Fulltext

Plant and system-level performance of combined heat and power plants
equipped with different carbon capture technologies
Downloaded from: https://research.chalmers.se, 2023-07-18 17:14 UTC
Citation for the original published paper (version of record):

Roshan Kumar, T., Beiron, J., Biermann, M. et al (2023). Plant and system-level performance of
combined heat and power plants equipped with different
carbon capture technologies. Applied Energy, 338. http://dx.doi.org/10.1016/j.apenergy.2023.120927
N.B. When citing this work, cite the original published paper.
research.chalmers.se offers the possibility of retrieving research publications produced at Chalmers University of Technology.
It covers all kind of research output: articles, dissertations, conference papers, reports etc. since 2004.
research.chalmers.se is administrated and maintained by Chalmers Library
(article starts on next page)

Applied Energy 338 (2023) 120927
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Plant and system-level performance of combined heat and power plants

equipped with different carbon capture technologies
Tharun Roshan Kumar *, Johanna Beiron , Maximilian Biermann , Simon Harvey ,
Henrik Thunman
Division of Energy Technology, Department of Space, Earth, and Environment, Chalmers University of Technology, SE-412 96, Gothenburg, Sweden
H I G H L I G H T S
• Process modeling and integration of CCS technologies with respect to a reference bio-CHP plant.
• Key technology performance indicators for the carbon capture technologies are quantified.
• The possibility of integrating heat pumps into the BECCS plant is investigated.
• Case-specific exergy performance of the BECCS plant at the plant and DH system levels.
• BECCS plant with higher power preservation preferable from a local DH system perspective.
A R T I C L E I N F O A B S T R A C T
Keywords: Installing carbon capture and storage (BECCS) capability at existing biomass-fired combined heat and power
Combined heat and power (bio-CHP) plants with substantial emissions of biogenic CO2 could achieve significant quantities of the negative
Carbon capture and storage CO2 emissions required to meet climate targets. However, it is unclear which CO2 capture technology is optimal
Exergy analysis
for extensive BECCS deployment in bio-CHP plants operating in district heating (DH) systems. This is in part due
District heating system
Hot potassium carbonate
to inconsistent views regarding the perceived value of high-exergy energy carriers at the plant level and the
Monoethanolamine extended energy system to which it belongs. This work evaluates how a bio-CHP plant in a DH system performs
when equipped with CO2 capture systems with inherently different exergy requirements per unit of CO2 captured
from the flue gases. The analysis is based upon steady-state process models of the steam cycle of an existing
biomass-fired CHP plant as well as two chemical absorption-based CO2 capture technologies that use hot po
tassium carbonate (HPC) and amine-based (monoethanolamine or MEA) solvents. The models were developed to
quantify the plant energy and exergy performances, both at the plant and system levels. In addition, heat re
covery from the CO2 capture and conditioning units was considered, as well as the possibility of integrating large-
scale heat pumps into the plant or using domestic heat pumps within the local DH system. The results show that
the HPC process has more recoverable excess heat (~0.99 MJ/kgCO2,captured) than the MEA process (0.58 MJ/
kgCO2,captured) at temperature levels suitable for district heating, which is consistent with values reported in
previous similar comparative studies. However, using energy performance within the plant boundary as a figure
of merit is biased in favor of the HPC process. Considering heat and power, the energy efficiency of the bio-CHP
plant fitted with HPC and MEA are estimated to be 90% and 76%, respectively. Whereas considering exergy
performance within the plant boundary, the analysis emphasizes the significant advantage the amine-based
capture process has over the HPC process. Higher exergy efficiency for the CHP plant with the MEA capture
process (~35%) compared to the plant with the HPC process (~26%) implies a relatively superior ability of the
plant to adapt its product output, i.e., heat and power production, and negative-CO2 emissions. Furthermore,
advanced amine solvents allow the BECCS plant to capture well beyond 90% of its total CO2 emissions with
relatively low increased specific heat demand.
* Corresponding author.
https://doi.org/10.1016/j.apenergy.2023.120927
Received 9 November 2022; Received in revised form 4 February 2023; Accepted 1 March 2023
Available online 17 March 2023
0306-2619/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
T.R. Kumar et al. Applied Energy 338 (2023) 120927
Nomenclature REF Reference

Solv. Solvent
Abbreviations SRD Specific reboiler duty
BECCS Bio-energy with carbon capture and storage
bio-CHP Biomass-fired combined heat and power Symbols
CAP Chilled ammonia process CHPPP Power preservation in the bio-CHP plant with and without
CC Carbon capture CCS units (PCHPCCS/PREF)
CCS Carbon capture and storage Ė Exergy flow
CDR Carbon dioxide removal e0j,ch Standard chemical exergy of gas component
CHP Combined heat and power m˙ Mass flow (kg/s)
CHPCC Combined heat and power plant with stand-alone carbon Pcc Electricity demand in the carbon capture and liquefaction
capture plant units
CHP-CCS Combined heat and power plant with carbon capture, PCHPCCS Net power delivered by the bio-CHP plant with CCS units
compression, and liquefaction units PREF Net power delivered by the reference bio-CHP plant
CHP-HPC CHP model integrated with the HPC model and CO2 without CCS units
conditioning models Q Heat duty (MW)
CHP-MEA CHP model integrated with the MEA model and CO2 QCC Steam demand in the carbon capture and liquefaction units
conditioning models QCHPCCS Net heat delivered by the bio-CHP plant with CCS units
CHP-MEA-HP Large-scale heat pump integrated into the modeled Q˙fuel Biomass fuel input to the bio-CHP plant
CHP-MEA process QREF Net heat delivered by the reference bio-CHP plant without
Cond. Drum Condensation drum CCS units
COP Coefficient of performance R Universal gas constant (8.3145 J/mol⋅K)
DACCS Direct air carbon capture and storage x Mole fraction
DH District heating α Loading (mol/mol)
FGC Flue gas condenser Subscripts
GSHP Ground source heat pump cfg cleaned flue gas
HDD Heat demand density (kWh/m2) ext-sys extended system
HEX Heat exchanger fg flue gas
HHV Higher heating value (MJ/kg) liq liquid
HP High pressure pen penalty
HPC Hot potassium carbonate q heat flows
IPCC Intergovernmental Panel on Climate Change reb reboiler
LHV Lower heating value (MJ/kg) ref reference
MEA Monoethanolamine source heat source (ground) for the heat pump at 10 ◦ C
PEN Penalty w work
PP Power preservation
1. Introduction Bio-CHP plants in Sweden operate, in most cases, within a district

