Sugar Decoloration Strategies and Solutions

João Luciano Amorim Azevedo
Sugar decolorization - strategies and solutions

Master’s Thesis Project in the scientific area of Chemical Engineering, Supervised by:
Professor Doctor Maria da Graça Bomtempo Vaz Rasteiro, Dr. Julien Arnold (AquaTech)
and Dr. José A. F. Gamelas, and submitted to the Department of Chemical Engineering, Faculty of Science and
Technology of the University of Coimbra
June 2017
João Luciano Amorim Azevedo
Sugar decolorization - strategies and

solutions
Thesis Project in the scientific area of Chemical Engineering, supervised by Professor Doctor Maria da
Graça Bomtempo Vaz Rasteiro and submitted to the Department of Chemical Engineering, Faculty of
Science and Technology, University of Coimbra
Supervisors:
Prof. Dr. Maria da Graça B. V. Rasteiro
Dr. Julien Arnold (AquaTech)
Dr. José A. F. Gamelas
In collaboration with AquaTECH (Switzerland) and RAR (Portugal)
Coimbra
2017
“We must accept finite disappointment, but never loose infinite hope.”
Martin Luther King Jr

.
Acknowledgements
Life is a constant pursuit for growth and for fulfillment, and the times we call “student
life” is one of the most beautiful stages of someone’s life, for in this period we have open minds,
in search for knowledge and for a definition for our lives and, at the same Time, our hearts
guide us with the strong flame of hope. I have dreamed a lot and, as life is, had days and nights,
fears and times we think about giving up facing the difficulties, but I loved the study and the
science field in general, and so, my life in Coimbra provided me much joy and also knowledge,
guided by few Professors, who fulfilled my desire for knowledge and consolidated my love for
Chemical Engineering, and some other people helped me to keep going and see life as a set of
different cultures, objectives and hopes. For that, I want here to express my deepest gratitude
and recognition to all of you, but can only name a few.
To Professor Doctor Maria da Graça Rasteiro, who supervised this project, being always
available and kind, without her support, knowledge, wisdom, patience and guidance surely this
work would not be possible and to take form in this dissertation. My profound gratitude, respect
and admiration for this wise and strong woman.
To RAR Portugal, for supporting this project and in special to the Engineer José Manuel
Chorão, whose support and guidance had a great deal in the realization of this project, thank
you for the hours on enlightenment dedicated to me and to this project.
To AquaTECH Switzerland, for the support provided to this project and to Doctor Julien
Arnold, for the guidance and support.
Have to be grateful to life thus there is always hope in it and opportunities, thanks to all
my Supervisors for providing this opportunity and for the patience to support me along the way.
To my family, in special to my beloved Mother, Lourdes, who is for me the strongest

representation of love and hope, to my Father Laurentino, who made me admire intelligence
and wisdom. To my brothers and sisters, Eliane, Luciana, Lussandra David, Sheilla and
Rogerio, for all love and support along this journey that is life. To Bruno Luigi, a friend who
has been a real brother and support.
To Portugal, this beautiful Country that I love very much and call home. To my
Portuguese friends and family, in special to Rodrigo and Rosa Silveirinha, who has a mother in
i
here, to Jolema Duarte and Ana Paula Azevedo, for all the philosophy shared, the love and
support.
To all Professors in the Chemical Engineering Department, in special to Professor Doctor

Lino Santos, Herminio Souza and Rosa Ferreira, for all the kindness since I first arrived in
Coimbra, and for all knowledge and inspiration you provided me.
To My Godfather in the department, Engineer João Nuno Simões, who has always been
a brother and strong a rock of support, for trusting and making me go on, even when I refused
to see the light.
To the Personal of the Department too, D. Dulce, Maria João, and To Sandra Silva, for
the availability and support.
And Last, but not less grateful, to my colleagues, in special to the PhD student Kinga
Grenda, you have been of so much support as friend and as guide in the laboratory work, making
us feel strong and willing to improve to work properly. To Anita for all the help in practical and
theoretical work knowledge shared with me, and for her friendship. To Ana Rita, Patricia Costa,
Luiza, Marta, Nuno, Silvia, Filipa and Mafalda Cardoso, thank you for supporting and for being
the kind persons and friends that you are.
To you all, my love and profound gratitude, may your lives be always of joy and peace.
ii
Abstract
Sugar industry comprises mainly of two products: raw and refined sugar. The second one is the
most required by the market and consumers, as the requirement for a better quality product for
a lower cost and environmental impact is a constant. Refining sugar is a process which main
objective is to remove impurities, which normally are related to color compounds, from sucrose.
This work was focused in the sugar refining process, more specifically in the color removal by
means of precipitation of color compounds from sugar solution in order to separate such
compounds by physical means like filtration or clarification. Based on the actual scenario and
information obtained in the literature, the study was conducted guided by the use of cationic
polymers as color precipitants through flocculation. The study comprises three steps: first is a
characterization of the polymers supplied by AquaTech Switzerland, using optical methods,
dynamic and static light scattering, to characterize the size and the molecular weight, and
electrophoretic light scattering, for zeta potential of the polymers. As second part of the work
the laboratorial decolorization tests were conducted in sugar carbonated liquor, supplied by
RAR Portugal, testing the different polymers at different dosages, in conditions, pH and
temperature, similar to that of the second stage of saturation in carbonatation process, and
measuring the decolorization achieved by the treatment with each polymer, following the
ICUMSA 420 color characterization method; as third and last part of the work continuous
monitoring of flocculation process was performed by using laser diffraction spectroscopy, as
technique for following the flocculation kinetics, this way to search for evidences of the
flocculation performed by the polymers selected after the laboratorial tests. The study was
performed using polymers individually and trying blends of two or three different polymers.
Conclusions were that the decolorization achieved was good, reaching levels of 41% of
additional color removal over carbonated sugar liquor, in the case of Amber 2009, and
Amber2001HMW, Amber 2013 and Amber MetQuat 6, all in the range of 30-40% of
decolorization, proving that cationic flocculants are effective decolorization agents when used
along the carbonatation process, at pH 8-8.5. The polymer performance is related to its
characteristics, needing to perform further studies to find a balance for molecular weight, size
of polymer chain and charge to find the most suitable relations in order to achieve the best
performance. LDS is used for first time to follow the flocculation kinetics in carbonated sugar
liquor using cationic polymers flocculants, and it has proven to be a good tool for monitoring
and study performance of the flocculation and the stability of the flocs in the system.
iii
Key-words: Sugar Refining, Carbonatation, Color Removal, Cationic Polymer Flocculation,
Laser Diffraction Spectroscopy, Light Scattering Spectroscopy.
iv
Resumo
A indústria do açúcar consiste principalmente em dois produtos: açúcar bruto e refinado. O

segundo é o mais procurado pelo mercado e pelos consumidores finais, sendo que a necessidade
por um produto de melhor qualidade, por menor custo e impacto ambiental é uma constante.
Refinar açúcar é um processo cujo objetivo principal é remover impurezas, que geralmente
estão relacionadas a compostos corados, da sacarose. Este trabalho foi focado no processo de
refinamento do açúcar, mais especificamente na remoção da cor por meio da precipitação de
compostos corados da solução de açúcar, para, então, separar tais compostos por meios físicos,
como a filtração ou clarificação. Com base no cenário atual e na informação obtida na literatura,
o estudo foi conduzido guiado pelo uso de polímeros catiónicos como agentes precipitantes por
meio de floculação. O estudo compreende três etapas: a primeira é uma caracterização dos
polímeros fornecidos pela AquaTech Suíça, utilizando métodos óticos, dispersão de luz
dinâmica e estática, para caracterizar o tamanho e peso molecular, e dispersão de luz
eletroforética para determinar o potencial zeta dos polímeros. A segunda parte do trabalho foi
realizada em licor carbonatado de açúcar bruto, fornecido pela RAR Portugal, e assim testando
os diferentes polímeros a diferentes dosagens e em condições, pH e temperatura, semelhantes
às da segunda fase de saturação no processo de carbonatação e, após a floculação, medindo a
descoloração atingida seguindo o método ICUMSA 420 de caracterização de cores de soluções
de açúcar; A terceira e a última parte do trabalho foi o monitoramento contínuo do processo de
floculação, utilizando a espectroscopia de difração a laser, para o acompanhamento da cinética
de floculação, buscando evidências da floculação realizada pelos polímeros selecionados após
os testes laboratoriais. O estudo foi realizado usando polímeros individualmente e testando
também misturas de dois ou três destes polímeros. As conclusões foram que a descoloração
alcançada foi boa, alcançando-se níveis de 41% de remoção de cor adicional sobre o licor de
açúcar carbonatado, no caso de Amber 2009 e, no caso do Amber2001HMW, Amber 2013 e
Amber MetQuat 6, alcançando níveis de 30-40% de Decoloração, provando que os floculantes
catiónicos são agentes de descoloração eficazes quando utilizados ao longo do processo de
carbonatação, a pH 8-8,5. O desempenho do polímero está relacionado às suas características,
necessitando de mais estudos para encontrar o equilíbrio do peso molecular, tamanho da cadeia
de polímero e carga para encontrar as dosagens mais adequadas para alcançar o melhor
desempenho. O LDS é uma boa ferramenta para monitorar e estudar a floculação em licor de
açúcar carbonatado diluído.
v
Palavras-chave: Refinamento De Açúcar, Carbonatação, Remoção De Cor, Floculação e
Polímeros Catiónico, Espectroscopia De Difração Laser, Espectroscopia De Dispersão Da Luz.
vi
Table of contents
Abstract .......................................................................................................................................iii
Resumo ........................................................................................................................................ v
Nomenclature..........................................................................................................................xviii
Chapter 1. Introduction ........................................................................................................... 1
Chapter 2. State of the Art ....................................................................................................... 5
2.1 Sugar refining .................................................................................................................... 5
2.2 Colorants in sugar .............................................................................................................. 6
2.2.1 Natural colorants......................................................................................................... 6
2.2.2 Colorants formed during sugar processing. .............................................................. 7
2.3 Measurement of color in sugar .......................................................................................... 8
2.4 Decolorization process ...................................................................................................... 9
2.5 Primary decolorization process ....................................................................................... 10
2.5.1 Phosphatation ........................................................................................................... 10
2.5.2 Carbonatation ........................................................................................................... 11
2.5.3 Sulphitation............................................................................................................... 12
2.6 Flocculants and color precipitants ................................................................................... 12
2.6.1 Polymer flocculation ................................................................................................ 13
2.6.2 Anionic flocculants ................................................................................................... 14
2.6.3 Cationic flocculants .................................................................................................. 15
2.7 Secondary color removal technics ................................................................................... 19
2.7.1 Bone char .................................................................................................................. 19
2.7.2 Activated carbon ....................................................................................................... 20
2.7.3 Ion exchange ............................................................................................................. 20
vii
Chapter 3. Materials and Methods ........................................................................................ 23
3.1 Materials .......................................................................................................................... 23
3.1.1 Sugar ......................................................................................................................... 23
3.1.2 Polymers ................................................................................................................... 25
3.2 Methods ........................................................................................................................... 26
3.2.1 Preparation of the polymer solutions ........................................................................ 26
3.2.2 Decolorization tests .................................................................................................. 26
3.4 Evaluation of IV on decolorization ................................................................................. 28
3.5 Polymers characterization................................................................................................ 29
3.5.1 Hydrodynamic diameter ........................................................................................... 30
3.5.2 Zeta potential measurement ...................................................................................... 32
3.5.3 Molecular weight ...................................................................................................... 34
3.5.4 Refractive index ........................................................................................................ 38
3.6 LDS flocculation monitoring tests ................................................................................... 39
Chapter 4. Results and discussion ......................................................................................... 43
4.1 Polymer Characterization ................................................................................................ 43
4.2 Standard decolorization tests results ............................................................................... 45
4.3 Determination of Index value .......................................................................................... 51
4.4 Flocculation monitoring using LDS ................................................................................ 53
Chapter 5. Conclusions and future works ............................................................................. 67
5.1 Conclusions ..................................................................................................................... 67
5.2 Future works .................................................................................................................... 68
References ................................................................................................................................. 71
Appendix ................................................................................................................................. 73
Appendix A . Polymer Characterization.................................................................................... 75
A.1 Size distribution by intensity for the polyelectrolytes .................................................... 75
viii
A.2 Zeta potential graphics .................................................................................................... 79
A.3 Molecular Weight Debye plots for the polymers ........................................................... 82
Appendix B . Decolorization tests results ................................................................................ 87
Tables and graphics of laboratorial decolorization tests results. ........................................... 87
Appendix C . Flocculation monitoring additional tests. ............................................................ 99
ix
x
List of figures
Figure 2.1: Block diagram representation of sugar refining process. [4] .................................... 5
Figure 2.2: Representation of bridging and patch flocculation, green balls are the anionic
impurities and the polymer chains [14]. .................................................................................... 14
Figure 2.3: Polymers tested By Bennett in his work (1971) adapted from [10]........................ 16
Figure 2.4: Representation of the interaction between surfactant and anionic color impurity,
adapted form [8] ........................................................................................................................ 17
Figure 3.1: Carbonatation process diagram, red circle represents point of sample extraction for
the laboratorial trail ( the scheme represents the unit of RAR). ................................................ 24
Figure 3.2: Schematic representation in flux diagram of the procedure for decolorization
laboratorial tests performed. ...................................................................................................... 27
Figure 3.3: Zeta Nano Sizer ZSN by Malvern. [18]. ................................................................. 29
Figure 3.4: Graphic representation of the correlation functions used in DLS treatment by the
Nano Sizer equipment [18]. ....................................................................................................... 31
Figure 3.5: Debye Plot [18] ....................................................................................................... 37
Figure 3.6: Flow diagram for molecular weight measurements methodology adopted. ........... 38
Figure 3.7: System mounted for LDS monitoring of the flocculation process, comprising the
Mastersizer and a water bath adapted to the system. ................................................................. 40
Figure 3.8: Diagram representation of steps sequence for the LDS trials. ................................ 42
Figure 4.1: Graphic representation of sugar liquor color (treated with polymer and blank) versus
polymer dosage for tests with polymer Amber 2006. ............................................................... 46
Figure 4.2: the filtrated liquor blank (on the left) and treated with the polymer Amber 2013
100ppm (on the right). ............................................................................................................... 48
Figure 4.3: Graphic representation of the by the best performing polymer at each dosage and
respective percentage of decolorization achieved. .................................................................... 49
Figure 4.4: Representation of Index Value and decolorization for filtrated liquor by
decolorization with Amber2001 HMW. .................................................................................... 52
Figure 4.5: Flocculation kinetics for carbonated sugar liquor using Amber 2013 at 100ppm
dosage: a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer. ........ 55
xi
Figure 4.6: Flocculation kinetics for carbonated sugar liquor using Amber 2001HMW, at
100ppm dosage: a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
................................................................................................................................................... 56
Figure 4.7: Flocculation kinetics for carbonated sugar liquor using Amber 2001HMW, at
200ppm dosage: a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
................................................................................................................................................... 57
Figure 4.8: Flocculation kinetics for carbonated sugar liquor using Amber MetQuat6 at 100 ppm
Figure 4.9 Flocculation kinetics for carbonated sugar liquor using Amber 2008 at 100 ppm
Figure 4.10: Flocculation kinetics for carbonated sugar liquor using Amber 2009 at 100 ppm
Figure 4.11: Flocculation kinetics for carbonated sugar liquor using the blending Amber 2001
HMW & Amber 2013 at 100 ppm dosage of each polymer: a) d10, b) d50 and c) d90. The bar
line corresponds to the addition of polymer. ............................................................................. 61
Figure 4.12: Flocculation kinetics for carbonated sugar liquor using Amber 2009 and Amber
2001HMW, both at 100ppm: a) d10, b) d50 and c) d90. The bar line corresponds to the addition
of polymer. ................................................................................................................................ 62
Figure 4.13: Flocculation kinetics for carbonated sugar liquor using Amber 2009, Amber 2013
and Amber 2001HMW, at 50 ppm each one, a) d10, b) d50 and c) d90. The bar line corresponds
to the addition of polymer. ........................................................................................................ 63
Figure 4.14: Flocculation kinetics for carbonated sugar liquor using Amber 2009, Amber 2013
and Amber 2001HMW, at dosage of 100ppm each polymer, a) d10, b) d50 and c) d90. The bar
line corresponds to the addition of polymer. ............................................................................. 64
Figure A.1: Size distribution by intensity for polymer Amber 2001 HMW at 0,02 and 0,05%
w/V ............................................................................................................................................ 75
Figure A.2: Size distribution by intensity for polymer Amber 2002 at 0,05% w/V ................. 75
Figure A.3: Size distribution by intensity for polymer Amber 2006 at 0,02 and 0,05% w/V ... 76
Figure A.4: Size distribution by intensity for polymer Amber 2008 at 0,05% w/V. ................ 76
Figure A.5: Size distribution by intensity for polymer Amber 2008HMW at 0,05% w/V. ...... 77
Figure A.6: Size distribution by intensity for polymer Amber MetQuat 6 at 0,02% w/V. ....... 77
Figure A.7: Size distribution by intensity for polymer Amber 2009 at 0,05% w/V ................. 78
xii
Figure A.8: Zeta potential graphics for Amber 2001HMW. ..................................................... 79
Figure A.9: Zeta potential graphics for Amber 2002. ............................................................... 79
Figure A.10: Zeta potential graphics for Amber 2006. ............................................................. 80
Figure A.11: Zeta potential graphics for Amber 2008. ............................................................. 80
Figure A.12: Zeta potential graphics for Amber 2008HMW .................................................... 81
Figure A.13: Zeta potential graphics for Amber MetQuat 6. .................................................... 81
Figure A.14: Debye plot for Amber 2001HMW 1. ................................................................... 82
Figure A.15: Debye plot for Amber 2001HMW 2. ................................................................... 82
Figure A.16: Debye plot for Amber 2002.1. ............................................................................. 83
Figure A.22: Debye plot for Amber MetQuat 6 1. .................................................................... 86
Figure A.23: Debye plot for Amber MetQuat 6 2. .................................................................... 86
Figure B.1: Graphic representation of color (IU) of the liquor, treated and untreated with the
polymer, versus polymer dosage for the test using Amber 2001HMW .................................... 87
polymer, versus polymer dosage for the test using Amber 2002 .............................................. 88
polymer, versus polymer dosage for the test using Amber 2008. ............................................. 90
polymer, versus polymer dosage for the test using Amber 2008HMW. ................................... 91
xiii
polymer, versus polymer dosage for the test using Amber MetQuat 6. .................................... 96
Figure C.1: Plot for flocculation kinetics using Amber2001HMWat 300 ppm, a) = d10, b) = d50
e c) = d90 ................................................................................................................................... 99
e c) = d90. ................................................................................................................................ 100
e c) = d90. ................................................................................................................................ 101
e c) = d90. ................................................................................................................................ 102
xiv
List of tables
Table 2-1: Type of sugar colorants present in sugar [4]. ............................................................. 8

