Swing Adsorption Unit (PSA)
The pressure swing adsorption unit adsorbs impurities like CO2 , CO, CH4 from
the hydrogen gas stream at 3–3.4 MPa [18]. In this unit two columns are in use
alternatively to adsorb methane and CO2 from the gaseous steam then releasing
impurities as tail gasses in desorption stage. Pressure swing units are capable of
reducing the impurities in H2 stream to less than 10 ppm, which is important for
fuel cells applications [16, 17]. Common adsorbents used in hydrogen PSA systems
include alumina, silica gel, zeolites and activated carbon [18]. Zeolite was chosen
in this study, with an effective pore diameter of 0.84 nm [19].
Since solid operation units cannot be simulated with the hysys software, a method
for PSA cost estimation by UOP Company [27] is used. The capital cost of the PSA
would be about $500,000 (gas flow rate in million scf/day)0.7 . The cost of this unit
is thus estimated at $2,704,000.
17.5.10 The Furnace Flue Gases Heat Recovery
A horizontal shell and tube heat exchanger in a countercurrent fashion was

employed for heat recovery from the furnace flue gases. It has a great benefit of
reducing the amount of fuel used in the process. It recovers the heat of the flue gases
leaving the heat exchanger C by reducing its temperature from 575 to 160 ı C to
increase the temperature of the fuel-air mixture from 25 to 462 ı C. Its heat transfer
area is 498 m2 . It uses 400 tubes with a Nominal pipe diameter of 1, STD schedule
[31] and 20 ft. long. Its cost is estimated at $57,229 [26]. Both tube and shell sides
are made of carbon steel.
17.5.11 Furnace
The furnace cost is estimated using the six tenth method and cost indices ratio
[10]. Its estimated cost is $2,785,970. Its cost and design will change with
feed concentration due to the need for higher heating requirement (and methane
quantities) to supply heat to the endothermic reaction.
17.5.12 Furnace Blower
Single stage axial flow blower is found to be a viable option to supply air to the
furnace [33]. It should supply 1.182 105 kg/h air at 160 kPa to the furnace. The
blower’s power is calculated at 1,540 kW.
17.5.13 Separator (F)
This separator is designed using the same method of separator E, but since it works
at atmospheric pressure and constructed using lined rubber carbon steel then its cost
will be around $25,000 only. It has a diameter of 2.58 m, and length of 7.74 m.
The plant components in the case of 25 wt.% glucose feed concentration were
also designed using the same method as described above. The reactor is a packed-
bed tubular reactor with 6 equal size catalyst beds. Each bed has D D 0.6 m and
L D 0.9 m. The total length of reactor is L D 7.5 m. The reactor is equipped with
a finned tube heating coil that has a Nominal diameter NPS D 1 with 20 fins 1=2
inch height by 0.035 in (20 BWG) width, with a total coil length of 610 ft. (183 m).
The total capital investment for the case of 25 % wt. glucose feed concentration is
calculated at $74,111,000.
17.6 Fuel Requirements in the Process
Since methane could be produced in this process if sewage is used as a feedstock

(185 kg/h is produced in case sewage is used), it is considered as part of the
fuel needed in the furnace. The heat is needed in two places in the process, heat
exchanger C and the gasification reactor. Total required methane in the case using
15 wt.% glucose feed stock is 1,306 kg/h, out of which 328 kg/h is burned to supply
heat to the reactor. In the case of 25 wt.% glucose solution feedstock, 1,524 kg/h
methane is needed.
17.7 Economic Analysis
A dynamic profitability analysis is done by computing the annual net income, annual
cash flow discount payback period, the net present value and discounted cash flow
rate of return [26, 29, 34].
Based on the above design and cost calculations, the cost of the SCWG equip-
ment is estimated and listed in Table 17.5 for 15 wt.% glucose feed concentration.
The equipment prices are given for feed flowrate of 30 ton/h. The estimated
equipment purchase prices are $10,343,000 and $12,352,000 for 15 wt.% and
25 wt.% glucose concentrations respectively.
As shown in Table 17.5, the scrubbing unit (scrubber and scrubbing pump),
the PSA unit, the gasification reactor and the burner play a major role in costs
estimation. Minimizing the cost of these units can reduce the total capital invest-
ment, depreciation and insurance costs. Recent studies indicate that water scrubbing
system could be replaced by more efficient physical absorbent (like Selexol)
[35, 36]. This absorbing system is cheaper and can work at lower pressures,
which reduces the cost of high pressure absorption system (absorber and water
absorption pump). Moreover, working at lower pressures would make it profitable
due higher pressure difference in the gas line, which may lead to higher power
generation. This could be achieved by using a turbine instead of the expansion
valve.
In addition, recent developments in adsorbents technology and adsorption oper-
ations promote using larger PSA units to adsorb all the CO2 and methane in gas
reforming streams [20, 21]. Extending this technology to SCWG could possibly
enable cancelling the scrubbing system by providing larger PSA unit. This could
reduce both fixed and operating costs. Recent advances in catalyst technology can
also reduce the cost of the reactor by lowering the severity of the operation and
reducing the residence time [37].
17.7.1 Direct and Indirect Costs
Direct costs include (installation, piping, instrumentation and control, electrical