heating (DH) system and are typically the primary providers of district
Carbon dioxide removal (CDR) technologies, such as bioenergy heat in the region. One such plant is Stockholm Exergiś CHP8 plant,
carbon capture and storage (BECCS) and direct air carbon capture and which is one of the world’s largest bio-CHP plants, with a production
storage (DACCS), are considered to be crucial climate change mitigation capacity of 280 MW of district heat and 130 MW of electricity, supplying
options. The recent Intergovernmental Panel on Climate Change (IPCC) over 80% of the DH demand in Stockholm [5]. This bio-CHP plant was
Special Report [1] has projected that a BECCS deployment of up to 8 also recently awarded funding by the EU Innovation Fund for the
GtCO2/yr by mid-century is needed to limit global warming to 1.5 ◦ C. demonstration and full-scale implementation of BECCS with a capture
Accordingly, the Swedish climate goal of net-zero emissions by the Year capacity of approximately 800 kt/yr of biogenic CO2 [6]. However,
2045 [2] is expected to be complemented with a proposed target of 3–10 retrofitting a CHP plant that operates in a DH system with CO2 capture
MtCO2 of annual negative emissions via BECCS to compensate for hard- and conditioning units (hereinafter referred to as ‘CCS units’) is ex
to-abate sector emissions. Existing biomass-fired combined heat and pected to incur an energy penalty on its overall energy efficiency. The
power (bio-CHP) plants, which are among the largest point-source extent and nature of the penalty will depend on the chosen CO2 capture
emitters of biogenic CO2 in Sweden, could be converted to large-scale technology. Furthermore, such a retrofit will negatively affect the
CO2 removal plants through the integration of carbon capture pro existing local energy system as it will have to compensate for the lost
cesses. A recent study that investigated the techno-economic potential of heat and power production.
BECCS from waste and biomass-fired CHP plants (hereinafter referred to Gustafsson et al. [7] have investigated and reported an energy pen
as ‘bio-CHP’ plants) in Sweden estimated a total potential for negative alty that ranges from − 3% to + 7% for a planned full-scale integration of
emissions of 10–16 MtCO2 annually, corresponding to 24%–40% of the hot potassium carbonate (HPC) CO2 capture process into the CHP
Sweden’s annual territorial fossil CO2 emissions (roughly 40 MtCO2) plant (CHP8) in Stockholm. The energy penalty represents the per
[3]. A recent study conducted by Fuss et al. [4] concluded that carbon centage change in the total useful energy output of the reference plant, i.
capture would be required at all of the large-scale bio-CHP plants in e., heat and power production, with the integration of the CCS units. The
Sweden to achieve the proposed Swedish BECCS target of 10 MtCO2/yr partly negative span for the energy penalty, reported in [7], indicating a
by the Year 2045, assuming that the target is to be met by bio-CHP plants higher total energy output compared to the reference plant without CCS
alone with their CO2 capture units capturing 90% of their total biogenic units, results from the assumption of slightly lower return (43 ◦ C) and
emissions. supply temperatures (78 ◦ C) for the DH water flows, compared to the
2
average Swedish temperatures of 47 ◦ C and 86 ◦ C, respectively [8]. combined cycle plants. Thus, it is crucial to evaluate different capture
Similarly, the partly positive span of energy penalty reported by Gus technologies integrated into a specific large-scale CHP plant that oper
tafsson et al. [7] corresponds to the case in which higher district heating ates in its local DH system in order to highlight the various operational
return (55 ◦ C) and supply temperatures (110 ◦ C) are considered. The differences and the impacts that such integration has on both the CHP
HPC process was originally developed and used as a gas purification plant and the consumers of its energy services, i.e., primarily DH and
technology [9], and it has only been investigated in a few academic power.
studies (summarized in the following paragraph) as a potential CO2 The amount of available excess heat that can be recovered from the
capture process due to its relatively high specific power consumption capture and conditioning units and delivered to a DH network will
(per tCO2 captured), as compared to other amine-based capture tech depend on the type of CCS technology retrofitted to the CHP plant. For
nologies. The HPC process is, however, commercially offered as a carbon example, Eliasson et al. [24] have estimated that roughly 25% of the
capture technology [10–12]. A high-power demand incurs a significant heat supplied to an amine-based capture plant can be recovered via
energy penalty in the context of fossil fuel power plants, which were the direct heat exchange and delivered to a DH network. In contrast, the
intended application in early CCS research and development. Further HPC process is expected to have a marginally higher heat recovery po
more, it is not realistic to recover waste heat in power plants due to the tential [7,14,15], mainly due to the heat that can be recovered from the
low-temperature level of such heat. Nevertheless, the HPC technology is flue gas compression section while incurring considerable losses to the
promising in the context of a CHP plant operating in a district heat electricity delivery capacity from a CHP plant. Hammar C. [15] quan
system, as argued by Levihn et al. [13]. tified and compared the DH and electricity delivery losses for a waste-
A few recent studies have evaluated the performances of different CHP plant equipped with MEA and HPC capture technologies. In com
CO2 capture technologies retrofitted to CHP plants and waste-to-energy parison to the DH delivery from the waste-CHP plant without CO2
plants in a DH context [14,15]. For example, Djurberg R. [14] compared capture, they concluded that the MEA process retains 99% of the DH
the HPC, monoethanolamine (MEA), and chilled ammonia (CAP) cap delivery capacity, whereas the capacity increases by 7% for the HPC
ture processes retrofitted to a CHP plant in Uppsala, assuming a 90% process due to the higher amount of recoverable excess heat from the
capture rate for the three capture technologies. They concluded that CO2 capture and liquefaction units. In addition, the electricity delivery
CAP is the economically feasible capture technology, followed by the losses related to the CHP plant without CO2 capture, alternatively, the
MEA and HPC processes, considering waste heat from the CCS units is retained or preserved electric power production capacity with the
utilized for heating DH water. However, their work did not consider integration of CCS units to the waste-CHP plant was estimated to be 80%
excess heat recovery from flue gas compression units in the HPC process, and 56% for the MEA and HPC processes, respectively.
which could have led to an overestimation of the specific capture costs Eliasson et al. [25] have investigated different strategies to utilize
for the HPC process. Furthermore, previous work on CAP [16] high available excess heat in industrial plants, such as an integrated steel mill,
lighted that this capture technology is only applicable in regions that to achieve cost-optimal heat supply to the CCS plant and the DH
have access to low-temperature cooling water (~5◦ C) and for flue gas network. Thus, incorporating the significant influence of seasonal vari
streams with a high CO2 concentrations (>15 vol.% CO2) to minimize ations on the availability of excess heat into the optimal design and
the cost associated with ammonia slip control and utility costs. How operational mode of the capture plant. Similar seasonal variations are
ever, many CHP plants operating inland have limited access to cooling expected in a large-scale CHP plant operating within a DH network. As
water at these low temperatures. In addition, experimental work [17] mentioned above, different exergy requirements of different types of
and process modeling studies [18–20] of the CAP process have shown capture technologies are expected to affect the operational flexibility of
that this capture process is limited up to a capture rate of 90%, due to a CHP plant in different ways. Higher preservation of electric power
increasing specific reboiler duty and increasing specific cooling re production capacity for a BECCS plant implies higher availability of
quirements to avoid ammonia emissions in excess of 10 ppm [21]. These high-exergy energy carrier, i.e., electricity, which inherently confers
aforementioned limitations of CAP were not considered in the techno- flexibility to the BECCS plant in that it could strategically distribute the
economic assessment by Djurberg R. [14]. electricity to the grid or utilize it within the plant to prioritize DH de
Several studies have evaluated the performance of a CHP plant livery. For example, integrating a heat pump that would upgrade low-
equipped with a CO2 capture unit, mainly applying the benchmark grade excess heat from a specific CO2 capture technology and its con
amine-based (MEA) solvent [22,23]. However, CO2 capture processes ditioning units could alleviate operational inflexibility or even further
with amine-based solvents are inherently different from the HPC process enhance the operational flexibility of the CHP plant. For example,
in that they do not require flue gas compression and have a higher heat Abrami G. [26] has estimated that integrating a heat pump into a waste-
demand per tonne of captured CO2 to regenerate the rich-amine solvent fired CHP plant retrofitted with CCS units would increase the total dis
(specific reboiler duty, SRD), which also results in a significant energy trict heat supply by 10% compared to the existing plant without these
penalty for the CHP plant. Wide-ranging overall energy penalties are units. Further details on the use of a large-scale heat pump in combi
reported for an amine-based capture plant retrofitted to CHP plants nation with a CHP plant (without CCS) [27] and strategies available for
evaluated in different contexts concerning their DH network and the the flexible operation of CHP plants can be found elsewhere [28,29].
regional market conditions. The differences are due to the very different District heating and system perspectives: One aspect often
technologies available to compensate for the loss of power and/or heat neglected in process-level technological comparisons of CCS capture
output from the CHP plant associated with implementing CCS. Pröll and technologies for CHP plants is the impact on the amount of heat deliv
Zerobin [22] have investigated large-scale BECCS at a bio-CHP plant ered to the DH system. In the literature, comparative techno-economic
with a maximum fuel capacity of 66 MWth,LHV, and an amine-based post- analyses are often limited to the plant system boundary, i.e., adopting
combustion CO2 capture technology and have found that it incurs en the perspective of investors in the CO2 capture technology who are also
ergy penalties of about 40.7% and 36.1%, respectively, depending on the owners of the CHP plants that emit CO2. For example, Kärki et al.
whether conditioning of the captured CO2 to pipeline specifications is [23] have reported that factors such as the electricity price and EU ETS
considered or not. Similarly, Kärki et al. [23] have investigated amine- carbon price are expected to be the dominant determinants of the
based CO2 capture for a 1,020-MWth,HHV CHP plant and estimated the viability of CCS from an investor’s point of view. It was also highlighted
energy penalty to be approximately 16%. In general, these studies have in the same study [23] that the most-feasible CO2 capture solution was
been limited to one specific CO2 capture technology applied to a CHP when there was heat recovery from the CCS plant and when the CHP
plant in a DH system. In some cases, the application of CCS at the CHP plant had the flexibility to operate in condensing mode so that it could
plant has been compared with other industrial CHP plants or dedicated respond to high electricity prices. Different CO2 capture technologies
electric power plants, such as condensing power plants or natural gas could result in altered electric power outputs from the CHP-CCS plant,
3
which could be consumed within the plant through large-scale heat in this work.
pumps or exported, depending on electricity prices. From the investor’s This work presents a detailed comparison of CO2 capture technolo
perspective, it is clear that the optimal CO2 capture technology is the one gies not only as stand-alone technological options but also considering
that minimizes the cost of implementation of BECCS or maximizes the impact on a process level within the existing bio-CHP plant bound
profit, factors that are dependent upon the market conditions of the local aries, as well as its DH system. The results indicate the optimal CO2
energy system in which the CHP plant operates. capture technology for different system boundaries. More specifically,
In Sweden, bio-CHP plants are one of the primary providers of DH. this work presents a detailed comparison of the HPC capture process (as
Since they operate in a closed DH market, they are regarded as local considered for planned full-scale implementation [43] by Stockholm
natural monopolies [30,31]. Thus, there are minimal limitations (apart Exergi) and the MEA capture process (i.e., a first-generation amine
from operational limitations such as minimum turbine load or power-to- solvent commonly considered as the benchmark solvent in the litera
heat ratio, as stated by Gustafsson et al. for the CHP8 plant [7]) on their ture) in the context of a reference bio-CHP plant in a DH system.
electric power production levels, which are rather dependent upon the Although advanced amine-based solvents could incur relatively lower
seasonal heating and cooling demands [27]. In this context, it is not clear SRD at a 90% capture rate or have capture rates > 90% with minimal
how to select the optimal end-use of the high-exergy energy carrier, i.e., increase in SRD compared to the benchmark MEA solvent, this work
the low-carbon electricity generated by the CHP-CCS plant. While the intentionally adopts a conservative approach. The MEA capture process
low-carbon electricity could be generated and consumed cost-optimally is therefore compared with the HPC process at the standard 90%-capture
by the energy-service provider, i.e., by the bio-CHP plant within its plant rate to clearly establish the optimal CO2 capture technology at the same
boundaries, the consumers of the low-carbon electricity could assign a absolute amount of CO2 captured from bio-CHP plants. To this end, this
higher value to the electricity due to its greater utility. More specifically, work contributes with a simplified method to evaluate BECCS systems in
the generated electric power could be consumed in the local DH system a DH system and presents the consequence of using energy or exergy as
to meet consumers’ heating needs using decentralized domestic heat the figure of merit for different system boundaries.
pumps. However, such power-to-heat technologies that compete with Specifically, the work seeks to:
DH are hindered by pre-existing market entry barriers, as discussed by
Åberg et al. [32]. The DH providers typically use pricing mechanisms to 1. compile technical performance data for the HPC and MEA capture
compete with decentralized heat pumps in the DH network [31,33]. In processes at the technology level with a design capture rate of 90%
addition, the tax imposed on electricity [34] disincentivizes the use of (stand-alone processes),
electricity for decentralized heat production and dissuades consumers 2. evaluate the exergy and energy performances of the two CO2 capture
from installing domestic heat pumps, such as ground source heat pumps processes within the boundary of a CHP-CCS plant, i.e., at the process
(GSHPs) or air-to-water heat pumps, instead of connecting to the DH level,
grid, thereby reducing the overall DH system costs. Therefore, the per 3. assess at the system level the exergy and energy performances of
spectives of the energy service provider and the consumer are CHP-CCS plants that are retrofitted with either the MEA (CHP-MEA)
discrepant, and it is unclear as to what the optimal choice of CO2 capture or HPC (CHP-HPC) capture technology. This analysis includes the
technology for a CHP-CCS plant would be, since the energy penalty CHP plant, as well as the consumers of the energy services that the
incurred by the two capture technologies discussed above is either in the CHP plant provides via the local DH system and the electricity grid,
form of loss of heat or electric power production capacity. Furthermore, 4. compare the local DH system with CHP-CCS plants that include
comparative studies of CO2 capture technologies could result in different electricity-driven heat pumps at two different locations: 1) central
conclusions depending on the chosen system boundary and whether ized heat pumps, i.e., large heat pumps that can upgrade residual
energy or exergy is used as a figure of merit. heat from the capture plant and deliver it to the DH grid; and 2)
Biermann M. [35] argued that specific conditions such as site-related decentralized heat pumps that use the ground as the heat source
conditions, policy landscape, product, and market-related factors could (GSHPs) to meet the end-use energy needs within the local DH sys
motivate partial capture in the process industry, where only a portion of tem, i.e., primarily space heating and domestic hot water
the total on-site emissions are captured in CO2 capture units. However, if
large-scale negative-CO2 emissions are to be achieved, bio-CHP plants The method (Section 2), results (Section 3), and discussion (Section
require a CO2 capture technology that not only incurs minimal energy 4) sections of the paper follow the same structure as the specific ob
penalty but also has the potential to go well beyond the ‘90%-capture jectives presented above with – i) technological comparison of CO2
rate’, previously considered as the techno-economic limit of post- capture technologies as a stand-alone process, followed by ii) compari
combustion CO2 capture. Both pilot plant tests [36] and techno- son of CHP-CCS plant cases within its plant boundaries, and, iii) within
economic studies [37–39] of the amine-based capture process in the its extended system boundary, i.e., the local DH system.
literature have established that capture rates up to 99% can be achieved
with a marginal increase in specific CO2 capture costs. However, the 2. Method
possibility of a capture rate higher than 90% does not hold for both the
CAP and the HPC processes without incurring a substantial increase in Fig. 1 shows an overview of the method used in this work, together
operational costs. For example, CAP pilot plant tests at different scales with the two system boundaries: 1) the CHP-CCS plant boundary, which
have shown an average capture rate of 75–85% [17], where increased includes the bio-CHP plant fitted with CO2 capture and conditioning
capture rates eventually lead to higher ammonia slip [40], thereby (CO2 compression and liquefaction for ship transport) units; and 2) the
leading to higher costs associated with controlling ammonia emissions extended system boundary, which includes the local DH system, where it
[21]. For these reasons, CAP was not considered in this work. In is assumed that the electric power delivered by the bio-CHP plant is
contrast, the HPC process with a relatively harmless capture solvent consumed by decentralized ground source heat pumps (GSHPs) installed
requires significantly higher operating pressures and, thereby, higher to meet the indoor-climate needs of end-users in residential buildings.
electricity consumption to achieve a capture rate beyond 90%. There The two system boundaries are chosen to allow evaluations of the plant-
fore, the HPC process was considered over the chilled ammonia process and system-level performances of a reference bio-CHP plant that is ret
due to its lower environmental impact [41,42] and recoverable excess rofitted with different carbon capture technologies. The comparison of
heat at higher temperatures that could be recovered for district heating. the two carbon capture technologies is performed for a large-scale
In line with the recommendations in Gustafsson et. al [7], the compar reference bio-CHP plant operated by Stockholm Exergi [6] in Stock
ison of HPC process with MEA process, and the possibility of integrating holm, Sweden. Table 1 shows the relevant data for the reference bio-
heat pumps to recover more heat from the BECCS plants was considered CHP plant. A steady-state model of the reference bio-CHP plant was
4
Fig. 1. Overview of the methods and models used with the two system boundaries. bio-CHP plant boundaries; and the extended system boundary (the local DH
system). The reference bio-CHP plant (indicated in red) data are fed to the developed numeric models, indicated in blue. Finally, the applied methods and results
obtained are highlighted in gray. 1Delivered energy services of the bio-CHP plant integrated with CCS (CHP-CCS), which includes the net district heating (DH) and
power delivered to the bio-CHP plant’s local DH system. Note that the delivered electricity to the grid is assumed to be consumed in decentralized GSHPs to meet
indoor-climate needs, i.e., domestic hot water and space heating. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
MEA and CHP-HPC) within the CHP-CCS plant boundary.