Table 3-1: List of Polymers used in the present study, provided by AquaTech Switzerland. .. 25
Table 4-1: Polymer Characterization results. ............................................................................ 44
Table 4-2: Decolorization performed by the polyamine Amber 2006. ..................................... 47
Table 4-3: Percentage of Decolorization per polymer and ppm dosage of polymer on Brix of the
sugar liquor. ............................................................................................................................... 47
Table 4-4: Blend Amber 2013 & Amber 2001HMW decolorization performance. ................. 50
Table 4-5: Blend Amber 2009 & Amber 2001HMW decolorization performance. ................. 50
Table 4-6: Blend using Amber 2013, 2009 & Amber 2001HMW decolorization performance.
................................................................................................................................................... 51
Table 4-7: Index value (IV) for filtrated sugar liquor by ppm of polymer used. ...................... 52
Table 4-8: List of polymers used in LDS tests, its dosage, time of flocculation and decolorization
percentage at the end of trials. ................................................................................................... 54
Table B-1: Decolorization results using Amber 2001HMW ..................................................... 87
Table B-2: Decolorization results for Amber 2002 ................................................................... 88
Table B-5: Decolorization results for Amber 2008HMW ......................................................... 91
Table B-10: Decolorization results for Amber MetQuat 6 ........................................................ 96
Table B-11: Decolorization results using Blend Amber 2009 & Amber 2001HMW ............... 97
Table B-12: Decolorization results using Blend Amber 2013 & Amber 2001HMW ............... 97
Table B-13: Decolorization results using Blend Amber 2013, Amber 2009 & Amber 2001HMW
................................................................................................................................................... 97
xv
xvi
xvii
Nomenclature
λ0 Laser Wavelength
µ Dynamic Viscosity
ɛ Dielectric constant
A2 2nd Virial Coeficient
C concentration
Dc Translational diffusion coefficient
Dh Hydrodynamic Diameter
Di Hydrodynamic Diameter of Class i
Dz Average Diameter of the Size Distribution
DLS Dynamic Light Scattering
dn/dc Differential Refractive Index Increment
ELS Electrophoretic Light Scattering
HMW High Molecular Weight
IA Residual Scattering Intensity of the Analyte
ICUMSA International Commission for Uniform Methods for Sugar Analyses
Ii Scattered Light Intensity of Class i
IT Toluene Scattering Intensity
IU ICUMSA color Unit
IV Index Value
K Optical Constant
k Boltzmann Constant
LDV Laser Doppler Velocimetry
MMW Medium Molecular Weight
MW Molecular Weight
NA Avogadro's Constant
no Solvent Refractive Index
nT Toluene Refractive Index
NaCl Sodium Chloride
Pθ Angular Dependence of the Scattering Intensity
xviii
DADMAC Dially Dimethyl Ammonium Chloride
Rθ_ Rayleigh Ratio
RT Rayleigh Ratio of Toluene
RI Refractive Index
SLS Static Light Scattering
T Absolute Temperature
UE Electrophoretic Mobility
ZNS Zetasizer Nano Series
Z Zeta Potential
xix
Chapter 1. Introduction
Sugar industry is one of the oldest in the industrial panorama. It had a very important role to
play in the development of that which is called the New World, the Americas, and being also very
relevant in the scenery of European history, and especially for Portugal itself, during the age of
Discovery and mercantilism. According to the sugar history, the cane sugar is originally from India
and the southern Asia, where the sugar cane plant comes from and also where it has been first
processed to produce the crystals obtained from the sugar cane juice[1]. Centuries later it meets its
largest production expansion, in the early 16th century, with the large developments of the sugar
mills, when sugar cane became one of the most important luxury commodities to be
commercialized in the world, and Portugal at this time became the strongest sugar producer and
exporter throughout all Europe. The sugar world production had then a high jump, due mainly to
the new American colonies where the sugar cane found favorable conditions of climate and soils
to develop its culture, promoting economic development and settling, this way, the age of the new
world and the slave system in the world.
Till nowadays sugar is still of great importance for the economy in many Countries around
the World. It plays a very important role in the food industry either as a direct consumer product
or as raw material for other industries, most of which require high purity white sugar.
The commodity that we call sugar in the everyday language is in truth the chemical
compound known as sucrose, a disaccharide formed by the bond of one molecule of glucose with
a molecule o fructose. But not only, there is, even if in very small amounts, a very complex mixture
of other chemicals in the commercial sugar, most of which are related to the color of the sugar.
Sucrose itself is a colorless compound, and the sugar market is mainly divided into raw sugar,
produced in the sugar mills and which has a high color index, and refined, or white, sugar [2]. By
that relation between color and the purity of sugar, the color removal process plays a very important
role in the sugar refining process. In truth the color removal is the main cost and the main goal of
a sugar refinery. This way the removal of colored impurities is a major effort of the sugar refineries
intending to produce the highest purity product required by the large market but taking into account
economic and environmental aspects.
1
This work objective is centered into the refining process of sugar, mainly in the use of
polyelectrolytes as color removal agents that act by flocculation, to enhance the color removal in
the refining process with the aim of reducing costs with chemicals and energy, taking into account
also environmental issues.
Many studies have been performed in the past decades in the use of polymers flocculants as
color removal agents in various different industrial fields, especially in the water treatment and
food industry [3]. The use of those materials have been also studied in the sugar industry, both in
the raw and refining sugar processes, where it can be observed that the use of these flocculants as
color removal agents can really enhance the removal of color compounds in sugar processing.
From the perspective of this work, the starting material is raw sugar, which has a large grade
of products, depending on the geographical area, season of the year, the process through which it
is produced, among others factors. Thereby the composition of the raw sugar can vary significantly.
The objective of this work is to evaluate the performance of a few novel cationic polymers,
provided by AquaTech Switzerland, as flocculants color removal agents in carbonatation process
sugar liquor, provided by RAR Portugal, through three main procedures, namely:
1) characterization of the polymers used in the tests;

2) laboratory trials on color removal flocculation;
3) Online flocculation monitoring by laser diffraction spectroscopy, to evaluate the kinetics
of flocculation.
This dissertation is then divided into five chapters, as following:
 Chapter one: introduction and main objectives of the work, giving a first view of the
problem of color in sugar industry and the importance of finding a suitable solution for that;
 Chapter two: in this an attempt to summarize the technology of color removal in the sugar
industry, as well a brief description of sugar refining process, presenting information on
previous studies conducted in this subject;
 Chapter three: describes the materials used in the study as well as the methodology and
technics through the study was conducted;
 Chapter four: presents the results achieved in this work;
2
 Chapter five: here the main conclusions reached with the present work are presented, along
with a perspective of new directions and complementary studies to be conducted in future
works.
3
4
Chapter 2. State of the Art
In this chapter I will try to explain how sugar is processed, its impurities content, how they
can be removed, and finally the use of polymers in the sugar industry as mean to promote
separations and color and impurities removal throughout the process of refining.
2.1 Sugar refining
The process to obtain white sugar from raw sugar is called refining, and it has the objective
of removing impurities and colored compounds from the sugar, aiming to achieve a product of as
near as possible of 100% purity of sucrose. This process is mainly comprised of a set of unit
operations, where the backbone is the clarification and color removal stages, and the specific
conditions on which these operations are to be performed depend on the quality and characteristics
of the raw sugar to be processed. However, usually the refining process is performed as described
in the following block diagram [4]:
Raw sugar
Affination Melting Clarification
Filtration
Decolorization
White sugar
Centrifuging Evaporation
crystalization
Drying
Refined sugar
Figure 2.1: Block diagram representation of sugar refining process. [4]
5
Chapter 2. State of Art
2.2 Colorants in sugar
The term “Color” in sugar industry is commonly used to refer to a wide range of components
that contribute to give color to the sugar. Sucrose, as mentioned before, is colorless although, as it
is a natural product, there is a complex mixture of compounds associated with sucrose when
extracted from cane, being those compounds associated with color themselves, originated in the
cane itself, or acting as precursors of colored compounds formed while processing the sugar. The
last ones are the most undesirable impurities in cane sugar production, and the removal of such
materials from the sugar is the main goal in order to produce a high quality white sugar [2].
Extensive research in order to identify and characterize those compounds have been
performed internationally, so better procedures to remove and avoid their formation could be
proposed and tested. Following will be given a brief description of those compounds, dividing it
into two main categories: natural colorants- present in the cane itself; and those known as hexose
degradation products, formed during the sugar production.
2.2.1 Natural colorants
Sugar cane contains plant pigments, and those are present in the juice after milling, some are
colored, Chlorophylls and carotenes, others will suffer reactions after milling to form colored
compound. Those pigments are mainly phenolics and flavonoids, and they represent up to two
thirds of the color present in raw sugar [3]. The phenolic compounds are in general colorless but
are oxidized or react with amine or iron forming colorants during the process. The flavonoids are
polyphenols present in the cane plant and they suffer enzymatic browning reactions. Those
compounds are well removed during clarification or carbonatation, but can also be incorporated
into the sugar crystals.
6
2.2.2 Colorants formed during sugar processing.
Many colorants can be formed during sugar processing, either in sugar mills or during the
refining process, being those compounds formed from precursors molecules present in the sugar
cane that undergo a complex range of reactions to form the colored compounds. In addition to that,
the degradation of hexoses, during the process due to high or low pH and temperature conditions,
and enhanced by the presence of amino and inorganic compounds in sugar liquors, can lead to
degradation products of the hexoses, which are highly colored compounds. Furthermore, sucrose
can be hydrolyzed to form inverted sugars, glucose and fructose, which also can be degraded
forming highly reactive compounds which can go further reactions to form high molecular mass
colored compounds. There are four main types of colored products that can be distinguished
depending on the mechanism by which they are formed [4, 5]:
 Melanins: those colorants are the result of the enzymatic oxidation, or enzymatic browning,
of the phenolic compounds originally present in the sugar cane. These compounds can be
formed at temperatures from -18ºC to 55ºC and a pH range from 4.5 to 8 [4]. The formation
of melanins is higher during the milling stage, after which it can be stopped, due to the
denaturation of the enzymes at the higher temperatures of following stages of the process.
 Melanoidins: produced by the condensation reaction between compounds having in its
chemical structure carbonyl groups and those having amino groups. In sugar solutions the
presence of inverted sugars is the main source of carbonyl compounds while amino groups
are related with the presence of proteins, amino acids and ammonia, etc. those are high
molecular mass compounds, with intense color.
 Caramels: it is the product of the thermal degradation of sucrose, comprising low anionic
charge compounds, having high molecular weight that increases with time and temperature
as a result of increasing polymerization degree, giving highly colored compounds.
Normally, caramel formation is slow and occurs at temperatures normally much higher than
those at which the process is performed, although it can happen due to contact with higher
temperatures or be catalyzed by inorganic compounds and other sub products of inverted
sugars;
 Hexose Alkaline Degradation Products (HADP): those are polymeric brownish yellow
compounds formed by the decomposition of monosaccharides in alkaline conditions.
7
The following table (Table 2.1) summarizes the main color compounds present in sugar
process [4].
Table 2-1: Type of sugar colorants present in sugar [4].
Sugarcane HDAP Melanin Melanoidins Caramels

colorants
Molecular Monomeric, Intermediate Very high, Polymeric, Polymeric,

<1200 <10000 >250000 >10000 >10000
Mass
Origin Flavonoids and Reactions of Enzymatic In the process by Thermal