systems, building and services, land and site development, and service facilities)
and indirect costs include (engineering and supervision, construction expenses, legal
expenses, contractor fees and contingency) representing the fixed capital investment
are determined with the help of ratio factors. Fixed capital investment is calculated
approximately by multiplying the total purchase price by a factor of 5 [26]. A
Lang factor of 6 is used to add the working capital and obtain the total capital
investment [26].
17.7.2 Manufacturing Costs
Manufacturing costs or annual total expenses include all expenses connected with
the manufacturing operation or the physical equipment of a process plant itself. The
total annual expenses can be divided into three categories:
Annual total expenses (ATE) D (Annual direct manufacturing costs) C (Annual
indirect manufacturing costs) C (Annual general expenses)
It should be noted that the plant works continuously for 24 h/day and 90 % of the
year.
17.7.2.1 Direct Manufacturing Costs
These costs represent expenses directly associated with the manufacturing opera-
tion, including expenditures for:
17.7.3 Raw Materials and Chemicals
Raw materials are hydrolyzed glucose solutions with different concentrations. The
cost is estimated to be $10.5/ton for 15 wt.% and $17.5/ton for 25 wt.% hydrolyzed
glucose solutions.
17.7.4 Labor Cost
3.6 skilled workers are assumed to operate the plant with 4 shifts per day due to the
intensity of the tasks. Supervision cost of 15 % of the labor cost is added. Average
salary is assumed to be $45,000 per year for each worker.
17.7.5 Utilities
Utilities needed in the pant are electricity, natural gas, cooling and scrubbing water.
Details of the utilities costs are calculated as follows:
Utility costs can be estimated by the method proposed by Ulrich [28]. It was
determined that utility costs generally contain two separately escalating compo-
nents: (1) – one comprised of materials and labor costs, which inflate at a rate
determined by the Chemical Engineering Plant Cost Index (CEPCI) [38]; (2) – the
others are energy or fuel costs, which can escalate at a much different rate.
Based on these components, a utility unit cost (CS,u ) is represented as follows.
CS;u D a CEPCI C b CS;f (17.3)
Where a and b D utility dependent parameters given in Ulrich, [28]; CS,f D cost of
fuel, normally heating oil ($/GJ), [39].
The values of factors a and b listed in Ulrich [28] depend on whether the utilities
are purchased from external supplier, which is usually more expensive, or are
generated on site by the company. In the second case the rates charged to a process
unit reflect the fixed capital and the operating costs.
The average price of electricity is assumed to be $0.1 per kW h [40]. Fuel (natural
gas) price is $3.82 MMBtu D 3.82/1.0546 D $3.622/GJ D $0.1802/kg [39]. The cost
of cooling water is assumed to be $1.54/m3.
17.7.5.1 Indirect Manufacturing Expenses
These include, overhead, depreciation, property taxes, insurance and rent that are
usually charged at a constant rate even when the plant is not in operation. Overhead
equals 60 % of the sum of all labor and maintenance costs. Depreciation is 10 % of
the fixed capital investment. Insurance and property taxes is approximately 2 % of
fixed capital cost.
17.7.5.2 Annual General Expenses
These costs include administrative costs, safety services, medical services, research
activities financing, etc. Administrative costs are 25 % of plant overhead. Research
and development costs 5 % of annual total expenses. Distribution and sales are 10 %
of annual total expenses [26].
Complete details of the calculations related to 15 and 25 wt.% glucose feed
concentration are provided in Tables 17.7 and 17.8. Comparison between the two
concentrations used is given in Fig. 17.2. It can be seen that annual total expenses
(ATE) is increased from $19,134,362 to $24,620,561 when increasing the feed
concentration from 15 to 25 wt.% glucose.
17.8 Profitability Analysis
Profitability analysis is summarized briefly below.