Table 1
An extended system boundary is adopted to enable a fair comparison
Reference bio-CHP plant data [7].
of the two CO2 capture processes at the system level (see Fig. 1), which
Parameter Unit Value includes the bio-CHP plant and the end-users of the distributed energy
Fuel type – Wood chips with 50% moisture service. Since DH is the primary energy service provided by the bio-CHP
content plant, we assume that GSHPs are used on the demand side to convert the
Fuel input (lower heating value, MW 362.1
delivered power to end-use heat, so as to meet the indoor-climate needs
LHV)
Fuel input (higher heating value, MW 451.1 related to space heating and domestic hot water. This assumption
HHV) mainly holds for regions in which DH and GSHPs compete or regions
District heat production capacity MW 280 where the DH network does not benefit from a technological monopoly,
Electric power production capacity MW 130 as defined by Åberg et al. [32]. It is worth noting that air-to-water heat
Live Steam temperature ◦
C 558.0
pumps could be an alternative solution with performance levels similar
Live Steam pressure bar 136.0
Flue gas flow rate kg/s 174.2 to those of GSHPs [45]. However, such heat pumps are not considered in
Wet flue gas composition this work due to their susceptibility to variable performance with vari
CO2 vol. 16.0 ations in the outdoor temperature. This somewhat simplifies the ana
%
lyses performed within the extended system boundary. Within the
O2 vol. 3.2
%
extended system boundary, the energy and exergy efficiencies are esti
H2O vol. 5.3 mated, assuming an end-use energy temperature range of 30◦ –60 ◦ C, for
% the supply of space heating and domestic hot water, respectively. Thus,
N2 and inert gases vol. 75.5 the extended system boundary shown in Fig. 1 includes the transmission
%
losses within the DH network and the electricity grid, as well as the
conversion losses associated with the technical systems (e.g., GSHPs)
developed in EBSILON Professional with available plant data, and the used to meet the indoor-climate needs.
MEA and HPC capture processes were modeled (steady-state) in Aspen
Plus V12.1 and simulated with the reported flue gas data (see Table 1).
2.1. Model descriptions
The detailed modeling and validation of the bio-CHP plant and the
carbon capture models are described in Section 2.1.
2.1.1. Descriptions of the CO2 capture and conditioning processes
The heat and power demands of the two CO2 capture plants were
Fig. 2 shows the process flowsheets of the two CO2 capture processes.
estimated using the capture plant models and thereafter used as inputs
Both capture process simulation models in Aspen Plus V12 adopt
to the bio-CHP plant model so as to represent a bio-CHP plant retrofitted
rigorous rate-based models with detailed reaction kinetics. The mass
with the CCS units. The reference bio-CHP plant retrofitted with CCS
transfer coefficients and interfacial areas in the packings were predicted
units is referred to as a ‘CHP-CCS plant’ hereinafter. Pinch analysis,
using the correlations described by Bravo et al. [46], whereas the liquid
which is an energy-targeting method [44], was used to quantify the
holdup was estimated using the correlations reported by Stichlmair et al.
recoverable process heat from the two carbon capture processes
[47]. Heat transfer coefficients were estimated via the Chilton and
assuming a minimum temperature difference of 5 ◦ C for heat exchange
Colburn analogy [48]. The specifications of the CHP flue gas fed to the
with the DH water. The recoverable process heat is categorized as low-
two CO2 capture processes were taken from Gustafsson et al. [7] (see
grade (47◦ –61 ◦ C) or high-grade (61◦ –86 ◦ C) heat, which is adapted
Table 1), with a flow of 174.2 kg/s, at ambient pressure and a temper
from the generic Swedish DH system defined by Gustafsson et al. [7],
ature of 160 ◦ C with a CO2 concentration of 16 vol.%wet (as compared to
with annual average representative temperature levels of 86 ◦ C (supply)
coal combustion with ~ 13 vol.%wet). The two capture models’ key as
and 47 ◦ C (return) [8]. Key performance indicators associated with the
sumptions and process parameters are compared in Table 2. The source
plant performance are then quantified for the CHP-CCS plants (CHP-
of data presented in these Table 2 is based on previous work, mainly
5
Fig. 2. Process flowsheet diagrams of a) the hot potassium carbonate (HPC) capture process; and b) the MEA capture process. The main differences between the two
processes, i.e., the flue gas compression train in the HPC process and the cross-heat exchanger in the MEA process, are highlighted in the red-shaded areas. The CO2
compression and liquefaction processes are common to both carbon capture models and therefore are highlighted in blue. Dashed lines depict the CO2-depleted flue
gas streams. Thick and thin solid lines depict the gaseous and liquid (solvent and water) streams, respectively. (Abbreviations. Lean-Rich Cross HEX–heat exchanger
for the CO2-rich and lean solvent, cond drum–condensation drum, FGC–flue gas condenser, HP – high pressure, Solv. – solvent). (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this article.)
Gustafsson et al. [7], for the HPC model and Biermann et al.[49,50], for model to avoid modeling complexity, as the rich and lean solvents are
the MEA model to obtain comparable results with the literature. typically at different pressures; alternatively, heat exchangers are used
Although the two simulation models were developed separately, to recover heat from the lean solvent that is pumped back to the absorber
necessary modifications were made to ensure design parameters and from the stripper; (3) a washer unit, which is not considered for the HPC
assumptions, e.g., capture rate, adiabatic and mechanical efficiencies, solvent owing to its much lower solvent volatility and water evaporation
and column heights, are kept consistent to allow for a reasonable at higher pressures; and (4) the optimal stripper pressure (minimum
comparison. SRD) in the HPC process is lower than in the MEA process, and this in
The fundamental differences between the HPC and MEA process fluences the reboiler temperature.
models include: (1) flue gas compression to allow the HPC to enhance The MEA model assumes an aqueous solution of MEA at a concen
the physical absorption of CO2 in the absorber; (2) a cross-heat tration of 30 wt.%. The model was based on previous work conducted by
exchanger for the MEA process model, which was omitted in the HPC Garđarsdóttir et al. [18] and Biermann et al. [50]. It was further
6
Table 2 stripper were set up according to Zhang and Chen [54], incorporating
Summary of the key process parameters and assumptions for the MEA and HPC the kinetic parameters described by Pinsent et al. [55] and Hikita et al.
CO2 capture process models. [56].
Parameter MEA HPC Comment/Reference The HPC model assumes a solution that contains 30 wt.% aqueous
Capture section
potassium carbonate (K2CO3) as the capture solvent. This is considered
Thermodynamic ENRTL- ELECNRTL MEA [49,50], HPC [7] to be the optimal solvent composition [57], being limited by a high
Properties RK solvent regeneration duty and salt precipitation in the lower and upper
Lean solvent 30 wt.% 30 wt.% MEA [49,50], HPC limits of the range for a 20–40 wt.% K2CO3 solution [57,58]. The
composition MEA in K2CO3 in [7,57,58]
operating conditions of the HPC model (Table 2) were chosen based on
H2O H2O
CO2 capture rate1 (%) 90 90 Note that a conservative the work of Gustafsson et al. [7], in which an equilibrium-stage model of
approach is taken here a full-scale capture plant (~90% capture rate) was developed using data
with the benchmark MEA obtained from a pilot-scale plant tested at the reference bio-CHP plant
solvent. Advanced amine (CHP8). In the present work, however, a rigorous rate-based model was
solvents have achieved
higher capture rates
developed that utilizes the ELECNRTL property method in Aspen Plus
(>90%) with a similar [59], with its associated reaction sets, as described by Ayittey et al. [60],
range of SRD [36] implemented with equilibrium and kinetic constants derived from
Isentropic efficiency 85 85 AspenTech and Wu et al. [61,62]. The HPC model developed in this
(pumps, compressors,
work was compared with published data from the full-scale models
and gas expander)2
(%) developed by Gustafsson et al. [7], based on their HPC carbon capture
Mechanical efficiency 95 95 test plant data, and was validated with respect to the solvent loading
(pumps, compressors, capacity and the corresponding SRD. The reaction set list and model
and gas expander)2(%) validation are provided in Supplementary Materials S.1.
Flue gas fan/compressor 1.16 7 [7,49,50]
The CO2 compression and liquefaction processes were modeled
discharge pressure1
(bar) based on the work carried out by Deng et al. [63], where the captured
Absorber gas outlet 1.06 7 [7,49,50] CO2 stream from the two different CO2 capture models was matched to
pressure1 (bar) the gas fed to compression and liquefaction train. This concentrated CO2
Absorber gas inlet 42.5 42.5 Set to a constant value
stream from the capture models was assumed to consist only of CO2,
temperature2 (◦ C) (taken from [7])
Absorber lean solvent 42.5 42.5 Set to a constant value in H2O, N2, and O2 (and trace amounts of MEA in the stream from the MEA
inlet temperature2 (◦ C) both models for model; no significant traces of K2CO3 are carried over in the HPC
comparable results model). This stream undergoes a three-stage intercooled compression
Direct contact cooler 7 7 [49] process in the CO2 compression train to attain the pressure required for
height1,2 (m)
liquefaction. The flue gas compression train was modeled with three
Absorber packing 20 20 Design factor (flooding
height1,2 (m) approach set to 80%) for compression stages with a constant pressure ratio per stage, increasing
estimating the column the pressure to 27.5 bar based on the specifications described by Deng
diameter [49,50] et al. [63], which assume some level of impurities that requires flashing
Water wash packing 2 Included in the MEA
–
and purging [31]. Thus, the compression outlet pressure is higher than
height1 (m) model with its diameter
equal to the estimated
the delivery pressure. Although we did not resolve these impurities in
absorber diameter our model, we included the elevated pressure to obtain a conservative
[49,50] estimate of the level of power consumption. Downstream of the
Stripper packing height1, 10 10 Design factor (flooding compression, a simple separator model mimics the removal of excess
2
(m) approach set to 80%) for
oxygen and water, which would otherwise violate the specifications for
estimating the column
diameter [49,50] liquid CO2 transport adopted from the Northern Lights project [64]. The
Stripper outlet pressure1 1.9 1.2 MEA [49,50], HPC [7] liquefaction process operates with an ammonia refrigeration cycle, as
(bar) described by Deng et al. [63], similar to the refrigeration cycle consid
Stripper CO2 gas outlet 20 20
ered in the carbon capture project at the waste-to-energy plant at Kle
temperature
(condenser
metsrud [65]. The specification of the outgoing liquid CO2 stream was
temperature)2 (◦ C) set to 16 bar and − 26.5 ◦ C, thereby resembling the specifications made
Minimum temperature 10 10 in the Northern Lights project [64], which considered ship transport of
difference (ΔTmin) in CO2 for permanent sub-seabed storage. Important assumptions
heat exchangers2 (◦ C)
regarding the process parameters of the CO2 compression and lique
Minimum temperature 5 5
difference (ΔTmin) in faction processes are listed in Supplementary Materials S.2. Finally, to
stripper reboiler2 (◦ C) evaluate the performances of the capture models, the following tech
1 nology performance indicators were considered, as defined in Eqs. (1)–
Constant design parameter.
2
Assumptions made to ensure comparable results from both CO2 capture models. (5):
[ ]
[HCO−3 ] mol
HPC loading (αHPC ) = (1)
developed regarding the MEA chemistry, and reaction sets based on the +
[K ] mol
most recent version of the MEA model by AspenTech [51], which has
[ ]
validated the model performance against pilot data by Notz et al. [52]. xCO2 mol
MEA loading (αMEA ) = (2)
The updated model has also been validated against large-scale pilot data xMEA mol
derived from testing CO2 capture from a steam reformer flue gas [ ]
[49,53]. Note that the property method ENRTL was used with the Qreboiler MJ
Specific reboiler duty(SRD) = (3)
Redlich-Kwong equation of state for the vapor phase, instead of the PC- ṁCO2,captured kgCO2
SAFT correlation, which is applied in the model developed by Aspen [ ]
Tech [51]. This was done to enhance convergence and entailed only Net power consumption MJ
Specific power demand = (4)
minor deviations in SRD (<0.3%). The reaction sets for the absorber and ṁCO2,captured kgCO2
7
[
Net cooling demand MJ
] flue gas stream from the absorber top, the lean solvent cooler, and the
Specific cooling demand = (5) heat of compression from the compression and liquefaction processes.
ṁCO2,captured kgCO2
The total heat recovered from these processes is shown in Fig. 4, denoted
as Qrecovered, which, together with the heat recovered from the bio-CHP
2.1.2. Bio-CHP plant steam cycle model
plant, is the total heat delivered by the CHP-CCS plant to the DH
A steady-state model of the reference bio-CHP plant steam cycle was
network, denoted as QCHPCCS.
developed in EBSILON Professional based on available plant data and
Steam and electricity are internally consumed to drive the carbon
general CHP design principles. Fig. 3 presents a schematic of the main
capture process, compression, and liquefaction, denoted as QCC and PCC,
steam cycle components modeled. The boiler, which is modeled as a
respectively. The net amount of electricity that is not used on-site is
steam generator with a fixed efficiency of 92% based on [7], produces
supplied to the grid, which is denoted as PCHPCCS. Finally, the remaining
live steam at a pressure of 136 bar and temperature of 558 ◦ C. The live
output streams are the liquefied CO2 stream at 20 ◦ C, the DH water
steam is expanded in a steam turbine in six stages, where steam is
supply at 86 ◦ C, and the CO2-depleted flue gas cooled to 42.5 ◦ C from the
extracted to feedwater preheaters, a deaerator, and two DH condensers.
carbon capture plant. With this system boundary, the energy and exergy
Input data are given for the turbine extraction pressures and DH target
flows of the CHP-CCS plant are evaluated for the three cases listed in
temperatures. The isentropic efficiency of the steam turbine is assumed
Table 3: CHP-MEA, CHP-HPC, and CHP-MEA-HP. Since the CHP-MEA
to be 88%. Based on the given fuel input, the model calculates the
plant is expected to consume less power than the CHP-HPC plant [13],
electricity generation and DH generation of the steam cycle.
the surplus power (net delivered electric power from case CHP-MEA
For the CHP-HPC plant, live steam is extracted to drive the flue gas
minus that from CHP-HPC) can be used instead to drive a centralized
compressor and to supply heat to the capture process (depicted with
heat pump so as to provide additional heat. This is done in the CHP-
black dashed lines in Fig. 3). For the CHP-MEA plant, low-pressure
MEA-HP configuration, which is modeled following two constraints:
steam for powering the capture process is extracted from the turbine
(i) the net power output must be the same as for the CHP-HPC case, and
at the deaerator stage. The steam condensate from the capture process is
(ii) it is assumed that the heat pump upgrades excess heat from tem
returned to the deaerator in both cases. The energy requirements of the
peratures lower than the DH return temperature of 47 ◦ C to the DH
capture process (reboiler duty and power demand of the flue gas
supply temperature of 86 ◦ C. Here, the lower temperature limit is set by
compressor) are given as inputs based on the capture plant process
the maximum temperature lift of 60 ◦ C, which is assumed in the heat
modeling, as described in Section 2.1.1.
pump. This assumption results in the lowest possible temperature of 26
◦
C. Thus, unutilized waste heat in the temperature range of 26◦ –47 ◦ C is
considered to be available for upgrading to the DH supply temperatures
2.2. Plant and extended system boundaries and performance evaluation
using centralized large-scale heat pumps in the CHP-MEA-HP case.
Therefore, the net delivered power to the grid in the CHP-MEA-HP case
2.2.1. CHP-CCS plant boundary
is equal to that of the CHP-HPC case.
The reference bio-CHP plant model is assumed to be retrofitted with
The cases listed in Table 3 are evaluated for their energy perfor
CO2 capture, compression, and liquefaction units. Heat is recovered
mances using the CHP plant performance indicators defined in Eqs. (6)–
within the bio-CHP plant through wet-gas cleaning of the exhaust gases
(9).
in a flue gas condenser and the turbine condenser. This is to provide heat
to the DH supply water (depicted with solid red lines in Fig. 4). With the
integration of the CCS unit, it is expected that additional excess heat can
be recovered through condensation of the CO2-rich stripper top gas, the
Fig. 3. Process schematic of the CHP steam cycle modeled for the reference plant, adapted from Beiron et al. [3]. Note that only one of the two CCS units is
considered when evaluating the CHP-MEA or CHP-HPC case. Note also that the FGC (shown in Fig. 2) is now placed outside the CCS unit blocks to represent the CHP
steam cycle more accurately. Black dashed lines – steam extracted from the steam cycle to drive the corresponding CCS unit; Red dashed lines – flue gases from the
boiler; Gray boxes – input data to the CHP steam cycle model. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
8
Fig. 4. Schematic of the CHP-CCS plant boundary. a) with the reference bio-CHP plant model integrated with carbon capture and CO2 conditioning process models;
and b) the exergy flows of the CHP-CCS plant boundary. Note that material streams, e.g., combustion air, makeup feedwater, and other makeup streams,
are neglected.
additional electric power consumed in the centralized large-scale heat