other phenolics hexoses at high browning of reactions decomposition of
from cane stalk. pH (>8.0) phenols and involving amino sucrose at high
Iron complexes reactions with acids and inverted temperatures.
with phenols amino acids. sugars.
Degree of Ionized at high Cationic below Anionic Cationic below Slightly charged.
ionization pH and neutral at pH 5, anionic pH 5, anionic
Low pH’s above pH 6. above pH 6.
2.3 Measurement of color in sugar
Color in sugar products is usually measured according to the International Commission for
Uniform Methods for Sugar Analyses, known by its initials, ICUMSA. The color index is then
given in ICUMSA units, IU, and is measured by the method that involves measuring the light
absorbance of the sugar solution at 420nm and pH adjusted to 7.0 at 20°C, after filtration though a
0.45µm membrane [4]. The absorbance value obtained is then adjusted to the thickness of the
measuring cell and expressed relatively to the concentration of solids dissolved in the liquid
solution, brix of the solution, the color index value is then calculated by equation 2.1 [4] presented
below:
𝐴420 ×1000
𝐼𝑈 = (2.1)
𝛿×𝑐
8
Where δ is the measurement cell length and c is the dissolved solids concentration in g/mL.
It is important here to notice that the pore size of the membrane is important as well as the
pH at measurements conditions, since they will both affect the result significantly. The membrane
filter retains high molecular weight color compounds which affect significantly the color measured
and the pH is of great importance since different colorants will present different behavior related
to pH changes, exhibiting sometimes a great variation on the color as result of small changes on
the pH of the solution.
The behavior of the colored substances in relation to variations in the pH is used as a mean
of differentiating between types of colorants, and this relation is what gives rise to what is called
Indicator Value, IV, which is an important parameter to characterize the types of colorants present
in certain sugar products. The indicator value is obtained by the ratio of the color measured at pH
9 and the color measured at pH 4, or simplifying we can use the relation of color to absorbance by
the previous method to obtain the color index, leading to the following equation (equation 2.2) [4]:
𝐴𝑝𝐻=9
𝐼𝑉 = (2.2)
𝐴𝑝𝐻=4
Notice that the color index calculated as well as the indicator value don’t give complete
information of the color present in the product, and one important contribute for the color
characterization in sugar industry was given by Bento [5, 6] by measuring the indicator value in
the ultraviolet zone.
2.4 Decolorization process
In order to achieve the desired degree of decolorization many refineries perform a two stages
decolorization process. The first stage is performed in a clarification process and there are mainly
9
two processes in use in the sugar refineries, carbonatation and phosphatation, there is also a process
called sulphitation but it is not usual in refineries, being more performed in sugar mills and in
unities of plantation white sugar. The secondary stage of the decolorization process is an adsorption
process, and can be adsorption in bone char, granular or powdered activated carbon, or by it can be
an ion exchange process. Or, as normally occur in the refineries, the two processes are used as
secondary decolorization, mainly ion exchange followed by adsorption in granular activated
carbon.
In this work it is of interest the understanding of the primary decolorization processes, where
is aimed the use of polymers flocculants and color precipitants targeting the reduction of color of
the liquor in order to prevent higher expenses in the secondary decolorization step.
2.5 Primary decolorization process
The choice of the primary decolorization process depends on the quality of the raw sugar to
be refined, but not only, is it also dependent on the overall refining process and the target quality
product to be obtained. It also has a great influence in the configuration of the next operation steps,
like filtration, flotation, centrifugation and the technology employed for the secondary
decolorization, when needed. The main objective in the primary decolorization is to reduce the
color of the liquor to an average of 200-250 IU, to avoid higher expenses in secondary color
removal to achieve the 45 IU, which is targeted in refining sugar [4].
2.5.1 Phosphatation
Phosphatation is a process in which lime and phosphoric acid are added to the sugar liquor
in a reaction vessel, being the neutralization reaction described by the following equation:
2𝐻3 𝑃𝑂4 + 3𝐶𝑎(𝑂𝐻)2 → 𝐶𝑎3 (𝑃𝑂4 )2 + 6𝐻2 𝑂 (2.3)
10
This process the anionic impurities are attached to the salt formed, Calcium Phosphate,, and
relies on a flocculation process to remove impurities, by the addition of flocculants that captures
the fine colloidal colored compounds and then removing it by flotation with dissolved air. In this
process the scums contain normally from 3 up to 7% of sugar, what leads to the need of a process
of desweetening the scums [4]. This is one of the strongest disadvantages of phosphatation along
with the lower color removal compared to that obtained by carbonatation process, but it can be
improved by the addition of cationic surfactants [7, 8].
2.5.2 Carbonatation
This is the most utilized process in sugar refineries, and was extensively studied by many
authors like Bennett [9, 10], that performed studies to describe the mechanism of this process and
it was concluded that the impurities are trapped within the crystals of calcium carbonate rather than
its adsorption onto the surface after precipitation. It was observed that any color body with anionic
character could be able to form a linkage with calcium, and so get incorporated into the precipitated
salt. For this reason the extension of color removal is not particularly sensitive to the precipitated
surface area, being more related to a function of the rate of growth of the crystals, making possible
the impurities to be incorporated in the salt. The conditions of operation in the carbonatation
process, temperature around 80°C and high pH, between 8.5 and 9.5, avoid the problems of inverted
sugars as well as the enzymatic browning while promoting the denaturation of the enzymes in this
conditions [4].
In this process lime is added to the melted sugar liquor in saturators where a gas containing
carbon dioxide is then bubbled in a saturation process, undergoing the following reaction (equation
2.4), where the calcium carbonate is formed:
𝐶𝑎(𝑂𝐻)2 + 𝐶𝑂2 → 𝐶𝑎𝐶𝑂3 + 𝐻2 𝑂 (2.4)
The carbonatation is performed normally in two or three stages. The first stage is the larger
and the pH value is kept around 9.5, being in this stage where most of the gas absorption occurs.
11
In the final stage the exiting liquor is at a pH value around 8.5 [4]. One of the issues of the
carbonatation process is the filterability of the liquor that exit the saturators, which depends very
much in the structure of the crystals formed during the carbonation, making very important to find
the condition in order to achieve a bigger and more compact crystal, to enable the pressure filtration
to be well performed.
2.5.3 Sulphitation
Sulphitation has proved to be more effective in color formation suppressing than as a mean
of removing color already formed in the sugar liquor [3], This was explained stating that the
sulphite ions can react with unsaturated carbonyl groups in the color molecules, shifting the light
absorption wavelength out of the visible zone. However, those new compounds formed remain in
the sugar liquor and can go further reactions later to form highly colored compounds. The process
avoids the formation of caramels and moreover the formation of melanoidins. This process is more
used in sugar juice clarification in the production of raw sugar, or in white sugar produced in sugar
mills, plantation white sugar. In refineries is sometime used after carbonatation process as pH-
adjusting, and maybe adding 10% more color reduction to the liquor [4].
2.6 Flocculants and color precipitants
It has been over forty years since the publication of some of the most important studies
performed in the use of polymer flocculants in sugar industry, which until nowadays are reference
to development in the art. That study was written by Michael Camm Bennett, by Tate & Lyle
Limited, London [10], but many others authors like Moodley [8] and Dhoerty [11] added new
insights to this process. Firstly, must be explained how polymers can induce flocculation in
particles suspensions, and so, it will be given next a brief explanation of flocculation mechanisms.
12
2.6.1 Polymer flocculation
Flocculation and coagulation are processes by which colloidal particles dispersed in solution
can be induced destabilization, promoting the aggregation of these particles in order to enhance the
separation process by physical means such as filtration or sedimentation. The polymers used in
flocculation processes can be divided into natural and synthetics, and even more important as
anionic, cationic or non-ionic, according to the charge character of the polymer.
There are mainly three mechanisms though which polymer flocculation can occur [12, 13]:
 Charge neutralization mechanism: the particle charge neutralization in colloidal

suspensions can occur by the addition of polyelectrolytes, through the compression of the
double layer, when the ionic strength of the solution is increased, allowing thereafter short-
range attractive forces to predominate over the electrostatic repulsive forces. In the
presence of opposite charges polyelectrolytes those can neutralize the charges of the
particles by adsorption onto its surface, reducing then the double layer repulsion allowing
the aggregation to occur due to Van Der Waals forces.
 Bridging mechanism: this flocculation mechanism occurs when a long chain polymer is
added to a colloidal suspension, the polymer chains are adsorbed onto the particles
suspended in a way that one single chain can be attached to two or more particles by kind
of “bridging” them together. This mechanism is very much dependent on an optimal
polymer dosage, at which it is assured that there are sufficient unoccupied particle surface
in order to allow attachment to other particles already attached to different polymer chains,
bridging them together. At same time those bridges should be made in a such extension to
guarantee that the distance over which particles aggregation is overcome. Following this,
too low dosages of polymers might not provide sufficient polymer chains to form good
bridging links between particles, but also too high dosages diminish particle surfaces for
the bridging attachment and steric repulsion may occur inhibiting aggregation of the flocs.
 Electrostatic patch mechanism: this mechanism has been proposed to explain the
resulted observation that polymers with high cationic charges added into anionic colloidal
suspensions the energy interactions favors a flattened adsorption configuration. The highly
charged polymer adsorbs onto the surface of the particle in a way that lead to areas, which
seem like “islands” or “patches” of charge, surrounded by areas of opposite charge. Thus,
13
when the polymer is adsorbed in this manner, it can interact with oppositely-charge areas
of a different particle promoting a strong electrostatic attraction between them. This
mechanism predominates in systems of anionic colloidal suspensions when cationic
polymer is added, even though bridging mechanism may occur if particle concentration is
high enough to promote collisions between them in a time similar to that needed for the
polymer to form the flattened configuration.
The next drawing (Figure 2.2), represents the bridge and patch mechanisms of flocculation,
which are, normally, expected to happen when using polymers flocculation:
Figure 2.2: Representation of bridging and patch flocculation, green balls are the anionic impurities and the polymer chains [14].
2.6.2 Anionic flocculants
The first reference flocculants used in sugar industry was the polyacrylamide [3], being that
polymer used mainly to improve the separation during the clarification process in the sugar mills,
in production of raw sugar, and in refineries using the phosphatation process to enhance the
separation of the calcium precipitates from the sugar solution. Those flocculants are normally high
molecular weight copolymers of acrylamide and sodium acrylate, with anionic character that react
14
forming a bridge with the calcium ions precipitates, promoting flocculation, increasing the size of
the precipitate and so promoting an easier and faster separation by filtration or flotation. This
anionic polymers are mostly used in phosphatation process, without that the process is much less
rentable, in this case those polyelectrolytes have a major importance [3]. The efficiency of those
polymers in the flocculation is very dependent on the molecular mass, the charge density and the
structure of the polymer chain. Gorjian [11] performed studies using cationic copolymers as
flocculation aids in raw sugar production, in the clarification process where only initially anionic
polyacrylamides were used, concluding that dosages 5-10ppms of copolymers containing
dimethylaminoethyl acrylate methyl chloride and DADMAC’s, enhanced the color removal in
sugar cane juice clarification.
2.6.3 Cationic flocculants
As the target color compounds to be removed by precipitation have an anionic character, as

concluded by Bennett [10], the surfactants to be used for targeting specifically those colored
compounds should have a strong cationic character.
Cationic polymers have been investigated since the 60’s in water treatment processes. In the
late sixties of the 20th century, Michael Camm Bennett [9,10], for Tate and Lyle Ltd., performed a
series of studies and research, on the field of flocculation in sugar solutions, which would compose
the basis for the research for the use of polyelectrolytes to enhance color removal in the sugar
industries processes. He concluded that the decolorization processes performed in the refineries,
namely ion exchange process and adsorption in bone char or activated carbon, were characterized
by their ability to remove anionic impurities from the sugar liquor.
Those anionic impurities are compounds that present a strong hydrophilic character, therefore
highly water soluble. For that, in order to remove them by physical means, like filtration or
clarification processes, those materials should react, forming chemical or physical bonds, with
substances which would turn them insoluble in water, and so allowing the separation. Bennett [10]
observed that those color compounds were trapped into calcium carbonate crystals during the
carbonatation process and that they were also well removed by adsorption process in bone char,
which also presents calcium ions, also with the capacity to adsorb negatively charged anionic
15
molecules. Thus he concluded that the most effective way to remove those color impurities was by
using materials with a strong cationic character in order to attract and promote the precipitation of
the colored compounds. Based on the use of cationic surfactants in water purifications processes,
he tested some of these polymers used for water purification in sugar liquor to see if it could
perform any decolorization, after what he concluded that flocculation with cationic polymers could
be used as a mean of color precipitation in sugar refining process.
Bennett [10] tested many cationic polymers, after these he concluded that those presenting
the best performance as color precipitants were those ones containing a cationic center configured
by a quaternary ammonium, attached to long or cyclic hydrocarbon chains, in this way the
ammonium cationic center would attract and bind to the anionic impurity while the hydrophobic
hydrocarbon chains proportionate the means to precipitate the complex formed, as it is insoluble
in water. He concluded by that time that the reaction between the polymer and the impurities has
an ionic character being, so, very fast reactions. He proposed that the chain length should be the
ones containing sixteen or eighteen carbon atoms, as it is present in the structure of the surfactant
named Talofloc (a trademark of Tate and Lyle Ltd., London), that is the dialkyl dimethyl
ammonium chloride, with the alky group containing eighteen carbon atoms. Below is an image
(Figure 2.2), containing the name and structure of cationic surfactants tested by Bennett in his trials
for the application of his patent submitted in 1972, which inspires the study of color precipitants in
sugar industry until nowadays.
Figure 2.3: Polymers tested By Bennett in his work (1971) adapted from [10].
16
After testing the polymers he concluded that the best one was the Talofloc. Bennett also
found that those anionic color compounds present in the sugar liquor could be precipitated from
water solutions with primary or secondary amines [15]. The figure presented next (figure 2.3)
shows the model of interaction through which the surfactant Talofloc interacts with the color bodies
in the sugar liquor.
Figure 2.4: Representation of the interaction between surfactant and anionic color impurity, adapted form [8]
Talomel [3], an acyclic dimethyl amine polymer, which was reported initially to have its use
more largely applied in the production of mill white sugar, or plantation sugar, has been reported
to increase the decolorization by 20-30% in a clarifying sugar syrup [16] by adding 100-200 ppm
of the polymer to the clarifying syrup. Talomel is in truth a Diallyldimethyl Ammonium Chloride
polymer, or as it is called, a poly (DADMAC). Another polymer, Talocarb, an Epichlorohydrin-
methylamine copolymer, approved by the FDA to use as color precipitant, was also reported that
adding 200ppms of this surfactant on brix of the sugar liquor, added immediately before filtration,
increased the decolorization by 25-30% in a carbonated liquor. Talofloc and Talocarb were the two
most used polymers in sugar decolorization process for a time, followed by the poly
(Diallyldimethyl ammonium chloride).
Talofloc was very successfully used in phosphatation refineries, because of its structure that
promotes very small flocs to be precipitated, which were difficult to be removed by filtration, while
used alone, but further studies proved that when used in a phosphatation process, it improved the
process and enhanced the color removal, jumping from an initial decolorization, promoted by the
phosphatation process alone, in the order of 25 %, to a total decolorization of 70%, when
17
phosphatation is performed using Talofloc during process. Additionally, using polyacrylamide in
the phosphatation, it has enhanced the flotation process, by promoting the formation of larger
aerated flocs [7]. But Talofloc has not been so successful in the carbonatation process, because it
is a wax at room temperature, difficult to handle and dosage. Besides, as carbonatation normally
uses pressure filtration as next step in the process it was not so useful, as it created a filtration
problem, due to large pressure drops during filtration.
The newer polymer Talocarb works as a color precipitant the can be used in carbonatation
refineries, which has proved to be effective in the process when added after the second stage of
carbonatation, along with the calcium carbonate precipitated. Also, it didn’t present major
problems for the filtration process and improved the decolorization performance of carbonatation
alone by the additional removal of color impurities till up to 300 IU [7].
M. Moodley [8], performed also tests in decolorization with commercial surfactants, Bulab
5031, a polyamine, Talocarb, dimethyl amine polymer, Primco 144, a polyamine and Ultrafloc
5000, also a polyamine, performing these tests in different plants, one using carbonatation and
other using phosphatation as clarification process. In his study, he proved the effective performance
of cationic surfactants as color removal agents, enhancing the clarification process by promoting
an additional color removal of 20 to 30% in the process, when adding around 200ppms of the
polymers. He concluded that polyamines could be added to carbonated liquor before filtration,
while in phosphatation refineries it should be added in raw melted liquor before the phosphatation
unity [3].
Runggas [16], performed some relevant studies in the sugar milling institute, university of
Natal, South Africa, which were presented in the proceedings of the Southern African sugar
technologists association, using a range of cationic flocculants in carbonatation refining process.
The polymers tested were produced locally, but various commercial polymers were also tested,
from trademarks like Talofloc, Flocotan, and a set of Superfloc trademark surfactants, which were
poly(DADMAC)’s of low, medium and high molecular weight, other comercial poly(amines), and
Basefloc C45, used in water clarification. This study conducted tests on color removal, turbidity,
ashes, starch and gums removal as well as on the influence in the filterability of the carbonated
liquor after polymer treatment. He concluded that the optimal color removal addition point of the
polymer in the process was after saturation, at a point prior to filtration. The most effective dosages
were 50ppm’s of polymer on sugar liquor brix. The Superfloc grade (a poly (DADMAC) of
18
medium molecular weight) had the best color removal performance. Additionally those surfactants
had good performance in gums and starch removal, as well as in the removal of suspended particles,
reducing the turbidity of the liquor. One important conclusion is also that for those cationic
polymers, the dosage have to be well tuned. This factor affects very much the performance: a low
dosage might not be so effective but high dosages can reverse the colored bodies bond to the
polymer, being prejudicial to the color removal, sending the color molecules back to the sugar
liquor. Furthermore, high dosages can produce foams in the liquor, what would be a big problem
in the next steps, the filtration and even in the secondary color removal process.
2.7 Secondary color removal technics
The final color in refined white sugar sometimes has to be below 60 IU, and for some markets
the requirement is that of 45 IU, and purity superior to 99.9% of sucrose [4]. The color achieved
normally by the primary decolorization process is in the range of 200-300 IU, so a secondary
decolorization is required, which can be mainly one or more of the following processes:
 Bone char adsorption;