Annual sales revenue is the sum of sales revenue from all products and
byproducts:
Annual sales .AS / D †SPi PRi (17.4)
Where, SPi is the sale price of product i and PRi is the production rate of product i.
Table 17.7 Profitability analysis for 15 wt.% glucose feed concentration at hydrogen price 3.5$/kg
Direct manufacturing cost
Maintenance and repair D 6 % fixed cost $3,102,914
Operating supplies D 0.15*maintenance cost $465,437
Laboratory expenses $111,780
Labor C supervision $745,200
Patents and royalties Expenses D 0.3 % of total annual expenses $574,030
Utilities
Cooling and scrubbing water $1,681,204
Electricity $1.85EC06
Methane needed (NG) D 1,306 kg/h 10,296,504 kg/y $1.86EC06
Raw material (hydrolyzed glucose) 2,483,460
Indirect manufacturing expenses
Overhead (60 % of all labor C maintenance costs) $2,308,868
Insurance and property taxes D 2 % of fixed capital cost $1,034,304
Annual general expenses
Administration costs 25 % of plant overhead $577,217
Research and development D 5 % of annual total expenses $956,718
Distribution and sales D 10 % of total annual expenses $1,913,436
Annual total expenses (ATE) D Annual Direct manufacturing cost C Annual Indirect manufacturing $19,134,362
expenses C Annual general expense
Annual sales (AS)
17 Techno-economic Analysis of Renewable Hydrogen Production via. . .
Assume hydrogen price for selling 3.5 $/kg

CO2 price D 0.02 $/kg 0.02 $/kg
Hydrogen production is 579 kg/h 4,564,836 Kg/y
CO2 production D 6,595 kg/h 51,994,980 Kg/year
Annual sales AS D $/year $17,016,825
Annual gross income (AGI) D AS ATE $2,117,537
465
Table 17.8 Profitability analysis for 25 wt.% glucose feed concentration at hydrogen price of $3.5/kg
466
Direct manufacturing cost

Maintenance and repair D 6 % fixed cost $3,705,560
Operating supplies D 0.15*maintenance cost $555,834
Laboratory expenses $111,780
Labor C supervision $745,200
Patents and royalties Expenses D 0.3 % of total annual expenses $738,616
Utilities
Cooling and scrubbing water $2,471,930
Electricity $2.37E C 06
Methane needed (NG) 1,524 kg/h per year D 12157128 kg/year $2.19E C 06
Raw material (25 wt.% hydrolyzed glucose) $4,139,100
Indirect manufacturing expenses
Overhead (60 % of all labor C maintenance costs) $2,670,456
Insurance and property taxes D 2 % of fixed capital cost $1,235,186
Annual general expenses
Administration costs 25 % of plant overhead $667,614
Research and development D 5 % of annual total expenses $1,231,028
Distribution and sales D 10 % of total annual expenses $2,462,056
Annual total expenses (ATE) D Annual Direct manufacturing cost C Annual Indirect manufacturing $24,620,561
expense C Annual general expense
Annual sales (AS)
Assume hydrogen price for selling 3.5 $/kg
CO2 price D 0.02$/kg 0.02 $/kg
Hydrogen production is 970 kg/h 7647480 Kg/year
CO2 production D 10,990 kg/h 86645160 Kg/year
Annual sales AS D $/year $28,499,083
Annual gross income (AGI) D AS ATE $3,878,521
ADP annual depreciation 10 % of fixed capital investment D 6,175,933
Annual gross income after depreciation changes (AGID) 2,297,412
D. Al-Mosuli et al.
Fig. 17.2 The variation of 20

annual net income with the 15 wt.% glucose
Annual net income (Million $/y)