Table 3
pump is included in the total power consumption levels of the associated
Cases investigated within the plant system boundary.
CO2 capture process, denoted as PCC. Equation (6) is further simplified,
Case Description as shown in Eq. (7), as the ratio of the sum of the net heat (QCHPCC) and
investigated
the net electric power production (PCHPCC), and the recoverable heat
CHP-MEA CHP model integrated with the MEA model and CO2 conditioning (Qrecovered) to the fuel input (QHHV,fuel). Furthermore, power preserva
models
tion (adapted from Gustafsson et al. [7]) is defined in Eq. (8) as the ratio
CHP-HPC CHP model integrated with the HPC model and CO2 conditioning
models of the retained (or preserved) electric power output by the CHP plant
CHP-MEA-HP Heat pump integrated into the modeled CHP-MEA process – to following the integration of the CCS units or the net electric power
balance the amount of net power delivered in the CHP-MEA case output (PCHPCC) from CHP-CCS plant to the net electric power output
with that in the CHP-HPC case (PREF) of the reference CHP. Finally, the energy penalty (CHPpen), as
defined by Gustafsson et al. [7], is defined as the percentage change in
(PREF − PCC ) + (QREF − QCC ) + Qrecovered the energy efficiency of the reference CHP plant following the integra
ηCHPCCS = × 100 [%] [6] tion of carbon capture processes.
mf ⋅HHVfuel
(6) The exergy performance of the three cases (Table 3) is compared
=
PCHPCC + QCHPCC + Qrecovered
× 100 [%] [7] within the bio-CHP plant system boundary. The exergetic efficiency is
Q̇HHV f uel defined in Eq. (10) as the ratio of all exergy flows exiting the plant
system boundary (denoted with the “-” sign) to all exergy flows entering
Power preservation (CHPPP ) =
PCHPCC
× 100 [%] (8) the plant system boundary (denoted with the “+” sign), under atmo
PREF spheric conditions (Tref = 298.15 K and P0 = 1.01325 bar) taken as the
reference environment [66]. The specific exergy (e) of each material
(PREF + QREF ) − (PCHPCC + QCHPCC + Qrecovered )
CHPpen = × 100[%] (9) stream (i) is defined as the sum of the thermal exergy, kinetic and po
(PREF + QREF ) tential exergy related to a reference environment, where thermal exer
The energy efficiency of the CHP-CCS plant (ηCHPCCS ) is calculated as gies include both the physical and chemical exergies of a material stream
shown in Eq. (6), which represents the heat and electric power output [66]. The kinetic and potential exergies are neglected, as is typically
from the plant in relation to the higher heating value of the fuel input done when evaluating stationary, steady-state processes. Work streams
(HHVfuel). Here, the higher heating value (HHVfuel) of fuel input is (ĖW) are considered to have pure exergy, i.e., the energy flow is equal to
considered, since the heat of condensation of the fuel moisture is the exergy flow, whereas for heat flows, the exergy (Ėq) is calculated
recovered as useful heat in the flue gas condenser at temperatures according to Eq. (11), in that it is related to the Carnot factor, where Tref
suitable for DH water. QREF and PREF terms are the heat and power (K) is the reference temperature, and T (K) is the temperature at which
production, respectively, of the reference CHP plant operating without heat is supplied or delivered to the system. In this case, the heat deliv
CO2 capture. QCC and PCC are the heat and power consumption, ered as DH supply water is at 86 ◦ C.
respectively, of the CO2 capture and conditioning processes, and Qre In Fig. 4b, all the associated exergy streams of the CHP-CCS plant
covered is the low-temperature heat that is recoverable (via direct heat boundary are shown, where the input exergy streams are the fuel input
exchange and centralized large-scale heat pumps, if applicable) from and DH return water, and the main output exergy streams comprise
these CCS units, as shown in Fig. 4. In the CHP-MEA-HP case, the liquefied CO2, CO2-depleted gas, and the net power and heat delivered
to the grid and DH network, respectively. Other material streams, such
9
as combustion air and makeup boiler feedwater, are neglected because the other main outputs from the extended system boundary are the
they typically contribute significantly less to the total exergy input than cleaned flue gas stream and the liquefied CO2 from the CHP-CCS plant.
the fuel exergy input. The Aspen Plus property set EXRGYFL is used to Fig. 5b shows the input and output exergy flows for the extended
estimate the physical exergies of material streams, while the chemical system boundary, where the heat is consumed at an end-use temperature
exergies of the material streams are computed manually. For example, in the range of 30◦ –60 ◦ C, thus incorporating the losses associated with
Eq. (12) is used for gaseous mixtures, such as a CO2-depleted flue gas decentralized GSHPs, conversion losses linked to the technical systems
stream leaving the CHP-CCS plant. In Eq. (12), xj is the mole fraction of of the buildings, and DH and electric power transmission losses. A
the gas component in the gas mixture, and e0j,ch is the standard chemical representative range for the coefficient of performance (COP) of the
exergy of the gas component j in the mixture, relative to its reference GSHPs was estimated (see Supplementary Materials S.3), assuming a
environment [66]. In this work, wood chips (moisture content ~ 50%, constant heat source temperature of 10 ◦ C and a Carnot efficiency in the
HHVas-received ~ 10.1 MJ/kg or LHVas-received ~ 8.1 MJ/kg [7]) are range of 40–60%. While the heat source temperature is kept constant,
assumed to be the primary fuel for the reference CHP plant. The stan the range of Carnot efficiency (40–60%) is chosen to represent any in
dard chemical exergy of biomass is computed using the correlation fluence of external conditions, such as seasonal changes in ground
factor proposed by Szargut et al. [66] for solid technical fuels, from temperatures, on the COP of the decentralized GSHPs The exergy effi
which the exergy flow is estimated. A factor of 1.06 is estimated between ciency of the CHP-CCS plant operating within its extended system
the standard chemical exergy and the lower heating value (LHV) for the boundary, denoted as ηex,ext-sys, is defined in Eq. (14) as the ratio of the
wood chips. Finally, the exergy efficiency of the CHP-CCS plant is exergy outflows, i.e., the primary DH delivered (ĖQ,DH), end-use heat
calculated as per Eq. (13), which relates the system exergy output to the delivered by the GSHPs (ĖQ,GSHPs), liquefied CO2 (Ėliq,CO2), and the clean
total exergy input to the defined CHP-CCS system boundary, as shown in flue gas stream (ĖCFG), to the main exergy inflow, i.e., the DH return
Fig. 4a. water and the biomass fuel input to the CHP-CCS plant. Note that the
∑ − ∑ − ∑ −
m ⋅ei + Ėw + Ėq exergy flows of the combustion air, feedwater makeup, and other
ηex = ∑i i+ ∑ + ∑ + (10) makeup streams are neglected, as they are expected to have lower input
i mi ⋅ei + Ėw + Ėq
exergy flows (estimated to be < 1.5% for steam power plants [66]) in
( ) relation to the fuel exergy and DH return water in the CHP-CCS plant
Tref
Ėq = 1− ⋅Q̇ (11)
T ĖQ,DH + ĖQ,GSHPs + Ėliq,CO2 + ĖCFG
ηex,ext− sys = (14)
∑ ∑ Ėbiomass + ĖDH,return
ech = xj e0j,ch + RTref xj ln(xj ) (12)
j j
3. Results
ĖQ,DH + ĖW,el + Ėliq,CO2 + ĖCFG