 Granular or powdered activated carbon adsorption;
 Ion exchange.
A brief explanation of those decolorization processes is given in the next subsections.
2.7.1 Bone char
The adsorption in bone char has been the main decolorization process for many years in the
sugar industry, until the development of ion exchange and granular activated carbon use in the
refineries to take place. It consists of a high surface area material, constituted basically calcium
phosphate, hydroxyapatite, which is coated with carbon. It comprises of only 10-12% carbon in
total. This kind of adsorption is no longer in use, mainly because of the disadvantage of large and
19
expensive unitis, along with the need of large char inventory as, with time operation, char degrades
and pores are blocked, increasing its density, in a way that it needs to be replaced. So it is not any
more cost effective [4, 17].
2.7.2 Activated carbon
Activated carbon is a non-polar highly porous material, with very large surface area, which
is used as color removal aid, by physical adsorption mechanism [4], where strong attraction
between the carbon surface and sugar molecules present in sugar liquor, promoted by van der
Waal’s or London forces, so occurs and the color molecules are held against many points in the
large surface of the activated carbon, thus being strongly adsorbed on it.
The color removal is mainly physical, but chemisorption may occur due to the activated
carbon still containing oxygenated functional groups which will adsorb polar molecules. It is not
specific to remove any type of colorant, giving a high overall decolorization, of around 80%, which
may go up to 90% [4]. It is very good to remove flavonoids and phenolic colorants [3]. Normally
granular activated carbon is used but powdered activated carbon may be used for small through
output applications. Besides, the use of granulated carbon adsorption allows the removal of odorous
compounds present in the sugar liquor.
One of the drawbacks of the use of activated carbon may be the sucrose losses by pH drop
and inversion reaction, which can be avoided by adding magnesite to the system during
revivification of the granular activated carbon [17]. One of the advantages of the use of activated
carbon instead of bone char is the impact in energy consumption for the decolorization process,
which dropped to 50% of that needed when used bone char [4].
2.7.3 Ion exchange
In this process, polymeric resins in macro reticular form and with strong basic character are
used to remove anionic colorants from the sugar liquor through ionic bounds with the cationic
center of the resin along with hydrophobic interactions with the resin matrix [4], if a styrenic resin
20
is used. Besides the styrenic, acrylic resins are used in ion exchange columns, and even though the
acrylic are generally able to remove more colorants they are more prone to fouling, thus, sometimes
the two resins are used in combination, with the styrenic ahead of the acrylic one, in a way that the
first one provides protection to the second.
Those resins are characterized by their high molecular weight, as well as a high degree of
cross-linking, presenting a quaternary ammonium functional group in their chemical structure. The
acrylic resins are less selective in the process than the styrenic, but on the other hand its
regeneration process is easier and more complete, this way they show slower rate of deterioration.
The main mechanism of this process is the ion exchange between the chloride ions to be
displaced from the resin matrix, which will be replaced by the anionic groups on the color
compounds. Competitive exchange occurs during the process [3], by smaller organic molecules,
sometimes not color related, or inorganic molecules, which may also occupy the active sites leading
to the exclusion of color molecules. In this process, besides the functional groups of the color
bodies, the size of those compounds also affect the performance of the process.
21
22
Chapter 3. Materials and Methods
This section refers to the methods by which the laboratory trials were performed, as well as
the main materials utilized for that purpose. The tests were divided into four stages: decolorization,
indicator valor evaluation, polymers characterization and last the monitoring of flocculation
process performance by the most effective polymers used.
3.1 Materials
3.1.1 Sugar
The starting point of the work is the raw sugar to be decolorized, that is sugar from sugarcane,
and it is important to mention that the raw sugar has different compositions in terms of color bodies
present on it, depending on its geographic origin, climate and soil conditions, season of harvesting
and process through which the raw sugar was produced, among other factors and aspects that may
influence the composition of raw sugar (the sugar used had its origin in central America).
The next figure represents the configuration of the carbonatation unity, from which the sugar
liquor was harvest, consistent of three saturators, where the major part of the calcium carbonate is
formed, representing the first stage of the carbonatation and where the liquor has an average pH of
9.5, followed then by a second stage saturator, where the pH range will be in the order of 8.5. It is
at this point that it is normally added the flocculants, as normally those polymers are pH sensitive
and it is known to be at this point that they are most effective in the carbonatation process [3, 6].
The sample collecting, to be used in the present study, is made then in the saturator B, pointed in
the diagram below by the red circle, before the before the polymer is added in the industrial process
conducted by RAR Portugal.
23
Lime
Melt liquor
Mixer
Addition of decolorization polymer

Saturator A
Extration of samples
before decolorization polymer be added
P-10
Saturator A1
Saturator B
Carbonated liquor to filtration
Saturator A2
Gas rich in CO2
Figure 3.1: Carbonatation process diagram, red circle represents point of sample extraction for the laboratorial trail ( the scheme
represents the unit of RAR).
For all the tests performed in this work the sugar source was raw sugar liquor collected in the
installations of the sugar refinery RAR, being the samples of 12 liters collected at the second stage
of the carbonatation process, where the normal conditions of the liquor are around 80°C and the
pH is 8-8.5 and with a degree of Brix( concentration of solids in the liquor) of approximately
65°Brix, and prior to the dosage of color precipitants that is added normally at this second stage,
and before the pressure filtration unit. The samples were immediately cooled in order to stop the
carbonatation process, which was still going on at the collecting time of the liquor. The sugar
carbonated liquor was then transported in thermal protection container, after what it was kept in a
refrigerator at temperatures close to zero, for the use in the decolorization tests planned in the
laboratory. The same process of collecting samples and storage was performed one second time, to
collect more liquor to perform tests of flocculation monitoring in the LDS (Laser Diffraction
Spectroscopy) system.
24
3.1.2 Polymers
After the research in the literature on the previous studies performed in color removal with
polyelectrolytes, and taking into account that the colored materials to be removed have an anionic
character it was decided to use cationic surfactants, which were provided by AquaTech
Switzerland.
Knowing that the most effective polymers to precipitate the color impurities present in the
sugar liquor are those with strongly cationic functional group [9, 3] to attract the anionic impurities,
and that those cationic centers must be bonded with hydrocarbon chains to provide the hydrophobic
character in order to allow the precipitation, and that, according to Bennett [10] the most effective
cationic character was provided by polymers containing quaternary ammonium compounds.
Following those information, it has been decided to test a set of polyamines and poly (diallyl
dimethyl ammonium chloride), or poly (DADMAC) as it is known. In addition, the novel polymer
poly [(2-methacryloxy) ethyl trimethyl ammonium chloride], commercial name Amber MetQuat
6. Those polymers were provided as solution with the concentrations given in weight of polymer
by volume of solution. The complete list of polymers used are shown next (Table 3.1).
Table 3-1: List of Polymers used in the present study, provided by AquaTech Switzerland.
Polymer Form, Chemistry Viscosity(MPas)

concentration(%w)
Amber 2001HMW Liquid, 50% Polyamine 2000-3000
Amber 2002 Liquid, 50% Polyamine 400-700
Amber 2006 Liquid, 50% Polyamine 100-300
Amber 2008 Liquid, 40% Poly(DADMAC) 500-1000
Amber 2008HMW Liquid, 40% Poly(DADMAC) 1000-2000
Amber 2013 Liquid, 35% Poly(DADMAC) >9000
Amber 2014 Liquid, 50% Poly(DADMAC) >10000
Amber MetQuat 6 Liquid, 40% MetQuat 6 1200
25
3.2 Methods
3.2.1 Preparation of the polymer solutions
The polymers to be used in the decolorization procedures were first prepared in solutions
of 1% (w/V) in ultrapure water. For each test, a solution of 50ml of polymer was prepared and kept
under agitation at 400 rpm’s in stirring plates for twelve hours, covered with parafilm to avoid
contamination with air particles. The solutions that were prepared took into account the
concentration of polymer by weight in the original solution provided by AquaTech.
3.2.2 Decolorization tests
The decolorization tests were performed using the carbonated sugar liquor brought from
RAR.
First the liquor was taken from the refrigerator were it has been kept, agitated and then taken
50 ml of the liquor measured in a graduated cylinder, and then transferred to a 100ml beaker which
was after covered with aluminum foil and then taken into a thermic bath, with a thermal sensor
placed inside the liquor to monitor and control the temperature, then it was heated to the
temperature of 80°C, in the intuit of lead to similar conditions of the carbonatation system. When
the temperature of the liquor reached 75°C the prepared polymer solution was then added with one
micrometric pipet, the volume of solution added to the system was that reached by calculations of
the referent ppm’s aimed to be used, those calculations were referent to the solids content in the
liquor, or as it is said in the sugar industry, based on the brix of the liquor.
In the literature [3, 8, 15] has been reported dosages from 10-600 ppm’s on brix for polymers, some
authors used the range till 150, in the work the tests were performed for: 25, 50, 100, 150, 200 and
250 ppm’s of polymers based on the brix of the liquor. After addition of the polymer the system
was kept at temperature maxim of 80°C, in agitation at 400 rpm’s for 10 minutes counting from
the addition of the polymer solution to the carbonating liquor.
After the ten minutes the agitations and heating are stopped, the liquor was diluted in one
part of liquor to two parts of ultrapure water, after what it was then cooled to temperatures around
20°C, proceeding then to the vacuum filtration with aid of a 0.45µm membrane filter. After that,
26
the filtrated was leaded to pH and temperature adjust for 20°C and pH 7.0, according to the
ICUMSA requirements for color measurement in sugar solutions. After this the measurement of
the absorbance at 420nm was carried out in three times for two solutions inserted in the measuring
covet, performing six absorbance measurements, from which an average absorbance was
calculated. After that the measurement of the Brix of the solution was performed in a refractometer.
After all that done, with the absorbance value, dimension of the cell, 1cm, and the brix, of the
solution, all measured at 20°C and pH 7.0 ± 0.02, the ICUMSA color index of the filtrated solution
was then calculated using the ICUMSA 420 method, by means of the equation 2.1 mentioned
previously.
The figure presented next (Figure 3.2), shows in block diagram the flow of the trial
procedure:
3- Polymer is then added

1- Solutions of 1% 2- Carbonated sugar
to the lliquor after it
weight of the polymers liquor brought from
reaches temperature of
were prepered in RAR, was then heated to
75ºC then left to react
ultrapure water; temperature of 80ºC;
for 10 minutes.
4- The licor was diluted

to 1:2 in ultrapure water, 5- pH adjustmment to 6- The brix of the
cooled to 20ºC, then 7.0±0.02, then followed solution was then
filtrated on vacum using by absorbance measured in a
membrane filters of 0.45 measument at 420nm. refractometer.
micros.
7- The color was then

calculated by the icumsa
color index method for
ph 7, 20⁰c and 420 nm.
Figure 3.2: Schematic representation in flux diagram of the procedure for decolorization laboratorial tests performed.
It has to be mentioned that at same time that the liquor was submitted to the decolorization
process with the polyelectrolyte, another beaker with liquor from the same collected sample was
taken and submitted to the same steps, but without addition of polymer. This second sample will
be referred as the blank of the test, to compare to the one to which the polymer was added. The
27
decolorization percentage for the test is then calculated based in the following formula (equation
3.1):
(𝐶𝐼𝑈𝑏𝑙𝑎𝑛𝑘 −𝐶𝐼𝑈𝑠𝑎𝑚𝑝𝑙𝑒 )
( ) 𝑥 100 = %𝑑𝑒𝑐𝑜𝑙𝑜𝑟𝑖𝑧𝑎𝑡𝑖𝑜𝑛 (3.1)
𝐶𝐼𝑈𝑏𝑙𝑎𝑛𝑘
Where:
 CIUblank is the color of the sample without polymer decolorization, or blank of the test, in
ICUMSA unities, which is the color of the filtrated liquor after carbonatation process;
 CIUsample is the color of the liquor submitted to the decolorization through flocculation
process with the polymers, in ICUMSA unities,
 %decolorization is the percentage of decolorization achieved.
All the procedure described above were repeated for all the tests of decolorization, which
were performed for each polymer and for each dosage of polymer tested individually.
In addition to the individual polymer decolorization tests, tests were also performed mixing
two and three of previous polymers simultaneously. The polymers having shown the best
performance in the individual decolorization tests have been chosen for using in the blending.
3.4 Evaluation of IV on decolorization
After the polymer decolorization tests in order to understand if the impurities removed by
this process were those characterized by high, medium or low molecular weight, evaluation of
changes in the Indicator Value( IV) were conducted. In order to do that, the procedure for individual
decolorization test was performed and the liquor submitted to the reaction with the polymer was,
after the filtration, split into two parts, both at 20°C temperature. The pH was adjusted to 7.0 in
one sample, which was again splinted in two parts, and the absorbance was measured for one of
those parts at the pH 7.0, the other part adjusted to pH 4.0, after which the absorbance was
measured. The other half of the filtrated liquor was adjusted to pH 9.0, and then the absorbance at
28
this pH was measured. As explained in the previous chapter, the indicator value (IV) can be
calculated by the ratio of the absorbance of the sugar solution at pH 9.0 and absorbance at pH 4.0,
or alternatively by the equation 3.2 [4]:
𝑐𝑜𝑙𝑜𝑟𝑝𝐻=9
𝐼𝑉 = (3.2)
𝑐𝑜𝑙𝑜𝑟𝑝𝐻=4
3.5 Polymers characterization
Characterization of the polymers used in the decolorization tests was also conducted to try to
understand their activity. It was decided to perform tests to determine three important aspects that
have influence in the flocculation process: the hydrodynamic diameter of the polymer (determined
by dynamic light scattering), the zeta potential of the polymer chains in deionized water
(electrophoretic light scattering), and the molecular weight of the polymers (static light scattering).
In order to perform the molecular weight measurement the determination of the refractive index of
the polymer solution was also measured.
The equipment used to perform the characterization of the polyelectrolytes was the high
performance Zetasizer Nano ZS (ZSN) commercialized by Malvern, installed in the facilities of
the solids technology laboratory, in the chemical engineering department of the university of
Coimbra.
Figure 3.3: Zeta Nano Sizer ZSN by Malvern. [18].
29
3.5.1 Hydrodynamic diameter
When referring to a hydrodynamic radius or diameter, this dimension is a hypothetical

dimension, considering that the easier shape to characterize three dimensionally a particle is a
sphere because its diameter is enough to determine the size of the particle. Following this idea the
hydrodynamic diameter determined by the Zetasizer is that one of the sphere that diffuses at same
speed as the particle that is being characterized. This parameter is determined using dynamic light
scattering (DLS), and the size is determined based on the Brownian motion of the particles in the
sample relating it to the measurement of the scattering light intensity fluctuation, when a laser beam
is projected into the sample [18].
The particles, below micrometric dimension, dispersed in a liquid solution are consider not
at rest, but in continuous Brownian motion, that is a motion due to the constant collision between
the particles and the molecule of the liquid that surrounds the particle. The speed of the particle in
this movement is dependent on the size of the particle, the larger is the particle slower it moves.
When a light source, in this case a laser beam, crosses the solution it will hit the particles and the
light will be scattered by them in all directions. Being the particles in constant motion, the scattering
pattern will change over time, presenting an intensity fluctuation as a function of time. The small
particles move quickly, producing a higher intensity fluctuation, on the other hand the larger
particles should move slower, presenting also lower intensity fluctuation. This intensity fluctuation
signal detected is then sent to a digital correlator, present within the Zetasizer equipment, which
will convert this signal into a particle size [18]. The intensity of two consecutive signals is higher
than two separated signals, so, the smaller the particles, the faster they move and the correlation
between the signals over time will decrease. The next figure is the graph that represents the
correlation functions on which the DLS is based in the Nano Sizer equipment (Figure 3.4).
30
Figure 3.4: Graphic representation of the correlation functions used in DLS treatment by the Nano Sizer equipment [18].
The equipment software relies in two algorithms to obtain a diameter value: Cumulants
analysis, that fits the correlation curve to an exponential function, or the CONTIN method, based
on the approximation of the correlation curve to a sum of exponential functions, which is more
adequate to complex multidisperse sample analysis. By one of those two algorithms the distribution
of the diffusion coefficient, Dc can be calculated, and for each value of Dc the hydrodynamic
diameter is calculated though the Stokes-Einstein equation presented below (equation 3.3) [18]:
𝑘𝑇
𝐷𝑐 = (3.3)
3𝜋𝜇𝐷ℎ
Where k is the Boltzmann constant, T is the absolute temperature, µ is the dynamic viscosity, and
Dh is the hydrodynamic diameter of the particle. From this calculation a distribution of diameters
is obtained from which an average diameter Dz is then calculated by the Zeta Sizer software,
assuming that the particles light scattering obey the Rayleigh theory. The average diameter Dz is
calculated by equation 3.4 [18] below:
∑ 𝐼𝑖
𝐷𝑧 = 𝐼 (3.4)
∑ 𝑖
𝐷𝑖
31
Where Ii is the scattered light intensity of the class i, and Di is the hydrodynamic diameter of the
same class i.
In the procedure for the measurements, some factors had to be considered, mainly preparation
and handling of the samples to be analyzed. The solution preparation had to take into account first
the concentrations of polymer in solution in order to have the adequate concentration to perform
the measurements, which has to be very low, but high enough to be clean the noise signal, but also
not too high that could cause aggregation of the particles or multiple scattering effects.
The polymer solutions were prepared in ultrapure water with addition of 0.05M of Sodium
Chloride, and concentrations of polymer from 0.1%to 0.01 % (w/V). The starting solution was
prepared in the day before, covered with parafilm, to avoid contamination, and left under agitation
through the night, being the dilutions prepared early in the morning and also left in agitation for at
least one hour before any measurement being performed. All the samples measured were filtrated
though a 0.45µm membrane filter before proceeding to measurements. In the present work, the
filling of the measurement cell was performed using a 5ml syringe coupled to the 0.45µm
membrane filter. After initial trials it has been concluded that the best concentration range for the
size measurement was that in the range between 0.02-0.1% w/V.
3.5.2 Zeta potential measurement
Zeta potential, or electrokinetic potential of colloidal particles is an important parameter used

to characterize electrochemical equilibrium on interfaces, and it is a very important key to the
stability of a particle suspension or solution. In the case of large molecules, zeta potential supplies
an indication of the charge of the molecules.
The zeta potential is a way to express the degree of electrostatic repulsion between particles
adjacent and similarly charged in a colloidal suspension, it is measured in mV (millivolts), thus
suspensions with high zeta potential, either negative or positive, are considered stable thus the high
repulsion between particles in the suspension.
To characterize the polyelectrolytes used in the present study, one important aspect was the
measurement of the zeta potential of the polymers in suspension, which was performed using the
32
equipment mentioned in previous topics, the Zetasizer Nano ZS, which is provided with
electrophoretic light scattering (ELS).
The ZSN equipment performs the sample’s zeta potential measurement based on the
electrophoretic mobility of the charged particles in the suspension when an electric field is applied
to the suspension, causing those charged particles to move towards the electrode of opposite charge,
and the velocity at which this migration occurs is measured by Laser Doppler Velocimetry,
included in the ZSN equipment. The velocity in measured will depend on the following factors
[18]:
 The strength of the electric field applied or voltage gradient;