price of hydrogen in $/kg for 25 wt.% glucose
15
different feedstocks. *Sewage 15 wt.% sewage *
revenue is taken to be
$0.55/kg dry mass 10
0
0 2 4 6 8
-5
-10 Hydrogen price $/kg
In this process, the main product is the renewable hydrogen for ammonia
production or for fuel cells, and the relatively pure CO2 may be considered as a low
value byproduct. Hydrogen production per year is increased from 4,565 to 7,647
ton/year by increasing the feed concentration from 15 % to 25 % wt. The effect of
CO2 is insignificant in cost analysis. Its recovery is mainly due to environmental
concerns and may be sent for industrial uses like urea or soda plants. The price of
hydrogen has a crucial effect on the net profit, as shown later in Fig. 17.2.
In Tables 17.7 and 17.8, the annual gross income (AGI) is calculated from
the difference between annual sales and annual total expenses, i.e., AGI D AS –
ATE, where ATE does not include depreciation. The amount of depreciation and
taxes are calculated considering that annual depreciation to be 10 % of the fixed
capital investment and the taxation rate factor is 0.35. Annual gross income after
depreciation charges (AGID) is thus calculated by: AGID D AS – ATE – ADP where
ADP is the annual depreciation.
Depreciation is subtracted as a cost before income tax calculations for annual
taxes (ATX) by: ATX D AGID TR, where TR is taxation rate factor.
Annual net income (ANI) is calculated by subtracting the depreciation and taxes
from annual cross income, i.e., ANI D AGID – ATX D AS – ATE – ADP – TR(AS –
ATE – ADP) D AS (1 TR) – ATE (1 TR) – ADP C ADP TR.
As shown in Table 17.7, a plant based on the glucose concentration of 15 wt.%
is not economically feasible with the present price of hydrogen (around 3.5$/kg)
because the annual gross income after depreciation becomes negative. Figure 17.2
shows the variation of annual net income with hydrogen price at different feed
concentrations. From Fig. 17.2, it is clear that no profit is available with feed
concentration of 15 % until the hydrogen price rises over $5/kg. If the price changes
in future to $7 per kg, the process will be feasible to generate an annual net income
of $5,647,111.
For 25 % feed concentration (Table 17.8), there is a small profit of about $1.0
million per year at hydrogen price of $4.0/kg. It is clear now that there are two
crucial parameters affecting the profit, which are hydrogen price and glucose feed
concentration.
The annual cash flow (ACF) was calculated from the annual net income taking
into account the annual depreciation as follows:
ACF D .AS–ATE/ .1 TR/ C ADP TR (17.5)
The payback period (PBP) time is the minimum period of time theoretically needed
to recover the original capital investment, which is estimated by: Payback period
() D (Total capital Investment)/ACF. For glucose feed concentration of 15 wt.%,
there will be a payback period of 5.7 years if the selling price of hydrogen becomes
7$/kg. Feed concentration of 25 wt.% glucose is more promising. It’s payback
period is only 3.3 years for the same selling price.
Another important factor to be considered in the profitability analysis is
the Return on Investment (ROI), as defined by ROI D Annual net income
(ANI)/(Fixed C Working Capital Investments). A typical minimum desired ROI
is about 15 %). In this study for the feed concentration of 25 wt.% glucose and
selling price of $4 for hydrogen, the value of ROI was found to be around 1.3 %,
Therefore it is important to have higher hydrogen prices and lower production costs
to make the operation profitable. For example, if hydrogen price reaches $7/kg then
there will be a payback period 3 years and ROI D 21 % for the feed concentration
of 25 wt.% glucose.
17.9 Economic Analysis for Sewage Sludge Waste
Equipment needed for SCWG of sewage sludge with up to 20 wt.% solids is