ηexCHP− CCS = (13) 3.1. Comparison of the technical performances of HPC-based and MEA-
Ėbiomass + ĖDH,return based CO2 capture processes
2.2.2. Extended system boundary The technology performance indicators for the two CO2 capture
Fig. 5a shows the extended system boundary with the local DH sys technologies, capturing 90% CO2 from a typical flue gas stream from a
tem. Note that the CHP-CCS plant boundary is within the extended biomass-fired boiler with approximately 16 vol.% (wet basis) CO2, are
system boundary, with the final energy service being end-use heat shown in Table 4. The grand composite curves (GCCs) of the MEA and
consumed in the local DH system. There are two components to the HPC capture processes, including the compression and liquefaction
delivered heat: i) heat supplied as DH supply water, and ii) heat supplied units, are shown in Fig. 6. Utility targeting was performed to estimate
by the decentralized GSHPs at the demand side, using the net electric the amounts of potentially recoverable excess heat in these units,
power delivered to the electricity grid. Apart from the delivered heat,
Fig. 5. A) extended system boundary with the energy service provider (chp-ccs plant) and the energy service consumers (local dh system), where it is assumed that
the electric power generated is consumed entirely in decentralized ground source heat pumps (gshps), thereby maximizing the heat delivered to the system. b) the
exergy flows of the extended system boundary.
10
Table 4 expanding the pressurized cleaned flue gas that exits the top of the
Comparison of the HPC and MEA processes for a capture rate of 90% with 16 vol. absorber. The recoverable excess heat in the flue gas intercoolers is
% CO2 (approximately 130–132 tCO2 captured per hour), which is representa approximately 1.09 MJ/kgCO2 (42.5◦ –199 ◦ C), a portion of which can
tive of a wet flue gas stream from a biomass-fired boiler. Note that only the CO2 be utilized for reducing the hot utility requirements (depicted with a
capture sections are compared (excludes the compression and liquefaction solid red line in Fig. 6b). The reduction in the hot utility requirement, via
units).
heat recovery, for the HPC process is more than offset by its higher
Technology performance indicators Unit HPC MEA specific power demand, as compared with the MEA process, resulting in
CO2 capture section total specific energy demand for CO2 capture that is only marginally
Specific reboiler duty (SRD) MJ/kgCO2 3.44 3.67 lower for the HPC process (3.46 vs. 3.73 MJ/kgCO2).
Heat recovered from flue gas compression MJ/kgCO2 1.09 – It is worth noting that the net hot utility demand accounts for 88%
(42.5 ◦ C–199 ◦ C)
and 99% of the total specific energy demand for the HPC and MEA
Net hot utility requirement MJ/kgCO2 3.02 3.67
Specific cooling demand MJ/kgCO2 3.06 3.61 processes, respectively. Furthermore, the reboiler temperature in the
Specific power demand MJ/kgCO2 0.44 0.05 HPC process is about 10 ◦ C lower than in the MEA process, owing to the
Reboiler temperature ◦
C 116.6 126.7 lower (optimal) stripper pressure. In addition, due to the higher
Lean loading mol/mol 0.21 0.20
absorber operating pressure in the HPC process, the specific packing
Rich loading mol/mol 0.78 0.50
Solvent makeup kg/tCO2 45.77 584.26 (absorber and stripper) volume requirement is around 59% smaller, and
Cyclic capacity molCO2/kg 1.26 1.37 the required solvent makeup is one order of magnitude lower (coupled
solvent with lower solvent volatility and water evaporation at higher pressures)
Liquid-to-gas ratio (absorber) kg/kg 5.23 3.48 than in the MEA process. However, the HPC process has a slightly lower
Solvent standard volumetric flow rate m3/h 1586 1369
cycling capacity and requires a 50% larger liquid flow in the absorber
Specific packing volume m3/tCO2/h 6.31 15.55
(absorber and stripper) (on a mass basis), which increases the pressure drop across the packing
Specific heat exchanger area m2/tCO2/h 364.6 333.6 (thereby increasing power consumption). The compression and lique
Total specific energy demand for CO2 capture MJ/kgCO2 3.46 3.73 faction processes add 0.37 MJ/kgCO2 in power demand to the CO2
Including a CO2 compression and liquefaction section (0.37 MJ/kgCO2)
capture processes.
Recoverable excess heat (47–86 ◦ C) from CO2 MJ/kgCO2 0.99 0.58
capture, compression, and liquefaction units
As shown in Fig. 6, the GCC of the HPC process is almost linear below
Total specific power demand MJ/kgCO2 0.82 0.41 the pinch point, as compared with the MEA process, which enables
Total specific cooling demand MJ/kgCO2 4.74 4.27 significantly higher levels of heat recovery for the DH network at the two
Total specific energy (heat and power) MJ/kgCO2 3.84 4.09 defined levels, i.e., high-grade heat (61◦ –86 ◦ C) and low-grade heat
demand for CO2 capture, compression, and
(46◦ –61 ◦ C). Note that the amount of high-grade heat recovered for DH
liquefaction
delivery is roughly 65% higher for the HPC process than for the MEA
process. The heat recovery estimated for each CCS process in Fig. 6 is
primarily for DH delivery. Stream data for the GCCs are listed in the then used to estimate the CHP-CCS plant performance in the following
Supplementary Material (see Section S.2). section.
As expected, the SRD is lower for the HPC capture process than for
the MEA process. Compared to the SRD, the net hot utility requirement 3.2. Performance of the CHP-CCS plant
(steam) remains unchanged for the MEA process, while it is reduced to
3.02 MJ/kgCO2 for the HPC process when considering the recoverable 3.2.1. CHP-CCS plant boundary
excess heat from the capture plant alone. The net steam requirement is Fig. 7 shows the energy performances of the CHP-CCS cases (see
lower for the HPC process due to the heat that can be recovered from the Table 3) within their plant boundaries, as compared to the reference
intercooler of the two-stage flue gas compressor. The first stage of the plant without CCS. The figure indicates significant differences between
compressor is driven by steam, whereas the second stage is driven by the two capture processes. For the MEA process, the energy required for
Fig. 6. Grand composite curves (with ΔTmin = 5 ◦ C) of the a) MEA process and b) HPC process, including the compression and liquefaction units. The solid blue line
indicates the cold utility curve, indicating the three cold utilities, i.e., cooling water and high- and low-grade heat recovered for the DH water. The solid red line
indicates the hot-utility curve (low-pressure steam) that is consumed at a slightly higher temperature in the MEA process (126.6 ◦ C) than in the HPC process (116.6
◦
C). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
11
Fig. 7. A) net electric power outputs and shares of the net heat outputs from different heat recovery units in the chp-ccs cases, and the corresponding energy ef
ficiencies (hhv) (eq.(7)) compared to the reference CHP plant without CCS units. b) Power preservation calculated as per Eq. (8), and energy delivered in the CHP-
CCS cases compared to the reference bio-CHP plant.
the capture process corresponds primarily to the internal steam con plants, which have COPs in the range of 3.3–3.5 [27]. Therefore, inte
sumption level, resulting in a lower turbine condenser heat output than grating a heat pump (CHP-MEA-HP) results in a higher energy output
in the CHP-HPC case. Similarly, increased internal consumption of than the CHP-HPC case and roughly the same energy output as the
electric power for flue gas compression in the CHP-HPC case results in reference plant without CCS, as shown in Fig. 7a. The energy delivered
lower electric power output than in the CHP-MEA case. As a result, the compared to the reference bio-CHP plant (%), shown in Fig. 7b, could be
total energy outputs for the CHP-MEA and CHP-HPC cases are similar, in alternatively presented as the energy penalty (as defined in Eq. (9)).
the range of 345–366 MW, with different power-to-heat ratios in the two Thus, the resulting energy penalty for the CHP-MEA-HP is approxi
cases. Fig. 7a and Fig. 7b also show the energy efficiencies of the CHP- mately − 1%, indicating a similar or slightly higher energy output
CCS cases and the energy penalties (as defined by Eq. (7)). The MEA compared with the reference CHP plant without CCS. The difference lies
case incurs the highest energy penalty (roughly ~ 15%). Thus, the HPC in the energy mix of the total energy output, where the total heat
process (with a significantly lower energy penalty) is more favorable for delivered is roughly 21% higher than in the reference CHP plant (and
a bio-CHP plant that operates as a baseload unit in a DH system. the electric power delivered is correspondingly lower). From the CHP
However, the higher levels of preserved power (roughly 80%, see plant operator’s perspective, using the energy penalty as a figure of
Fig. 7b) in the CHP-MEA plant could be used internally to upgrade the merit to compare different capture technologies, it is clear that the CO2
low-grade heat (i.e., unutilized below the DH return temperature of 46 capture technology (HPC process) that minimizes DH delivery losses is
◦
C) to the desired DH supply temperature of 86 ◦ C using a heat pump the preferred option.
(CHP-MEA-HP). The heat pump COP was estimated at 2.54, considering In contrast, the electric power output retained by the MEA process in
that the unutilized low-grade heat (in the range of 26◦ –47 ◦ C) is the CHP-MEA plant could be consumed on-site to increase DH delivery
upgraded entirely using the preserved electric power in the CHP-MEA or delivered to the electricity grid. The choice between the two CO2
case. Furthermore, a heat pump with higher efficiency would result in capture technologies comes down to electricity consumption in heat
a lower specific power demand to upgrade the same limited amount of pumps versus flue gas compressors in the HPC process, followed by heat
unutilized low-grade heat, resulting in an increased total energy output recovery in the new CCS units. In any case, the CHP-MEA-HP would be
from the CHP-CCS plant. Thus, the estimated COP is deemed to be rather the most-effective solution to maximize heat delivery to the DH system,
conservative compared to similar heat pumps operating within CHP as shown in Fig. 7. The exergy performances of the CHP-CCS cases
Fig. 8. Exergy efficiencies of the CHP-CCS plant cases within their plant boundaries a) and extended system boundaries b), with GSHPs operating with Carnot
efficiencies ηCarnot in the range of 40%–60%.
12
within their plant boundaries are presented in the following section to is consumed locally in decentralized GSHPs (CHP-MEA), followed by the
highlight the differences that emerge when the system boundary is case where it is consumed on-site in centralized large-scale heat pumps
extended to the end-users of the delivered heat. (CHP-MEA-HP), and finally, the CHP-HPC case, which exhibits the
lowest system exergy efficiency owing to its lower power-to-heat ratio.
3.2.2. Extended system boundary These results show that high levels of DH delivery can be achieved with
Fig. 8a shows the exergy performances of the CHP-CCS cases within the uptake of GSHPs into the local energy systems, which can use the
their plant boundaries, where the CHP-MEA case has the highest exergy retained electric power output in power-to-heat technologies to avoid
efficiency (approximately 35%), primarily due to its higher power DH distribution losses. Thus, further emphasizing that the CO2 capture
preservation compared to the other two cases. It is important to note technology (MEA) that retains the electric power output of a bio-CHP
here that only a fraction of the preserved electric power output in the plant is preferable to the HPC process. It is evident here that using
CHP-MEA case is consumed (see Fig. 7) to drive large-scale heat pumps exergy as a figure of merit leads to divergence in the optimal CHP-CCS
on-site (CHP-MEA-HP case) to have the same exergy output as the CHP- plant configuration (indicated in Section 3.2.1) when viewed from
HPC case (see Fig. 8a). As a result, the total DH output of the CHP-MEA- both the CHP plant operator’s (or investor’s) perspective and the local
HP case is about 12% and 41% higher than CHP-HPC and CHP-MEA DH system’s perspective. In addition, consumers of end-use heating
cases, respectively. This is an example of the product flexibility inher services in this local DH system could optimize their costs by connecting
ently offered to a BECCS plant with higher exergy efficiency. The bio- to the DH network or choosing competing domestic heat pump tech
CHP plant fitted with the MEA process and large-scale heat pumps nologies. Alternatively, they could optimize their indoor climate de
within their plant boundary has greater availability of higher-exergy mand patterns by complementing their existing heating systems with
energy carrier, electricity, that could be strategically used in heat heat pumps.
pumps to increase the total DH output or delivered to the electricity grid.
Here, product flexibility is defined as the ability of the bio-CHP plant to 4. Discussion
vary the output load of a specific product by adapting the product ratios,