 The dielectric constant of the suspension medium;
 Viscosity of the suspension;
 The related zeta potential.
The velocity of migration, so referred as electrophoretic mobility, is measured and the

software calculates the zeta potential by the Henry’s equation [18] (equation 3.5):
2ɛ𝑧𝑓(𝑘𝑎)
𝑈𝐸 = (3.5)
3µ
where UE is the electrophoretic mobility, ɛ is dielectric constant, z is the zeta potential, µ is the
medium viscosity and f(ka) is the Henry’s function, which may assume two approximation values-
1.0 or 1.5, depending on the liquid media. In the case of this study since the suspensions were
prepared in ultrapure water, the value of 1.5 was used, which is called Smoluchowski
approximation [18].
For the trials the polymers were prepared in ultrapure water in concentrations similar to those
used to measure the particle size. The zeta software analysis reported as good quality measurements
those made at concentrations on the range of 0.02-0.1% w/V. In this case the, an initial solution
was prepared in the day before and left under continuous agitation at 400 rpm, over a period of
twelve hours, then dilutions were made and left for also under agitation, for a minimum of one hour
before proceeding to the measurements. The solutions were submitted to ultrasound treatment for
33
two minutes prior to cell filling. Here the filling of the measurement cells were also conducted
using a 5ml syringe coupled to a membrane filter of 0.45µm.
3.5.3 Molecular weight
The measurement of the molecular weight (MW) of the polyelectrolytes used in the study
has been performed seeking to understand the influence of this variable on the performance of
those polymers as color precipitants in the sugar liquor. It is known that the MW of the polymer,
along with the charge density and the molecule’s architecture and its length [10, 16] are some of
the major parameters that may influence the flocculation mechanism and performance of such
materials.
The molecular weight of the polymers was measured by means of the Static Light Scattering
spectroscopy technique, which is one of the most widely used procedures to obtain absolute
molecular weight of macromolecules [13]. In this technique, the particles in a diluted suspension
are illuminated by a source of light, such as a laser, being then the light scattered by the particles
in all directions. But in this case, differently to DLS, that measures the fluctuation of the intensity
of scattered light as a function of time, the static light scattering is based on the time-averaged
intensity of the scattered light.
The Zetasizer, ZNS, equipment, used in the previous size and zeta potential characterization,
has also been to conduct the molecular weight measurements of the polymers. In order to guarantee
accurate measurements it is important that the size of the particles to be measured is small enough
to avoid the angular dependence of the scattering intensity, which leads to multiple scattering by
the same particles, and by doing so, it can be assumed the Rayleigh scattering model, letting the
angular disperse term Pθ to be equal to one, allowing then the use of the Rayleigh approximation
for a single angle light scattering, the application of the Rayleigh equation, becoming the equation
3.6 [18] showed next.
𝐾𝐶 1
=( + 2𝐴2 𝐶) 𝑃𝜃 (3.6)
𝑅𝜃 𝑀𝑊
34
Where:
 R θ is the Rayleigh ratio (ratio of scattered light to incident light of the sample);
 MW is the sample’s molecular weight;
 A2 represents the second virial coefficient;
 C is the sample’s concentration;
 Pθ is the angular dependence of the sample scattering intensity, which in the case of
Rayleigh scattering will assume the value of 1;
 K is the optical constant, which is obtained as shown in the next equation, (equation 3.7)
[18];
2𝜋 2 𝑑𝑛 2
𝐾= (𝑛0 𝑑𝑐 ) (3.7)
𝜆4𝑜 𝑁𝐴
Being:
 NA is the Avogadro’s constant;

 λo is the laser wavelength;
 no is the solvent’s refractive index;
 dn/dc is the differential refractive index increment, representing the change in the refractive
index of the sample as a function of the polymer concentration.
The molecular weight measurements are normally performed using standards approach. This
means that, in order to measure a sample’s MW, it should be related to a “standard”, that is a highly
pure substance that has a well-known Rayleigh’s ratio, and for which the scattering intensity will
be first measured in order to compare and relate the measurements of the analyte’s sample to be
measured. In light scattering MW measurements, the most widely used standard is Toluene, which
presents Rayleigh’s ration suitably high to perform precise measurements for a good range of
temperatures and wavelengths. Based on that approach [18], the next equation (equation 3.8) is
used to calculate the Rayleigh ratio for a sample using toluene as standard.
35
𝐼 𝑛2
𝑅𝜃 = 𝐼𝐴𝑛2𝑜 𝑅𝑇 (3.8)
𝑇 𝑇
Where:
 IA is the residual scattering intensity of the analyte;

 IT the toluene scattering intensity;
 no is the solvent’s refractive index;
 nT the toluene refractive index;
 RT is the Rayleigh ratio for toluene.
And so, taking into account that we will assume no angular dependence on the scattering
intensity by the particles, supposing they are small enough to avoid multiple scattering behavior in
the samples; supposing that in the Rayleigh theory the parameter Pθ assumes the value of one as a
consequence, the intensity of the scattered light by a particle is to be proportional to the product of
the weight-average molecular weight and the concentration of the particles in the sample. Thus,
the Rayleigh equation will lead, when plotting the scattering intensity of the light as function of the
sample concentration, to a graphic representation of a straight line, which intercepts the Y axis at
zero concentration on an ordinate that represents the value of 1/MW that is given in Daltons.
Thus, the Zetasizer Nano measures the intensity of scattered light for the various
concentrations (C) of the sample, at one angle (Pθ), which is, then, compared with the scattering
produced by the standard, in this case toluene, and the graphical representation (of the intensity of
scattered light as function of the sample’s polymer concentration) called Debye plot and it leads
to the determination of the molecular weight and the second virial coefficient, A2, which represents
the interaction strength between the particles and the solvent. A typical Debye plot is shown in the
next figure (Figure 3.5).
36
Figure 3.5: Debye Plot [18]
The Debye plot comprises two lines, as shown in the figure 3.5 [18]. The Debye line, in blue,
from which the interception with the Y axis provides the 1/MW, and the green line, defined as
intensity line, that gives information about the quality of the scattering measurements, mainly if
the phenomena of multiple scattering is happening, which should be avoided.
The measurements of the molecular weight in the Zetasizer Nano were performed preparing
first samples with appropriate concentrations, based on the previous size measurements for the
same samples, from which the first range of concentration was defined, and preparing then a
minimum of six samples with different concentrations within the range from 1% to 0.02% w/W. A
table containing the concentrations used for each polymer will be shown in the appendix A, but
those are normally the following: 1.0; 0.5; 0.2; 0.1; 0.05; 0.02, in percentage of weight of polymer
by volumes of solution. The results for the tests showed that the concentration of 1%, and
sometimes 0.5%, presented a behavior of possible multiple scattering, and so, later the 1% was cut
and the 0.08% was added, as is shown in the tables presented in the appendix A. The flow diagram
of the measurement process is shown next (Figure 3.6).
37
after agitation period,

Preparation of 0.05M preparation of 1%w/V of
preparation of dilutions of the
solution of NaCL in polyemer in the NaCl 0.05M
main solution for the
ultrapurewaterm to be used main solution, left for
pretended concentrations,
to prepare the polymer agittation for a minimun of
also in NaCL 0.05M, left for
solutions; 12hours, at 400RPM ;
minimum 1hour agitation;
after dilutions ready proceed

configuration of the ZSN to
to refractive index
proceed to molecular weight
measurements of the polymer
measument, introducction of
solutions, plot it in order to
value of dn/dc, and proceed
the concentratios to obtain
to measuremnts.
dn/dc;
Figure 3.6: Flow diagram for molecular weight measurements methodology adopted.
To perform MW measurements the following order was followed:
 Standard measurement, in this case introduction of the toluene cuvette and measure;
 Introduction of a solvent pure cuvette to measure;
 Beginning of samples measurements, first the extreme concentrations, followed by the
central and intermediate ones.
For all samples measurements, the filling of the measurement cell was made using a 5mL
syringe coupled to a membrane filter of 0.45µm, in order to filtrate the solutions before perform
the measurements. In addition, to low MW polymers the use of ultrasounds prior to measurements
was not needed, but in case of medium to high molecular weight it was applied for two minutes
before proceeding with the measurements.
3.5.4 Refractive index
To perform MW measurements, it is also necessary to know the differential increment for

the refractive index as a function of the increasing range of concentrations of the sample’s solution,
38
which, as mentioned in the molecular weight Rayleigh theory, is an important parameter to be

introduced in the ZSN software in order to calculate the sample’s molecular weight. This was done
by using the Atago RX-5000-CX Refractometer.
The Refractive Index (RI) was determined for each solution concentration, by first
performing calibration with the solvent, solution of NaCl 0.05M, with fixed temperature of 25°C.
After this step, the RI of the solutions, at the given temperature, was measured. RI was plotted as
a function of the sample’s concentration, which presented a linear behavior, from which the dn/dc
value, slope of the straight line, was calculated. The RI results are shown in the appendix A.
3.6 LDS flocculation monitoring tests
After performing the decolorization trials using the standard procedure and the
characterization of the polyelectrolyte’s physical parameters that could most influence their
performance as decolorization agents, it was decided to perform some trials, with the best
decolorization agents, to try to follow continuously the flocculation process in carbonated sugar
liquor, which was performed by Laser Diffraction Spectroscopy, For that end the Mastersizer 2000
by Malvern Instruments, Ltd (Figure 3.7). With adaptation of a thermal bath to allow to perform
tests at temperature above the room temperature was used [13].
39
Figure 3.7: System mounted for LDS monitoring of the flocculation process, comprising the Mastersizer and a water bath adapted
to the system.
By using laser diffraction spectroscopy (LDS) it is possible to obtain information about

particle size and its distribution through measuring the scattering intensity as a function of the
scattering angle and the wavelength of the light source [20]. This technique is able to measure
particles and aggregates in a size range that goes from nanometers to few millimeters.
Briefly explaining, this characterization technic uses angular scattering patterns generated by
a particle when it is illuminated by a light source, like a laser beam. Those angular patterns are then
converted into intensity patterns by Fourier transforming, detected and processed as intensity flux
pattern that will further be converted, using appropriated scattering theories, into a particle size
distribution. LDS is suitable to give information about size, size distribution, shape and, in the case
of the Mastersizer equipment used here, able to give those information in order to the time, what
can give information about the evolution of the particles size and shape, allowing this way to
monitor the flocculation kinetics [13].
40
The Mastersizer 2000 was used to evaluate the flocculation kinetics for four of the best
performing polyelectrolytes used in decolorization of the carbonated sugar liquor, measuring the
size distribution of the suspension along time, before the addition of the polymer for 30 minutes
after polymer addition. Normally the polymer is allowed to react for a period of 15 minutes, in the
decolorization process, but here is was intended to see what happens for longer periods of
flocculation time. The results are presented in the next chapter and in the appendix C.
The monitoring trials were performed after preliminary tests in order to tune the best
experimental conditions. First was tested with filtrated liquor, to try to detect the interaction of the
polymer with the color compounds present in the liquor after submitting it to a filtration though a
1µm membrane. Although this was not successful due to the inappropriate laser obscuration. Then
was tried to perform the tests with the carbonated liquor as it came, in a brix degree of around 65º,
what was also not successful due to the very high obscuration of the laser and blue light in the
Mastersizer system. After trying some dilutions, just enough to make achieve obscuration suitable
to perform the tests it was decided to use a dilution of 1:4 of the original carbonated liquor in
distillated water, which permit to have obscuration between 91% and 94% during the
measurements, which is still a high obscuration, but it allowed the system could so perform the
monitoring. The next diagram (figure3.8) shows the steps for followed while performing the LDS
continuous monitoring of the flocculation for all the polymers tested.
41
preparation of carbonated while heating solution,

sugar liquor solution diluted at proceed to equipment and
1:4 in ultrapure water and software preparatios and
agtitating for minimum 1hr at specifications for the
500 RPM, and heating up to measurement(material,
80°C; refractive index etc.)
peform the background proceed to placement of

measurement, in this case sample solution, 600ml, on
distilled water was used as the sample dispersion unity
background; and then begin measurement;
after certain time, on which after trial stops, a sample of

particle size seams stable, the liquor was take and
polymer solution was added submitted to color
to the system and allowed to measurement according to
react at 75-85°; the ICUMSA 420 method.
Figure 3.8: Diagram representation of steps sequence for the LDS trials.
For the trials the polymers solutions were prepared and dosed as done in the decolorization
trials, described in the previous subsection, speed of the pumping system of the Mastersizer was
set to 1400 RPM to avoid settling and the solution samples were kept in a 600ml beaker, and placed
inside a jacket where hot water was circulating in order to keep the sample at temperatures between
75°C and 80°C, which is the temperature of the carbonatation process. The equipment has always
been clean, lenses and tubes, before and after all trials. By agitating for a period of time before the
data acquisition. The use of ultrasounds in the system has been avoided during trials, to prevent
interfering with the decolorization process or aggregation of the calcium carbonate, present in the
solution.
42
Chapter 4. Results and discussion
In this chapter the results obtained in the study will be presented, being first shown the
polymers characterization results, performed using the Zetasizer Nano, ZSN, comprising of the
hydrodynamics diameter, zeta potential and molecular weight for the polyelectrolytes tested as
color precipitants in this work. In the second part the decolorization achieved with each polymer
in laboratorial tests are presented, keeping in mind that those results were performed according
to the methods used in the sugar industry to measure the color of sugar solutions, which were
presented in the previous chapter. A few tests to evaluate the change in the index value of the
filtrated sugar liquor, the ratio between liquor color measured at pH 9 and the color at pH 4,
were also performed and the results presented in the third part of this chapter. As final studies,
the flocculation process online monitoring by using laser diffraction spectroscopy (LDS) is
presented, as kinetic curves of the flocs size distribution during the flocculation process.
4.1 Polymer Characterization
One aspect to be observed, that should be of great importance for the present study, is the
attempt to relate the performance of the polymers in the color removal process with its
characteristics, namely the Molecular weight, hydrodynamics radius and the zeta
potential(related to the polymer charge).
Those characteristics of the polyelectrolytes are known to be very important when

studying flocculation. For that reason, the polymers used in the study were characterized by
using the Zetasizer Nano ZSN equipment, according to the procedures presented in the previous
chapter. The results obtained are summarized in table 4.1, where Dz is the hydrodynamic
diameter (the average-intensity weight average diameter), presenting three different
measurements results, followed by the Dzav., which is the average of the three previous
measurements; Z is the zeta-potential of the polymer solution, also presenting three values
measured and the Zav., that is the average zeta-potential and the last column is the average
molecular weight of the polymer, MWav.
43
All procedures in the measurements were conducted with attention to avoid errors due to
measurements procedures, namely contamination and manipulation of samples and testing
cells. The solutions were all kept covered with parafilm, and filling of the cell was done by
using syringes and a 0.45µm membrane filter, assuring always that no air bubbles were present
in the measurement cells.
Table 4-1: Polymer Characterization results.
Polymer Dz Dz av. Z Z av. MWav.