the same as that needed for glucose (15 wt.%). Although physical properties
and quantity of methane produced will change slightly the size of the equipment
will be practically the same. However additional equipment will be required for
feed conditioning (stirring and milling purposes). Increasing concentrations above
20 wt.% will need a hydrothermal liquefaction step and different heat balance and
equipment sizing. Cost of the conditioning step was estimated to be $333,250
based on literature [10]. Less hydrogen and CO2 are produced from sewage.
From mass/heat balance calculation, hydrogen and CO2 productions are 3,792
and 46,507 ton/year, respectively. On the other hand, more methane is produced.
Methane is separated in the PSA unit and sent to the furnace to offset the total
methane required for heating the reactor. Methane production in this process is about
185 kg/h providing about 14 wt.% of its fuel demand at the furnace and the balance
(1,108 kg/h) should be supplied from external sources. Hydrogen production is
481 kg/h, which is less than 579 kg/h in the case of 15 wt.% glucose.
For SCWG of sewage at 15 wt.% feed concentration, the fixed capital investment
is found to be equal to $53,381,495, and total capital investment is $64,057,794,
close to those in the case of 15 wt.% glucose feed concentration.
The profitability analysis for SCWG process for sewage at 15 wt.% feed
concentration (data not included in this study) shows that the process becomes
more attractive if there are some incentives from industry or government for sewage
sludge collection and treatment. As per some literature, the incentive can be as high
as $550 per ton of dry material [10]. The process will be more attractive when the
sewage treatment incentive exceeds 481$/ton.
Similar procedure, as in the case of glucose, was performed for the profitability
analysis of sewage SCWG process. Similar to what concluded in the profitability
analysis for glucose, increasing sewage concentration makes the process more
profitable. On the other hand the increase in concentration will result in a less yield
and formation of other byproducts such as tar, which may have a negative impact on
the capital and operating costs. Selection of the best sewage concentration requires
optimization of the process.
If a hydrogen selling price of $3.5/kg is employed, assuming the sewage sludge
treatment incentive of $550/t-dry sewage, the annual net income will be as high
as $7,060,284. The annual cash flow will be $12,398,433. Payback period will be
5.16 years, which are much more promising than the condition of using glucose
solution as a feedstock. The return on investment would be 11.2 %, which is slightly
lower than the minimum desired level (15 %). This shows the need of more effort
to reduce the capital and operating cost of the plant. Comparison of the change of
annual net income with selling price of hydrogen for the sewage sludge (assumed
incentive of $550/t t-dry sewage) is also shown in Fig. 17.2, comparing with the
cases of 15 wt.% and 25 wt.% glucose feedstock. It is to mention that using
Selexol as the scrubbing agent in the absorption tower can reduce the production
cost substantially. Less quantity of Selexol is required compared to water due
to its higher capacity for carbon dioxide absorption. Selexol can also be used at
much lower pressures resulting in lower equipment and maintenance costs for the
absorption process.
17.10 Conclusions
Supercritical water gasification is a promising technology due to its capability of

handling wet biomass eliminating the costly drying process as required for other
gasification processes. Detailed design and profitability analysis show that the
feasibility of the process depends on the type of the feed and its concentration.
The feasibility (profitability) of SCWG processes suffers from the existing prices
of hydrogen and utility prices. The price of raw material also plays an important
role in economics of the process. SCWG of industrial grade glucose was feasible
only at high concentrations (25 wt.% in this study) and a higher hydrogen price,
however SCWG of sewage sludge was feasible at all concentrations above 10 wt.%
provided that the incentive for sewage sludge treatment is taken into account (e.g.,
$550/t-dry sewage in this study). Higher concentrations of sewage sludge required
hydrothermal processing, which will result in higher yields of hydrogen. Further
process optimization and research may improve the yield and profitability in the
future.
Biodegradation and biochemical methods for hydrogen production are at their

infancy period and require more research to become commercially available. It is
to mention that the hydrogen production yield by these methods is also quite low.
The existing biochemical methods of conversion of biomass to fuel are all about
production of methane by anaerobic digestion, which is not of the main concern in
SCWG process.
Acknowledgements The authors are grateful for the financial support from NSERC/Innovations
Industrial Research Chair Program in Forest Biorefinery and the Ontario Research Fund-Research
Excellence (ORF-RE) from Ministry of Economic Development and Innovation. Support from
NSERC Discovery Grants to authors (C.X. and S.B.) are also acknowledged.
References
1. SRI Consulting, Menlo Park, California, USA. Chemical economics handbook:2001.