i.e., between heat, power, and negative CO2 emissions. It is important to Bio-CHP plants operating in a DH system are considered highly
recall that this study adopts a conservative approach by evaluating the suitable for extensive BECCS deployment to enable carbon dioxide
MEA capture technology with a 90% capture rate, which could, in re removal from the atmosphere. However, in a DH system, the values
ality, also have the possibility to capture beyond 90%, if required, with a allocated to the different energy services/carriers of an existing bio-CHP
marginal increase in specific capture costs. In addition, the benchmark plant, i.e., heat and electricity, are different at the plant level than at the
amine solvent considered here incurs a much higher SRD (3.67 MJ/kg system level, which includes the consumers. This difference in value
CO2, see Table 4) at a 90% capture rate than state-of-the-art amine allocation is expected, as an energy carrier with higher exergy, i.e.,
solvent blends (for example, the blend of 2-amino-2-methyl-1-propanol electricity could be utilized for several different purposes (heating and
(AMP) and piperazine (PZ), incurred roughly 3.41–3.54 MJ/kgCO2 at a cooling, but also any other power demands in households or industrial
capture rate of 97–99% [36]). However, a higher capture rate (>90%) is applications) in the local energy system. In contrast, from the bio-CHP
not an option applicable to the HPC process because significant amounts plant operator’s perspective, district heat delivery is prioritized over
of exergy need to be expended to drive the flue gas compressors and electricity power production due to the pre-existing local monopolies of
operate the capture units at higher pressures. CHPs in the Swedish DH system [30], while electric power is traded in
Fig. 8b shows the exergy efficiencies of the three CHP-CCS cases the power market and is subject to competition and price volatility [27].
estimated with the extended system boundary approach. The maximum Thus, from the perspective of a bio-CHP plant operator in a DH system,
values in the box plots indicate the exergy efficiencies estimated for a the optimal carbon capture technology for large-scale BECCS imple
DH supply temperature of 86 ◦ C. The minimum and median values mentation would be the technology that minimizes loss of heat sales.
represent the exergy efficiencies corresponding to typical end-use supply Therefore, several different options arise regarding how such a DH
temperatures, in the range of 30◦ –60 ◦ C for space heating and domestic system could be optimized to implement BECCS efficiently and cost-
hot water, respectively. Overall, the CHP-MEA case yields the highest effectively to meet national or regional climate goals.
exergy efficiency when heat is assumed to be delivered within the end-
use supply temperature range. This result is expected due to the higher 4.1. Carbon capture processes
retained electric power output within the CHP-MEA plant boundary,
resulting in higher end-use energy output in the extended system The rigorous numeric modeling of HPC and MEA capture processes
boundary. While the difference in exergy efficiency between CHP-HPC reveals the inherent differences between the two processes when applied
and CHP-MEA-HP cases is minimal, the total DH heat delivered at the to the same flue gas stream with the same capture rate. The differences
DH supply temperature of 86 ◦ C is roughly 12% higher in the CHP-MEA- were quantified using the technology performance indicators listed in
HP cases with the centralized heat pump, as depicted by the maximum Table 4. These technology performance indicators could be used to es
value in the box plots. Fig. 8b illustrates the expected system exergy timate specific investment and operational costs per tonne of CO2
efficiency range, with assumed Carnot efficiency (ηCarnot) in the range of captured from the two capture processes, which could be retrofitted to
40–60% for the GSHPs. similar bio-CHP plants. The results indicate a marginally higher specific
At an end-use temperature of 60 ◦ C, the system exergy efficiency is heat exchanger area requirement for the HPC process than the MEA
roughly 3 percentage points higher for the CHP-MEA plant than for the process (due to a higher rate of solvent circulation), while smaller col
CHP-HPC and CHP-MEA-HP cases at 60% (ηCarnot). However, this dif umn volumes are required due to the higher operating pressures applied
ference in system exergy efficiency for the CHP-MEA plant diminishes to in the HPC process. Thus, it is recommended to quantify the spatial
1 percentage point at 40% (ηCarnot), indicating lower levels of end-use footprints of these two capture technologies when performing more-
heat delivered. Nevertheless, the COP estimated for the upper bound advanced feasibility studies.
of the end-use supply temperature of 60 ◦ C is in the range of 2.7–4.0 Advanced process configurations, such as those involving absorber
(ηCarnot ~ 40–60%) (see Supplementary Materials S.3), which is intercooling, rich-solvent splitting, and vapor compression, although not
reasonable for commercial GSHPs. Additional exergy efficiency plots are considered in this work, could reduce the specific energy requirements
presented in Supplementary Materials S.4, for all the cases, within their and, thereby, the operational costs of individual capture plants. In
plant boundaries, and with the extended system boundaries, with addition, the HPC process could operate at higher temperatures, i.e., the
varying supply temperatures and Carnot efficiencies (ηCarnot). The sys absorber and stripper could operate close to the solvent’s boiling point,
tem exergy efficiency is highest when the retained electric power output thereby reducing the SRD of the stripper. However, a high-temperature
13
configuration is typically suitable for industrial processes with high flue seasonal variations in electricity and DH demands is expected to play a
gas temperatures at the stack and no flue gas condensers in place, so it is significant role in minimizing the specific carbon capture costs. For
deemed unsuitable for a CHP plant with flue gas condensers in place to example, a recent study carried out by Eliasson et al. [25] has reported
maximize heat delivery to its DH system. comparable specific capture costs (in €/tCO2) for a capture plant with
The MEA capture process modeled in this work is a conventional seasonally varying (partial) capture rate that utilizes excess heat alone
amine-based solvent with an estimated SRD of 3.67 MJ/kgCO2, which is while maintaining the DH supply and for a capture plant that operates at
significantly higher than proprietary advanced amine solvents tested in constant load (full capture) at the expense of the DH supply. Thus, the
recent years. For example, proprietary advanced amine solvents such as design and operation (utilization hours, capture rate) are expected to
Shell Cansolv [67], Hitachi H3-1 [68], and MHI – KS1 [69] are reported confer further flexibility when the CO2 capture plant is operated,
to have specific heat demands in the range of 2.3–2.5 MJ/kgCO2 for a depending on varying heating, cooling, and low-carbon electricity de
typical capture rate of 90%. As mentioned previously, capture rates of mands, as well as the scale of the reference plant. Aspects mentioned
up to 99% have been demonstrated with the CESAR-1 solvent, which is a above, such as the operational flexibility of a CHP-CCS plant in relation
blend of 2-amino-2-methyl-1-propanol (AMP) and piperazine (PZ), to its local DH system, with seasonal variations in heat loads and tem
resulting in a specific heat demand in the range of 3.41–3.54 MJ/kgCO2 peratures, were outside the scope of this work. However, it is expected
[36]. The higher capture rates (~97–99%) achieved with the CESAR-1 that the overall conclusion of this work will be unaffected, owing to the
solvent at a relatively similar range of specific energy demands as the various possibilities that inherently come with the availability of high-
benchmark MEA solvent highlights the development of amine-based exergy energy-carrier, i.e., electricity, in a BECCS plant with higher
capture processes. In contrast, the HPC process is very limited with preservation of electric power production capacity.
respect to capture rates >90%, which would require significantly higher Comparing the cases (see Table 3) investigated in this work, it is clear
levels of energy being expended, in the form of heat and electricity, in that the CHP-MEA-HP plant could achieve a higher level of heat delivery
the reboiler and the flue gas compressor to attain high operating pres by including a heat pump, which could alternatively switch to a higher
sures. The possibility to capture beyond 90% with advanced amine- power output by shutting down heat pumps, depending on fluctuations
based solvents, with minimal increase in operational costs, could be in the heating and cooling demands over the year. Thus, the CHP-CCS
essential for a large-scale BECCS plant when accounting for the actual plant could be operated as either a net-electric power exporter or
amount of CO2 ultimately removed from the atmosphere over its importer depending on high or low electricity price periods, as previ
extended cradle-to-grave system boundary that includes the impact of ously discussed by Levihn [27]. Furthermore, higher operational and
indirect land use change [70]. product flexibility, as defined by Beiron et al. [72], is expected for the
CHP-MEA-HP plant. For example, the higher level of preserved electric
4.2. CHP-CCS plant boundary power output in the CHP-MEA plant could be utilized optimally with
heat pumps to meet the peak DH demand during winter months, and this
At the plant level, the results presented in this work extend the work could partially reduce dependence on heat-only boilers and, alterna
of Gustafsson et al. [7], who investigated retrofitting a bio-CHP plant for tively, provide low-carbon electricity, district cooling, and CO2 removal,
CO2 capture using the HPC capture technology. Additional aspects as services during the summer months. An analysis to identify a cost-
considered in this work include: i) a comparison of the performance of optimal capture solution that minimizes specific CO2 removal costs
the HPC process with that of an amine-based (MEA) capture technology; and maximizes profit with corresponding operational strategies for the
ii) an evaluation of the integration of heat pumps to upgrade low-grade CHP-CCS plant cases (see Table 3) was outside the scope of this work.
heat that is rejected from the capture process. A comparison of the en Future work could focus on cost-optimization of operational and product
ergy penalties for the two capture processes at the plant level indicates a (heat, power, and negative-CO2 emissions) flexibility measures available
higher penalty for the MEA process (~24%) than for the HPC process to a future BECCS plant. Cost-optimization based on these aspects could
(~19%). However, the results presented in Section 3.2 indicate that the be valuable to the future BECCS plant operator, as the proposed state aid
choice of capture technology is likely to be sub-optimal if the capture for BECCS via the reverse auctioning system [73] in Sweden would be
technology is chosen based on energy efficiency, as defined in Eq. (7), awarded to an industry or a bio-CHP plant that incurs the lowest costs.
within the plant system boundary. Although it is clear that the HPC
process has higher levels of available excess heat in the CO2 capture and 4.3. Extended system boundary
conditioning units compared to the MEA process, the higher power
output (with low-carbon intensity) of the plant equipped with an MEA Several factors in the extended system boundary could influence the
process could be exploited within the plant boundary to drive heat optimal choice of carbon capture technology for the CHP-CCS, as
pumps, which would ultimately deliver more heat to the DH system. For compared to the rational choice of adopting the plant boundary. One of
example, within the plant boundary, the CHP-MEA-HP case with an on- the external factors explored in this work is the performance of GSHPs
site heat pump delivers more heat to the DH system than the CHP-HPC that are operating within the extended system boundary. The results
case while having the same electric power output to the electricity grid. indicate a diminishing gap between the exergy efficiency of the CHP-
These conclusions are not in accordance with the case made for the HPC MEA case with higher power preservation (80.6%) and its counterpart
capture process within plant boundaries by Gustafsson et al. [7] and cases (CHP-HPC and CHP-MEA-HP) with lower power preservation
highlight the limitations of the isolated comparisons presented in the (43.2%), as shown in Fig. 7b, with decreasing COP values for the GSHPs
literature for different carbon capture technologies applied to CHP in the extended system boundary. In general, this means that, compared
processes that operate in a different context concerning their local en to a reference plant that would typically incur DH transmission losses in