(nm) (nm) (mV) (mV) (kDa)
80,7 51,3
80,3 52,6
Amber 2001HMW 80,6 56,6 235,0
79,7 49,9
75,1 49,3
74,7 51,2
Amber 2002 74,1 48,2 158,0
75,0 56,1
47,6 39,9
Amber 2006 48,5 50,3 37,4 36,4 175,0
54,7 31,8
37,3 49,1
Amber 2008 38,7 36,8 42,1 48,0 138,0
34,4 52,7
36,3 73,2
Amber 2008HMW 35,4 35,8 68,1 72,0 120,0
35,8 74,7
37,8 48,2
Amber 2009 38,4 38,1 48,6 48,0 66,2
38,0 47,1
46,1 20,8
Amber 2013 45,1 45,3 20,6 20,1 160,0
44,7 18,9
38,4 73,7
72,3
Amber 2014 37,9 37,9 73,0 53,1
37,3 70,1
44,4 40,5
Amber MetQuat 6 44,8 44,4 45,8 52,6 152,0
44,1 42,1
In the appendix A, the graphs for size and zeta potential distributions and the molecular
weight plots are presented.
For what can be observed in the results, the polyamines have the larger diameters (Amber
2001HMW, Amber 2006 and Amber 2002), and relatively higher molecular weight; all those
results are obtained using a suitable concentration of the polymers in the solution, which was
the most important, and also difficult, part of the characterization, finding out a proper range of
44
concentrations to avoid polymer aggregation and to allow the equipment to perform good
measurements.
For size and zeta potential the concentrations were of 0.1%, 0.05% and 0.02% w/w, being
applied two minutes of ultrasound to samples before preceding to measurements. For the
molecular weight, the range of concentrations used were from 1% to 0.02% w/w, but for most
cases 1% was too high. In the end the most appropriated concentrations used in the trials were,
for some cases beginning at 0.5%, and for all polymers the values of 0.2%, 0.1%, 0.05% and
0.02% were used. In some trials the 1% concentration was cutted off, and a new point added at
0.08%. For molecular weight measurements trials ultrasounds were just applied to the higher
viscosity polymers, Amber 2001HMW, Amber 2006, and also Amber 2008HMW, the others
the measurement achieved good results without needing ultrasound. The values for polymers
Amber 2013 and 2014 were provided by Patricia Costa [21].
All the results met quality criteria in the Zetasizer software. As expected all the polymers
presented a positive zeta-potential.
4.2 Standard decolorization tests results
The decolorization achieved in the laboratory decolorization trials by each polymer is here
presented. These results are in offline laboratory flocculation tests in carbonated sugar liquor,
provided by RAR Portugal, and proceeding along to the ICUMSA 420 method for color
measurements of sugar solutions.
The decolorization was evaluated by comparing the color of the liquor submitted to the
flocculation tests using the polymers, to the color presented in a carbonated liquor from the
same batch of carbonatation, maintained at the same temperature for the same period of time,
but without polymer addition, indicated in the results as blank. The results are presented taking
into account the pH for the filtrated liquors, decolorized and blank, as required by the ICUMSA
420 method, this variable should be in the range of 7.00 ± 0,02, due to possible large changes
in the color when pH of the solutions undergo significant variations. Decimal changes in the
pH could imply changes of dozens ICUMSA units in the color. It should also be mentioned
here that every trial for the same polymer was performed in a liquor from the same batch of
carbonatation, to make a proper comparison between only carbonated liquor color and
carbonated liquor using color precipitants agents.
45
An example of the plots of the ICUMSA color index for both, decolorized and blank liquor,
is presented next (figure 4.1).
AMBER 2006
color index( ICUMSA color index on brix) BLANK
900
800
700
COLOR INDEX (IU)
600
500
400
300
200
100
0
0 50 100 150 200 250 300
POLYMER PPM'S ON BRIX
Figure 4.1: Graphic representation of sugar liquor color (treated with polymer and blank) versus polymer dosage for tests
with polymer Amber 2006.
As shown in the plot above, the color index of the filtrated decolorized liquor has
decreased to a certain amount as the dosage of the polymer increased, but at one point it goes
up again, and this behavior has been presented in all trials results. Thus, we can conclude that
the dosage of the polymer has a maximum, after which the color bodies return to the liquor, and
in some cases it may even increase the color of the liquor [3], when flocculation stops being
effective and the polymer can even contribute to redisperse the color agents by stabilizing them
and contributing to keep them disperse in the liquor.
The following table presents all the decolorization performed by the polymer Amber 2006
(Table 4.2) showing that the pH of the liquor was always within the expected range. All
measurements of color were performed complying to the ICUMSA 420 method, at temperature
of 20°C and pH 7.00 ± 0.02, and filtrated through a 0.45µm membrane before absorbance
measurements at 420 nm. The original pH of the carbonated liquor was in the range of 8.1 to
8.3, which is as expected at the second stage of carbonatation processes, being all the polymers
active at this pH.
46
Table 4-2: Decolorization performed by the polyamine Amber 2006.
pH (carbonated pH Blank ppm's pH Color (treated liquor ) %

liquor) blank. color (poly (decolorize (IU) Color
(IU) mer d) Removed
on
Brix)
8,24 6,99 769 25 7,02 693 9,9
8,17 6,99 778 50 7,00 641 17,6
8,18 6,99 752 100 6,99 549 27,0

8,31 6,99 818 150 7,00 567 30,7
8,24 6,98 801 200 7,01 683 14,7
8,31 6,98 780 250 6,98 709 9,1
The complete tables of the results obtained in the decolorization tests will be presented in
the Appendix B of this dissertation. Therefore the results of the decolorization trials performed
are summarized in the next table (Table 4.3), presenting only the final decolorization percentage
achieved for each polymer and for the different dosages. Another polyelectrolyte used was the
Amber 2012, but its decolorization performance was not significant, so the results are not
presented here but are also shown in the Appendix B.
Table 4-3: Percentage of Decolorization per polymer and ppm dosage of polymer on Brix of the sugar liquor.
25 ppm 50 ppm 100 ppm 150 ppm 200 ppm 250 ppm
Polymer
(%) (%) (%) (%) (%) (%)
Amber 2001HMW 15,3 25,5 36,4 28,7 35,4 27,7
Amber 2002 11,6 18,4 25,6 14,0 1,6 -6,2
Amber 2006 9,9 17,6 27,0 30,7 14,7 9,1
Amber 2008 15,7 20,1 31,7 23,9 14,7 11,9
Amber 2008HMW 14,4 20,5 12,9 8,2 -2,3 -10,5
Amber 2009 12,1 28,4 41,0 14,4 6,1 -3,3
Amber 2010 3,6 19,8 21,7 -8,6 -7,9 -12,6
Amber 2013 20,5 22,9 35,2 13,9 12,4 -6,7
Amber 2014 9,2 23,8 13,7 7,9 -11,2 -15,8
Amber MetQuat 6 18,6 20,5 24,1 34,2 38,9 30,1
47
The complete results, including temperature and pH of flocculation, filtrated liquor pH at
which color was measured and the respective ICUMSA color index are presented in the
appendix B. the next picture (Figure 4.2) shows the filtrated liquor colors after carbonatation
and after carbonatation using polymer Amber 2013 aid as color precipitant.
Figure 4.2: the filtrated liquor blank (on the left) and treated with the polymer Amber 2013 100ppm (on the right).
All polymers tested present good decolorization performance, having its optimal in the
range of 50 up to 100 ppm, decolorization decreasing for dosages much higher than 100 ppm
for most of the polymers, except for Amber 2001HMW and the novel polymer Amber MetQuat
6, that seemed to present good decolorization activity for higher dosages.
Next, the graphic representation presented (Figure 4.2) summarizes the final results
achieved in the decolorization tests, presenting, for each dosage, the best performing polymer.
Four of these polyelectrolytes have shown very good results in the tests, achieving, in the case
of Amber 2009 at 100 ppm dosage, up to 41% decolorization when compared with the color of
the filtrated liquor untreated with the polymer.
48
%color removal
AMBER 2009; METAQUAT 6;

50
41,04 METAQUAT 6; 38,87
METAQUAT 6;
% of color removed
40 AMBER 2009; 34,16

28,36 30,14
AMBER 2013
30
; 20,32
20
10
0
25 50 100 150 200 250
ppm's polymer on brix of liquor
%color removal
Figure 4.3: Graphic representation of the by the best performing polymer at each dosage and respective percentage of
decolorization achieved.
At the dosages of 50 and 100 ppm the polymers Amber 2009, Amber 2013 and Amber
2001 HMW presented the highest percentages of decolorization. Thus, a few tests using a blend
of these polymers were performed, to evaluate if in combination of two or three of them, they
would present higher global decolorization.
Amber 2001HMW presents a high molecular weight and a large hydrodynamic diameter,
but low polymer charge when comparing the MW and size of the other polymers, nevertheless,
this polymer showed a good performance as color removal and flocculant. The polymers Amber
2013 showing a low charge and relatively high molecular weight showed very good
decolorization performance at low dosages, 25-100 ppm, also from the flocculation monitoring
presented a good result forming not much of large flocs, but resistant to breakage what can be
understood as reason of its higher color removal performance. Amber 2009 is the polymer
presenting lower molecular weight but shows an intermediate high charge, and low
hydrodynamic diameter, showed very good decolorization action when used in dosages of 50
and 100 ppm, forming also small flocs but with good stability, from what can be seen from the
continuous monitoring study. As for the novel polymer Amber MetQuat 6, this polymer
49
presented intermediate characterization values, for molecular weight, size and charge, having
good decolorization performance for a large range of polymer dosages.
The next table presents the decolorization achieved by using the blend of Amber
2001HMW, a polyamine, with Amber 2013, a poly (DADMAC), at three different dosages, 25,
50 and 100ppm, of each polymer.
Table 4-4: Blend Amber 2013 & Amber 2001HMW decolorization performance.
ppm's color index Blank %

(polymer concentration on Brix) on brix (IU) Color (IU) Color Removed
25+25 609 772 21,1
50+50 542 783 30,8
100+100 607 794 23,6
Comparing the values for using the polymers alone, Amber 2001HMW and Amber 2013,
presented in table 4.3, and the ones obtained with the blend of the two polymers, no significant
increase in decolorization is noticed. Moreover, when thinking about economic aspects is
should not be suitable to use a blend in this case, but yet further studies should be performed,
to understand if the polymers interfere with each other or if this output is due to the polymers
targeting the same color compounds and thus competing with each other. Refining the
polymer’s dosage, should also be considered, especially studying the range below 50 ppm.
Table 4-5: Blend Amber 2009 & Amber 2001HMW decolorization performance.

(polymer concentration on Brix) on brix (IU) color (IU) Color Removed
25+25 614 772 20,5
50+50 562 790 28,9
100+100 598 821 27,2
In table 4.5, the decolorization results reached by using the blend of Amber 2001HMW
and Amber 2009 are shown. For the low dosage of 25 ppm, of each polymer, the mixture shows
better results than using the polymers alone, comparing to table 4.3 results, but for higher
dosage this does not happen. The fact that a low dosage leads to a color removal around 20%
may be economically interesting.
Next, the result of testing a blend of the three polymers are presented in the following
table (Table 4.6):
50
Table 4-6: Blend using Amber 2013, 2009 & Amber 2001HMW decolorization performance.

(polymer concentration on Brix) on brix (IU) Color (IU) Color Removed
25+25+25 642 887 27,6
50+50+50 568 847 32,9
100+100+100 620 834 25,7
Again using a 25 ppm dosage, of each polymer, presents higher decolorization with the
mixture than with the polymers alone. Also the 50 ppm dosage, of each one, shows higher
decolorization than when using of the polymers alone.
The objective of testing these blends was to see if they improved the color removal,
considering possible that the different polymers will present affinity to different color
compounds in the solution, increasing, this way, the decolorization of the liquor in global.
The range around 25 ppm dosage should be better studied along with the economic aspects, to
analyze the convenience or not of this methodology. Moreover, for this dosage, the
decolorization achieved comes close to the best results achieved with the single polymers for a
dosage of 100 ppm.
4.3 Determination of Index value
As said in the previous chapter, the Index Value is an important parameter to evaluate the
types of colorants present in the liquor, related to the molecular mass of such color bodies. As
an attempt to observe, using this parameter, if the color precipitants are removing more of the
high or the lower molecular mass color components of the sugar liquor, a few trials were
performed with the four best decolorization performing polymers. An example of the variation
of the IV with polymer dosage is presented in the figure 4.3, in this case for the polymer Amber
2001HMW.
51
IV & %decolorization versus dosage

amber 2001HMW
Título do Eixo
0 50 100 150 200 250 300
3,5 35,00
% DEcolorization (IU)
3 30,00
2,5 25,00
2 20,00
IV
1,5 15,00
1 10,00
0,5 5,00
0 0,00
0 50 100 150 200 250 300
Polymer Dosage (ppm)
IV % decolorization
Figure 4.4: Representation of Index Value and decolorization for filtrated liquor by decolorization with Amber2001 HMW.
The IV for the filtrated sugar liquor after decolorization by the polymers is presented in
the next table (Table 4.7). The index value of the carbonated liquor, before decolorization
treatment with the polymers, was also measured being that the IV for the liquor used on the
Amber 2001HMW tests was of 3,03, while the IV for the liquor used on trails with the other
polymers was 2,89. The difference of those two IV is related to the origin of the liquor, which
came from different carbonatation batches, as well as different raw sugar origins, as they were
collected in different periods.
Table 4-7: Index value (IV) for filtrated sugar liquor by ppm of polymer used.
Dosage (ppm) Amber 2001HMW Amber 2008 Amber 2009 MetQuat 6

Blank 3,03 2,89 2,89 2,89
25 3,06 2,99 3,19 3,10
50 2,99 3,01 3,14 3,04
100 2,82 3,05 3,28 3,26
150 2,43 3,14 2,93 3,36
200 2,47 2,94 2,71 3,12
250 2,73 2,78 2,90 2,95
The index value should be evaluated along with the decolorization performed by the
polymers, and so, from the previous section, the decolorization was most effective till 150 ppm
52
dosage for all polymers, except for Amber 2001HMW and MetQuat 6, that still showed good
decolorization at higher dosage than 150 ppm (table4.3).
Looking at the results in table 4.7, the polymer Amber 2001HMW presented a reduction
on the index value, along with the increase of the dosage, the same happening to the
decolorization percentage up to 100 ppm (see Table 4.3). The opposite occurred for the
polymers Amber 2008, Amber 2009 and the novel polymer MetQuat 6, for which no reduction
on the IV was observed, instead of it, the IV increased, but for high dosages it decreased again,
reaching values around that presented for the untreated liquor.
As mentioned by Bento [6, 7], high molecular weight color particles exhibit low Index
Value, while low molecular weight color bodies show IV higher than five. It can be then
understood that, if the index value of the overall solution reduces, higher IV colored compounds
are being removed while if it grows lower values IV compounds are removed. So, here there is
a possibility that the polymer Amber 2001HMW is removing low molecular weight colored
compounds, while the others may remove better the higher molecular weight color bodies in
the solution and thus the increase in the IV. In the case of Amber 2001HMW this agrees with
the maximum decolorization within the range of 100-150 ppm.
This study alone is not enough to state surely if the polymers are removing better low or
high molecular weight color compounds, thus a more profound study is required, using other
analytical techniques, such as chromatographic and other techniques used by Bento [2, 6, 7].
4.4 Flocculation monitoring using LDS
In this section will be presented the results obtained by LDS (Laser Diffraction
Spectroscopy) online monitoring of the flocculation process using some of the best performing
polymers tested in the decolorization laboratorial trials. The results are presented as the kinetics
plots of the flocculation process, showing the variation of the particles size indicator, d10, d50
and d90 for the suspension. The first one implying information about the evolution of the
smaller particles, while the last one is related to the evolution of the large particles resulting
from the flocculation process.
53
As mentioned earlier, in the previous chapter, those trials were performed using
carbonated liquor, the same used in the decolorization trials, but here it had to be diluted in 1:5
in distillated water, in order to reduce the obscuration of the laser light in the Mastersizer 2000,
allowing then the measurement of particle size and the size distribution in the colloidal
suspension.
The next table (Table 4.8) presents the list of the polymers used in those trials, the dosage
in ppm based on the degree of Brix of the sugar solution, the time for the monitoring and the
decolorization percentage obtained at the end of the flocculation, being the decolorization based
also on the color measured following the ICUMSA 420 analytical method [4].
Table 4-8: List of polymers used in LDS tests, its dosage, time of flocculation and decolorization percentage at the end of trials.
Polymer Dosage ( ppm on Time Decolorization