2. Rajesh JK, Gupta SK, Rangaiah GP, Ray AK. Multi-objective optimization of steam reformer
using genetic algorithm. Ind Eng Chem Res. 2000;39(3):706–17.
3. Garcia L, French R, Czernik S, Chornet E. Catalytic steam reforming of bio-oils for the
production of hydrogen: effects of catalyst composition. Appl Catal A. 2000;201:225–39.
4. Ni M, Leung DYC, Leung MKH, Sumathy K. An overview of hydrogen production from
biomass. Fuel Process Technol. 2006;87:461–72.
5. McKendry P. Energy production from biomass (Part 1): overview of biomass. Bioresour
Technol. 2002;83:37–46.
6. Abu El-Rub Z, Bramer EA, Brem G. Review of catalysts for tar. Elimination in biomass
gasification processes. Ind Eng Chem Res. 2004;2004:6911–19.
water. Energy Fuels. 1993;7:574–7.
8. Byrd AJ, Pant KK, Gupta RB. Hydrogen production from glucose using Ru/Al2 O3 catalyst in
production from glucose via catalytic supercritical water gasification. Int J Hydrogen Energy.
2011;36(2011):9591–601.
10. Gasafi E, Reinecke M-Y, Kruse A, Schebek L. Economic analysis of sewage sludge gasification
in supercritical water for hydrogen production. Biomass Bioenergy. 2008;32:1085–96.
11. Adschiri T, Hirose S, Malaluan R, Arai K. Noncatalytic conversion of cellulose in supercritical
and subcritical water. J Chem Eng Jpn. 1993;26:676–80.
12. Larson ED. Technology for electricity and fuels from biomass. Annu Rev Energy Environ.
1993;1993(18):567–630.
13. Unger J. Druckwechseladsorption zur gastrennung; model- lierung, simulation und prozessdy-
namik. Dü sseldorf, VDI Verlag; 1999. VDI Reihe 3, Verfahrenstechnik Nr. 602.
14. Boukis N, Galla U, Müller H, Dinjus E. Biomass gasification in supercritical water. Exper-
imental progress achieved with the VERENA pilot plant. In: Proceedings of 15th European
biomass conference & exhibition. Berlin, Germany; 2007, p. 1013–6.
15. Perry RH, Chilton CH. Chemical engineering handbook. 7th ed. McGraw-Hill, New York,
1997.
16. Divilio RJ. Modeling of the supercritical water pyrolysis process. In: Proceedings of the 1999
US DOE hydrogen program review, NREL/CP-570–26938; 1999.
17. Hydrogen and clean fuels, Central hydrogen production, U.S. Department of Energy; 2013.
http://www.netl.doe.gov/technologies/hydrogen_clean_fuels/central.html. Retrieved on 29 Oct
2013.
18. General Atomics. Hydrogen production by supercritical water gasification of biomass. Techni-
cal and business feasibility study. Technical progress report fur das US Department of Energy,
No. DE-FC36–97GO010216. US Department of Energy; 1997.
19. Thomas WJ, Crittenden B. Adsorption technology and design. Amsterdam: Elsevier Science
& Technology Books; 1998.
20. Pant KK, Gupta RB. Hydrogen fuel: production, transport and storage, Fundamentals and Use
of Hydrogen as a Fuel. Boca Raton: CRC Press/Taylor & Francis Group; 2009.
21. Boyce CA, Crews MA, Ritter R, CB&I Howe-Baker, USA. Time for a new hydrogen plant?
2013. http://www.cbi.com/images/uploads/technical_articles/CBI_HydrocarbonEngineering_
Feb04.pdf. Retrieved on 25 Oct 2013
22. Hemmes K, Beld VD, Kersten SRA, Vergassing van natte biomassa/reststromen in superkritiek
water (SCWG), voorde productie van “groen gas”(SNG), SNG/H2 mengsela, basis chem-
icalien en puur H2. www.ecn.nl/docs/library/report/2004/c04107.pdf. Retrieved on 28 Oct
2013.
biomass utilization in Japan. Energy Conversion Manag. 2002;43:1301–10.
24. Carpenter Technology Corporation – stainless steels and specialty alloys; 2013. http://www.
cartech.com/ssalloysprod.aspx?id=1926. Retrieved on 29 Oct 2013.
25. High performance alloys for resistance to aqueous corrosion, Special Metals Corporation;
2013. http://www.parrinst.com/wpcontent/uploads/downloads/2011/07/Parr_Inconel-Incoloy-
Monel-Nickel-Corrosion-Info.pdf. Retrieved on 29 Oct 2013.
26. Peters MS, Timmerhaus KD, West RE. Plant design and economics for chemical engineers.
5th ed. New York: McGraw-Hill; 2003.
27. Ogden JM, Kreutz T, Kartha S, Iwan L. Hydrogen energy systems studies. Princeton
University, Center for Energy and Environmental Studies, NJ, USA. Technical report submitted
to NREL, DOE/GO/10061–T1; 1996.
28. Ulrich GD. A guide to chemical engineering process design and economics. New York: Wiley;
1984.
29. Dhanuskodi R, Arunagiri A, Anantharaman N. Analysis of variation in properties and its
impact on heat transfer in sub and supercritical conditions of water/steam. Int J Chem Eng
Appl. 2011;2(5):320–5.
30. Cheng X, Schulenberg T. Heat transfer at supercritical pressures: literature review and
application to an HPLWR, Institut fur Kem- und Energietechnik Programm Nuklear Sicher-
heitsforschung Forscungazentrum Karlsruhe GmbH, Karlsruhe 2001.
31. The Engineering Toolbox. Nominal wall thickness seamless and welded carbon and alloy steel
pipes; 2013. http://www.engineeringtoolbox.com/Nominal-wall-thickness-pipe-d_1337.html.
Retrieved on 29 Oct 2013.
32. Behnia I. Treatment of aqueous biomass and waste via supercritical water gasification for the
production of CH4 and H2 . MSc thesis, University of Western Ontario, London, Ontario; 2013.
33. Sinnott RK, Coulson JM. Coulson & Richardson’s chemical engineering, vol. 6. 3rd ed.
Oxford: Linacre House; 2003.
34. Bejan A, Tsatsaronis G, Moran M. Thermal design and optimization. New York: Wiley; 1996.
35. Sweny JW. Gas treating with a physical solvent. In: Acid and sour gas treating process; 2013.
www.rschendel.com/PDF/book.PDF. Retrieved on 28 Oct 2013.
36. Rubin ES, Antes M, Yeh S, Berkenpas M. Estimating the future trends in the cost of CO2
capture technologies. Report no. 2006/6. IEA greenhouse Gas R&D Programme 2006 (IEA
GHG), Cheltenham, UK; 2006.
37. Nakamura A, Eiji K, Yamamura Y, Shimizu Y, Minowa T. Detailed analysis of heat and mass
balance for supercritical water gasification. J Chem Eng Jpn. 2008;41(8):817–28.
38. Cost indices. Chem Eng J. www.che.com, June 2012;119 6:80.
39. Bloomberg markets, www.bloomberg.com/energy. Retrieved on 29 Oct 2013.
40. Wilson L., Average electricity prices around the world; 2013. www.shrinkthatfootprint.com/
average-electricity-prices-kW.h. Retrieved on 29 Oct 2013.
Editors’ Biography
Prof. Dr. Zhen Fang is leader and founder of biomass group, Xishuangbanna
Tropical Botanical Garden, Chinese Academy of Sciences. He is also an adjunct full
Professor of Life Sciences, University of Science and Technology of China. He is
the inventor of “fast hydrolysis” process. He is specializing in thermal/biochemical
conversion of biomass, nanocatalyst synthesis and its applications, pretreatment of
biomass for biorefineries. He obtained his PhDs from China Agricultural University
(Biological & Agricultural Engineering, 1991, Beijing) and McGill University
(Materials Engineering, 2003, Montreal).
for Biorefineries, Biofuels and Biorefineries 2, DOI 10.1007/978-94-017-8923-3,
474 Editors’ Biography
Dr. Chunbao (Charles) Xu is currently an Associate Professor of Chemical