ergy system. the DH network, with this hypothetical extended system boundary, the
It should also be noted that the reference bio-CHP plant in Stockholm losses would occur locally in the GSHPs, in the form of conversion losses.
is an exception in relation to other bio-CHP plants operating in many DH Nevertheless, the lower-bound COP values estimated in this work (range
systems in Sweden since it is a very large-scale CHP plant that delivers of 2.7–4.0, for ηCarnot of 40–60%) are fairly conservative compared to the
roughly 588 GWh of electricity and 1,468 GWh of heat to its local DH COPs reported for commercial GSHPs, which are in the range of 3.1–3.8
system annually, with roughly 6,000 full-load hours [71]. Thus, for a temperature difference of 50 ◦ C [74]. Therefore, the inference
compared to other smaller bio-CHP plants in the region, this reference made above holds for the extended system boundary defined in this
CHP plant, with large CO2 flows (approximately 800 ktCO2/yr) and a work (see Section 2.2.2). Alternatively, if the delivered electric power is
higher number of full-load hours, is expected to incur relatively lower consumed elsewhere for an energy consumption end-use other than
specific capital costs for the capture plant [3]. In addition, the effect of indoor space heating, the exergy efficiency of the overall system would
14
always be higher for the case with the highest power preservation, i.e., 90% capture rate) from the two capture technologies, the results of this
the CHP-MEA plant. Furthermore, from a Swedish perspective, elec work indicate the following:
tricity consumers are subject to considerably higher energy taxes than
those levied on the power producers, who use the power themselves. • From an energy perspective, for a bio-CHP plant, the MEA capture
Thus, providing further incentives to opt for a BECCS plant that operates process incurs an energy penalty (defined as the change in the total
with an MEA capture unit, complemented with a large-scale heat pump energy efficiency of a CHP plant, considering heat and electric power
on-site to recover excess heat. This configuration could then be operated output, resulting from the integration of CO2 capture processes) of
optimally based on the seasonal variations in the heat and power de 15%–16%, which can be attributed to the low-temperature steam
mands of the local DH system. required to provide the heat need of the reboiler in the capture
In the extended system boundary strategy, the uptake of GSHPs is plantr, which in turn reduces the total heat output of the BECCS
also expected to face technical and regulatory challenges owing to the plant. In contrast, the HPC process results in an energy penalty of
risk of net cooling of the ground in urban areas, as argued by Åberg et al. 9%–10%, albeit with significant electricity consumption in the flue
[32], whereby the regulations regarding distances between wells in gas compressors prior to the capture process.
urban areas are identified as critical for the potential competition be • The MEA process has a higher level of power preservation (defined as
tween GSHPs and DH. Nevertheless, despite the current regulations on the ratio of the electric power production capacity of a CHP plant
distances between wells, zones with competition between GSHPs and with CCS to that of the same plant without CCS) than the HPC pro
DH still exist in urban areas, such as Stockholm. In addition, the heat cess. Thus, introducing an on-site heat pump to recover low-
demand density (HDD), or the ratio of the buildings’ DH demand to the temperature heat with a conservative COP (~2.5) using the pre
land area that the buildings cover (kWh/m2), could determine which served power will lead to a significantly higher heat output (~12%).
CO2 capture technology is optimal or, instead, which energy service The corresponding energy penalty is − 0.6%, indicating that the heat
should be prioritized in a future BECCS plant. For example, CHP plants output is larger than that of the existing bio-CHP plant without the
operating in sparsely built areas yield a larger competitive region for CCS units.
GSHPs, as defined by Åberg et al. [32], and therefore have considerable • The exergy efficiency of the bio-CHP plant is around 23% higher in
incentive to preserve exergy with an MEA process, as compared to CHP the MEA process (~35%) than in the HPC process (~26%) due to the
plants operating in densely built areas with high HDD, where there is higher level of power preservation in the former within the plant
limited access to heat sources for GSHPs. In densely built areas, within system boundary. The same trends are observed when adopting the
regions in which there is competition with DH, other heat pumps such as extended system boundary perspective for the BECCS plant with a
air-source and ventilation or exhaust air heat pumps could be alternative decentralized DH system, considering the use of GSHPs in the local
solutions to meeting end-use heating needs with the preserved electric DH system to maximize the amount of heat delivered to the system.
power output from the CHP-MEA plant. Another aspect that could be In this context, the difference in the exergy efficiencies of the two
crucial in deciding on the optimal capture technology for a bio-CHP carbon capture technologies depends on how efficiently the GSHPs
plant in the extended system boundary strategy, from the consumer’s operate. Assuming a Carnot efficiency of 60% and a desired end-use
perspective, is the environmental risk associated with the capture sol target temperature in the range of 30◦ –60 ◦ C (estimated COP ~ 4.0 at
vent used. A comparative lifecycle assessment between MEA and po 60 ◦ C), the MEA process would have a 10%–15% higher exergy ef
tassium carbonate solvent [41] has indicated the significantly superior ficiency than the HPC process, which would decrease to 4%–8% if a
performance of the potassium carbonate solution for all environmental conservative Carnot efficiency of 40% (COP ~ 2.7 at 60 ◦ C) were to
indicators considered in the study, including global warming, water use, be assumed.
ecotoxicity, and carcinogenic emissions.
According to these results, the MEA capture process clearly demon
5. Conclusions strates a significant advantage over the HPC process for enabling BECCS,
as it inherently offers greater product flexibility (i.e., the ability of the
This study compared the performance of two well-established but plant to vary a specific product output) with higher preservation of the
inherently different carbon capture processes (the hot potassium car electric power production capacity. Although a conservative approach
bonate (HPC) process and the amine-based (MEA) process) for the with benchmark amine solvent was taken in this work, recent studies
capture of 90% of the biogenic CO2 emissions from a typical flue gas with advanced amine solvents have demonstrated that these solvents
stream of a bio-CHP plant in a DH system that mainly operates to pro can also provide an opportunity to have capture rates up to 99% with
vide heat as its primary energy service. In this context, a future trans minimal increase in specific energy demand, unlike the HPC process that
formation to a BECCS plant that additionally provides carbon dioxide reaches a techno-economic threshold at a capture rate of 90%. In
removal (CDR) as a societal climate service will require careful selection addition, these state-of-the-art amine-based solvents have relatively
of the carbon capture technology that minimizes the costs associated lower environmental impacts than the benchmark MEA solvent.
with implementing BECCS, for both the energy service provider and the Considering energy performance as the primary figure of merit for
consumers of this energy service. different carbon capture technologies for a future BECCS plant could
A detailed process model of a large-scale bio-CHP plant in a DH lead to sub-optimization of the societal services of a bio-CHP plant,
system was developed and integrated with rigorous, rate-based carbon incurring a higher economic cost for CDR as a service to Society. How
capture process models for the two capture processes, including the CO2 ever, the optimal choice of carbon capture technology for a bio-CHP
compression and liquefaction train. The energy and exergy performance plant in a DH system will also depend on factors other than those
of the respective BECCS plants were evaluated within the plant bound considered here, such as techno-economics, the DH network density,
ary as well as its extended system boundary, considering that the seasonal variation in DH and power demands, and safety factors such as
delivered heat and power are used to meet the end-use heating needs in ecotoxicity, as well as other environmental repercussions of the capture
the local DH system. In addition, the integration of a centralized heat solvents used in these carbon capture technologies.
pump at the BECCS plant site to recover low-temperature heat from the
BECCS plant was compared with a decentralized heat pump setting, CRediT authorship contribution statement
using ground heat as the heat source. This was based on the possibility
that the energy consumer might opt for increasing electricity usage to Tharun Roshan Kumar: Conceptualization, Methodology, Soft
meet their indoor-climate needs, such as space heating and domestic hot ware, Data curation, Visualization, Validation, Writing – original draft.
water. Considering the same level of negative-emissions capacity (at Johanna Beiron: Methodology, Software, Visualization, Validation,
15
Writing – original draft, Writing – review & editing. Maximilian Bier [14] Djurberg R. Practical implementation of Bio-CCS in Uppsala: a techno-economic
assessment, M.Sc. Thesis. Royal Institute of Technology, Stockholm, 2020. http://
mann: Methodology, Software, Visualization, Validation, Writing –
urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-277820.
original draft, Writing – review & editing. Simon Harvey: Funding [15] Hammar C. Heat integration between CO2 Capture and Liquefaction and a CHP
acquisition, Project administration, Supervision, Writing – review & Plant Impact on Electricity and District Heating Delivery at Renova’s CHP Plant in
editing. Henrik Thunman: Conceptualization, Methodology, Funding Sävenäs, M.Sc. Thesis, Chalmers University of Technology, Gothenburg 2022.
https://hdl.handle.net/20.500.12380/304511 (accessed June 22, 2022).
acquisition, Supervision, Writing – review & editing. [16] Jilvero H. Ammonia as an Absorbent of Carbon Dioxide in Post-Combustion
Capture-an Experimental, Technical and Economic Process Evaluation, Doctoral
Thesis 2014. https://research.chalmers.se/en/publication/201381.
[17] Augustsson O, Baburao B, Dube S, Bedell S, Strunz P, Balfe M, et al. Chilled
Declaration of Competing Interest Ammonia Process Scale-up and Lessons Learned. Energy Procedia 2017;114:
5593–615. https://doi.org/10.1016/j.egypro.2017.03.1699.
The authors declare that they have no known competing financial [18] Garđarsdóttir SÓ, Normann F, Andersson K, Johnsson F. Postcombustion CO2
capture using monoethanolamine and ammonia solvents: The influence of CO2
interests or personal relationships that could have appeared to influence concentration on technical performance. Ind Eng Chem Res 2015;54:681–90.
the work reported in this paper. [19] Jilvero H, Normann F, Andersson K, Johnsson F. Heat requirement for regeneration
of aqueous ammonia in post-combustion carbon dioxide capture. Int J Greenh Gas
Control 2012;11:181–7. https://doi.org/10.1016/j.ijggc.2012.08.005.
Data availability
[20] Jilvero H, Normann F, Andersson K, Johnsson F. Thermal integration and
modelling of the chilled ammonia process. Energy Procedia 2011;4:1713–20.
Data will be made available on request. https://doi.org/10.1016/j.egypro.2011.02.045.
[21] Jilvero H, Normann F, Andersson K, Johnsson F. Ammonia-based post combustion
– The techno-economics of controlling ammonia emissions. Int J Greenh Gas
Acknowledgment Control 2015;37:441–50. https://doi.org/10.1016/J.IJGGC.2015.03.039.
[22] Pröll T, Zerobin F. Biomass-based negative emission technology options with
combined heat and power generation. Mitig Adapt Strateg Glob Chang 2019;24:
This work was carried out within the framework of the project - 1307–24. https://doi.org/10.1016/B978-0-88415-220-0.X5000-9.
Transformative change towards net negative emissions in Swedish refinery [23] Kärki J, Tsupari E, Arasto A. CCS feasibility improvement in industrial and
and petrochemical industries (FUTNERC), which is a collaboration between municipal applications by heat utilization. Energy Procedia 2013;37:2611–21.
https://doi.org/10.1016/j.egypro.2013.06.145.
Chalmers University of Technology, Borealis AG, and Preem AB, with [24] Eliasson Å, Fahrman E, Biermann M, Normann F, Harvey S. Integration of
funding provided by the Swedish Energy Agency (Project 49831-1), industrial CO2 capture with district heating networks: a refinery case study. TCCS-
Borealis AG and Preem AB. 11 2021:197–201. https://www.sintef.no/globalassets/project/tccs-11/tccs-11/
sproceedings-no-7.pdf (accessed January 16, 2023).
[25] Eliasson Å, Fahrman E, Biermann M, Normann F, Harvey S. Efficient heat
Appendix A. Supplementary data integration of industrial CO2 capture and district heating supply. Int J Greenh Gas
Control 2022;118:103689. https://doi.org/10.1016/j.ijggc.2022.103689.
[26] Abrami G. Energy targeting for heat recovery from carbon capture processes using
S.1 Hot potassium carbonate model–reactions list and model vali hybrid absorption heat pumps, M.Sc. Thesis, Politecnico Di Milano, Milan, 2022.
dation. S.2 CO2 compression and liquefaction processes. S.3 Estimation http://hdl.handle.net/10589/187439 (accessed July 8, 2022).
of the coefficients of performance of heat pumps. S.4 Comparison of [27] Levihn F. CHP and heat pumps to balance renewable power production: Lessons
from the district heating network in Stockholm. Energy 2017;137:670–8. https://
exergetic performances of the CHP-CCS cases. Supplementary data to doi.org/10.1016/J.ENERGY.2017.01.118.
this article can be found online at https://doi.org/10.1016/j. [28] Beiron J, Göransson L, Normann F, Johnsson F. Flexibility provision by combined
apenergy.2023.120927. heat and power plants – An evaluation of benefits from a plant and system
perspective. Energy Convers Manag X 2022;16:100318. https://doi.org/10.1016/
J.ECMX.2022.100318.
References [29] Beiron J. Combined heat and power plants in decarbonized energy systems,
Doctoral Thesis. Chalmers University of Technology, Gothenburg 2022. https://
research.chalmers.se/en/publication/531778.
[1] Shukla PR, Skea J, Slade R, Khourdajie A Al, van Diemen R, McCollum D. IPCC,
[30] Werner S. District heating and cooling in Sweden. Energy 2017;126:419–29.
2022: Climate Change 2022: Mitigation of Climate Change. Contribution of
https://doi.org/10.1016/j.energy.2017.03.052.
Working Group III to the Sixth Assessment Report of the Intergovernmental Panel
[31] Egüez A. District heating network ownership and prices: The case of an
on Climate Change. Cambridge University Press, Cambridge, UK and New York,
unregulated natural monopoly. Util Policy 2021;72:101252. https://doi.org/
NY, USA.; 2022. 10.1017/9781009157926.
10.1016/J.JUP.2021.101252.
[2] Ministry of the Environment and Energy. The Swedish Climate Policy Framework
[32] Åberg M, Fälting L, Lingfors D, Nilsson AM, Forssell A. Do ground source heat
2020:1–5. https://doi.org/10.5040/9781509941742.ch-007.
pumps challenge the dominant position of district heating in the Swedish heating
[3] Beiron J, Normann F, Johnsson F. A techno-economic assessment of CO2 capture in
market? 2020;254. 10.1016/j.jclepro.2020.120070.
biomass and waste-fired combined heat and power plants – A Swedish case study.
[33] Swedish Energy Agency. Värmepumparnas roll på uppvärmningsmarknaden -
Int J Greenh Gas Control 2022;118:103684. https://doi.org/10.1016/j.
Utveckling och konkurrens i ett föränderligt energisystem (Heat pumps’ role in the
ijggc.2022.103684.
heat market - Development and competition in a changing energy market) 2015.
[4] Fuss S, Johnsson F. The BECCS Implementation Gap – A Swedish Case Study. Front
[34] Skatteverket. Energy tax on electricity n.d. https://www.skatteverket.se/foretag/
Energy Res 2021;8:385. https://doi.org/10.3389/fenrg.2020.553400.
skatterochavdrag/punktskatter/energiskatter/
[5] IEA Bioenergy. Large biomass CHP plant in Stockholm Sweden 2018. https://www.
skattpael.4.15532c7b1442f256bae5e4c.html (accessed September 6, 2022).
ieabioenergy.com/wp-content/uploads/2018/02/8-LargeCHP-Värtaverket_SE_
[35] Biermann M. Partial CO2 capture to facilitate cost-efficient deployment of carbon
Final.pdf (accessed June 22, 2022).
capture and storage in process industries, Doctoral Thesis. Chalmers University of
[6] Stockholm Exergi’s project for negative emissions receives EU support - Stockholm
Technology, Gothenburg 2022. https://research.chalmers.se/en/publication/
Exergi n.d. https://www.stockholmexergi.se/blogg/news-stockholmexergi/
531680.
stockholm-exergis-project-for-negative-emissions-receives-eu-support/ (accessed
[36] Hume S, Shah MI, Lombardo G, Kleppe ER. Results from CESAR-1 testing with
June 22, 2022).
combined heat and power (CHP) flue gas at the CO2 Technology Centre Mongstad.
[7] Gustafsson K, Sadegh-Vaziri R, Grönkvist S, Levihn F, Sundberg C. BECCS with
TCCS-11 2021:355–61.
combined heat and power: assessing the energy penalty. Int J Greenh Gas Control
[37] Brandl P, Bui M, Hallett JP, Mac DN. Beyond 90% capture: Possible, but at what
2021;110:103434. https://doi.org/10.1016/j.ijggc.2021.103434.
cost? Int J Greenh Gas. Control 2021:105. https://doi.org/10.1016/j.
[8] Gadd H, Werner S. Achieving low return temperatures from district heating
ijggc.2020.103239.
substations. Appl Energy 2014;136:59–67. https://doi.org/10.1016/J.
[38] Danaci D, Bui M, Petit C, MacDowell N. En Route to Zero Emissions for Power and
APENERGY.2014.09.022.
Industry with Amine-Based Post-combustion Capture. Environ Sci Technol 2021:
[9] Kohl AL, Nielsen R. Gas Purification, Elsevier. Science 1997. https://doi.org/
55. https://doi.org/10.1021/acs.est.0c07261.
10.1016/B978-0-88415-220-0.X5000-9.
[39] Feron P, Cousins A, Jiang K, Zhai R, Shwe Hla S, Thiruvenkatachari R, et al.
[10] Giammarco-Vetrocoke 2022. http://www.giammarco-vetrocoke.com/aboutus_
Towards Zero Emissions from Fossil Fuel Power Stations. Int J Greenh Gas Control
history.htm (accessed October 24, 2022).
2019;87:188–202. https://doi.org/10.1016/j.ijggc.2019.05.018.
[11] CO2CAPSOL 2022. https://www.co2capsol.com/ (accessed October 24, 2022).
[40] Pérez-Calvo JF, Sutter D, Gazzani M, Mazzotti M. Pilot tests and rate-based
[12] Carbon Capture CCUs - CATACARB 2022. https://catacarb.com/applications/
modelling of CO2 capture in cement plants using an aqueous ammonia solution.
carbon-capture-ccus/ (accessed October 24, 2022).
Chem Eng Trans 2018;69:145–50. https://doi.org/10.3303/CET1869025.
[13] Levihn F, Linde L, Gustafsson K, Dahlen E. Introducing BECCS through HPC to the
[41] Grant T, Anderson C, Hooper B. Comparative life cycle assessment of potassium
research agenda: The case of combined heat and power in Stockholm. Energy Rep
carbonate and monoethanolamine solvents for CO2 capture from post combustion
2019;5:1381–9. https://doi.org/10.1016/J.EGYR.2019.09.018.
16
flue gases. Int J Greenh Gas Control 2014;28:35–44. https://doi.org/10.1016/J. method. Chem Eng J 1977;13:7–12. https://doi.org/10.1016/0300-9467(77)
IJGGC.2014.06.020. 80002-6.
[42] Matin NS, Flanagan WP. Life cycle assessment of amine-based versus ammonia- [57] Kohl AL, Nielsen RB. Alkaline Salt Solutions for Acid Gas Removal. Gas Purification
based post combustion CO2 capture in coal-fired power plants. Int J Greenh Gas 1997. https://doi.org/10.1016/b978-088415220-0/50005-7.
Control 2022;113:103535. https://doi.org/10.1016/j.ijggc.2021.103535. [58] Berrouk AS, Ochieng R. Improved performance of the natural-gas-sweetening
[43] Gustafsson K. “Spearheading negative emissions in Stockholm’s multi-energy Benfield-HiPure process using process simulation. Fuel Process Technol 2014;127:
system,” Bioenergy with carbon capture and storage. From global potentials to 20–5. https://doi.org/10.1016/j.fuproc.2014.06.012.
domestic realities. Eur Lib Forum 2017:69–87. 10.1016/b978-0-12-805423- [59] Technology A. Physical Property Methods and Models 2011;11(1):1–19.
9.00004-1. [60] Ayittey FK, Obek CA, Saptoro A, Perumal K, Wong MK. Process modifications for a
[44] Kemp IC. Pinch Analysis and Process Integration: A User Guide on Process hot potassium carbonate-based CO2 capture system: a comparative study. Greenh
Integration for the Efficient Use of. Energy 2019. https://doi.org/10.1016/B978- Gases Sci Technol 2020;10:130–46. https://doi.org/10.1002/GHG.1953.
075068260-2.50007-9. [61] AspenTech. ELECNRTL Rate-Based Model of the CO2 Capture Process by K2CO3
[45] Nikitin A, Deymi-Dashtebayaz M, Muraveinikov S, Nikitina V, Nazeri R, using Aspen Plus - Version 12.1. n.d.
Farahnak M. Comparative study of air source and ground source heat pumps in 10 [62] Wu Y, Wu F, Hu G, Mirza NR, Stevens GW, Mumford KA. Modelling of a post-
coldest Russian cities based on energy-exergy-economic-environmental analysis. combustion carbon dioxide capture absorber using potassium carbonate solvent in
J Clean Prod 2021;321:128979. https://doi.org/10.1016/J. Aspen Custom Modeller. Chinese J Chem Eng 2018;26:2327–36. https://doi.org/
JCLEPRO.2021.128979. 10.1016/J.CJCHE.2018.06.005.
[46] Bravo JL, Rocha JA, Fair JR. Mass Transfer in Gauze Packings. Hydrocarb Process [63] Deng H, Roussanaly S, Skaugen G. Techno-economic analyses of CO2 liquefaction:
1985;64:91–5. Impact of product pressure and impurities. Int J Refrig 2019;103:301–15. https://
[47] Stichlmair J, Bravo JL, Fair JR. General model for prediction of pressure drop and doi.org/10.1016/J.IJREFRIG.2019.04.011.
capacity of countercurrent gas/liquid packed columns. Gas Sep Purif 1989;3: [64] Northern lights 2022. https://norlights.com/ (accessed August 18, 2022).
19–28. https://doi.org/10.1016/0950-4214(89)80016-7. [65] Klemetsrudanlegget and Project CCS Carbon Capture Oslo 2017.https://
[48] Chilton TH, Colburn AP. Mass transfer (absorption) coefficients prediction from ccsnorway.com/app/uploads/sites/6/2019/09/fortum_oslo_varme.pdf (accessed
data on heat transfer and fluid friction. Ind Eng Chem 1934;26:1183–7. August 18, 2022).
[49] Biermann M, Normann F, Johnsson F, Hoballah R, Onarheim K. Capture of CO2 [66] Szargut J, Morris DR, Steward FR. Exergy Analysis of Thermal, Chemical, and
from Steam Reformer Flue Gases Using Monoethanolamine: Pilot Plant Validation Metallurgical Processes. 1988.
and Process Design for Partial Capture. Ind Eng Chem Res 2022;61:14305–23. [67] Singh A, Stéphenne K. Shell Cansolv CO2 capture technology: Achievement from
https://doi.org/10.1021/acs.iecr.2c02205. first commercial plant. Energy Procedia 2014;63:1678–85. https://doi.org/
[50] Biermann M, Normann F, Johnsson F, Skagestad R. Partial Carbon Capture by 10.1016/j.egypro.2014.11.177.
Absorption Cycle for Reduced Specific Capture Cost. Ind Eng Chem Res 2018;57: [68] Ridge B. Testing of Hitachi H3-1 Solvent at the National Carbon Capture Center :
15411–22. Final Report Prepared by : National Carbon Capture Center Wilsonville , AL Hitachi
[51] AspenTech. ENRTL-RK Rate-Based Model of the CO2 Capture Process by MEA using Power Systems America , Ltd . Babcock Hitachi K . K . n.d.
Aspen Plus - Version 10.0. 2017. [69] Zheng RF, Barpaga D, Mathias PM, Malhotra D, Koech PK, Jiang Y, et al. A single-
[52] Notz R, Mangalapally HP, Hasse H. Post combustion CO2 capture by reactive component water-lean post-combustion CO2 capture solvent with exceptionally
absorption: Pilot plant description and results of systematic studies with MEA. Int J low operational heat and total costs of capture-comprehensive experimental and
Greenh Gas Control 2012;6:84–112. https://doi.org/10.1016/J. theoretical evaluation. Energy Environ Sci 2020;13:4106–13. https://doi.org/
IJGGC.2011.11.004. 10.1039/d0ee02585b.
[53] Biermann M, Harvey S, Kjärstad Rahul Anantharaman J, Fu C, Jordal K, Reyes Lúa [70] Tanzer SE, Ramírez A. When are negative emissions negative emissions? Energy
A, et al. Preem CCS Synthesis of main project findings and insights 2022. https:// Environ Sci 2019;12:1210–8. https://doi.org/10.1039/c8ee03338b.
research.chalmers.se/en/publication/528685 (accessed January 16, 2023). [71] Stockholm Exergi. Värtaverket, Stockholm Exergi, miljörapport. Stockholm: 2019.
[54] Zhang Y, Chen CC. Modeling CO2 Absorption and Desorption by Aqueous [72] Beiron J. Combined heat and power plant flexibility, Licentiate Thesis. Chalmers
Monoethanolamine Solution with Aspen Rate-based Model. Energy Procedia 2013; University of Technology, Gothenburg 2020. https://research.chalmers.se/en/
37:1584–96. https://doi.org/10.1016/J.EGYPRO.2013.06.034. publication/516671.
[55] Pinsent BRW, Pearson L, Roughton FJW. The kinetics of combination of carbon [73] State aid for BECCS n.d. https://www.energimyndigheten.se/en/sustainability/
dioxide with hydroxide ions. Trans Faraday Soc 1956;52:1512–20. https://doi.org/ carbon-capture-and-storage/state-aid-for-beccs/ (accessed July 8, 2022).
10.1039/tf9565201512. [74] Staffell I, Brett D, Brandon N, Hawkes A. A review of domestic heat pumps n.d.
[56] Hikita H, Asai S, Ishikawa H, Honda M. The kinetics of reactions of carbon dioxide 10.1039/c2ee22653g.
with monoethanolamine, diethanolamine and triethanolamine by a rapid mixing
17

Fulltext

Uploaded by

Copyright:

Available Formats

Fulltext

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

Fulltext

Uploaded by

Copyright:

Available Formats

Plant and system-level performance of combined heat and power plants

equipped with different carbon capture technologies

Downloaded from: https://research.chalmers.se, 2023-07-18 17:14 UTC

Citation for the original published paper (version of record):

(article starts on next page)

Contents lists available at ScienceDirect

Plant and system-level performance of combined heat and power plants

Nomenclature REF Reference

1. Introduction Bio-CHP plants in Sweden operate, in most cases, within a district

MEA and CHP-HPC) within the CHP-CCS plant boundary.

additional electric power consumed in the centralized large-scale heat

ĖQ,DH + ĖW,el + Ėliq,CO2 + ĖCFG

You might also like