brix) (min.) %
Amber 2013 100 30 24,9
Amber 2001HMW 100 15 20,2
Amber 2001HMW 100 30 18,4
Amber 2001HMW 200 30 16,6
Amber MetQuat6 100 30 16,8
Amber 2008 100 30 17,2
Amber 2009 100 30 24,3
Blend Amber 2001HMW+2009 100+100 30 30,2
Blend Amber2013 + 2001HMW 110+100 30 30,9
Blend Amber2013 +2009+ 2001HMW 50+50+50 30 24,7
Blend Amber2013 +2009+ 2001HMW 100+100+100 30 29,2
For the Amber 2001HMW at 100 ppm dosage, the color of the filtrated liquor was
measured after fifteen and thirty minutes of the polymer addition to the liquor, to evaluate if
significant difference is shown in the color of filtrated liquor, following the ICUMSA 420
method.
The kinetic curve for the flocculation in the diluted carbonated sugar liquor, using the
different polymers and for 30 minutes of monitoring will be shown next. The kinetic curves are
presented for the three size indicators, d10, d50 and d90.
54
a) b)
c)
Figure 4.5: Flocculation kinetics for carbonated sugar liquor using Amber 2013 at 100ppm dosage: a) d10, b) d50 and c) d90.
The bar line corresponds to the addition of polymer.
The red lines in the graphics represent the point where the polymer was added to the
system, after what it can be seen that the size has a step increase, short time after the polymer
addition. Even if the polymer was added careful and slowly to the system, it induces
immediately formation of flocs, which are bigger at beginning. As time goes the size reduces,
reaching a stabilization value, this happens for all the polymers tested, and can be understood
as result of the hydrodynamics of the system. As time passes, flocs break and reconfirm,
forming smaller ones. The time and speed of agitation are the parameters which have a
significant influence in the flocculation process. At the end of 30 minutes the size parameters
d10, d50 and d90, for all the polymers tested, show always larger particles than before the
addition of the polymer, confirming that flocculation occurred, even if this increase is more
pronounced for some of the polymers (Amber 2013, MetQuat 6 and Amber 2009).
The figure 4.5 presented before refers to the polymer Amber 2013, at 100 ppm on brix
dosage, and for what can be seen after 30 minutes of test monitoring the final size shown in the
parameters d10, d50 and d90 is bigger than in the begin, and after this time color measured
55
shows a good decolorization in the filtrated liquor (Table 4.8). This polymer presents an
relatively high molecular weight and low charge(table 4.1), and from the kinetic curve seems
like the flocs formed are not so much bigger but compact enough to be noticed even after the
30 minutes of monitoring.
Tests with higher dosage, 200 and 300 ppm, for this polymer were performed but as the
results does not show significant differences comparing to the 100 ppm dosage, and the color
removal is lower, the graphics will be presented in the appendix C.
Following will be presented the graphics for Amber 2001HMW (Figure 4.6), a high
molecular weight polyamine, for a dosage of 100 ppm on brix of the sugar solution. Test was
conducted at the same conditions and specifications as the previous trial with Amber 2013.
a) b)
c)
Figure 4.6: Flocculation kinetics for carbonated sugar liquor using Amber 2001HMW, at 100ppm dosage: a) d10, b) d50 and
c) d90. The bar line corresponds to the addition of polymer.
Comparing to the previous polymer, Amber 2013, this polyelectrolyte presents larger
hydrodynamic diameter, as well as higher molecular weight and zeta potential (see table 4.1
56
previous section), and for what can be seen in the plots the size of flocs formed initially after
polymer addition are larger as expected. The larger initial flocs might be related to possible
bridging flocculation with calcium carbonated bonded to color bodies, which later undergoes
breakage e reorganization. Thus, in the end, since breakage may be more pronounced for this
polymer due to its higher molecular weight, the final flocs size decreases a lot, the final values
being similar or smaller than the ones obtained with Amber 2013, also decolorization is higher
with Amber 2013.
A trial for Amber 2001HMW at 200 ppm dosage was also performed, and the resulting
kinetic curve is plotted next, as decolorization is yet good at this dosage, this test aimed to
evaluate if the final size was also affected (Figure 4.7).
a) b)
c)
Figure 4.7: Flocculation kinetics for carbonated sugar liquor using Amber 2001HMW, at 200ppm dosage: a) d10, b) d50 and
Those results are similar to the ones obtained for the100 ppm dosage of the same polymer,
even if the color removal is slightly lower. The flocs at the beginning of flocculation are slightly
larger with this dosage, but after the thirty minutes of flocculation the sizes are similar (probably
breakage was more pronounced and thus the slightly lower color removal).
57
For the polymer Amber 2001HMW more tests were performed, using dosages of 300,
400 500 and 600 ppm, and longer time of flocculation
Above 300 ppm color removal was ineffective, and flocculation kinetic curves did not show
appreciable differences (see appendix C).
Next the result for the flocculation kinetics using the novel polymer, Amber MetQuat6,
is presented (Figure 4.8). The flocculation was proceeded in the same conditions as for the
previous polymers.
b)
a)
c)
Figure 4.8: Flocculation kinetics for carbonated sugar liquor using Amber MetQuat6 at 100 ppm dosage: a) d10, b) d50 and
The polymer Amber MetQuat 6 presents significant shifting in the particle size, for all
parameter monitored, that is indicative of higher flocculation, especially evident when looking
at the kinetics of d90 evolution, as well as appreciable decolorization for the time of the test.
Apparently the initial step increase is not as large as for Amber 2001HMW, but it stabilizes for
58
a much higher value of d90. In the offline trials for decolorization this polymer showed good
decolorization activity also at different dosages higher than 100 ppm.
Similar to Amber 2013 this polymer has a lower molecular weight, and a higher charge
than Amber 2013, and thus the flocs obtained must be more compact and not so exposed to
breakage.
In the following graphic, (Figure 4.9), the kinetic curve for the poly (DADMAC) Amber
2008 is presented for 100 ppm dosage and the same flocculation conditions as for the previous
polymers.
a) b)
c)
Figure 4.9 Flocculation kinetics for carbonated sugar liquor using Amber 2008 at 100 ppm dosage: a) d10, b) d50 and c) d90.
The bar line corresponds to the addition of polymer.
This polymer presented good decolorization activity in the offline decolorization

laboratory tests for the time of fifteen minutes, but for thirty minutes it is not as good result as
the previous one, the color removal activity decreasing with time of flocculation. In fact the
size values presented in the kinetic curves after 30 minutes of flocculation are not so
59
significantly different from the initial ones, which agrees with the low decolorization achieved
for 30 minutes. The behavior of this polymer and the polymers characterization is not clear.
In the next graphic set(Figure 4.10), the curves for Amber 2009, a poly(DADMAC), are
shown. This is one of the best polymers tested for the dosage of 100 ppm, according to the
results showed before, table 4.3 of this chapter and table 4.8. The kinetic flocculation curves
shown in Figure 4.9 present an atypical behavior, comparing to the other polymers used,
showing that for d10 and d50 after twenty minutes of the test those values are smaller than in
the beginning of the test, where no polymer was added. This may result from some
inhomogeneity of the initial liquor, comprising sucrose, color compounds and calcium
carbonate crystals, which were later redispersed. Still flocculation obvious when looking at the
d90 evolution, which agrees with higher decolorization obtained with this polymer.
Even if flocculation may be difficult with this polymer due to its lower molecular weight,
considering the polymer charge the flocs obtained must be compact and resistant to breakage,
which agrees with the higher decolorization percentage.
a) b)
c)
Figure 4.10: Flocculation kinetics for carbonated sugar liquor using Amber 2009 at 100 ppm dosage: a) d10, b) d50 and c)
d90. The bar line corresponds to the addition of polymer.
60
After presenting the individual polymers kinetic tests, the next step is to present the
graphic results when blends of two and three of the previous polymers were used, to compare
the kinetic curves and size indicators obtained.
The first combination was a blend of 100 ppm of Amber 2001HMW polyamine with 100
ppm of Amber 2013, poly (DADMAC). From the IV tests, it is seems that those polymers may
target different color bodies, what supposedly will increase decolorization, still it does not
imply a decolorization performance equal to the sum of that achieved by each polymer
individually. The kinetic curves are shown if figure 4.11:
a) b)
c)
.
Figure 4.11: Flocculation kinetics for carbonated sugar liquor using the blending Amber 2001 HMW & Amber 2013 at 100
ppm dosage of each polymer: a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
Comparing to the graphics presented before for the individual polymers at 100 ppm
(Figures 4.5 and 4.6), the values of D50 and D90, obtained with blend are between those for
Amber 2013 and Amber 2001HMW, what is here interpreted as that both polymers are active
as color precipitation agents. The final color removed is in the order of 30%, which is a good
61
value, higher than the polymers alone, but also not equal to the sum of the decolorization
achieved individually. The fact that, in total, the sum of the concentrations of the two polymers
has increased twice the value used previously with the individual polymers, may have enhanced
flocculation.
A similar test was then performed for another blend, using this time another poly
(DADMAC), Amber 2009, and again Amber 2001HMW, both at a 100 ppm dosage referred to
the content of sucrose in the solution. The results are shown in the next graphic, figure 4.12:
a) b)
c)
Figure 4.12: Flocculation kinetics for carbonated sugar liquor using Amber 2009 and Amber 2001HMW, both at 100ppm: a)
d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
Here it can be observed again that the size parameters are similar to that for the blend
with Amber 2013. However, it seems that flocculation is not yet fully stabilized, aggregation
still going on after 20 minutes. Also the values of d10, d50 and d90 at the begin of flocculation
are in the range between those for the flocculation using the two polymers individually. Finally
decolorization percentage is also in the order of 30%, higher than when suing the polymers
62
individually, suggesting that the two polymers are active as color precipitants in the flocculation
process. The combination of a low and a higher molecular weight polymer may beeffective in
color removing.
Finally, the blending of the three polymers, Amber 2001HMW, Amber 2013 and Amber
2009 was tested, this time for lower dosages, 50 and 100 ppm based on the brix degree of the
filtrated solutions.
First the kinetic curve for using 50 ppm of each polymer in the blend is presented in the
figure 4.13, the test having been carried out in the same conditions as in all the others before.
a) b)
c)
Figure 4.13: Flocculation kinetics for carbonated sugar liquor using Amber 2009, Amber 2013 and Amber 2001HMW, at 50
ppm each one, a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
Looking at the graphic representations, what first calls the attention is that the size
parameters d10, d50 and d90 values all go up to a sharp pic, more like a pulse, decreasing more
quickly to what seems a more stable size. The d90 shows a much higher pulse after addition of
polymer than in the previous tests, but decreases in the next point to a value in the range of the
63
trials with two polymers blends, below 20µm. The pulse intensity is related to the addition of
the three polymers blend but the following points are between the values for Amber 2001HMW
and the two poly(DADMAC), confirming the activity of the polymers used. The final
decolorization was in the order of 24% after 30 minutes of the test, lower than when using the
two polymers blends.
The nest graphics, figure 4.14, presents the resulting kinetic curve for the flocculation
using the blend of the same three polymers, Amber 2001HMW, 2013 and 2009, again but this
time using higher dosage, 100ppm of each polymer.
a) b)
c)
Figure 4.14: Flocculation kinetics for carbonated sugar liquor using Amber 2009, Amber 2013 and Amber 2001HMW, at
dosage of 100ppm each polymer, a) d10, b) d50 and c) d90. The bar line corresponds to the addition of polymer.
The kinetic curve presented above are similar to that presented in the figure 4.12, but the shape
is a little different, showing slower decreasing in the size values. Also in the d90 curve the
larger pic presented before at around 40µm does not appear, it is here more smooth, decreasing
more slowly to a stable value which maybe attribute to an overall higher concentration of the
polymers blend. Values for the final diameters are slightly higher than that of the poly
64
(DADMAC)’s alone, but below the values presented for Amber 2001HMW individually. The
d90 value is slightly higher in this case than when a lower concentration was used, suggesting
that 50 ppm of each polymer maybe too low concentration for effective flocculation. In fact,
the decolorization achieved here, after 30 minutes was in the order of 29%, higher than when
using only 50 ppm of each polymer.
Overall analyzing the results of the polymers blends it may be interesting to use blends
of two complementary polymers, while the three polymers blends do not appear very effective.
Also, in general, it seems possible to conclude, from the results in this section, that for
decolorization to be effective it is not necessary to produce very large flocs. The results shown
that decolorization is more effective with not too large, but more compact flocs not so much
affected by the breakage.
65
66
Chapter 5. Conclusions and future works
5.1 Conclusions
The present study had as main objective to evaluate the performance of cationic polymers
as color removal agents in the sugar refining process, and to try establish correlations between
performance and the polymers characteristics, namely, molecular weight, zeta potential and
hydrodynamics radius.
After a tentative to understand and correlate the results, regarding polymers characteristics,
their performance as color removal and also the continuously monitoring of the flocculation
process induced by those polymers in sugar carbonated liquor, the following conclusions were
reached:
 Regarding the color removal, all the polymers showed activity, being some of them able
to perform good color removal, up to 41% over carbonated liquor color, being the polyamine
Amber 2001HMW, the poly(DADMAC)’s Amber 2013 and Amber 2009 and the novel
polymer Amber MetQuat 6 the the ones showing the better decolorization performance;
 The dosage most effectives were between 25 and 100 ppm on brix of solution, used in
the second stage of carbonatation process conditions, temperature around 80ºC and pH 8-
8,5 and after 15 minutes of flocculation it was possible to reach higher color removal in the
tests performed;
 For what could be observed about the changings in the Index Value of the solutions
after decolorization using the polymers, the Polyamine showed a reduction in the value of
IV, while the poly(DADMAC) increased the IV, as well as the MetQuat 6, based on this it
is possible infer that the Amber 2001HMW removes more high IV colorants from the liquor,
while the DADMAC’s and MetQuat remove lower IV colorants;
 After testing some blends of polymers to perform decolorization, in dosages of 50 and
100 ppm of each polymer. It showed higher decolorization than using one polymer alone,
50 ppm of Amber 2001HMW with 50 ppm Amber 2009 or Amber 2013 showed good color
removal action in the liquor. By the IV tests performed for the polymers the
poly(DADMAC) and polyamines seem to have different affinities with the different types
of colorants;
67
Chapter 5. Conclusions and future work
 Relating the polymer’s characteristics and color removal performance, it showed that
best performing polymers are characterized by having a high molecular weight with a
moderate polymer charge, and polymers having a medium molecular weight that present
higher polymer charge;
 The online monitoring of the flocculation process using LDS, with a stirring speed of
1200 rpm, showed evidence of flocculation activity of the polymers. Observing the kinetics
of the flocculation, after 10 minutes of flocculation the flocs breakage and reflocculation
are predominant, and after 15 minutes flocs breakage went on increasing and, with this,
color molecules seam to return to the liquid, as the liquor showed higher ICUMSA color
index. This strategy allowed to perform a better understanding of the polymer action,
enlightening some parameters to be improved in the process;
 The total color removal promoted by the addition of the polymer was in the range of
30-41% in the carbonated sugar liquor, with dosage of polymers between 50 and 100 ppm
on brix of liquor;
 Some major factors affecting the color removal in the carbonated liquor were identified
as the pH, time of reaction of the flocculants with the solution, besides stirring speed,
polymer molecular weight and charge;
 Poly(DADMAC) seems to perform good decolorization for all the polymer
tested(different molecular weight polymers).
5.2 Future works
After the present study and observations along it, some points should be further addressed,
which can have a deeper impact in the technology in the future, to know:
 Perform deeper study with high molecular weight polyamines having high-moderated
charge for the polymer, and medium molecular weight poly(DADMAC) that present
medium to high polymer charges;
 The dosage of those polymers should be tested and screened from 25 to 100 ppms on
brix dosage, and blending could be tested in range of 10-50 ppm dosage of each polymer;
 From the IV information obtained in this study maybe the blending of polyamines and
poly(DADMAC) could be an interesting solution, and promote higher decolorization but,
yet, economic impact should be evaluated;
68
 Testing slower stirring speeds and in the time interval of 10 to 15 minutes in the
flocculation monitoring process, to see if any improvement in decolorization can be
achieved;
 An economic study to screen the polymers used, their color removal activity, and the
impact on the saving in secondary decolorization processes, as well as to estipulate the most
appropriated dosage relating decolorization activity and economical cost should be
performed;
 The LDS technic can be used to enlighten some factors of the process, so further study
on the flocculation kinetics should be performed in the carbonated liquor, to try to
understand better the flocculation mechanisms involved in relation with the polymers
characteristics. For that a better control of the system refining temperature, agitation and
dilution will be important, or better yet, to find a way to perform simultaneously a
carbonatation flocculation monitoring, to understand the influence of the carbonatation
along the flocculation promoted by the polymers, where agitation would be promoted by
gas bubbling in the system, what can reduce flocs breakage, even if some other
monitoring technique rather than LDS would have to be used;
 Understand better the relation between the characteristics of the flocs obtained and the
performance influence in the subsequent operations in the refining process.
69
Chapter 5. Conclusions and future work
70
References
[1]. Galloway, J.H., “The Sugar Cane Industry. An Historical Geography from its Origins to
1914”. Cambridge University Press, New York, 2005.
[2]. Bento, L.S.M., Colorants through cane sugar production and refining part 2, Sugar Industry,
134, nº3, pp.168-176, 2009.
[3]. Davis, S.B., “The Chemistry of Colour Removal: A Processing Perspective. Proc. S. Afr.
Sugar Technol. Assoc, 75, pp. 328–336, 2001.
[4]. Rein, P. W. “Cane Sugar Engineering”, Verlag Dr. Albert Bartens KG, Berlin, 2007.
[5]. Bento, L.S.M., Colorants through cane sugar production and refining, Sugar Industry, 134,
nº4, pp. 230-236, 2009.
[6]. Elvin, J.R- Decolorization and Clarification of sugar solutions using Color precipitants and
Flocculants, proceedings of the workshop on separation process in sugar industry, pp 159-
181, New Orleans, 1996.
[7].Bento, L.S.M.; Pereira, M.E.; Sá, S., Study of sugar colorants through gel permeation
chromatography using UV and evaporative light scattering detectors. Proceedings SIT, pp.
383–392, Montreal 1997.
[8]. Moodley, M.. The application of cationic flocculants as Decolourising agents in the sugar
industry, Proc. S. Afr. Sug. Technol. Ass. 67, pp. 155-169, 1993.
[9]. Bennett, M.C., Gardener, F.J., Abram, J.C. and Rundell, J.T., The Talofloc decolorization
process, Proc int. Soc. Sug. Cane Technol., 14, pp. 1569-1588, 1971.
[10]. Bennett, M. C., London, patent US3698951, Oct., 1972.
[11]. Doherty, W.O.S., Fellows, C. M., Gorjian, S., Senogles, E., Cheung, W. H., Flocculation
and sedimentation of cane sugar juice particles with cationic homo- and copolymers,
Journal of Applied Polymer Science, vol. 90, pp 316-325, John Wiley & Sons, Sydney Au,
2003.
[12]. Tripathy, T. and De, B. R., flocculation: a new way to treat the waste water, J. of Physical
Sciences, vol. 10 pp 93-127, 2006.
[13]. Rasteiro, M. G., Pinheiro, I., Garcia, F. A. P., Ferreira, P. and Hunckeler, D. “using light
scattering to screen polyelectrolytes performance in flocculation”, Polymers journal 3, pp
915-927, 2011.
[14]. Salim S., Bosma R., Vermuë M.H., Wijffels R.H., Harvesting of microalgae by bio-
flocculation, J Appl Phycol.23(5), pp. 849–855. 2011.
71
[15]. Chen, J. C. P. and Chou, C. C., “Cane Sugar Handbook”, 12th ed. Wiley Interscience,
New York, 1993.
[16]. Runggas, F. M., “Applications of flocculants in the carbonatation refining process”, Proc.
South African Sugar Techl. Association, South Africa,1978.
[17]. Chou, C. C., “Pros and Cons” of various decolorization processes for production of
refined sugar, S.I.T. Sixty Ninth Annual Technical Conference Savannah, Georgia, USA,
2010.
[18]. Malvern Instruments, “Zetasizer Nano Series User Manual." http://
www.biophysics.bioc.cam.ac.uk/_les/Zetasizer Nano user manual Man0317-1.1.pdf
[Accessed 26th September 2016], 2004.
[19]. Goddard, S., Van Zyl M., new Polymer Chemistry for Refinery Liquor Decolourization
Proc S Africa Sug TEchnol. Ass, 86: 457-463, Natal, South Africa, 2013.
[20]. Costa, P.D. C., “chemical Stabilization of soils using carbon nanotubes”, master
dissertation FCTUC, university of Coimbra, 2016.
[21]. Xu, R., Particle Characterization: Light Scattering Methods, kluwer academic publishers,
N.Y. USA, 2002.
[22].Favero, D. M., “Clarification of sugar juice thought carbonatation process”, master thesis
in chemical engineering, University of Paraná, Brazil, 2011.
[23]. Rundell, J.T., Pottage, P.R., patent US3926662, Original London, 1975.
72
Appendix
73
74
Appendix A. Polymer Characterization
A.1 Size distribution by intensity for the polyelectrolytes
Figure A.1: Size distribution by intensity for polymer Amber 2001 HMW at 0,02 and 0,05% w/V
Figure A.2: Size distribution by intensity for polymer Amber 2002 at 0,05% w/V
75
Figure A.3: Size distribution by intensity for polymer Amber 2006 at 0,02 and 0,05% w/V
Figure A.4: Size distribution by intensity for polymer Amber 2008 at 0,05% w/V.
76
Figure A.5: Size distribution by intensity for polymer Amber 2008HMW at 0,05% w/V.
Figure A.6: Size distribution by intensity for polymer Amber MetQuat 6 at 0,02% w/V.
77
Figure A.7: Size distribution by intensity for polymer Amber 2009 at 0,05% w/V
78
A.2 Zeta potential graphics
Figure A.8: Zeta potential graphics for Amber 2001HMW.
Figure A.9: Zeta potential graphics for Amber 2002.
79
80
Figure A.12: Zeta potential graphics for Amber 2008HMW
Figure A.13: Zeta potential graphics for Amber MetQuat 6.
81
A.3 Molecular Weight Debye plots for the polymers
Figure A.14: Debye plot for Amber 2001HMW 1.
Figure A.15: Debye plot for Amber 2001HMW 2.
82
Figure A.16: Debye plot for Amber 2002.1.
83
84
85
Figure A.22: Debye plot for Amber MetQuat 6 1.
Figure A.23: Debye plot for Amber MetQuat 6 2.
86
Appendix B. Decolorization tests results
Tables and graphics of laboratorial decolorization tests results.
Table B-1: Decolorization results using Amber 2001HMW
T. pH pH polymer color Blank % color