Engineering and NSERC/FPInnovations Industrial Research Chair in Forest Biore-
finery at Western University (Canada), and is leading the Industrial Bioproducts
Laboratory funded by Canada Foundation for Innovation. His research is centered
on development of high-value bioproducts from bioresources via thermo-chemical
and catalytic conversion processes. Typical processes investigated by Dr. Xu and his
group include catalytic conversion of sugars into platform chemicals and biomate-
rials, renewable hydrogen/methane production via supercritical water gasification
of wet biomass/wastes, hydrothermal liquefaction of biomass for bio-crude pro-
duction and upgrading of bio-crude via hydro-de-oxygenation, depolymerisation of
lignin into bio-phenols and bio-polyols, and synthesis of bio-based resins/adhesive,
polyurethane and epoxy resins. In addition, Dr. Xu is active in upgrading of heavy
residual oil via hydro-treatment and recovery of organics from oil sands tailing
water. He was awarded the Japan Institute of Energy Outstanding Young Scientists
Award in 1999, and more recently the prestigious Syncrude Canada Innovation
Award in 2011 from Canadian Society of Chemical Engineers (presented to a young
Canadian chemical engineer under the age of 40 who has made a distinguished
contribution to the field of chemical engineering while working in Canada). Dr. Xu
is currently serving as an editor-in-chief for the International Journal of Chemical
Reactor Engineering, an international peer reviewed journal published by De
Gruyter.

(Biofuels and Biorefineries 2) Zhen Fang, Chunbao (Charles) Xu (Eds.) - Near-Critical and Supercritical Water and Their Applications For Biorefineries-Springer Netherlands (2014)

Uploaded by

Copyright:

Available Formats

(Biofuels and Biorefineries 2) Zhen Fang, Chunbao (Charles) Xu (Eds.) - Near-Critical and Supercritical Water and Their Applications For Biorefineries-Springer Netherlands (2014)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Biofuels and Biorefineries 2) Zhen Fang, Chunbao (Charles) Xu (Eds.) - Near-Critical and Supercritical Water and Their Applications For Biorefineries-Springer Netherlands (2014)

Uploaded by

Copyright:

Available Formats

What is the scope and aims of the book series?

What is the scope and aims of the book series?

What are some of the advantages of using biomass compared to coal?

What are some of the advantages of using biomass compared to coal?

Biofuels and Biorefineries 2

Professor Zhen Fang, Chinese Academy of Sciences, Kunming, China

Editorial Board Members:

Professor Liang-shih Fan, Ohio State University, USA;

For further volumes:

ISSN 2214-1537 ISSN 2214-1545 (electronic)

© Springer Science+Business Media Dordrecht 2014

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

readers with comprehensive information of co-solvents interactions. Chapter 4 uses

Kunming, China Zhen Fang

Dr. Selhan Karagöz, Karabuk University, Turkey;

Zhen Fang, January 28, 2014 in Kunming, China (Zhen Fang)

Chunbao (Charles) Xu, January 28,

Part I Fundamentals of Supercritical Water

1 Hydrogen Bonding in Supercritical Water . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 3

Part II Reactor Design

5 Optical Reactors for Microscopic Visualization

Part III Near-Critical and Supercritical Water Applications

9 Production of Renewable Solid Fuel Hydrochar

Editors’ Biography .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 473

Mesut Akgün SCFT—Supercritical Fluid Technologies Research Group,

Victor Mazur Institute of Refrigeration, Cryotechnologies, and Eco-Power

Tau Len-Kelly Yong Department of Mechanical Science and Engineering,

Abstract Hydrogen bonding (HB) in supercritical water (SCW) is much less

Keywords Hydrogen bonding • Criteria for hydrogen bonds • Hydrogen bonds

< nHB >D =2: (1.1)

Ordinary ice has a regular tetrahedral hydrogen bond network with D 4, so

The percolation threshold is another quantity that describes the hydrogen

1.2 Criteria for Hydrogen Bonds to Exist

MD computer simulations of water at ambient conditions and ab initio theoretical

Fig. 1.1 Schematic

1.3 The Extent of Hydrogen Bonding in SCW

The extent of hydrogen bonding in SCW can be described in terms of several

r is the segment length, a dimensionless quantity, Vr 00 is the reduced

p D 1  exp ŒE.P /=R .T  T0 / (1.5)

the value of f0 being obtained by difference, because f0 C f1 C f2 D 1.

and 1.6: T0 D 637 K and E(P)/Jmol–1 D 18 C 0.028(P/ MPa)2. In order to

Kirkwood dipole orientation parameters g:

1.4 Hydrogen Bonding in SCW Deduced

The arrangements of water molecules in SCW are described by the radial

dN .r; dr/ D g.r/4 r 2 dr (1.11)

As no correlation exists between water molecules at large distances from each

S.k/  ŒI.k/  I.0/ = ŒI .1/  I.0/ (1.12)

k D 1 4 sin .=2/ (1.13)

f4 D 0:851  8:68  104 .T =K/ (1.15)

1.5 Hydrogen Bonding in SCW Deduced from Spectroscopy

Luck [35, 36] applied infrared absorption spectroscopy to just-supercritical water,

were de-convoluted into bands corresponding to small clusters: monomer ( free

" .!/ D " .1/ C .".0/  " .1// Œ.1  fb / = .1 C i ! D f / C fb = .1 C i ! D b /

1.6 Hydrogen Bonding in SCW Deduced

In the computer simulations, a specified number of water molecules are placed in

f0 D 0:948  0:886 C 0:2862  0:0303 (1.18)

1.7 Ionic Dissociation of SCW

The self-dissociation of water in the supercritical state has to be written as a

p D 1 exp ŒE.P /=R .T T0 / (1.5)

S.k/ ŒI.k/ I.0/ = ŒI .1/ I.0/ (1.12)

f4 D 0:851 8:68 104 .T =K/ (1.15)

" .!/ D " .1/ C .".0/ " .1// Œ.1 fb / = .1 C i ! D f / C fb = .1 C i ! D b /

f0 D 0:948 0:886 C 0:2862 0:0303 (1.18)

Mesmer et al. [71] obtained the standard thermodynamic functions H ı , Sı ,

P D RT= .V b/ NA 2 ˛a=V .V C b/ (1.23)

ıh D ıh ref exp f0:00132 .T Tref / C 0:5 ln .V =Vref /g (1.28)

D .0:71 ˙ 0:06/ C .1:77 ˙ 0:09/ (1.31)

Pexact .T; V / Pmod .X; T; V / D 0;