flocculation initial adjusted dosage ICUMSA ICUMSA removed
(°C) (ppm) (IU) (IU)
76,2-78 8,18 6,99 25 643 761 15,51
76,1-78,5 8,19 7,00 50 664 894 25,54
76-78,4 8,16 7,00 100 570 896 36,38
75,8-77 8,25 6,99 150 571 801 28,71
76-78,6 8,24 7,00 200 512 793 35,44
75,8-77,2 8,20 7,02 250 552 763 27,65
AMBER 2001 HMW POLYAMINE

1000
800
LIQUOR COLOR (IU)
600
400
200
0
0 50 100 150 200 250 300
POLYMER DOSAGE(PPM)
Figure B.1: Graphic representation of color (IU) of the liquor, treated and untreated with the polymer, versus polymer dosage
for the test using Amber 2001HMW
87
Table B-2: Decolorization results for Amber 2002

77,9-79 7,80 6,94 25 622 902 11,64
77,8-80,1 7,86 6,99 50 728 849 18,37
78,8-80,4 7,97 7,03 100 657 883 25,59
78,9-80,2 8,31 6,98 150 631 734 14,03
78-80,6 8,22 7,03 200 734 746 1,61
78-79,7 8,17 7,01 250 784 738 -6,23
AMBER 2002 POLYAYME

1000
800
LIQUOR COLOR (IU)
600
400
200
0
0 50 100 150 200 250 300
POLYMER DOSAGE (PPM)
for the test using Amber 2002
88

77,2-79,2 8,24 7,02 25 693 768 9,77
77,7-79,1 8,17 7,00 50 641 778 17,61
76,8-79,2 8,18 6,99 100 549 752 26,99
77,1-80,6 8,31 7,00 150 602 797 24,47
76,7-79,4 8,24 7,01 200 683 801 14,73
77,1-79 8,31 6,98 250 709 780 9,10
AMBER 2006
900
800
700
LIQUOR COLOR (IU
600
500
400
300
200
100
0
0 50 100 150 200 250 300
89

77 7,80 7,01 25 638 757 15,72
77,5 7,86 7,02 50 682 781 20,10
78 7,97 7,01 100 538 787 31,73
78,2 8,31 6,99 150 610 802 23,94
77,4 8,22 7,01 200 681 798 14,66
75,4-79,8 8,17 7,00 250 750 851 11,87
AMBER 2008 POLYDADMAC

900
800
LIQUOR COLOR (IU)
700
600
500
400
300
200
100
0
0 50 100 150 200 250 300
for the test using Amber 2008.
90
Table B-5: Decolorization results for Amber 2008HMW
T. pH pH polymer color Blank color

flocculation initial adjusted dosage ICUMSA ICUMSA ICUMSA
(°C) (ppm) (IU) (IU) (IU)
76-79,6 7,80 6,94 25 667 779 14,38
77-80,2 7,86 6,99 50 696 875 20,46
76-80,1 7,97 7,03 100 771 885 12,88
77-79,2 8,31 6,98 150 820 893 8,17
76-79,6 8,22 7,03 200 923 902 -2,33
78-79,1 8,17 7,01 250 968 876 -10,50
AMBER 2008 HMW POLYDADMAC

1200
1000
LIQUOR COLOR (IU)
800
600
400
200
0
0 50 100 150 200 250 300
for the test using Amber 2008HMW.
91

77-79.6 8,31 7,01 25 721 748 3,61
77-80,1 8,22 7,02 50 593 739 19,76
77-78 8,23 7,00 100 589 752 21,68
76-79 8,19 7,00 150 870 801 -8,61
78-79 8,22 7,00 200 846 784 -7,91
77-79,1 8,17 7,03 250 878 780 -12,56

1000
800
LIQUOR COLOR (IU)
600
400
200
0
0 50 100 150 200 250 300
92

76-77 8,27 6,91 25 586 667 12,14
76-78 8,26 7,01 50 518 723 28,35
76,5-77,9 8,23 7,02 100 477 809 41,04
76-77 8,25 6,99 150 659 770 14,42
76-78,2 8,22 7,00 200 734 782 6,14
76-78,5 8,27 6,99 250 722 699 -3,29

900
800
LIQUOR COLOR (IU)
700
600
500
400
300
200
100
0
0 50 100 150 200 250 300
93

75,4-77,8 8,23 7,00 25 674 848 20,52
76-78 8,29 7,01 50 600 778 22,88
76-77,8 8,30 7,01 100 518 799 35,17
75,6-78,5 8,31 6,98 150 599 630 13,94
76,5-79 8,31 7,00 200 545 645 12,40
76-78,7 8,34 6,99 250 734 631 -6,69

900
800
LIQUOR COLOR (IU)
700
600
500
400
300
200
100
0
0 50 100 150 200 250 300
POLYMER DOSAGE(PPM)
94

76,9-78 8,37 7,00 25 562 619 9,21
76,1-78,4 8,37 7,00 50 468 614 23,78
76-78,2 8,32 6,99 100 547 634 13,72
75,9-77,9 8,38 7,01 150 604 656 7,93
76-77,9 8,41 7,02 200 724 651 -11,21
76,1-78,7 8,35 6,99 250 778 672 -15,77

900
800
LIQUOR COLOR (IU)
700
600
500
400
300
200
100
0
0 50 100 150 200 250 300
POLYMER DOSAGE(PPM)
95
Table B-10: Decolorization results for Amber MetQuat 6

76-77,8 8,20 7,02 25 595 731 18,60
75,8-78,1 8,19 7,00 50 674 848 20,52
76-77,8 8,24 7,02 100 610 804 24,13
76-78 8,15 7,02 150 505 767 34,16
76,2-78,9 8,20 7,02 200 497 823 39,61
76-78,4 8,23 6,99 250 554 793 30,14
AMBER METAQUAT 6
900
800
LIQUOR COLOR (IU)
700
600
500
400
300
200
100
0
0 50 100 150 200 250 300
for the test using Amber MetQuat 6.
96
Table B-11: Decolorization results using Blend Amber 2009 & Amber 2001HMW

78-80 8,32 7,00 25 614 772 20,47
78-79 8,39 7,00 50 562 790 28,86
78-80 8,30 6,99 100 598 821 27,16
Table B-12: Decolorization results using Blend Amber 2013 & Amber 2001HMW

78-80 8,27 7,01 25 609 772 21,11
78-79 8,31 7,00 50 542 783 30,78
78-80 8,29 7,00 100 607 794 23,55
Table B-13: Decolorization results using Blend Amber 2013, Amber 2009 & Amber 2001HMW

78-80 8,20 7,00 25 642 887 27,62
77-79 8,26 7,00 50 568 847 32,94
76-79 8,31 7,00 100 620 834 25,66
97
98
Appendix C. Flocculation monitoring additional tests.
a) b)
c)
Figure C.1: Plot for flocculation kinetics using Amber2001HMWat 300 ppm, a) = d10, b) = d50 e c) = d90
99
a) b)
c)
Figure C.2: Plot for flocculation kinetics using Amber2001HMWat 400 ppm, a) = d10, b) = d50 e c) = d90.
100
a) b)
c)
101
a) b)
c)
102

Sugar Decoloration Strategies and Solutions

Uploaded by

Copyright:

Available Formats

Sugar Decoloration Strategies and Solutions

Uploaded by

Document Information

Original Description:

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

Sugar Decoloration Strategies and Solutions

Uploaded by

Copyright:

Available Formats

João Luciano Amorim Azevedo

Sugar decolorization - strategies and solutions

Technology of the University of Coimbra

Sugar decolorization - strategies and

In collaboration with AquaTECH (Switzerland) and RAR (Portugal)

Martin Luther King Jr

To my family, in special to my beloved Mother, Lourdes, who is for me the strongest

To all Professors in the Chemical Engineering Department, in special to Professor Doctor

A indústria do açúcar consiste principalmente em dois produtos: açúcar bruto e refinado. O

Chapter 1. Introduction ........................................................................................................... 1

Chapter 2. State of the Art ....................................................................................................... 5

2.1 Sugar refining .................................................................................................................... 5

2.2 Colorants in sugar .............................................................................................................. 6

2.2.1 Natural colorants......................................................................................................... 6

2.2.2 Colorants formed during sugar processing. .............................................................. 7

2.3 Measurement of color in sugar .......................................................................................... 8

2.4 Decolorization process ...................................................................................................... 9

2.5 Primary decolorization process ....................................................................................... 10

2.5.1 Phosphatation ........................................................................................................... 10

2.5.2 Carbonatation ........................................................................................................... 11

2.6 Flocculants and color precipitants ................................................................................... 12

2.6.1 Polymer flocculation ................................................................................................ 13

2.6.2 Anionic flocculants ................................................................................................... 14

2.6.3 Cationic flocculants .................................................................................................. 15

2.7 Secondary color removal technics ................................................................................... 19

2.7.1 Bone char .................................................................................................................. 19

2.7.2 Activated carbon ....................................................................................................... 20

2.7.3 Ion exchange ............................................................................................................. 20

3.1 Materials .......................................................................................................................... 23

3.1.1 Sugar ......................................................................................................................... 23

3.1.2 Polymers ................................................................................................................... 25

3.2 Methods ........................................................................................................................... 26

3.2.1 Preparation of the polymer solutions ........................................................................ 26

3.2.2 Decolorization tests .................................................................................................. 26

3.4 Evaluation of IV on decolorization ................................................................................. 28

3.5 Polymers characterization................................................................................................ 29

3.5.1 Hydrodynamic diameter ........................................................................................... 30

3.5.2 Zeta potential measurement ...................................................................................... 32

3.5.3 Molecular weight ...................................................................................................... 34

3.5.4 Refractive index ........................................................................................................ 38

3.6 LDS flocculation monitoring tests ................................................................................... 39

Chapter 4. Results and discussion ......................................................................................... 43

4.1 Polymer Characterization ................................................................................................ 43

4.2 Standard decolorization tests results ............................................................................... 45

4.3 Determination of Index value .......................................................................................... 51

4.4 Flocculation monitoring using LDS ................................................................................ 53

Chapter 5. Conclusions and future works ............................................................................. 67

5.1 Conclusions ..................................................................................................................... 67

5.2 Future works .................................................................................................................... 68

Appendix A . Polymer Characterization.................................................................................... 75

A.1 Size distribution by intensity for the polyelectrolytes .................................................... 75

A.3 Molecular Weight Debye plots for the polymers ........................................................... 82

Appendix B . Decolorization tests results ................................................................................ 87

Tables and graphics of laboratorial decolorization tests results. ........................................... 87

Appendix C . Flocculation monitoring additional tests. ............................................................ 99

Table 2-1: Type of sugar colorants present in sugar [4]. ............................................................. 8

1) characterization of the polymers used in the tests;

This dissertation is then divided into five chapters, as following: