FACULTY OF SCIENCES AND ENGINEERING SCHOOL OF ENGINEERING PhD Thesis PROCESS MODELLING, SIMULATION AND OPTIMISATION OF NATURAL GAS COMBINED CYCLE… [628922]
THE UNIVERSITY OF HULL
FACULTY OF SCIENCES AND ENGINEERING
SCHOOL OF ENGINEERING
PhD Thesis
PROCESS MODELLING, SIMULATION AND
OPTIMISATION OF NATURAL GAS COMBINED CYCLE
POWER PLANT INTEGRATED WITH CARBON CAPTURE,
COMPRESSION AND TRANSPORT
Xiaobo Luo
Supervisor: Prof essor Meihong Wang
May 2016
This thesis is submitted in fulfilment of the requirements for the
degree of Doctor of Philosophy (PhD)
i
The candidate confirms that the work submitted is his/her own, except where work
which has formed part of co -authored publications has been included. Th e contribution
of the candidate and the other authors to this work has been explicitly indicated below.
The candidate confirms that appropriate credit has been given within the thesis where
reference has been made to the work of others.
The work in Chapt er 5 of the thesis has appeared in publication as follows:
Luo, X., Wang, M., Chen, J., 2015. Heat integration of natural gas combined cycle
power plant integrated with post -combustion CO 2 capture and compression. Fuel, 151,
110-117.
Prof M. Wang is my supervisor who provided suggestions regarding the
research direction , analysis methods and commented the paper writing . Prof J.
Chen provided suggestions on the research direction.
The work in Chapter 6 of the thesis has appeared in publication as follows:
Luo, X., Wang, M., Oko, E., Okezue, C., 2014. Simulation -based techno -economic
evaluation for optimal design of CO 2 transport pipeline network. Applied Energy , 132,
610-620.
Prof M. Wang is my supervisor who provided suggestions regarding the
research dir ection , analysis methods and commented the paper writing. Dr
E.Oko and Mr. C.Okezue discussed and commented on this research topic.
The work in Chapter 7 of the thesis has appeared in publication as follows:
Luo, X., Wang, M., 2016. Optimal operation of M EA-based post -combustion carbon
capture for natural gas combined cycle power plants under different market conditions.
International Journal of Greenhouse Gas Control, 48, Part 2 , 312-320.
Prof M. Wang is my supervisor who provided suggestions regarding the
research direction , analysis methods and commented the paper writing.
ii
This copy has been supplied on the understanding that it is copyright material and that
no quotation from the thesis may be published without proper acknowledgement.
The right of Xiaobo Luo to be identified as Author of this work has been asserted by
him in accordance with the Copyright, Designs and Patents Act 1988.
© 201 6 University of Hull and Xiaobo Luo
iii
Acknowledgements
I wish to express my gratitude to all who have supported me throughout my PhD study.
Firstly, I would like to thank Prof Meihong Wang for supervising me through this
project. With his guidance, I enjoyed th is research work and was able to publish several
journal and conference papers in different stages of this project. Moreover, his diligence
and being responsible encourage me to overcome the challenges from both the research
work and the life. I am also thankful to all the members of the Thesis Advisory Panel
(TAP) meetings, including Prof Ron Patton, Dr Ming Hou, Dr Chunfei Wu and Dr
Dipesh Patel, for their support and suggestions for this project.
Secondly, I am very grateful to my wife Hongling Sui. She has made great contribution
to take care of the family during this period. With her sacrifice and support, I was able
to focus on the research work and finish ed this project. I am also glad to see the
growing of my son Junhao Luo and his great understanding about my being busy for
this project . I am als o thankful to my parents for their support and forgiving me being
far away from them to pursue my dreams.
Third ly, I would like to thank all friends around me for your love and supports. We
shared the experience and the joys of our life, which is very pre cious for me during this
period and in the future.
Finally, I would like to thank the funding bodies involved this project. They are: (1)
Natural Environment Research Council (NERC Ref: NE/H013865/2) ; (2) EU FP7
Marie Curie ( Ref: FP7-PEOPLE -2013 -IRSES) ; (3) 2013 China -Europe small -and
medium sized enterprises energy saving and carbon reduction research project (No. :
SQ2013ZOA100002), and (4) Department of Energy and Climate Change (DECC) for
the project Process intensification for carbon capture with new solvents .
iv
Abstract
Reducing CO 2 emission s from fossil fuel -fired power plants is a significant challenge ,
technically and economically. Post-combustion carbon capture (PCC) using amine
solvents is widely regarded as the most promising technology that can be commercially
deployed for carbon capture from fossil fuel -fired power plants . However, for its
application at full commercial scale, th e main barrier is high cost increment of the
electricity due to high capital costs and significant energy penalty. This thesis presents
the studie s on optimal design and operation of Monoethanolamine ( MEA )-based PCC
process and the integrated system with natural gas combined cycle (NGCC) power plant
through modelling , simulation and optimisation , with the aim to reduc e the cost of PCC
commercial deployment for NGCC power plants.
The accuracy of optimi sation depends on good prediction s of both process model and
economic model. For the process model ling, the philosophy with its framework was
analy sed for this reactive absorption (RA) process . Then the model was developed and
validated at three stages . In the first stage, the predictions of thermodynamic modelling
were compared with experimental data of CO 2 solubility in aqueous MEA solutions .
The results show the combi nation of correlations used in this study has higher accuracy
than other three ke y published contributions. Then key physical properties of MEA –
H2O-CO 2 system were also validated with experimental data from different
publications. Lastly, a steady state process model was developed in Aspen Plus® with
rate-based mass transfer and kinetic -controlled reactions. The process model was
validated again st comprehensive pilot plant experiment data, in terms of absorption
efficiency and thermal perform ance of the integrated system.
The cost model was developed based on the major equipment costs provide d by vendors
after detailed engineering design in a benchmark report . The uncertainty of this method
could be in the range of from −15% to 20%, instead of other empirical methods with
uncertainty of from −30% to 50%. The cost model was integrated into the process model
by coding Fortran subroutine in Aspen Plus®. Using this integrated model , the
optimisation studies were carried out for the PCC process only. The impact of key
variables variation was also analyse d.
v
Subsequently , the scope of this study was extended to cover different sections of the
integrated system including a 453MW e NGCC power plant, PCC process, CO 2
compression trains and CO 2 transport pipeline network. For the integration of NGCC
power plant with PCC process and CO 2 compression , exhaust gas recirculation (EGR)
technology was investigated and showed significant economic benefit . A specific
supersonic shock wave compress or was adopted for the CO 2 compression and its heat
integration options with power plant and PCC process were studied.
For the study on the CO 2 transport pipeline network planned in the Humber region of
the UK , a steady stat e process model was developed using Aspen HYSYS®. The
process model was integrated with Aspen Pro cess Economic Analy zer® (APEA), to
carry out techno -economic evaluations for different options of the CO 2 compression
trains and the trunk onshore \offshore pipelines respectively. The results show the
optimal case h as an annual saving of 22.7 M€ compare d with the base case.
In the end, o ptimal operation s of NGCC power plant integrated with whole carbon
capture and storage ( CCS ) chain under different market conditions were studied.
Leveli sed cost of electricity (LCOE) is formulated as the objective function. The
optimal op erations were investigated for different carbon capture level under different
carbon price, fuel price and CO 2 transport and storage ( T&S ) price. The results show
that carbon price needs to be over €100/ton CO 2 to justify the total cost of carbon
capture from the NGCC power plant and needs to be €120/ton CO 2 to drive carbon
capture level at 90%. The results outline the economic profile of op erating an NGCC
power plant integrated with CCS chain. It could help power plants operators and
relevant govern ment organizations for decision makings on the commercial deployment
of solvent -based PCC process for power plans .
Keywords: Process modelling, P rocess simulation, Process optimisation , Post-
combustion c arbon capture, Gas-fired power plant, NGCC, CO 2 pipeline transport,
CCS
vi
Table of Contents
Acknowledgements ………………………….. ………………………….. ………………………….. ……… iii
Abstract ………………………….. ………………………….. ………………………….. ………………………. iv
Table of Contents ………………………….. ………………………….. ………………………….. ………… vi
List of Figures ………………………….. ………………………….. ………………………….. ……………… xi
List of Tables ………………………….. ………………………….. ………………………….. …………….. xiv
Nomenclature ………………………….. ………………………….. ………………………….. …………….. xvi
Abbreviations ………………………….. ………………………….. ………………………….. ………….. xviii
Chapter 1: Introduction ………………………….. ………………………….. ………………………….. 1
1.1 Background ………………………….. ………………………….. ………………………….. ……….. 1
1.1.1 Global warming and greenhouse gases emissions ………………………….. …… 1
1.1.2 CO 2 emission and its reduction ………………………….. ………………………….. .. 2
1.1.3 CCS technology ………………………….. ………………………….. …………………….. 5
1.1.3.1 CO 2 capture ………………………….. ………………………….. ……………………. 6
1.1.3.2 CO 2 transport ………………………….. ………………………….. ………………….. 7
1.1.3.3 CO 2 storage ………………………….. ………………………….. ……………………. 7
1.2 Motivations ………………………….. ………………………….. ………………………….. ……….. 7
1.3 Aim and objectives of this study ………………………….. ………………………….. ………. 9
1.4 Novel contributions ………………………….. ………………………….. ………………………. 10
1.5 Scope of the study ………………………….. ………………………….. ………………………… 12
1.6 Tools to be used in this study ………………………….. ………………………….. …………. 14
Chapter 2: Literature Review ………………………….. ………………………….. ………………… 16
2.1 NGCC power plant and its modelling ………………………….. ………………………….. 16
2.1.1 Combined cycle gas turbine (CCGT) ………………………….. ………………….. 16
2.1.2 Modelling of gas turbine ………………………….. ………………………….. ………. 16
2.1.3 Simulation of NGCC power plants ………………………….. ……………………… 18
2.2 PCC based on chemical absorption process ………………………….. ………………….. 19
2.2.1 Experimental studies ………………………….. ………………………….. ……………. 19
2.2.1.1 Thermodynamic and physical properties ………………………….. ………. 19
2.2.1.2 Mass transfer and thermal performance ………………………….. ………… 20
2.2.2 Model -based studies ………………………….. ………………………….. …………….. 21
2.2.2.1 Thermodynamic modelling of MEA -H2O-CO 2 system ……………….. 21
vii
2.2.2.2 Rate-based model for solvent -based PCC Process ………………………. 23
2.2.2.3 Model -based optimal design and opera tion of PCC processes ……… 25
2.3 NGCC integrated with solvent -based PCC ………………………….. …………………… 30
2.3.1 NGCC integrated with PCC ………………………….. ………………………….. ….. 30
2.3.2 Energy penalty ………………………….. ………………………….. …………………….. 30
2.3.3 Exhaust gas recirculation (EGR) technology ………………………….. ……….. 31
2.4 CO 2 transport pipeline network ………………………….. ………………………….. ………. 33
2.4.1 CO 2 transport pipeline ………………………….. ………………………….. ………….. 33
2.4.2 EOS selection ………………………….. ………………………….. ……………………… 33
2.4.3 Modelling and simulation studies ………………………….. ……………………….. 34
2.4.4 The cost of CO 2 pipeline transport ………………………….. ……………………… 35
2.5 The studies on whole CCS chain ………………………….. ………………………….. …….. 36
2.6 Concluding remarks ………………………….. ………………………….. ………………………. 37
Chapter 3: Model Development of Solvent -based PCC Process …………………………. 40
3.1 Framework of modelling of solvent -based PCC process ………………………….. … 40
3.2 Thermodynamic modelling of MEA -H2O-CO 2 system ………………………….. …… 41
3.2.1 EOSs and relevant model parameters ………………………….. ………………….. 41
3.2.1.1 PC-SAFT EOS for vapour phase ………………………….. …………………. 42
3.2.1.2 Electrolyte -NRTL for liquid phase ………………………….. ………………. 43
3.2.2 Physical solubility and Henry’s constant ………………………….. ……………… 43
3.2.3 Chemical reaction equilibrium ………………………….. ………………………….. . 45
3.2.4 Validation of CO 2 solubility prediction ………………………….. ……………….. 46
3.2.4.1 Case setups ………………………….. ………………………….. …………………… 46
3.2.4.2 Experimental data ………………………….. ………………………….. ………….. 47
3.2.4. 3 Validation results ………………………….. ………………………….. …………… 48
3.3 Physical property of MEA -H2O-CO 2 system ………………………….. ………………… 53
3.3.1 Physical property model ………………………….. ………………………….. ……….. 53
3.3.2 Available experimental data for validation ………………………….. ………….. 54
3.3.3 Validation results ………………………….. ………………………….. …………………. 54
3.4 Process model development and validation at the pilot scale ………………………. 58
3.4.1 Introduction of the pilot plant ………………………….. ………………………….. … 58
3.4.2 Process model development ………………………….. ………………………….. ….. 59
3.4.2.1 Model flowsheet and process description ………………………….. ……… 59
viii
3.4.2.2 Kinetics -controlled reactions ………………………….. ……………………….. 59
3.4.2.3 Rate-based mass transfer ………………………….. ………………………….. … 61
3.4.3 Model validation ………………………….. ………………………….. ………………….. 62
3.5 Model scale -up ………………………….. ………………………….. ………………………….. … 65
3.6 Concluding remarks ………………………….. ………………………….. ………………………. 68
Chapter 4: Optimal Design of Solvent -based PCC Process ………………………….. ……. 69
4.1 Development of cost model ………………………….. ………………………….. ……………. 69
4.1.1 Cost breakdown ………………………….. ………………………….. …………………… 69
4.1.2 CAPEX ………………………….. ………………………….. ………………………….. ….. 70
4.1.2.1 Equipment type and material ………………………….. ……………………….. 70
4.1.2.2 Direct cost of equipment ………………………….. ………………………….. … 71
4.1.2.3 Annualized CAPEX ………………………….. ………………………….. ………. 71
4.1.3 Fixed OPEX ………………………….. ………………………….. ………………………… 73
4.1.4 Variable OPEX ………………………….. ………………………….. ……………………. 73
4.1.5 The costs of the base case ………………………….. ………………………….. ……… 74
4.2 Optimisation methodology ………………………….. ………………………….. …………….. 75
4.2.1 Sequential quadratic programming (SQP) ………………………….. ……………. 75
4.2.2 Objective function ………………………….. ………………………….. ……………….. 76
4.2.3 Optimisation constraints ………………………….. ………………………….. ……….. 77
4.2.3.1 Equality constraints ………………………….. ………………………….. ……….. 77
4.2.3.2 Inequality constraints ………………………….. ………………………….. …….. 77
4.2.4 Optimisation variables ………………………….. ………………………….. ………….. 77
4.2.4.1 Key design variables ………………………….. ………………………….. ……… 77
4.2.4.2 Operating pressure and temperature ………………………….. ……………… 78
4.2.4.3 Key operational variables ………………………….. ………………………….. .. 79
4.3 Optimisation result s ………………………….. ………………………….. ………………………. 80
4.4 Optimisations in response to variations of key variables ………………………….. … 81
4.4.1 Variation of MEA concentration in solvent ………………………….. …………. 82
4.4.2 Variation of CO 2 concentration in flue gas ………………………….. ………….. 85
4.4.3 Variation of flue gas flow rate ………………………….. ………………………….. .. 88
4.5 Concluding remarks ………………………….. ………………………….. ………………………. 91
Chapter 5: Integration of NGCC Power Plant and Solvent -based PCC Process and
CO 2 Compression Train ………………………….. ………………………….. ………………………….. . 92
ix
5.1 NGCC power plant model ………………………….. ………………………….. ……………… 92
5.1.1 Modelling of gas turbine ………………………….. ………………………….. ………. 92
5.1.2 Model development for NGCC power plant ………………………….. ………… 94
5.1.3 Model validation ………………………….. ………………………….. ………………….. 96
5.2 Integration of NGCC with PCC process and CO 2 compression ……………………. 97
5.2.1 General interfaces of the integration ………………………….. …………………… 97
5.2.2 EGR technology ………………………….. ………………………….. ………………….. 99
5.3 Heat integration options based on supersonic shock wave compression ……… 101
5.3.1 CO 2 compression technology ………………………….. ………………………….. . 101
5.3.2 Heat integration case setups ………………………….. ………………………….. … 102
5.3.3 Results and discussion ………………………….. ………………………….. ………… 104
5.4 Concluding remarks ………………………….. ………………………….. …………………….. 106
Chapter 6: Optimal Design of CO 2 Transport Pipeline Network ……………………….. 107
6.1 Pipeline network system ………………………….. ………………………….. ………………. 107
6.2 Process model development and economics evaluation methodology …………. 108
6.2.1 Process simulation model development for the base case …………………. 109
6.2.1.1 Physical property method ………………………….. ………………………….. 109
6.2.1.2 Assumptions, constraints and inputs ………………………….. …………… 110
6.2.1.3 Model validation ………………………….. ………………………….. …………. 111
6.2.2 Economic evaluation methodology ………………………….. …………………… 112
6.3 Techno -economic evaluation of CO 2 compression ………………………….. ………. 114
6.3.1 Compression configuration options ………………………….. …………………… 114
6.3.2 Results and analysis ………………………….. ………………………….. ……………. 116
6.3.3 Comparison with other studies in the literature ………………………….. …… 117
6.4 Techno -economic evaluation of trunk pipelines ………………………….. ………….. 118
6.4.1 Calculation of pipeline diameter ………………………….. ……………………….. 119
6.4.2 Results and analysis ………………………….. ………………………….. ……………. 121
6.4.3 Comparison with other studies in the literature ………………………….. …… 122
6.5 Overall cost of CO 2 transportation pipeline network ………………………….. ……. 123
6.5.1 Comparison of the base case and the optimal case ………………………….. . 123
6.5.2 Comparison with other studies in the literature ………………………….. …… 124
6.6 Concluding remarks ………………………….. ………………………….. …………………….. 125
x
Chapter 7: Optimal operation under different market conditions based on whole CCS
chain consideration ………………………….. ………………………….. ………………………….. ……. 127
7.1 Optimisation methodology update ………………………….. ………………………….. … 127
7.1.1 Objective function ………………………….. ………………………….. ……………… 127
7.1.2 CO 2 emission cost ………………………….. ………………………….. ………………. 128
7.1.3 CO 2 T&S cost ………………………….. ………………………….. ……………………. 128
7.1.4 Equality constraints ………………………….. ………………………….. ……………. 129
7.1.5 Inequality constraints ………………………….. ………………………….. ………….. 129
7.2 Techno -economic evaluation of the base case ………………………….. …………….. 130
7.3 Optimal operation ………………………….. ………………………….. ……………………….. 132
7.3.1 Optimal capture level under different carbon price ………………………….. 132
7.3.2 The effect of NG price ………………………….. ………………………….. ………… 137
7.3.3 The effect of CO 2 T&S price ………………………….. ………………………….. .. 139
7.4 Concluding remarks ………………………….. ………………………….. …………………….. 141
Chapter 8: Conclusions and recommendations for future research …………………….. 143
8.1 Conclusions ………………………….. ………………………….. ………………………….. …… 143
8.2 Recommendations for future research ………………………….. ………………………… 146
Appendix A: Publications from this thesis ………………………….. ………………………….. … 148
Reference ………………………….. ………………………….. ………………………….. …………………. 149
xi
List of Figures
Figure 1.1 Global temperature departure from long -term average (NOAA, 2015) ……… 1
Figure 1.2 Total annual anthropogenic GHG emissions by groups of gases 1970 –2010:
CO 2 from fossil fuel combustion and industrial processes; CO 2 from Forestry and Other
Land Use (FOLU); methane (CH 4); nitrous oxide (N 2O); fluorinated gases covered
under the Kyoto Protocol (F-Gases) (IPCC, 2015) ………………………….. …………………….. 2
Figure 1.3 CO 2 concentration in atmosphere, (a) indirect measurements for historical
CO 2 level, (b) direct measurements from Jan 2005 to Feb 2016 (NASA, 2016) …………. 3
Figure 1.4 World CO 2 emissions by sector in 2013 (IEA, 2015) ………………………….. ….. 3
Figure 1.5 Key technologies for reducing CO 2 emissions, (IEA, 2010) …………………….. 5
Figure 1.6 Schematic of whole CCS infrastructure (NERC, 2014) ………………………….. .. 5
Figure 1.7 Processes of CO 2 capture technologies (IPCC, 2005) ………………………….. …. 6
Figure 1.8 Processes of CO 2 capture technologies (DECC, 2013) ………………………….. … 8
Figure 1.9 Study scope of each chapter ………………………….. ………………………….. ………. 13
Figure 2.1 CCGT power plant schematic (Adapted from blog.gerbilnow.com (2012)) 17
Figure 2.2 Diagram for a gas -fired power plant with a triple -pressure HRSG (Godoy et
al., 2011) ………………………….. ………………………….. ………………………….. ……………………. 18
Figure 2.3 Model complexities for reactive absorption process (Kenig et al., 2001). … 24
Figure 2.4 Discretized liquid film for counter current flow (Zhang et al., 2009) ……….. 25
Figure 2.5 Process flow diagram of solvent -based PCC (IPCC, 2005) …………………….. 26
Figure 2.6 BLUE map emission reduction plant (IEAGHG, 2010) …………………………. 30
Figure 2.7 Impact of EGR, (a) on O 2 concentration in combustion air feed, and (b) on
exhaust gas compositions (Canepa et al., 2013) ………………………….. ……………………….. 32
Figure 2.8 Schematic of a full CCS chain (SCCS, 2016) ………………………….. …………… 37
Figure 3.1 Framework of modelling of a solvent -based PCC process ……………………… 41
Figure 3.2 CO 2 partial pressure as function of CO 2 loading with 15 wt% MEA ……….. 49
Figure 3.3 CO 2 partial pressure as function of CO 2 loading with 30 wt% MEA ……….. 49
Figure 3.4 CO 2 partial pressure as function of CO 2 loading with 45 wt% MEA ……….. 50
Figure 3.5 CO 2 partial pressure as function of CO 2 loading with 60 wt% MEA ……….. 50
Figure 3.6 Total pressure as function of CO 2 loading with 15 wt% MEA solvent …….. 51
Figure 3.7 Total pressure as function of CO 2 loading with 30 wt% MEA solvent …….. 51
Figure 3.8 Total pressure as function of CO 2 loading with 45wt% MEA solvent ……… 52
Figure 3.9 Total pressure as function of CO 2 loading with 60 wt% MEA solvent …….. 52
Figure 3.10 Liquid density of MEA -H2O-CO 2 at 30 wt% MEA ………………………….. … 55
Figure 3.11 Liquid density of MEA -H2O-CO 2 at 40 wt% MEA ………………………….. … 55
xii
Figure 3.12 Liquid density of MEA -H2O-CO 2 at 50 wt% MEA ………………………….. … 56
Figure 3.13 Liquid density of MEA -H2O-CO 2 at 60wt% MEA ………………………….. …. 56
Figure 3.14 Liquid heat capacity of MEA -H2O-CO 2 at 298.15K ………………………….. .. 56
Figure 3.15 Liquid viscosity of MEA -H2O-CO 2 at 298.15K ………………………….. ……… 57
Figure 3.16 Surface tension of MEA -H2O-CO 2 at 30 wt% MEA at 298.15K …………… 57
Figure 3.17 Process flowsheet in Aspen Plus® ………………………….. …………………………. 59
Figure 3.18 Validation results between model predictions and experimental data, (a)
temperature profile of the absorber, (b) temperature profile of the stripper, (c) CO 2
composition profile inside the absorber, (d) CO 2 composition profile inside the absorber
………………………….. ………………………….. ………………………….. ………………………….. …….. 64
Figure 3.19 Generalized pressure drop correlation (Stichlmair and Fair, 1998) ……….. 66
Figure 4.1 Optimisation results with variation of MEA concentration in solvent …….. 84
Figure 4.2 Optimisation results with variation of CO 2 concentration in flue gas ……… 87
Figure 4.3 Optimisation results with variation of flue gas flow rate ……………………….. 90
Figure 5.1 Schematic of gas turbine developed in Aspen Plus® ………………………….. …. 93
Figure 5.2 The flowsheet of NG CC power plant standalone ………………………….. ……… 95
Figure 5.3 The flowsheet of NGCC power plant with EGR integrated with PCC process
and compression ………………………….. ………………………….. ………………………….. …………. 98
Figure 6.1 The pipeline sketch for the Humber case study ………………………….. ………. 108
Figure 6.2 The flowsheet of pipeline network in Aspen HYSYS® ………………………… 110
Figure 6 .3 The work flow of the techno -economic evaluation ………………………….. …. 113
Figure 6.4 Comparison of levelised costs of different compression options …………… 117
Figure 6.5 The comparison of levelised cost of different cost model …………………….. 118
Figure 6.6 Annua l cost comparison for different diameters of the pipelines …………… 122
Figure 6.7 Comparison of capital cost of different cost models ………………………….. … 123
Figure 6.8 Comparison of annual costs of base case and optimal case …………………… 124
Figure 6.9 Comparison of levelised cost of the optimal case and COCATE project … 125
Figure 7.1 LCOE of different capture level with carbon price of 7 €/ton CO 2 ………… 133
Figure 7.2 LCOE of different capture levels with carbon price of 50 €/ton CO 2 ……… 133
Figure 7.3 LCOE of different capture levels with c arbon price of 100 €/ton CO 2 ……. 134
Figure 7.4 LCOE of different capture levels with carbon price of 150 €/ton CO 2 ……. 134
Figure 7.5 Optimal lean loading and L/G ratio for different capture levels …………….. 135
Figure 7.6 Optimal reboiler duty and specific duty for different capture levels ………. 136
Figure 7.7 Thermal efficiency of the NGCC with PCC at different capture levels ….. 136
xiii
Figure 7.8 LCOE of different capture level with carbon price of €100/ton CO 2 and NG
price of €2/GJ ………………………….. ………………………….. ………………………….. …………… 137
Figure 7.9 LCOE of different capture l evel with carbon price of €100/ton CO 2 and NG
price of €6.58/GJ ………………………….. ………………………….. ………………………….. ………. 137
Figure 7.10 LCOE of different capture level with carbon price of €100/ton CO 2 and NG
price of €12/GJ ………………………….. ………………………….. ………………………….. …………. 138
Figure 7.11 Required carbon price for driving 90% capture level in response to different
fuel prices ………………………….. ………………………….. ………………………….. ………………… 139
Figure 7.12 LCOE of different capture level with carbon price of €100/ton CO 2 and
T&S price of €9.32/ton CO 2 ………………………….. ………………………….. ……………………. 139
Figure 7.13 LCOE of different capture level with carbon price of €100/ton CO 2 and
T&S price of €39.54/ton CO 2 ………………………….. ………………………….. ………………….. 140
Figure 7.14 LCOE of different capture level with carbon price of €100/ton CO 2 and
T&S price of €102.5/ton CO 2 ………………………….. ………………………….. ………………….. 140
Figure 7.15 Carbon price for driving 90% capture level in response to different CO 2
T&S price ………………………….. ………………………….. ………………………….. ………………… 141
xiv
List of Tables
Table 1.1 CO 2 emissions from industry emitters (IPCC, 2005) ………………………….. ……. 4
Table 1.2 CO 2 emissions from different fuels (EIA, 2016) ………………………….. ………….. 8
Table 2.1 Correlations for the calculation of Henry’s constants ………………………….. … 22
Table 2.2 Literature review of key parameters of optimal PCC process for NGCC
power plant at the industrial scale ………………………….. ………………………….. ……………… 28
Table 3.1 PC-SAFT parameters of pure components ………………………….. ………………… 42
Table 3.2 Binary parameters for PC -SAFT EOS ………………………….. ……………………… 42
Table 3.3 Model parameters for eNRTL ………………………….. ………………………….. …….. 43
Table 3.4 Correlations for the calculation of Henry’s constants (on the Molality Scale)
………………………….. ………………………….. ………………………….. ………………………….. …….. 45
Table 3.5 Correlations for chemical equilibrium constants (on the Molality Scale) ….. 46
Table 3.6 Different combinations of correlations for validation ………………………….. … 47
Table 3.7 Chos en experimental data for solubility of CO 2 in MEA aqueous solution … 47
Table 3.8 MAPE of validation with CO 2 partial pressure of MEA -H2O-CO 2 system . 48
Table 3.9 Correlations used for property calculation of the mixture ……………………….. 53
Table 3.10 Available experimental data for physical properties of liquid phase ……… 54
Table 3.11 MAPE of validation results of physical properties in liquid phase ………… 55
Table 3.12 Main specifications of the pilot plant ………………………….. ……………………. 58
Table 3.13 Parameters k and E in Equation (3.15) (Zhang and Chen, 2013) …………… 60
Table 3.14 Parameters and correlations selection for mass transfer in RateSep model 61
Table 3.15 Validation results of model predictions against experimental data ………… 63
Table 3.16 Boundary conditions of solvent -based PCC process ………………………….. … 65
Table 3.17 Design parameters of the absorber and the stripper a t the base case ……….. 67
Table 3.18 Overall performance of PCC process at the base case ………………………….. . 67
Table 4.1 Equipment type and material selection of PCC process ………………………….. . 70
Table 4.2 Direct material costs and the scaling factor of equipment ……………………….. 72
Table 4.3 Factors for total project cost calculation ………………………….. …………………… 72
Table 4.4 Key economic evaluation cost inputs ………………………….. ……………………….. 74
Table 4.5 Costs of the base case ………………………….. ………………………….. ……………….. 74
Table 4.6 Comparison of the optimal case and the base case ………………………….. …….. 81
Table 4.7 Optimisation results with variation of MEA concentration ……………………. 83
Table 4.8 Optimisation results with variation of CO 2 concentration in flue gas ………. 86
Table 4.9 Optimisation results with variation of flue gas flow rate ……………………….. 89
xv
Table 5.1 GT modelling assumptions and tuning parameters ………………………….. …….. 94
Table 5.2 Model validation with manufactory product data sheet ………………………….. . 94
Table 5.3 Model parameters of NGCC power plant ………………………….. ………………….. 96
Table 5.4 Comparison of the simulation results for model validation ……………………… 97
Table 5.5 Optimal design of the integration without EGR and with EGR ………………. 100
Table 5.6 Process boundary conditions and parameters of CO 2 compress ion …………. 102
Table 5.7 Performance comparison results of different cases ………………………….. ….. 105
Table 6.1 Parameters of the pipelines ………………………….. ………………………….. ……….. 108
Table 6.2 APE between experimental data and PR -EOS for corresponding values
………………………….. ………………………….. ………………………….. ………………………….. …… 110
Table 6.3 Input and boundary conditions of the base case ………………………….. ……….. 111
Table 6.4 Comparison of the simulation results ………………………….. ……………………… 112
Table 6.5 Economic evaluation cost inputs ………………………….. ………………………….. .. 113
Table 6.6 Compression technology options and their process definition ………………… 115
Table 6.7 Energy and utilities requirements of c ompression technologies ……………… 116
Table 6.8 Input and boundary conditions ………………………….. ………………………….. …. 118
Table 6.9 Overview of the different diameter calculation methods in literature ………. 120
Table 6.10 The calculation results of different diameter models ………………………….. . 121
Table 6.11 Techn ical performance of trunk pipelines system in different diameters .. 121
Table 7.1 Key economic evaluation cost inputs ………………………….. ……………………… 131
Table 7.2 Cost comparison ………………………….. ………………………….. ……………………… 131
xvi
Nomenclature
Symbols Unit Description
Letters
m2 Area
Parameter of bypass air calculation for gas turbine
cooling
Correlations for property calculations
The vector of the coefficients
m Column diameter
The vector of the design variables
kJ/mol Activation energy
The objective function
Packing factor
kg/s Total vapour flow rate in mass
J/kmol Reference state Gibbs energy change for reaction
Henry’s constant
m Packing height of a column
The interest rate
Cost index
The pre–exponential factor
Parameter of bypass air calculation for gas turbine
cooling
Parameter for size calculation of column
Chemical equilibrium constants of reaction
kg/s Total liquid flow rate in mass
Specific factor for different type of equipment
Bypass air mass flow rate for gas turbine cooling
Mass fraction in a mixture
The economic life of plant
The vector of operational variables
bar Pressure
cost Direct material cost
Ideal g as constant
mol/(min. m3) Reaction rate of reaction
Parameter of bypass air calculation for gas turbine
cooling
K Temperature
m3/mole Partial molar volume of molecular solute i at infinite
dilution in solvent
Weight factor for the calculation of Henry’s constant
xvii
Mole fraction in a mixture
Selected scaling factor
Greek
CO 2 loading in lean solvent or rich solvent
the reaction order of component in reaction
the infinite dilution activity coefficient of molecular
solute in solvent
Lagrange multipliers
kg/m3 density
Lagrange multipliers
CO 2 capture level
MEA concentration in solvent
Superscript
o Standard state
T Vector transpose
Subscript
Absorber
Cooling
CO 2 component
EG Exhaust gas
FG Flue gas
Gas
Lean solvent
L Liquid phase
MEA
Stripper
Rich solvent
V Vapour phase
xviii
Abbreviations
AAD Average absolute deviations
APE Absolute percentage error
ACAPEX Annualised capital expenditure
APEA Aspen Process Economic Analyser®
ASNI American National Standards Institute
BBC British Broadcasting Corporation
BP British Petroleum
CAPEX Capital expenditure
CCA The cost of CO 2 avoidance
CCGT Combined cycle gas turbine
CCS Carbon capture and storage
CL Capture level
CO 2 FOLU CO 2 from forestry and other land use
CRF Capital recovery factor
DCC Direct contact cooler
DECC Department of Energy and Climate Change (of the UK)
DOE Department of Energy (of the USA)
EGR Exhaust cas recirculation
EIA Energy Information Administration (of the USA)
EU The European Union
eNRTL Electrolyte Non -Random Two Liquid
EOR Enhanced oil recovery
EOS Equation of state
ETP Energy Technology Perspective
ETS Emission Trading System
F-Gases Fluorinated gases covered under the Kyoto Protocol
FOPEX Fixed operational expenditure
GHG Greenhouse gas
GCCSI The Global CCS Institute
HICP The Harmonised Consumer Price Index
HP-ST High pressure steam turbine
HRSG Heat recovery steam generator
IEA International Energy Agency
IEAGHG International Energy Agency Greenhouse Gas Team
IGCC Integrated gasification combined cycle
IGHAT Integrated gasification humid turbines
IPCC Intergovernmental Panel on Climate Change
IP-ST Intermediate pressure steam turbine
LCOE Levelised cost of electricity
LHV Low heat value
xix
LP-ST Low pressure steam turbine
L/G Liquid gas ratio
NASA National Aeronautics and Space Administration
NERC Natural Environment Research Council
NG Natural gas
NGCC Natural gas combined cycle
NOAA National Oceanic and Atmospheric Administration (of the USA)
MAPE Mean absolute percentage error
MEA Monoethanolamine
O&M Operating and maintenance cost
OEM Original equipment manufacturers
OPEX Operational expenditure
PCC Post-combustion carbon capture
PR Peng -Robinson
QP Quadratic Programming
SAFT Statistical Associating Fluid Theory
SCCS Scottish Carbon Capture and Storage Research Group
SQP Sequential Quadratic Programming
SW Span and Wagner
TAC Total annualised cost
T&S Transport and storage
UNFCCC The United Nations Framework Convention on Climate Change
VLE Vapour -liquid equilibrium
VOPEX Variable operational expenditure
WEO World Energy Outlook
1
Chapter 1: Introduction
This chapter introduces the background of this research. Section 1.1 updated current
status of global warming , CO 2 emissions and CCS technologies. Section 1.2 stated the
motivation s for of this study. Section 1.3 summarized the aim and objectives and
Section 1.4 justified the novel contributions. Section 1.5 outlined the study scope of
each ch apter . In Section 0, the selections of tools used in this study w ere explained.
1.1 Background
1.1.1 Global warming and green house gas es emission s
Figure 1.1 Global temperature departure from long -term average (NOAA, 2015)
Global w arming including its anthropogenic causes is one of the most contentious
scientif ic issues in the last two decades (Easterling and Wehner, 2009) . The evidence of
a warming world including changes in surface, atmospheric and oceanic temperatures;
glaciers; snow cover; sea ice; sea level and atmospheric water vapour comes from
multiple independent climate indicators and have been independently verif ied many
times by scientists all over the wor ld (Hartmann et al., 2013) . Althou gh the hiatus of
global warming has been widely discussed, one recent res earch (Lin and Franzke, 2015)
suggest ed that the hiatus is due to natural fluctuations imposing a decreasing
temperature trend and, thus, temporally overshadowing the global warming trend.
Another study by Karl et al. (2015) pointed out global warming does not slow down at
all (see Figure 1.1). Therefore, anthropogenic global warming still exerts a strong signa l
and is worth world -wide concern.
‘Global warming hiatus’
2
The scientific understandings of the causes of global warming are the changes of
external forcing including increased concentrations of greenhouse gases, more intensive
solar luminosity, more volcanic eruptions, and vari ations in Earth's orbit around the Sun
(Hegerl et al., 2007) . Among th ese factors, increasing concentrations of greenhouse
gases (GHGs) caused by anthropogenic activities are responsible for most of global
warming and this is more than 95% proven by scientists (Pachauri et al., 2014) . The
main GHGs include water vapour, carbon dioxide (CO 2), methane (CH 4) and ozone
(O3). Figure 1.2 shows that total anthropogenic GHG emissions increase d significantly
in the last decade and CO 2 is the main anthropogenic GHG.
Figure 1.2 Total annual anthropogenic GHG emissions by groups of gases 1970 –2010 :
CO 2 from fossil fuel combustion and industrial processes; CO 2 from Forestry and Other
Land Use (FOLU); methane (CH 4); nitrous oxide (N 2O); fluorinated gases covered
under the Kyoto Protocol (F -Gases) (IPCC, 2015)
1.1.2 CO 2 emission and its reduction
Since the Industrial Revolution started in 1760s, atmospheri c CO 2 concentration has
increase d more than 40%, from 280 ppm in 17 60 to about 402 ppm in 201 6 (Blasing,
3
2013 ; NASA, 2016) . Recent research es (Buizert et al., 2014; Sime, 2014) into
Greenland's ice sheets explain ed one of the mysteries of the climatic past , confirming
the importance of carbon dioxide on global temperature changes. The indirect
measurements by reconstruction analysis from i ce cores show the CO 2 concentration in
the atmosphere has increased rapidly in recent decades ( refer to Figure 1.3(a)) and it has
not slow ed down (see Figure 1.3(b)).
(a) (b)
Figure 1.3 CO 2 concentration in atmosphere, (a) indirect measurements for historical
CO 2 level, (b) direct measure ments from Jan 2005 to Feb 2016 (NASA, 2016)
Figure 1.4 World CO 2 emissions by sector in 2013 (IEA, 2015)
In the successful UNFCCC 2015 Paris meeting , a common goal was agreed by all the
196 parties to keep global warming to less than 2 °C compared to pre -industria l levels
(UNFCCC, 2015) . Anthropogenic CO 2 emission sources cover every aspects related
with human activities, including electricity and heat generation, industr ial
4
manufactur ing, transport sector, residence , agriculture, forestry, fishing and so on ( see
Figure 1.4). Except for transport and residential emissions which are normally very
small and distributed , industrial processes are major individual CO 2 sources . Table 1.1
lists w orldwide large stationary CO 2 sources with emissions of more than 0.1 million
tons of CO 2 per year (IPCC, 2005) which shows fossil -fired power generation is the
biggest CO 2 emitter .
Table 1.1 CO 2 emissions from industry emitters (IPCC, 2005)
Process Number of emitters Emission
(MtCO 2/yr)
Fossil fuels
Power 4,942 10,539
Cement production 1,175 932
Refineries 638 798
Iron and steel industry 269 646
Petrochemical industry 470 379
Oil and gas processing – 50
Other sources 90 33
Biomass
Bioethanol and bioenergy 303 91
Total 7,887 13,468
Reduc ing CO 2 emission is a big challenge both technically and commercially because
large amount of CO 2 emissions needs to be cut down to ensure global temperature rise
remains below 2 degrees Celsius (UNFCCC, 2015) . Internat ional Energy Agency ( IEA)
set up a BLUE Map scenario with 14 GT CO 2 emissions in 2050 compared with the 57
GT CO 2 emissions in Baseline scenario. In order to achieve this target , significant
efforts are required to develop and improve the technologies rela ted with energy
efficiency and low -carbon energy supply ( see Figure 1.5). CCS technology will play a
vital role in delivering 1 9% of the cumulative carbon dioxide emissions reductions
between 2015 and 2050 in the power sector .
5
Note: WEO 2009 is World Energy Ou tlook 2009, ETP 2010 is Energy Technology Perspective 2010
Figure 1.5 Key technologies for reducing CO 2 emissions , (IEA, 2010)
1.1.3 CCS technology
CCS is a process of capturing CO 2 from large industrial sources and transporting it to a
storage site, to mitigate CO 2 emission to the atmosphere (see Figure 1.6). CO 2 in flue
gases from industrial process es, such as fossil -fired power plants and refineries, is
captured by physical adsorption or chemical absorption and then is compressed to a
high pressure (over 100 bar) for transport. Except for a small part of CO 2 reutilized for
other industry such as dri nk, food production and agriculture , concentrated CO 2 is
finally injected into either a saline aquifer or depleted oil and gas reserves for storage.
Figure 1.6 Schematic of whole CCS infrastructure (NERC, 2014)
6
1.1.3.1 CO 2 capture
Three main approaches can be envisaged for CO 2 capture from power plants: pre –
combustion capture, post -combustion capture and oxy-fuel capture (Kanniche et al.,
2010) . Figure 1.7 illustrates the process diagram of main capture techn ologies. For p ost-
combustion technology , CO 2 is separated from flue gas after combustion by using
techniques such as chemical absorption (Wang et al., 2011) , adsorption (Samanta et al.,
2011) , and membrane separation etc (Merkel et al., 2010) . In p re-combustion
technology, fuel react s with steam and air or oxygen to produce syngas . In a shift
reactor, CO component o f syngas is converted to CO 2 which can then be separated by
adsorption or physical absorption or membrane separation . Remaining H2 component is
then used to generate power and heat. Oxy-fuel combustion process employs an air
separation unit to provide n ear pure oxygen for combustion (Buhre et al., 2005; Dillon
et al., 2004) , resulting in a high concentration of CO 2 around 80% in the flue gas. CO 2
can be enriched by cooling down t he flue gas es to condense the water vapour
component . Among these technologies, solvent -based chemical absorption post-
combustion capture is regarded as the preferred technology for carbon capture from
fossil -fired power plants (Wang et al., 2011) because of its relatively low cost and easy
implemen tation for existing power plant (POST, 2009) .
Figure 1.7 Processes of CO 2 capture technologies (IPCC, 2005)
7
1.1.3.2 CO 2 transport
In the transport section of CCS, CO 2 is compressed and transported from captur e plants
to storage sites (or EOR sites) by pipeline, ship or tanker trucks mainly depending on the
amount and the distance. Pipelines are the preferred method for onshore and offshore
transport of large volumes of CO 2 (Svensson et al., 2004; IPCC, 2005) . The CO 2
pipeline transport for enhanced oil recovery (EOR) is a mature technology. Sev eral
millions of tons of CO 2 are already transported for EOR purpose by pipelines in the
USA and Canada. In 2050, to achieve the carbon emission target, the global CO 2
captured will be about 7Gt/y (IEA, 2012) . This is a much larger amount than about 50
Mt/y transported in pipelines for EOR in the USA (USDOE, 2010b) .
1.1.3.3 CO 2 storage
Underground geological storage is a main method of CO 2 sequestration (Szulczewski et
al., 2012) before major breakthroughs are achieved for CO 2 utilization (Aresta et al.,
2013) . Information and experience gained from CO 2 injection for EOR applicati on
indicates that CO 2 can be safely injected and stored in well-characterized and properly
managed sites (IPCC, 2005) . However, there is a grea t deal of uncertainty in the
quantification of storage potential (Boot -Handford et al., 2014) and its costs.
1.2 Motivation s
Although fossil -fired power generation is the biggest single CO 2 emitter , fossil energy is
projected to remain a major source of energy in the near future with the advantages of
high energy density and high reliability (Mac Dowell and Shah, 2013) . Natural gas is a
major source for electricity generation and it currently accounts fo r around 22% of
global electricity generation capacity (BP, 2014) . This number is expected to increase in
the n ext coming decades because of the advent of cheap natural gas , high thermal
efficiency and carbon emission mitigation policies (BBC, 2015) . Natural gas can burn
more clea nly than other fo ssil fuels such as coal and oil . Another remarkable advantage
of natural gas power generation is its high net low heat value (LHV) efficiency (close to
60%) with the application of combined cycle gas turbine (CCGT) technology. Table 1.2
shows CO 2 emission s from different fuels in the world for 2015, reported by the US
Energy Information Administration (EIA) (2016) . To generate the same amount of
electricity, burning natural gas emits about 42% less carbon dioxide than burning coal.
8
Table 1.2 CO 2 emissions from different fuels (EIA, 2016)
Fuel CO 2 content
(kg/GJ) Heat rate
(kg/kWh) CO 2 emission
(kg/kWh)
Bituminous Coal 98.25 10,644.50 0.94
Sub-bituminous Coal 101.79 10,644.50 0.98
Lignite Coal 103.08 10,644.50 0.98
Natural gas 56.03 10,924.09 0.55
Distillate oil 77.23 10,902.99 0.76
Residual oil 83.23 10,902.99 0.82
Note: IGCC is integrated gasification combined cycle power plant
Figure 1.8 Processes of CO 2 capture technologies (DECC, 2013)
Obviously, NGCC is not carbon neutral technology. In IEA BLUE map, 5% of NGCC
power plants will be equipped with solvent -based PCC process to achieve the carbon
reduction target (IEA, 2010) . Using a mine solvent such as MEA to absorb the CO 2 from
the flue gases is mature technology (Rochelle, 2009) . But for the full commercial scale
application of this carbon captu re technolog y, the main barrier is the high increment in
electricity cost due to its massive capital costs and high thermal energy penalty
(Rochelle, 2009; Marchioro Ystad et al., 2013) . It is reported that the cost of electricity
9
from an NGCC power generation w ill increase to £144.1 from £66 per MWh (see
Figure 1.8 ) when it is integrated with a PCC process (DECC, 2013) . Considering the
great amount of total electricity consumption, it’s a great cost increment for both
industry production and living expense. Therefore research efforts are required for
potential improvements to reduce both the capital cost as well as the energy penalty to
gain a better economic profile of commercial deployment of carbon capture.
1.3 Aim and objectives of this study
The research presented in this thesis is aimed to achieve optimal design and operation of
NGCC power plant integrated with CO 2 capture and transport process to help reduce the
CCS deployment co st. To achieve the stated aim, the follow ing objectives have been
identifi ed:
i. To develop the process model for the optimisation studies, including (1)
select ing and validating correlation s for thermodynamic model and physical
properties prediction of MEA -H2O-CO 2 system , (2) develop ing a steady state
rate-based process model of solvent -based PCC process in Aspen Plus® at the
pilot scale and validat ing the model with experimental data, and (3) scaling up
the model to match power plants at industry scale.
ii. To conduct optimal design of the solvent -based PCC process, including (1)
develop ing the cost model of the PCC process , (2) integrating the cost model
with the process model, and (3) conduct ing optimisation of the PCC process.
iii. To explore the integration of NGCC power plant with PCC process, including ( 1)
developing a steady state process of NGCC power plant , (2) analysing the
integration of NGCC power plant with PCC process and EGR, and (3) studying
heat integration options based on a specific supersonic shock wave compression
technology
iv. To carry out the study on optimal design of the CO 2 transport pipeline network
consisting of the compression trains and onshore and offshore trunk pipelines,
that was planned in the Humber region of the UK.
v. To carry out the study on optimal operation for an assumed existing NGCC
power plant integrated with MEA -based PCC process based on whole CCS
10
chain consideration in respon se to different market conditions including
different carbon price, fuel price and CO 2 T&S price.
1.4 Novel contributions
Compared with the literature on NGCC power plants integrated with solvent -based PCC
process, novel contributions of the studies in this thesis are claimed as follows :
i. The model used for optimisation studies in this thesis is expected to have better
accuracy than previous studies . It could be justified by (1) the correlations of the
prediction for liquid density of mixture and effective vapour liquid interfacial
area were improved by coding Fortan subroutines in Aspen Plus®; (2) different
value s were input to the kinetic s of reverse reactions for bicarbonate formation
in the absorber and the stripper respectively , which reflects the nature of
different operating conditions of the absorber and the stripper ; (3) the process
model was validated at three different stages , incl uding thermodynamic
modelling, physical property calculation and a rate -based process model of the
close -loop system ; and ( 4) the cost model of PCC process was developed based
on the major equipment costs provided by vendors after detailed engineering
design in a benchmark report (IEAGHG, 2012) . The uncertainty of this method
could be in a range of from −15% to 20%, instead of empirical methods with
uncertainty in a range of from −30% to 50%.
ii. In this thesis, non-linear optimisation s were implemented in Aspen Plus® and
solved by Sequential Quadratic Programming (SQP) method, which is a quasi –
Newton nonlinear programming algorithm . In one recent similar study
(Agbonghae et al., 201 4) using Aspen Plus®, the optimal designs were obtained
by comparing different options at the specified values of several key variables,
which may exhibit local minimum solution.
iii. For study on the integration of NGCC power plant and solvent -based PCC
process, a specific supersonic shock wave compression (Lawlor, 2009)
technology was adopted for the CO 2 compression train. For conventional multi –
stage compressors, the discharge temperature of each stage (around 90–110oC)
is lower than the lowest pinch temperature so that the compression heat canno t
11
be used directly (Gibbins et al., 2004) . The discharge temperature of the
compressed CO 2 is as high as 2 14.5–230.5oC (Witkowski and Majkut, 2012) ,
which provide s opportunities for directly integrating compression heat with
NGCC/PCC processes. Therefore, t he study on h eat integrations based on this
compression technology with NGCC/PCC process is novel.
iv. For the study on optimal design of CO 2 transport pipeline network, s imulation –
based techno -economic evaluation method is used in this study , compared with
empirical methods used in p revious studies (IEAGHG, 2002; McCoy and Rubin,
2008; McCollum and Ogden, 2006) . Detailed steady state model about the CO 2
transport pipeline network was developed including compression train and
collecting system for multi -sources, onshore and offshore trunk pipelines with
booster pump station. Most of previous process simulation models (Zhang et al.,
2006; Nimtz et al., 2010; Liljemark et al., 2011; Chaczykowski and Osiadacz,
2012) for CO 2 pipeline network are about single emitter and pipelines without
booster station.
v. The novelties of the study on optimal operation under different market
conditions based on whole CCS chain consideration are (1) in the cost model,
the total annual cost of CO 2 T&S was regarded as an operating expense charged
by the operators of the CO 2 T&S infrastructure, which avoids heavy
computationally demanding for the CO 2 T&S with many uncertainties. With this
method, the cost model was developed to cover the cost o f the whole integrated
system. Thus the results and insights obtained from this study present the
optimal operation for the NGCC power plant equipped with a whole integrated
CCS chain , and (2) the optimisations were carried out for the optimal carbon
captu re level under different carbon price, natural gas (NG) price and CO 2 T&S
price. It is found that t he coactions of carbon price, NG price and CO 2 T&S
price will significantly affect the decision making about the optimal carbon
capture level for operating carbon capture process for a n NGCC power plant.
12
1.5 Scope of the study
As can be seen in Figure 1.9, the integrated system in this thesis consists of 5 sections
including the NGCC power plant, the solvent -based PCC process, the CO 2 compression
process, the CO 2 transport pipeline network and the CO 2 storage section . However, th is
study focuses on solvent -based PCC process first and then extended to cover o ther
sections and whole CCS chain.
Chapters 3 and 4 present model development and optimisation studies on solvent -based
PCC process. In Chapter 3 , different combinations of correlations were examined and
validated for the thermodynamic model and physical property calculations of MEA –
H2O-CO 2 mixture. The steady state rate -based process model was developed in Aspen
Plus® and validated with the pilot plant data. The model was then scaled up to match
industrial power plant scale. In Chapter 4, the cost model i s developed and is integrated
into process model by coding Fortran subroutine . Optimal design was carried out for the
solvent -based PCC process. These were illustrated in Figure 1.9 with red box.
Chapter 5 presents the study on the integration of a 453MW e NGCC power plant with
solvent -based PCC process and CO 2 compression train. The process model inthis
chapter includes the NGCC power plant , the solvent -based PCC process and the specific
supersonic shock wave compressor s. These were illustrated in Figure 1 .9 with blue box.
Chapter 6 presents model -based techno -economic evaluations for optimal design the
CO 2 transport pipeline network. The models developed in Aspen HYSYS® include
compression trains and collecting system for multi -sources, onshore and offshor e trunk
pipelines with booster pump station. The process models were integrated with APEA
for techno -economic evaluations. These were illustrated in Figure 1.9 with green box.
In Chapter 7, the optimal operation was studied for an assumed existing NGCC pow er
plant integrated with whole CCS chain. The process models include the NGCC power
plant, the solvent -based PCC process and the CO 2 compression train. The CO 2 transport
and storage sections were considered as an operating expense (CO 2 T&S cost). These
were illustrated in Figure 1.9 with purple box.
13
Figure 1.9 Study scope of each chapter
NGCC power plant
Process model
CAPEX as a whole unit
VOPEX based on simulation results
Solvent -based PCC process
Rate-based process model
CAPEX in bottom up approach
VOPEX based on simulation results
CO2 compression train
Process model
CAPEX in bottom up approach
VOPEX based on simulation results
CO2 transport pipeline network
Process model
CAPEX in bottom up approach
VOPEX based on simulation results
CO 2 storage
Cost as a whole unitIntegration
Flue gas (EGR)
Stream extraction
Electricity supply
Compression heat integration
Integration
Impurity specification
Temperature
PressureChapter 7
Chapter 3
Chapter 4
Chapter 5
Chapter 6
14
1.6 Tools to be used in this study
The study on solvent -based PCC process is a core part of this thesis. Aspen Plus® was
chosen for model development and optimisation study of the PCC process. The reasons
are: (1) Aspen Plus® has various physical property methods and comprehensive
property databank , which makes it can well support thermodynamic modelling and
process simulation involving complex electrolyte system. Different routes can be
chosen by users for different physical properties. For example, in this stud y, PC -SAFT
EOS was used to calc ulate major properties in vapour phase whilst eNRTL was used for
liquid phase. (2) R ateSep model in Aspen Plus® is proven to be capable to simulate the
absorber and the stripper in solvent -based PCC process (Zhang et al., 2009; Zhang and
Chen, 2013) . It employ s rate -based mass transfer and kinetic -controlled reactions to
describe the chemical phenomenon happening in th is reactive absorption process. The
correlations and kinetics in the model could be adjusted by comparing model
predictions against the experimen tal, to improve the model accuracy. (3) Aspen Plus®
has optimisation function with SQP method, which has been one of the most successful
general methods for solving large -scale nonlinear constrained optimization problems
(Boggs and Tolle, 2000) . (4) Aspen Plus® opens accesses of parameters and correlations
of major equations to users and well support user defined model by coding Fortran
subroutines. In this study, there are three Fortran subroutines linked into Aspen Plus®,
including Han et al. (2012) correlations for density of liquid mixture , Tsai et al. (2011)
correlations for interfacial area for packed column and the cost model. Aspen Plus® was
also used for model d evelopment of NGCC power plant and CO 2 compression in
Chapters 4, 5 and 7.
Although APEA has been embedded into Aspen P lus® for economic evaluation, t he cost
model used in Chapters 4, 5 and 7 was developed in Fortran and was integrated into the
Aspen Plus® model. The reasons are: (1) APEA needs to re -map and re -size the
equipment in the process for cost estimate in each case. Currently APEA cannot
automatically run with iterating of optimi sation in Aspen Plus®. (2) APEA uses a
bottom -up approach for the cost estimate based on historic real project data but it hardly
handle some special equipment . For example, the absorber is a rectangular column
15
constructed in concrete with epoxy lining inside surface. Its cost cannot be accurately
estimated from historical cylindrical column with metal mate rial.
In Chapter 6, Aspen HYSYS® was used for process model development for CO 2
pipelines network (including CO 2 compression) because it s pipe model provides more
detailed engineering specifications such as elevation changes and heat transfer between
the pipeline and surroundings , which is important for th is study involving CO 2 dense
phase transport . As a comparison, the pipeline model in Aspen Plus® is relatively too
simple for this study . It should also be notice d that, i n this chapter, ‘optimal design ’ is
not strictly derived from optimisation study, just by comparing several specific options.
In this way, APEA is capable for economic evaluation for pipelines and compressors.
16
Chapter 2: Literature Review
This chapter presents literature review of previous experimental and model -based
studies on NGCC power plant (Section 2.1), solvent -based PCC process(Section 2.2),
NGCC integrated with capture process and CO 2 compression (Section 2.3), CO 2
transport pipeline network (Section 2.4) and whole CCS chain (Section 2.5). The
research gaps were identified and discussed in Section 2.6.
2.1 NGCC power plant and its modelling
2.1.1 Combined cycle gas turbine (CCGT)
For gas -fired power plant, CCGT is a prevailing technology because of its high thermal
efficiency (IEA, 2008) . The thermal efficiency of the CCGT power plant at the Irsching
Power Station has reached a 60.75% net efficiency with a 578 megawatts SGT5 -8000H
gas turbine (SIEMENS, 2016) .
CCGT uses a combination of Brayton cycle (gas turbine) and Rankine cycle (steam
turbine) for electricity/heat generation. Figure 2.1 displays a typical schematic of a
CCGT power plant which is a dual -cycle process. A ir and fuel combust to generate heat
and then gas mixture expands through gas turbine to generate a part of electricity.
Exhaust gas enters heat recovery steam generator (HRSG) by which waste heat of the
exhaust gas is recovered to create steam . In the steam cycle, steam s at different pressure
levels enter multi steam turbines to generate another part of electricity.
2.1.2 Modelling of gas turbine
Although g as turbine is integrated equipment , it could be separated into three sections
including air compression, combustion and gas expansion from process view (Refer to
Figure 2.1). Analysis of gas turbine is complicated due to large number of parameters
and their interactions. For modelling of gas turbines and power plants, some
professional software package developed by the gas turbine manufacturers, such as GE
Gate -cycle® and Thermoflow GT Pro®, are normally used to predict the performance.
But accurate modelling of power plants in generic process software packages such as
Aspen Plus® is also required in the case of the integration of power plant with chemical
absorption PCC process in this study.
17
Figure 2.1 CCGT power plant schematic (Adapted from blog.gerbilnow.com (2012) )
In the study by Ong'iro et al. (1995) , the model in Aspen Plus® has been developed to
analyse the performance of integrated gasification combined cycle (IGCC) and
Integrated gasification humid turbines (IGHAT) power plants. COMPR block in Aspen
Plus® was used to simulate the compressors, fans and turbines. COMPR calculates the
power required for some certain pressure ratios and the accuracy depends on the
efficiencies specified. The gas combustor was simulated with a Gibbs type reactor
(RGIBBS) in Aspen Plus®, by which the equilibrium could be calculated by Gibbs free
energy minimization method.
One important factor affecting the whole gas turbine performance is the modelling of
cooling of gas turbine blades (Jonss on et al., 2005) . Its calculation in the professional
power plant software is very complex and requires rigorous heat transfer calculation for
the blades stage by stage . Jonsson et al. (2005) proposed a g eneric cooling model for
heavy -duty gas turbines in a joint -project with the gas turbine manufacturers. In their
model, three adjustable parameters (i.e. could be tuned to represent a gas turbine
by comparing the thermal performance (Canepa e t al., 2013) .
Cooling TowerWater
Make UpGenerator 1
Generator 2Steam
Turbine
Steam
CondenserSteam
boilerStackElectricity Grid
Electricity
ElectricitySteam
Condensate
PumpCompressor Turbine
CombustorNatural Gas
Hear Recover Steam
Generator (HRSG)UnitIntegrated Gas
Turbine UnitFresh
Air
18
2.1.3 Simulation of NGCC power plants
Aside gas turbine, other two important parts of a whole NGCC power plant are HRSG
and stream turbines. In order to achieve thermodynamic optimisation , there are many
studies (Valdés and Rapún, 2001; Bassily, 2007; Vargas and Bejan, 2000; Godoy et al.,
2010) on synthesis and design of different parts of NGCC power plants. For large scale
power plants, a triple -pressure HRSG ( see Figure 2.2) is employed to i mprove the
overall efficiency (Bassily, 2007; Godoy et al., 2011) , which is more complex than one
pressure HRSG for small scale power plants.
Figure 2.2 Diagram for a gas -fired power plant with a triple -pressure HRSG (Godoy et
al., 2011)
For its simulation in Aspen Plus®, HRSG could be regarded as the combination of
multiple heat exchangers including the economizer, evaporator, super -heater and water
pre-heater, which reflects the functions of different sections in HRSG (Canepa et al.,
2013) . Those sections could be simulated us ing HeatX blocks in Aspen Plus® (Ong'iro
et al., 1995; Canepa et al., 2013) . Three kinds of steams at different pressures are
produced from HRSG. The typical pressure and temperature are about 120 bar and
556 °C for high pressure steams , about 30 bar and 550 °C for intermediate pressure
steams and 4.15 bar and 290°C for low press ure steams (Marchioro Ystad et al., 2013;
Jordal et al., 2012) . These steams are lined to the high pressure steam turbine (HP -ST),
the intermediate pre ssure steam turbine (IP -ST) and the low pressure steam turbine (LP –
ST) respectively to generate another part of electricity. The three steam turbine sections
are simulated by Compr block in Aspen Plus® (Canepa et al., 2013) .
19
For large scale power plan ts, there are few public data about design features, process
and operating conditions because of security measures for intellectual properties . For
model validation purpose, the simulation results from those professional software
packages are used to valid ate the simulation results from generic process software tools
(Canepa et al., 2013) . In a benchmark report of IEA GHG (2012) , the reference NGCC
power plant with a net power output of 910.3MW e comprises two gas turbines, two
HRSGs and one steam turbines generator. GT PRO® was used to implement thermal
performance modelling for the design cases and GT MASTER® was used for the part –
load cases. It is noticed that these steam conditions are higher in both temperature and
pressure than what is currently typical, which are 170 ba r and 600 °C for the high
pressure steam and 40 bar and 600 °C for the intermediate pressure steam. It was
explained that original equipment manufacturers (OEMs) consider ed that utilizing these
similar conditions in NGCC plant will be common practice by 20 20 (IEAGHG, 2012) .
2.2 PCC based on chemical absorption process
2.2.1 Experimental studies
2.2.1.1 Thermodynamic and physical properties
Using MEA solvent to absorb CO 2 is a mature technology (Rochelle, 2009) . However,
complex electro lyte aqueous solvent is involved in this reactive absorption process
(Kenig et al., 2001) , which requires accurate thermodynamic modelling and physical
properties calculations for modelling this process. Generally, 30 wt% MEA solution is
considered a benchmark solvent for this process . Thermodynamic data, especially about
the CO 2 solubility in MEA aqueous solutions, around this condition have been reported
(Jou et al., 1995; Harris et al., 2009; Tong et al., 2012) . In addition, data covering wider
MEA solution concentration range have also been reported (Mason and Dodge, 1936;
Lee et al., 1974; Lee et al., 1976; Dang and Rochelle, 2003; Hilliard, 2008; Aronu et al.,
2011; Xu and Rochelle, 2011; Wagner et al., 2013) . Mason and Dodge (1936) presented
CO 2 partial pressure of different CO 2 loaded MEA aqueous solutions with 0 –100 wt%
MEA with the temperature from 0°C to75°C under atmosphere pressure. Aronu et al.
(2011) produced experimental data of vapour -liquid equilibrium (VLE) of CO 2 in MEA
aqueous solutions with 15, 30, 45 and 60 wt% MEA and at temperatures from 40 to 120
°C. A low temperature equilibrium apparatus was first used to measure CO 2 partial
20
pressures over loaded MEA solutions with 1 bar, and then a high temperature
equilibrium apparatus was operated to measure the total pressures from 1 bar to 10.5
bar. Wagner et al. (2013) published new experimental data for the CO 2 solubility of in
aqueous 15 and 30 wt% MEA aqueous solutions at 313, 353, and 393K with a wider
range of partial pressures of CO 2 between 0.001 and 8.6 MPa. With those data, it is
possible to develop and validate a reliable thermodynamic model for MEA -H2O-CO 2
system.
For parameterization and validation of properties calculation methods of MEA -H2O-
CO 2 mixture, the experimental data of aqueous MEA solution are valuable especially
with CO 2 loaded. The correlations for the calculation of density and viscosity of MEA –
H2O-CO 2 mixture at different temperature s and MEA concentration s can be found in the
literature (Cheng et al., 1996; Hartono et al., 2014; Littel et al., 1992; Weiland et al.,
1998; Han et al., 2012) . In the study of Han et al. (2012) , liquid densities of CO 2 loaded
aqueous MEA solutions were measured with 30, 40, 50, and 60 wt% MEA and at
temperatures from 298.15 to 413.15 K. Surface tensions of unloaded MEA solutions
were also measured at temperatures from 303.15 to 333.15K with MEA concentration
ranged from 0 wt% (pure water) to 100 wt% (Pure MEA) . Ying and Eimer (2012)
measured the diffusivities of N 2O in MEA aqueous solution and calculated the
diffusivities of CO 2 in MEA aqueous solution by N2O analogy method. They found that
the diffusivities of CO 2 in MEA aqueous solution decrease with increas ing of MEA
concentration and increase when solution temperature rises.
2.2.1.2 Mass transfer and thermal performance
For the mass transfer and thermal performance of the integrated PCC based on chemical
absorption process , there are many research projects having been implemented world –
wide (CCP, 2000; CASTOR, 2004; CO2CRC, 2003; BIGCO2, 2007; CESAR, 2008) . In
most of these studies, MEA is chosen as a reference solvent for validation of the models
and scale -up, sometimes also for comparison with new solvents investigations. The
information about pilot plants experimental data obtained for the solvent MEA can be
found in some reports (Chapel et al., 1999; Faber et al., 2011; Mang alapally and Hasse,
2011; Tobiesen et al., 2007; Dugas, 2006; Notz et al., 2012) .
21
Dugas (2006) present ed a great number of experimental data about separation
performance and mass transfer of the absorber and the stripper respectively. Their pilot
plant consists of the absorber and the stripper with same diameter of 0.42 7 m and same
packing height of 6.1 m. But their study did not investigate the impact of process
parameters on heat requirement for solvent regeneration of the closed -loop absorption
and desorption process. Tobiesen et al. (2007) published various experimental data for
validation purpose of rigorous modelling for the absorber and the stripper individually.
In one recent contribution of Notz et al. (2012) , the pilot plant with a closed -loop
absorption/desorption process was continuously running . The diameter is 0.125m for
both the absorber and the stripper and packing height (packing type: Sulzer® Mellapak
250YTM) is 4. 2 and 2.25 m respectively. Comprehe nsive experimental studies were
conducted about the impact of several key process conditions and operational variables
such as CO 2 concentration in flue gases, CO 2 capture level, hydraulic parameter of the
absorber, lean solvent flow rate, stripper pressur e and MEA concentration in solvents on
the process behaviour.
2.2.2 Model -based studies
2.2.2.1 Thermodynamic modelling of MEA -H2O-CO 2 system
Accurate thermodynamic modelling and physical properties prediction of pure
components and mixtures is one of the basic prere quisites for the process modelling and
simulation (Lee et al. 1975 ). For high ly non-linear electrolyte MEA -H2O-CO 2 solution,
the electrolyte Non -Random Two Liquid ( eNRTL) model (Song et al., 1996; Chen and
Evans, 1986) are the most widely adopted models . For example, Austgen et al. (1989)
applied eNRTL to correlate CO 2 solubility in aqueous MEA solution. Hilliard (2008)
impoved the model by regressing correlations of phase equlibrium , heat of absorption
and heat capacity and predicted composition concentrations in MEA aqueous solutions
loaded with CO 2 (Hilliard, 2008; Böttinger et al., 2008) . Hessen et al. (2010) improved
the eNRTL model from Bollas et al. (2008) to correlate CO 2 solubility in MEA aqueous
solutions and to predict the composition in MEA -H2O-CO 2 system . PC-SAFT EOS was
used for vapour phase fugacit y coefficients of CO 2 with system temperature up to 500 K
and system pressure up to 150 MPa . The results was compared with REFPROP EOS
22
(Span and Wagner, 1996) developed specifically for the property prediction of pure
CO 2.
The prediction of VLE of MEA -H2O-CO 2 system largely depends on the accurate
calculation of CO 2 solubility in MEA aqueous solutions, which is determined by both
its physical solubility and chemical equilibrium in aqueous solutions (Zhang et al.,
2011) . Physical solubility is the equilibrium between CO 2 molecules in vapour phase
and liquid phase and it can be calculated by Henry's law . The available binary Henry’s
constants are summarized i n Table 2.1. In the system of MEA -H2O-CO 2 mixture, early
studies (Austgen et al., 1989; Yan a nd Chen, 2010) only cons idered Henry’s constants
for CO 2 with H 2O and regressed its value from extensive amounts of experimental VLE
data for the CO 2-H2O binary system. Some of them also considered binary Henry’s
constants for the CO 2-MEA. In one recent study (Wagner et al., 2013) , MEA was
regarded as a Henry component because they claimed MEA could evaporate in the
column resulting in higher solvent make -up requirement.
Table 2.1 Correlations for the calculation of Henry’s constants
Henry
constants Unit C1 C2 C3 C4 T (°C) Source
CO 2 , H2O Pa 170.7126 −8477.711 −21.95743 0.005781 0–100 Chen et al.
(1979)
CO 2, H2O Pa −9624.4 −28.749 0.01441 192.876 273–
473 Rumpf and
Maurer (1992)
CO 2, H2O Pa 100.650 −6147.7 −10.191 0 273–
473 Yan and Chen
(2010)
CO 2, MEA Pa 20.1759 −1183.5 0 0
Aspen Databank
(2012b)
MEA, H 2O MPa −11803.5 −10.617 0 84.599 288–
408 Wagner et al.
(2013)
The chemical equilibrium in aqueous solution of MEA -H2O-CO 2 systems can be
presented by a series of equilibrium reactions in an acid -base buffer mechanism
(Austgen et al., 1989) . The chemical equilibrium constants of those reactions can be
23
estimated in two ways. Most models (Austgen et al., 1989) used a po lynomial
correlation with parameters regressed using experimental data as in Equation (2.1).
(2.1)
where is the chemical equilibrium constants for each equation , is system
temperature, are correlations for chemical equilibrium constants .
Another method is to calculate chemical equilibrium constants from the reference state
Gibbs free energies of the participating components (Zhang et al., 2011) , as in Equation
(2.2).
(2.2)
where is the chemical equilibrium constant of reaction j , is the reference state
Gibbs energy change for reaction j, is the universal gas constant, and is the system
temperature.
2.2.2.2 Rate -based model for solvent -based PCC Process
Using MEA solvent to absorb CO 2 is a reactive absorption process. A rate-based
approach for both mass transfer and reactions ( see Figure 2. 3) offers a more accurate
prediction than equilibrium -stage approach (Kenig et al., 2001; Lawal et al., 2009) .
Gas absorption into liquid in the absorber and gas desorption from liquid in the stripper
are fundamental for solvent -based PCC process . Various theories , including two -film
theory (Whitman, 1962) , penetration theory (Higbie, 19 35), surface renewal theory
(Danckwerts, 1951) and Eddy diffusivity theory (King, 1966) could be used to explain
the phenomenon of mass transfer inside columns, Two-film theory (Whitman, 1962) is
widely used to describe the mass transfer of components across the gas phase and the
liquid phas e in packed columns . In each phase, the thickness of the film is determined
as the ratio of the average diffusivity to average mass transfer coefficient, the
24
calculation of film resistance is improved by discretizing the films (see Figure 2.4 ) and
reaction s are considered in the liquid phase (Austgen et al., 1989) .
Note: G represents gas phase ; L represents liquid phase ; represents chemical potential; ® represents
kinetic -controlled reaction model .
Figure 2.3 Model complexities for reactive absorption process (Kenig et al., 2001) .
Zhang et al. (2009) published the details of a rate -based model development for the
absorber in Aspen Plus®. The mo del was validated by comparing model predictions of
lean solvent loading, rich solvent loading, capture level and the temperature profile with
the experimental data from University of Texas at Austin (Dugas, 2006) . The study
showed that the rate -based model using Aspen P lus® was proven to be capable of
providing acceptable accuracy for performance prediction of solvent -based PCC plant.
In their recent study (Zhang and Chen, 2013) , the kinetics of forward and reverse
reactions for carbamate formation and bicarbonate formation were improved with new
experimental data (Mangalapally and Hasse, 2011) . The significant contribution is that
the value of kinetic of reverse reactions for bicarbonate formation is different for the
absorber and the stripper, which reflects the nature of different operating conditions of
the absorber and the stripper.
25
Note: V represents vapour phase; Y represents mole fraction in vapour phase; L represents liquid phase;
represents mole fraction in liquid phase; T represents chemical potential .
Figure 2.4 Discretized liquid film for counter current flow (Zhang et al., 2009)
The correlation selection for rate -based mass transfer also ha s large impact on the
prediction accuracy (Kvamsdal et al., 2011a; Razi et al., 2012) . It mainly includes mass
transfer coeffici ents, interfacial area, liquid holdup and pressure drop inside packing
beds. Razi et al. (2012) discovered large differences of the model prediction resu lts for
different correlations used in the model and they recommended that model validation
using pilot plant or commercial data is required for accurate prediction.
2.2.2.3 Model -based optimal design and operation of PCC processes
The process flow diagram can b e seen in Figure 2.5. The flue gas is treated by a
preconditioning process (desulfurizing and cooling) and then enters the absorber, in
which, lean amine solvent reacts with the CO 2. The scrubbed flue gas is emit ted to the
atmosphere and the CO 2-rich solvent is discharged from the bottom of the absorber and
enters the stripper. The CO 2-rich solvent is regenerated inside the stripper with heat
input to the reboiler. The regenerated solvent is cooled and recirculate d to the absorber
for reuse.
26
Figure 2.5 Process flow diagram of solvent -based PCC (IPCC, 2005)
In order to describe this process better , several technical items are defined as follows.
CO 2 capture level (CL) is defined in Equation (2.3) .
(2.3)
where and are mass flow rates of flue gas and exhaust gas respectively,
and are CO 2 mass fraction s in the flue gas and exhaust gas
respectively.
CO 2 loading in lean solvent (lean loading) and rich solvent (rich loading) in mole basis
are defined in Equation (2.4) .
(2.4)
Specific duty is defined by Equation (2.5).
(2.5)
where is heat duty of the reboiler, is mass flow rate of CO 2 captured.
Because of significant energy requirement for solvent regeneration of the solvent -based
PCC processes (Rochelle, 2009) , the cost of carbon capture is high when PCC is
27
equipped to the emitters such as power plants, refineries and cement plants (DECC,
2013) . One of the most important engineering tools for addressing these cost issues is
optimisation (Edgar et al., 2001) . Optimisation of a large process, such as NGCC power
plant integrated with PCC process in this study, can involve several levels such as
process configurations (Amrollahi et al., 2012; Oyenekan and Rochelle, 2007; Sipöcz
and Tobiesen, 2012) , equipment designs (Agbonghae et al., 2014; Mores et al., 2014;
Canepa and Wang, 2015) (see Table 2.2), controlled variables of plant operations (Abu –
Zahra et al., 2007a; Kvamsdal et al., 2011b; Mac Dowell and Shah, 2013) as well as
control strategies (Panahi and Skogestad, 2011; Schach et al., 2013) .
Most earl y studies were carried out for parametric studies of solvent -based PCC
processes in the context of coal -fired power plants, which forms the base for late r
researches on PCC process in context of gas -fired power plants. Abu -Zahra (2007b)
investigated carbon capture from the flue gas of a 600 MW e bituminous coal fired
power plant using Aspen Plus®. The results proved that several key variables, such as
CO 2 capture level, MEA concentration, CO 2 loading in lean solvent, stripper operating
pressure and lean solvent temperature, have significant impact to energy requirement for
solvent regeneration. A minimum specific duty of 3.0 GJ/ton CO 2 was achieve at lean
loading of 0.3 mol CO 2/mol MEA , a 40 wt% MEA solvent and a 2.1 bar stripper
operating pressure, compared to 3.9 GJ/ton CO 2 in the base case with 30 wt% MEA .
However, in this study, equilibrium -based approach was used for modelling both the
absorber and stripper and this adds big uncertainty to the results.
Temperature bulges in the absorber were demonstrated by Kvamsdal et al . (2008) with
variations of L/G ratio , solvent type , height of packing, and flue gas CO 2 concentration.
In their later publication (Kvamsdal et al., 2011b) , they discovered that flue -gas cooling
(30–50°C) has benefits for both coal -fired case and natural gas -fired case. Inter-cooling
only has a positive effect for the coal case but a negative cost effect for the natural gas
case.
28
Table 2.2 Literature review of key parameters of optimal PCC process for NGCC power plant at the industrial scale
Description Kvamsdal et al.
(2010) Sipocz
and
Tobiesen (2012) Biliyok
and
Yeung (2013) Agbonghae et
al. (2014) Mores et al .
(2014) Canepa et al .
(2015)
Power plant size (MW e) 540 410.6 440 450 788 427
Flue gas flow rate (kg/s) 1045.6 639.61 693.6 725 – 702
CO 2 concentration (mol%) 3.5 4.2 3.996 4.00 3.99 –
CO 2 capture level (%) 90 90 90 90 90 90
MEA concentration ( wt%) 30 30 30 30 30 32.5
Liquid/Gas ratio (g/g) 0.87 0.68 1.04 0.96 – 0.97
Lean loading (mol CO 2/mol MEA) 0.216 0.132 0.234 0.2 0.159 0.2
Rich loading (mol CO 2/mol MEA) 0.47 0.473 0.495 0.483 0.451 0.477
Specific duty (GJ/ton CO 2) 3.77 3.97 4.003 3.96 4.35 4.1
Absorber Number 4 1 4 2 4 3
Dameter (m) 9.6 9.13 10 12.88 11.9 10.3
Packing height (m) 13.6 26.9 15 19.99 30.6 25
Packing type Mellapak 250X Mellapak 250a Mellapak 250X Mellapak 250Y IMTPa IMTP no. 40
Stripper Number 1 1 1 1 1 1
Diameter (m) 6.2 5.5 9 7.74 4.2 7.4
Packing height (m) 7.6 23.5 15 28.15 8.2 15
Packing type Mellapak 250X Mellapak 250a Mellapak 250X Mellapak 250Y IMTPa Flexipack 1Y
Pressure(bar) 1.912 1.92 1.5 1.62 2 2.1
Economics LCOE(€/MWh) – 80.30 70.00 20.84 60.82 68.00
CCA (€/ton CO 2) – 99.67 51.00 63.79 63.38 –
a. The detailed size of the packing was not given in the publication.
29
The study by Mores et al. (2012) carried out different cost optimisations including both
investments and operating costs for a typical chemical absorption PCC process. The
cost model was developed based on empirical equation with correlations. Using the
model, detailed investigations were performed about the impacts of different CO 2
capture level on the total annual cost, operating variables and equipment sizes. Later,
they (Mores et al., 2014) develop ed an equations -oriented optimisation model for power
plants coupled to CO 2 capture process. The electricity cost, CO 2 avoidance cost, energy
penalties, as well as the optimal values of decision variables were investigated. In the
context of a 731 MW e NGCC power plant with the PCC process, the optimal overall
CO 2 capture level of 82.1% was achieved with three capture trains with 94.8% capture
level of each train, whilst 13.4% of the flue gas stream is bypassed. The avoidance cost
is €63.38 per ton of CO 2 captured.
Razi et al. (2013a) applied Aspen RateSep to study alternative absorber designs for a
gas-fired power plant and a coal -fired power plant respectively, both with a power
output of 400 MW e and a 90.0% of CO 2 capture level. Large electrical energy savings in
the flue gas blower (decreasing from 4493 kW to 2223kW) was found follow ing 52%
decrease of the pressure drop when the diameter of the absorber increased from 16 m to
18m. However, the investment is slightly increased because of increase in the column
cross sectional area. The optimal values of the flooding factor of the absorber in the two
cases were 71.0 and 74.0% respectively.
In a recent study (Agbonghae et al., 2014) , optimal designs have been carried out for
two MEA -based carbon capture plants for gas -fired power plants of 400 MW e and 450
MW e respectively. Mellapak 250Y structured packing was used in t he absorber and the
stripper. The optimal lean CO 2 loading is about 0.2 mol CO 2/mol MEA. The optimal
L/G ratio for a NGCC power plant with a flue gas with 4 mol % CO 2 is about 0.96,
while it is from 2.68 to 2.93 for a flue gas.
Arias et al. (2016) conducted comprehensive sensitivity analyses of main parameters of
a solvent -based PCC process. The results revealed that the temperature of flue gas feed,
lean solvent, rich solvent have high sensitivities to the specific total cost.
30
2.3 NGCC integrated with solvent -based PCC
2.3.1 NGCC integrated with PC C
For a typical NGCC power plant, the LHV efficiency approaches 60% while the CO 2
per kW h electricity generated is only about half of the coal -fired power plant of
equivalent capacity . These advantages will promote more NGCC power plants to be
built in th e next decade, especially in developed countries. However NGCC is not a
neutral carbon technology. The blue map of IEAGHG (2010) shows NGCC equipped
with carbon capture will contribute 5% electricity supply in 2050 to achieve the target
of CO 2 emission reduction (see Figure 2.6).
Figure 2.6 BLUE map emission reduction plant (IEAGHG, 2010)
MEA -based post -combustion chemical absorption is the most likely technology to be
implemented for carbon capture from fossil fuel power plants (Wang et al., 2011;
IEAGHG, 2012) . This is because: (1) amine scrubbing, typically using MEA as a
solvent, is a proven technology for CO 2 separation from flue gas (Rochelle, 2009) ; (2)
they easily retrofits to existing power plants; and (3) it is easy to bypass the carbon
capture process if need be .
2.3.2 Energy penalty
In amine scrubbing, large amount of thermal energy is required f or rich solvent
regeneration in the stripper (Rochelle, 2009) . When NGCC power plant is integrated
with PCC capture plant, an efficiency penalty was reported with a reduction of net plant
efficiency from 58.5% to 50.6% (Marchioro Ystad et al., 2013) . This energy penalty
31
comprises: (1) steam extraction from power plant for solvent generation; (2) power
consumption of CO 2 compression and (3) au xiliary power consumption for PCC capture
plant.
For MEA -based PCC process, t he typical thermal energy required to regenerate 1 ton of
CO 2 is between 3.4 GJ and 4.2 GJ (Abu -Zahra et al., 2007b; Kvamsdal et al., 2007; Mac
Dowell and Shah, 2013; Marchioro Ystad et al., 2013; Canepa and Wang, 2015) . Recent
research efforts focus on how to improve the capture plant efficiency to reduce the
energy requirem ent for CO 2 captured. In the study of Abu -Zahra et al (2007b) several
key parameters such as lean solvent loading, CO 2 capture level, MEA concentration in
solvent and stripper operating pressure were examined. The lean solvent loading was
found to have a major effect on the the rmal energy requirement. The economic range of
lean solvent loading is 0. 29–0.32 mol CO 2/mol MEA for MEA concentration in solvent
of 30–40 wt%. High operating pressure in the stripper would lead to a reduction of
energy requirement of both solvent regeneration and CO 2 compression. Mac Dowell
and Shah (2013) conducted a cost optimisation study of a capture plant. The results
showed when the capture level increase from 85% to almost 100%, the optimal energy
requirement of per ton of CO 2 decrease to 3.8 from 4 .2 GJ with optimal lean loading
0.18–0.22 mol CO 2/mol MEA , which is obviously lower than the result of Abu -zahra et
al. (2007). In a recent study by Canepa and Wang (2015) , a sensitivity analysis was
conducted for a capture plant scaled up to meet a 427MW e CCGT power plant. The
optimal specific duty was approximately 4 .1 GJ/ton CO 2 with a 0.2 lean loading,
0.97mol/mol L/G ratio (Table 2.2).
2.3.3 Exhaust gas recirculation (EGR) technology
Compared with coal fired power plants, NGCC power plant emits only half CO 2 per
unit power. Consequently, the CO 2 concentration in flue gas from a n NGCC power
plant is as low as 3 .5–4.5 mol% whilst it is 11 –13 mol % for flue gas from a coal fired
power plant (Agbonghae et al., 2014) . Low CO 2 concentration causes low absorption
efficiency whilst large flow rate of inert gas requires big equipment size in PCC capture
plant (Jonshagen et al., 2011) . EGR is regarded as an effective solution (Biliyok and
Yeung, 2013) . The flue gas leaving the HRSG is split into two streams. One is lined to
the PCC process and the other is cooled and recirculated to compressor inlet where it is
32
mixed with fresh air. Thus the flow rate of fresh air intake reduces greatly.
Consequently, the flow rate of flue gas going to be treated by the PCC process would
decrease largely whilst the CO 2 concentration in the flue gas increase obviously
(Canepa et al., 2013) . Sipöcz and Tobiesen (2012) presented thermodynamic and
economic analyses of a 440 MW e NGCC plant integrated with a n MEA -based PCC
process , combin ing absorber intercooling , lean vapour recompression and EGR options
together . The results showed that EGR adds significant benefits for reducing the
operating and investment costs.
EGR ratio is defined as Equation (2.6):
(2.6)
Figure 2.7 Impact of EGR, (a) on O 2 concentration in combustion air feed, and (b) on
exhaust gas compositions (Canepa et al., 2013)
691215182124
0 10 20 30 40 50O2concentration (mol%)
EGR (%)
110100
0 10 20 30 40 50Exhaust gas composition (mol%)
EGR (%)O2
H2O
CO2N2(a)
(b)
33
The impacts of EGR can be seen in Figure 2.7. Figure 2.7(a) illustrates the change of O 2
concentration in combustion air when EGR ratio varies. In Figure 2.7(b), exhaust gas
composition is shown as a function of the EGR ratio. With the increase of EGR ratio,
the concentrations of N 2, H 2O and CO 2 increase. But O 2 concentration decreases
because less oxygen is available in the recirculated stream. The maximum EGR ratio of
flue gas recirculation is limited by combustion performance. It is believed that the
changes in turbomachinery performance may be very small with an oxygen
concentration in combustion air of minimum 16 –18 mol % (Ulfsnes et al., 2003; Canepa
et al., 2013) .
2.4 CO 2 transport pipeline network
2.4.1 CO 2 transport pipeline
CO 2 transportation is one important section of whole CCS chain. Captured and purified
CO 2 is compressed and transported from the capture plant to other sites for storage or
reutilization by pipeline, ship or tanker trucks mainly depending on the distance.
Pipelines are the preferred method for onshore and offshore transport of large volumes
of CO 2 (Svensson et al., 2004; IPCC, 2005) . Pipelines have been used to transport CO 2
in gaseous and dense (i.e. sub -cooled liquid or supercritical) phases. The dense phase is
regarded as the most energy -efficient condition due to its high density and low viscosity
(Zhang et al., 2006; McCoy and Rubin, 2008) . Consequently, current oper ating practice
for CO 2 pipelines is to maintain the pressure well above the critical pressure.
2.4.2 EOS selection
The cubic equation of state (EOS) such as Soave -Redlich -Kwong (SRK) (Soave, 1972)
and Peng -Robinson (PR) (Peng and Robinson, 1976) has been widely used to calculate
the physical properties of the CO 2 and impurit ies (Li and Yan, 2009) . More complex
EOS such as Lee -Kesler (Lee and Kesler, 1975) , the Statistical Associating Fluid Theory
(SAFT) (Wertheim, 1984; Wertheim, 1986) , Span and Wagner (SW) (Span and Wagner,
1996) and GERG (Kunz and Wagner, 2012) were used in recent studies . SW is accurate
for p ure CO 2 as it was specially developed for pure CO 2. But it is difficult to generalize
for multi -component mixture (Diamantonis et al., 2013) because it contains many terms,
34
some of which are complex expo nential for computation (Kim, 2007) . Molecular -based
SAFT is an attractive EOS for CO 2 including impurities because of better per formance
than other models for predicting thermodynamic properties of several complex mixtures.
SAFT -VR, one of modifications of original SAFT, is used for CO 2 capture process (Mac
Dowell et al., 2009; Mac Dowell et al., 2011) . But SAFT is not yet used in published
literatures focusing on the dense phase pipeline transport of the CO 2 and impurities.
GERG is the international reference equation of state for natural gas. The accuracy of
GERG EOS cl aims to be very high covering a large part of the T/P range for CCS
application. GERG was used in recent studies emphasizing the transient behaviours of
the CO 2 and impurities in dense phase pipeline transport (Liljemark et al., 2011;
Chaczykowski and Osiadacz, 2012) . However the average absolute deviations (AAD) of
the liquid vo lume of CO 2 mixtures could reach up to 18% (Li et al., 2011) .
There is no consensus in literature regarding the best EOS for design of CO 2 transport
pipeline. PR EOS was chosen in some studies (Zhang et al., 2006; Seevam et al., 2008;
Mahgerefteh et al., 2008) giving reasonable results for properties of the CO 2 and
impurities. Li et al. (Li and Yan, 2009; Li et al., 2009) conclu ded that calibrating the
binary interaction parameters ( ) based on experimental data improve s the accuracy of
EOS after comparing results generated with the from literature and obtained
through calibration . Their later study (Li et al., 2 011) indicated that SAFT have better
accuracy than PR for volume calculation, but PR is better for VLE calculations.
(Diamantonis et al., 2013) compared the results of several EOS with experimental data
and found that PR EOS is of reasonable accuracy, even when compared with more
advanced EOS such as SAFT and PC -SAFT, when calibrated binary interaction
parameters are used.
2.4.3 Modelling and simulation studies
The impurities in CO 2 stream have great impacts on th e design, operation and
optimisation studies (Li et al., 2009; Li and Yan, 2006; Race et al., 2012) . Seevam et al.
(2008) studied the impact of the impurities on phase behaviour and density of CO 2. The
presence of the impurities may result in the formation of gaseous CO 2 or two -phase CO 2
flow inside the pipelines. The water content in the CO 2 stream may cause hydrate
formation, which results in flow assurance problems involving phase transient and
35
pipeline blockage (Race et al., 2012; Chapoy et al., 2011; Kvamme et al., 2014) .
Therefore, before the pipeline transport, the CO 2 stream has to be con ditioned to remove
impurities such as water vapour, H 2S, N 2, methane, O 2, hydrocarbons and free water
(Aspelund and Jordal, 2007; Koornneef et al., 2010) .
Steady state simulations and analysis were carried out to calculate pressure drop,
temperature profile and mass flow in the pipelines. Zhang et al. (2006) studied the
density and pressure profiles of CO 2 stream along the length of the pipeline with
different inlet temperatures. Maximum length of pipeline, in which CO 2 stream stays in
dense phase, is determined for different inlet temperatures. In the study of Nimtz et al.
(2010) , the model includes the pipeline and an injection well for pure CO 2 stream. The
profiles of pressure, temperature, density and flow velocity were presented for several
cases and the phase change was found and discussed. Re garding the dynamics of
pipeline systems, there is little work reported in the literature. Liljemark et al.
(2011) developed a pipeline transfer function model to evaluate phase transition of the
transported CO 2 mixture. Ope ration scenarios of pipeline cooling, load change, start -up,
shut-down and compressor trip were simulated. Chaczykowski and Osiadacz (2012)
built a first principle single -phase compressible flow model, suitable for supercritical
and dense -phase calculations, to examine the hydraulic parameters of the CO 2 pipeline.
However, these simulations were per formed for a single CO 2 emission source without
intermediate boosters. This may not reflect realistic operating scenarios for a typical
CO 2 pipeline network system.
2.4.4 The cost of CO 2 pipeline transport
The cost of the CO 2 pipeline transport can become signif icant when the distances
between the storage locations and the emission sources are more than a few hundred
kilometres. Collecting CO 2 mixture from several emitters into trunk pipelines is more
cost-effective than the use of separate pipelines (Chandel et al., 2010; IPCC, 2005) . As
a part of economic evaluation of CCS deployment, some research efforts were given on
the cost estimate of CO 2 pipeline transport. Van den Broek et al. (2010) , Heddle et al.
(2003) and Pershad et al. (2010) used a linear cost related with diameter and length of
the pipelines to calculate investment costs. Gao et al. (2011) developed a cost model
based on the weight of the pipeline , which is specific for the Chinese market. In the
36
report of IEAGHG (2002) , six different kinds of coefficients, for 600#, 900# and 1500 #
ASNI class and onshore/offshore pipelines, were used for the operating and
maintenance costs calculation of CO 2 transport pipelines. McCoy and Rubin (2008)
developed a cost equation for pipeline transport with different parameters for each cost
category (material, labour, right of way and miscellaneous costs) for different regions of
the USA. Dahowski et al. (2009) and McCollum and Ogden (2006) built their linear
cost equations only based on the flow rate of CO 2 stream and the length of the pipelines .
The cost of transporting (without the compression) CO 2 by a 100 km onshore pipeline
was estimated by the Global CCS Institute (GCCSI) at 0.46 –1.55 €/t CO 2 (GCCSI,
2011) . However, large ranges for capital and levelise d costs of CO 2 transportation were
found for a given diameter (Ogden et al., 2004; Wildenborg et al., 2004; McCollum and
Ogden, 2006; Knoope et al., 2013) . For example, Knoope et al. (2013) came up with a
cost range of 0.6 –11 M€/km for a 0.91 m diameter pipeline after comparing seven
different models.
2.5 The studies on whole CCS chain
In terms of power plant integrated with whole CCS chain , like the schematic in Figure
2.8, most of studies focus on the overall performance combined with the cost
performance of the power plants integrat ed with carbon capture process. Few of them
considered the CO 2 transport section and geologic storage section.
In the study from Rao and Rubin (2002) , CO 2 dense phase pipeline transport and
geologic storage was taken into account in the integrate system of power plants and
MEA -based PCC process. It is f ound that the design assumptions for all sections of
whole CCS chain significantly affect the cost of CO 2 avoided. For the optimal operation
of capture process with power plant, Rao and Rubin (2005) found that the relationship
between the cost and carbon capture level is non -linear and venting a fraction of flue
gas to keep low capture level less than 75% could achieve a significant cost saving.
37
Figure 2.8 Schematic of a full CCS chain (SCCS, 2016)
Mores et al. (2012) found that the total annual cost of carbon capture plant varies
linearly for carbon cap ture level within a range of 70 –80% but it increases exponentially
when carbon capture level increases from 80% to 95%. Cohen at al . (2012) investigated
the economic benefits of a 50 0 MW e coal-fired power plant with CO 2 capture for a
carbon pricing from 0 to 200 US$/ton CO 2 and concluded that CO 2 capture investment
is unjustifiable at low CO 2 prices. In the study by Mac Dowell and Shah (2013) , optimal
CO 2 capture level is 95% for £30/ton CO 2 and £90/MWh scenario and is around 70%
for £8/ton CO 2 and £55/MWh scenario for a 660 MW e coal fired power plant integrated
with a capture plant. Their result shows that carbon price should be more than £40/ton
CO 2 to justify the total cost of carbon capt ure for an objective of capture level greater
than 90% without considering the costs of CO 2 compression, transport and storage.
2.6 Concluding remarks
From the above literature review, several research gaps have been identified towards the
readiness of solven t-based PCC process for commercial deployment for power plants.
Firstly, most of the studies (Abu -Zahra et al., 2007b; Kvamsdal et al., 2011b; Lawal et
al., 2012; Mac Dowell and Shah, 2013) on solvent -based PCC process were carried out
in the context of coal-fired power plants and only a few papers (Kvamsdal et al., 2010;
38
Mores et al., 2014; Canepa et al., 2013) focus on its application for NGCC power
plants. Compared to coal-fired power plant s, CO 2 concentration in the flue gas is much
lower for a gas fired power plant which causes some significantly different features in
terms of the economic performance such as bigger equipment size and lower L/G ratio.
Thus the research outputs of carbon capture for a coal -fired power plant may not be
applied directly to NGCC power plant.
Second ly, in current studies on solvent -based PCC process for gas -fired power plants,
the optimal ranges are very large for key equipment design features (such as diameters
and packing heights of the absorber and the stripper) and key operational variables
(such as lean loading and L/G ratio). For example, the packing height varies from 13.6
m (Kvamsdal et al., 2010) to 30.6 m (Mores et al., 2014) for the absorber and from 7.6
m (Kvamsdal et al., 2010) to 28.15m (Agbonghae et al., 2014) for the stripper for
similar captur e tasks, which has large impact to the capital cost. The optimal lean
loading range is equally wide from 0.132 mol CO 2/mol MEA (Sipöcz and Tobiesen,
2012) to 0.234 mol CO 2/mol MEA (Biliyok et al., 2013) with correspondi ng specific
duty at a range of 3.77–4.35GJ/ton CO 2. The significant inconsistencies in the literature
cause confusions for future researches in this field . It also may cause some troubles for
feasibility studies of industr ial design of solvent -based PCC process.
Finally, most of current studies focused on the solvent -based PCC process itself, and
some of them explored the integration of the power plants with PCC process. Few of
them considered the optimal design and operation of the power plants int egrated whole
CCS chain. In fact , CO 2 transport and storage sections are strongly linked with carbon
capture process via entry requirements of temperature, pressure and purities for CO 2
stream . Their capacity and costs significantly influence the optimal d esign and operation
of carbon capture for power plants.
The main reasons for above gaps may be related to the conflicts between the complexity
of the integrated system and the accuracy requirement for both technical and economic
performance prediction of the modeling and simulation studies. It could be analysed as
follows . (1) The models were relatively simple in some early publications. The papers
published by Abu -Zahra et al. (Abu -Zahra et al., 2007b; Abu -Zahra et al., 2007a) are
two of most cited papers in this field but the equilibrium models were used for both the
39
mass transfer and reaction in the absorber and the stripper. (2) Impropriate correlations
were wrongly used in the models. Several publications (Agbonghae et al., 2014; Lawal
et al., 2012; AspenTech, 2008b) were found that using Brav o et al. (1985) corre lation
for Mellapak 250 X/Y and Flexipak. Actually, Brava et al. (1985) correlation was
obtained for wire gauze structured packing whilst Mellapak 250 X/Y and FlexipakTM
are metal sheet structured packing. There are obvious difference s between gauze
structured packing and sheet structured packing in terms of the hydraulic performance
such as effective wetted area, liquid hold -up and pressure drop (Sulzer, 2015; Koch –
Glitsch, 2015) . (3) Lacking of engineering experience caused some unrealistic designs,
especially for the studies towards industrial applications. For large -diameter absorption
column , structured packing is prefer red considering serious maldistribution of both
liquid and vapour phase inside random packing bed (Hoek, 1983; Harriott, 1989) . But in
some papers (Mores et al., 2014; Canepa and Wang, 2015) , random packing was cho sen
for the absorber and the stripper with the diameters larger than 10 m. Low absorption
efficiency of random packing required higher packing height which resulting higher
CAPEX cost. Because of the above reasons, those designs may be suboptimal.
The new studies should be carried out by carefully check ing most updated correlations
for the model s, such as new correlations (Yan and Chen, 2010) for Henry’s constant of
CO 2-H2O for thermodynamic model and new reaction kinetics (Zhang and Chen, 2013)
in rate -based model. The models should be validated with updated experimental data
(Aronu et al., 2011; Han et al., 2012; Notz et al., 2012) to ensure the predictions
accuracy. On the other hand, detailed design with vender quotes for the solvent -based
PCC process (IEAGHG, 2010; IEAGHG, 2012) provided a solid base for dev eloping
accurate cost model, rather than using empirical correlations developed on the basis of
historical cost data. By implementing non -linear optimization programming based on
the above process model and cost model, the study on optimal design and opera tion for
solvent -based PCC process as well as the integrated system could be expected to be
more realistic to support the decision making for the commercial deployment at the
industrial scale.
40
Chapter 3: Model Development of Solvent –
based PCC Process
This chapte r presents the model development and validation of PCC process. Section
3.1 analysed t he framework of modelling of a PCC based on chemical absorpti on
process . In Section 3.2, correlations of thermodynamic modelling were examined and
validated against the experimental data of CO 2 solubility. In Section 3.3 calculation
methods of main physical properties were examined. In Section 3.4, a rate -based steady
state process model was developed and validated with experimental data from a
continual operation pilot plant . In Section 3.5 , the process model was scaled up to
match the capacity requirement for carbon capture from a 453MW e NGCC power plant .
The process model developed and validated in this chapter provides a solid base for the
optimisation studies in Chapter 4, 5, 7.
3.1 Framework of modelling of solvent -based PCC process
Using amine solvent to absorb CO 2 from exhaust gases is a reactive absorption process
involving electrolyte aqueous solvent (Rochelle, 2009) . The numerical modelling of
such a non -ideal multi -components system is a systematic work in different levels.
Figure 3.1 outlines the framework of modelling of such a PCC process. Although the
software package Aspen Plus® was used for the modelling and simulation of the
process, it is important to check the calculation methods with their corrections in order
to ensure the accuracy of process simulation and optimisation .
Accurate calculati ng of physical properties of pure components and mixtures is one of
the basic prerequisites in process modell ing and simulation. As the first step, the
thermodynamic model should be developed to present vapour -liquid phase equilibrium
(VLE) and to calculate the state parameters of the MEA -H2O-CO 2 mixture, such as
temperature, pressure and composition of the liqui d and vapour phase. The solubility of
CO 2 in MEA -H2O-CO 2 mixture is one key parameter and is normally used for
validation purpose for the correlations calibration or selection for VLE calculation. The
acid gas solubility in aqueous amines solutions is determined by both its physical phase
equilibrium and the ch emical equilibrium for the aqueous phase reactions among acid
gas, water and amines (Zhang et al., 2011) .
41
For the simulation of solvent -based PCC process, the absorption and desorption in the
packed column s are the key processes. Rate-based model offers bett er accuracy than
equilibrium model for absorption efficiency and costs of the column s. This accuracy is a
function of the appropriate correlations used for liquid and vapour phase mass transfer
coefficients, the effective gas -liquid interfacial area and th e pressure drop in rate -based
model (Agbonghae et al., 2014) .
Figure 3.1 Framework of modelling of a solvent -based PCC process
3.2 Thermodynamic modelling of MEA -H2O-CO 2 system
3.2.1 EOSs and relevant model parameters
The selection of appropriate property methods is crucial to ensure the accuracy of the
modelling and simulation. In this chapter, the e NRTL (Song and Chen, 2009; Chen and
Evans, 1986; Chen et al., 1982) is used to model the electrolyte system of MEA -H2O-
Thermodynamic
modelling
Physical Property
Prediction
Mass , Heat and
Momentum TransferDensity (molar volume ) (L,V)
Heat Capacity (L,V)
Viscosity (L, V)
Thermal conductivity (L,V)
Diffusivity (L,V)
Surface tension (L)Physical phase
equilibrium
Chemical equilibrium Henry’s law constant
Chemical equilibrium constantFugacity
Activity
Thermodynamic
properties
Transport propertiesEnthalpy (L,V)
Rate-based mass
transfer
Kinetic -controlled
reactionGas phase mass transfer coefficient
Liquid phase mass transfer coefficient
Gas liquid interfacial area
Reaction kinetics
Mass , Heat and
Momentum BalanceBalance inside a unit
Balance of the integrated systemMomentum Pressure dropHeat transfer coefficient
42
CO 2 mixture and the PC -SAFT EOS (Gross and Sadowski, 2001; Gross and Sadowski,
2002) is used to calculate the properties of vapour phase.
3.2.1.1 PC-SAFT EOS for vapour phase
Compared with some typical cubic EOS such as PR and SRK, PC -SAFT EOS is able to
accurately estimate vapour phase fugac ity coefficients at high pressures (Zhang and
Chen, 2011; Zhang et al., 2011) , which is an important advantage for accurate
performance predictions of CO 2 compression section. The PC-SAFT parameters of pure
components were summarized in Table 3.1. The PC-SAFT pure component parameters
for H 2O and CO 2 are taken from Gross and Sadowski (2002) and Aspen Databank
(AspenTech, 2012b) . The parameters of MEA are obtained from Zhang’s regression
work (Zhang et al., 2011) . Table 3.2 listed the PC -SAFT binary interaction parameters
of the binary pairs.
Table 3.1 PC-SAFT parameters of pure components
Component H2O CO 2 MEA
Source Gross and
Sadowski
(2002) Aspen
Databank
(2012b) Zhang et al .
(2011)
segment number parameter, 1.0656 2.5692 2.9029
segment energy parameter, 366.51 152.1 306.2
segment size parameter, 3.0007 2.5637 3.1067
association energy parameter, 2500.7 0 2369
association energy parameter, 0.034868 0 0.01903
Table 3.2 Binary parameters for PC -SAFT EOS
Component pair MEA -H2O CO 2-H2O
Source Fakouri Baygi and Pahlavanzadeh (2015) Yan and Chen (2010)
kij C −0.052 0
43
3.2.1.2 Electrolyte -NRTL for liquid phase
The liquid phase of MEA -H2O-CO 2 mixture is a typical electrolyte solution (Austgen et
al., 1989) . The eNRTL method was validated and used for modelling of electrolyte
solution in many publications (Austgen et al., 1989; Liu et al., 1999; Zhang et al., 2009;
Zhang et al., 2011; Zhang and Chen, 2013) .
Table 3.3 summarized the model parameters and their sources for this study. Most of
the parameters were obtained from Aspen Databank (2012b) . Some of them were
updated by recent studies either by regression using new experimental data (Yan and
Chen, 2010; Mangalapally and Hasse, 2011) . Because of the large numbers, the values
of the parameters were not listed in this thesis but can be obtained from the references .
Table 3.3 Model parameters for eNRTL
Model parameters Component Source
Antoine equation parameters Aspen Tech (2012b)
AspenTech (2012b)
Dielectric constant AspenTech (2012b)
NRTL binary parameters binary Yan and Chen (2010)
– binary Zhang et a l. (2011)
Molecule -electrolyte
binaries Zhang et al . (2011)
, , , , AspenTech (2012b)
, , , , AspenTech (2012b)
, Zhang et al . (2011)
, AspenTech (2012b)
, Zhang and Chen (2011)
, Zhang et al . (2011)
3.2.2 Physical solubility and Henry’s constant
Physical solubility is the equilibrium between CO 2 molecules in vapour phase and in the
liquid solutions , which is calculated by Henry's law, as Equation (3.1):
(3.1)
where is the system pressure, is the CO 2 mole fraction in vapour phase, is
the CO 2 fugacity coefficient in vapour phase which is calculated using the Redlich –
44
Kwong equation of state as modified by Soave (1972) . is the Henry's law
constant of CO 2 in aqueous amine solution , is the CO 2 mole fraction in liquid
phase,
is the activity coeffici ent of CO 2 in aqueous amine solution.
The Henry's constant of the mixture ( ) can be calculated from the binary Henry’s
constants of pure solvents in Equation (3.2):
(3.2)
where is the Henry's constant for binary pairs (i.e., CO 2-H2O, CO 2-MEA ), is
the infinite dilution activity coefficient of molecular solute in the mixed solvent, is
the infinite dilution activity coefficient of molecular solute in pure solvent .
Weighting factor is calculated by Equation (3.3).
(3.3)
where is the mole fraction of solvent on solute -free basis, is the partial molar
volume of molecular solute i at infinite dilution in pure solvent and its detailed
calculation method could refer to Brelvi -O’Connell model (1972) .
The Henry's law constants for CO 2 with water and with MEA are required. They can be
calculated by Equation (3.4).
(3.4)
where is the binary Henry's constant between pure component and , is system
temperature, are correlations for Henry’s constants . The available binary
Henry’s constants of MEA -H2O-CO 2 mixture were summarized in Table 3.4. In the
system of MEA -H2O-CO 2 mixture, most of publications only take gases components
such as CO 2, N 2 as Henry component. Most studies only considered Henry’s constants
for CO 2 with H 2O in their study (Austgen et al., 1989) . The Henry’s law constants for
CO 2 with H 2O have been well studied by Yan and Chen (2010) by examin ing extensive
amounts of experimental VLE data for the CO 2-H2O binary system . Recent studies (Liu
45
et al., 1999) consider ed Henry’s constants for CO 2 with MEA. Normally, MEA is
assumed to be mutual solution with H 2O so MEA is not considered as a Henry
component. One recent study (Wagner et al., 2013) regressed the Henry’s constants
correlations of MEA -H2O then MEA solvent loss in the process could be more
accurately estimated. But their study used Pitzer equation for electrolyte system, not
like eNRTL used in this thesis.
Table 3.4 Correlations for the calculation of Henry’s constants (on the Molality Scale)
Binary pairs CO 2-H2O CO 2-MEA
Unit Pa Pa
Source Yan and Chen (2010) Liu et al . (1999)
C1 100.650 89.452
C2 −6147. 7 −2934.6
C3 −10.191 −11.592
C4 0 0.01644
T (°C) 273–473 280–600
3.2.3 Chemical reaction equilibrium
Aqueous phase chemical reactions involved in the MEA -H2O-CO 2 system can be
expressed as follow s:
R1: water dissociation
(3.5)
R2: Dissociation of CO 2
(3.6)
R3: Dissociation of carbonate
(3.7)
46
R4: Dissociation of the protonated amine
(3.8)
R5: Carbonate formation
(3.9)
Chemical equilibrium constants of those reacti ons are calculated by Equation (2.1) and
the related correlations can be seen in Table 3.5. Once the chemical equilibrium
constants are determined, the chemical equilibrium of each reaction is determined by
Equation (3.10) (Austgen et al., 1989) .
(3.10)
Table 3.5 Correlations for chemical equilibrium constants (on the Molality Scale)
Reaction C1 C2 C3 C4 T (°C) Source
R1 132.899 −13445.90 −22.4773 0 0–225 Edwards et al . (1978)
R2 231.465 −12092.10 −36.7816 0 0–225 Edwards et al. (1978)
R3 216.049 −12431.70 −35.4819 0 0–225 Edwards et al . (1978)
R4 −4.9074 −6166.12 0 −0.00098482 0–50 Bates and Pinching (1951)
R5 2.8898 −3635.09 0 0 25–120 Austgen et al. (1989)
3.2.4 Validation of CO 2 solubility prediction
3.2.4.1 Case set ups
In order to compare and select out appropriate correlations for this study, several
correlation combinations (Austgen et al., 1989; Liu et al., 1999; Zhang et al., 2011)
were chosen for car rying out the validation against the experimental data. The model
details can be seen in Table 3.6.
47
Table 3.6 Different combinations of correlation s for validation
Combinations
of correlation s This study Zhang et al.
(2011) Liu et al.
(1999) Austgen et al.
(1989)
EOS for
vapour PC-SAFT PC-SAFT SRK SRK
EOS for liquid eNRTL eNRTL eNRTL eNRTL
Dielectric
Constants Zhang et al.
(2011) Zhang et al.
(2011) – Ikada et al.
(1968)
NRTL binary Zhang et al.
(2011) Zhang et al.
(2011) Liu et al.
(1999) Austgen et al.
(1989)
Electronic –pair Zhang et al.
(2011) Zhang et al.
(2011) Liu et al.
(1999) Austgen et al.
(1989)
Henry’s
Constants CO 2
in H 2O Yan and Chen
(2010) Yan and Chen
(2010) Chen et al.
(1979) Chen et al.
(1979)
Henry’s
Constants CO 2
in MEA Liu et al.
(1999) Zhang et
al.(2011) Liu et al.
(1999) –
Chemical
equilibrium
constants Liu et al.
(1999) Zhang et al.
(2011) Liu et al.
(1999) Austgen et al.
(1989)
3.2.4.2 Experimental data
The experimental data of CO 2 solubility are normally used for validation of
thermodynamic modelling . It is typical VLE validation. CO 2 partial pressure and/or
total pressure of vapour phase at the different CO 2 loading in MEA aqueous solution
were compared between model predictions and experimental data.
In this study, the experimental data from Aronu et al. (2011) were chosen for the
validation purpose (see Table 3.7) because these data cover a wider range of the MEA
concentration than other publications as well as system temperature and pressure.
Table 3.7 Chosen experimental data for solubility of CO 2 in MEA aqueous solution
Source Temperature
(oC) Pressure
(bar) MEA concentration
(wt%) CO 2 loading
(mol/mol)
Aronu et al. (2011) 40–120 0.001–10.5 15, 30, 45, 60 0–saturated
48
3.2.4.3 Validation r esults
The comparisons of partial pressure of CO 2 in the vapour phase of MEA -H2O-CO 2
mixture between the model predictions and the experimental data for different
concentration MEA could be seen in Figures 3.2 –3.9, in which the lines present the
model predictions whilst the blocks present the experimental data . The abbreviation
representing experimental data in the legends in the figures of is ‘ Exp’.
Table 3.8 presents the mean absolute percentage error (MAPE) of validation results at
the different MEA concentration. Generally, the deviations between experimental data
and model predictions become b igger at the lower (15 wt%) and higher ( 45–60 wt%)
MEA concentration, compared with 30 wt% MEA concentration. It also shows that
model predictions of this study is more accurate than other three models for MEA
concentration range of 15–45 wt% (20–40 wt% ME A concentration used in
optimisation study in Chapter 4) . It is notice that t he model predictions of this study at
15wt% concentration are worse than Liu et al. (1999) . The direct reason is that some
correlations used in this study inherit from Zhang et al. (2011) , which can be seen in
Table 3.6. Further, none of these four combinations could have good predictions
covering low to high MEA concentrations , which reflects one inherent limitation of
correlation method , which is that correlation should not go beyond the conditions of
the data for its regression . However, many correlations used for thermodynamic
modelling of MEA -H2O-CO 2 system were regressed based on the experimental data at
the 30% MEA concentrations.
Table 3.8 MAPE of validation with CO 2 partial pressure of MEA -H2O-CO 2 system
MAPE (%) This study Zhang et al.
(2011) Liu et al.
(1999) Austgen et al.
(1989)
Abbreviation in the
legends in the figures TS Zhang Liu Austgen
15wt% MEA 23.86 43.33 7.97 11.06
30wt% MEA 7.63 6.09 6.4 8.72
45wt% MEA 10.62 11.57 38.76 36.47
60wt% MEA 17.97 20.86 61.9 51.56
15–45wt % MEA 14.04 20.33 17.71 18.75
49
Figure 3.2 CO 2 partial pressure as function of CO 2 loading with 15 wt% MEA
Figure 3.3 CO 2 partial pressure as function of CO 2 loading with 30 wt% MEA
0.0010.010.1110100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8Partial Pressure of CO 2, KPa
CO 2loading, mol CO 2/mol MEAExp_40 °C
Exp_60 °C
Exp_80 °C
TS_40 °C
TS_60 °C
TS_80 °C
Zhang_40 °C
Zhang_60 °C
Zhang_80 °C
Liu_40 °C
Liu_60 °C
Liu_80 °C
Austgen_40 °C
Austgen_60 °C
Austgen_80 °C
0.0010.010.1110100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8Partial Pressure of CO 2, KPa
CO 2 loading, mol CO 2/mol MEAExp_40 °C
Exp_60 °C
Exp_80 °C
TS_40 °C
TS_60 °C
TS_80 °C
Zhang_40 °C
Zhang_60 °C
Zhang_80 °C
Liu_40 °C
Liu_60 °C
Liu_80 °C
Austgen_40 °C
Austgen_60 °C
Austgen_80 °C
50
Figure 3.4 CO 2 partial pressure as function of CO 2 loading with 45 wt% MEA
Figure 3.5 CO 2 partial pressure as function of CO 2 loading with 60 wt% MEA
0.0010.010.1110100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8Partial Pressure of CO 2, KPa
CO 2loading, mol CO 2/mol MEAExp_40 °C
Exp_60 °C
Exp_80 °C
TS_40 °C
TS_60 °C
TS_80 °C
Zhang_40 °C
Zhang_60 °C
Zhang_80 °C
Liu_40 °C
Liu_60 °C
Liu_80 °C
Austgen_40 °C
Austgen_60 °C
Austgen_80 °C
0.0010.010.1110100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8Partial Pressure of CO 2, KPa
CO 2loading, mol CO 2/mol MEAExp_40 °C
Exp_60 °C
Exp_80 °C
TS_40 °C
TS_60 °C
TS_80 °C
Zhang_40 °C
Zhang_60 °C
Zhang_80 °C
Liu_40 °C
Liu_60 °C
Liu_80 °C
Austgen_40 °C
Austgen_60 °C
Austgen_80 °C
51
Figure 3.6 Total pressure as function of CO 2 loading with 15 wt% MEA solvent
Figure 3.7 Total pressure as function of CO 2 loading with 30 wt% MEA solvent
020040060080010001200
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9Total pressure, KPa
CO 2loading, mol CO 2/mol MEAExp_60 °C
Exp_80 °C
Exp_100 °C
Exp_120 °C
TS_60 °C
TS_80 °C
TS_100 °C
TS_120 °C
Zhang_60 °C
Zhang_80 °C
Zhang_100 °C
Zhang_120 °C
Liu_60 °C
Liu_80 °C
Liu_100 °C
Liu_120 °C
Austgen_60 °C
Austgen_80 °C
Austgen_100 °C
Austgen_120 °C
020040060080010001200
0.2 0.3 0.4 0.5 0.6 0.7Total pressure, KPa
CO 2loading, mol CO 2/mol MEAExp_100 °C
Exp_120 °C
TS_100 °C
TS_120 °C
Zhang_100 °C
Zhang_120 °C
Liu_100 °C
Liu_120 °C
Austgen_100 °C
Austgen_120 °C
52
Figure 3.8 Total pressure as function of CO 2 loading with 45wt% MEA solvent
Figure 3.9 Total pressure as function of CO 2 loading with 60 wt% MEA solvent
020040060080010001200
0.2 0.3 0.4 0.5 0.6 0.7Total pressure, KPa
CO 2loading, mol CO 2/mol MEAExp_60 °C
Exp_80 °C
Exp_100 °C
Exp_120 °C
TS_60 °C
TS_80 °C
TS_100 °C
TS_120 °C
Zhang_60 °C
Zhang_80 °C
Zhang_100 °C
Zhang_120 °C
Liu_60 °C
Liu_80 °C
Liu_100 °C
Liu_120 °C
Austgen_60 °C
Austgen_80 °C
Austgen_100 °C
Austgen_120 °C
020040060080010001200
0.2 0.3 0.4 0.5 0.6 0.7Total pressure, KPa
CO 2loading, mol CO 2/mol MEAExp_100 °C
Exp_120 °C
TS_100 °C
TS_120 °C
Zhang_100 °C
Zhang_120 °C
Liu_100 °C
Liu_120 °C
Austgen_100 °C
Austgen_120 °C
53
3.3 Physical property of MEA -H2O-CO 2 system
3.3.1 Physical property model
This study will use rate -based model to simulate the MEA -H2O-CO 2 system . Thus it is
required to calculate quantitative values of physical propert ies. Those physical
properties are part of the correlations for heat transfer, mass transfer, interfacial area,
liquid holdup and pressure drop, etc. It is important to choose the right property models
to ensure the success of process modelling and simulation.
The physical properties include (1) thermodynamic properties such as density and heat
capacity, (2) transport properties such as viscosity, surface tension, thermal
conductivity, and diffusivity. The chosen models for property calculation for mixture in
this study were listed in Table 3.9. It should be noticed that the correlations of density
of liquid mixture is from Han et al. (2012) by coding Fortran subroutine in Aspen Plus®.
Table 3.9 Correlations used for property calculation of the mixture
Property Phase Correlation
Thermodynamic
Properties Density liquid Han et al. (2012)
vapor PC-SAFT
Enthalpy liquid eNRTL
vapor PC-SAFT
Heat capacity liquid Calculated from Enthalpy
vapor Calculated from Enthalpy
Transport
Properties Viscosity liquid Jones -Dole model
vapor Chapman -Enskog -Brokaw
Diffusivity liquid
(molecule) Wilke -Chang
liquid (ion) Nernst -Hartly
vapor Dawsom -Khoury -Kobayashi
Thermal
conductivity liquid Sato-Reidel
vapor Stiel-Thodos
Surface tension Liquid Hakim -Steinberg -Stiel
54
3.3.2 Available experimental data for validation
The available literature experimental data of physical properties validation of MEA –
H2O-CO 2 can be seen in Table 3.10. The vapour phase of MEA -H2O-CO 2 mixture under
operating temperature (20 –150oC) and pressure (1 –2 bar) of the absorber and stripper is
not an issue so there is no available experimental data for those properties of vapour
phase. Available experimental data for the thermal conductivity of liqu id phase were not
found currently. Further, d irect measurement of CO 2 diffusivity in MEA aqueous
solution is impossible because CO 2 reacts with MEA. The NO 2 analogy method was
used to produce the data of CO 2 diffusivity (Ying and Eimer, 2012) .
Table 3.10 Available experimental data for physical properties of liquid phase
Property Temperature
(oC) MEA concentration
(wt%) CO 2 loading
(mol/mol) Source
Density 25–140 30,40,50,60 0.1–0.6 Han and Eimer (2012)
Heat capacity 25 10,20,30,40 0–0.5 Weiland et al. (1997)
Viscosity 25 10,20,30,40 0–0.5 Weiland et al. (1998)
Surface tension 25 10,20,30,40 0–0.5 Weiland (1996)
3.3.3 Validation results
The comparisons of different properties of MEA -H2O-CO 2 mixture between the model
predictions and experimental data for different concentration MEA could be seen in
Figure 3.10 – Figure 3.16. In these figures, the lines present the modelling results whilst
the blocks present the experimental data. The names for short representing experimental
data and model predictions in the legends in the figures of are ‘ Exp’ and ‘Model’
respectively.
Table 3.11 presents the devi ations of validation results of physical properties. Both
MAPE and maximum absolute percentage error (APE ) are given. For liquid density
(Figure 3.10–Figure 3.13), model predictions are in good agreement with the
experimental data in full range of system conditions. For the heat capacity ( see Figure
3.14), the deviation s gradually increases when CO 2 loading rises up. For surface
tension, the experimental data themselves have large deviations ( see in Figure 3.16 ).
55
Table 3.11 MAPE of v alidation results of physical properties in liquid phase
Property Temperature
(oC) MEA
concentration
(wt%) CO 2 loading
(mol/mol) MAPE
(%) Max. APE
(%)
Density 25–140 30, 40, 50, 60 0.1–0.6 0.348 1.48
Heat capacity 25 20, 30, 40 0–0.5 3.74 10.74
Viscosity 25 20, 30, 40 0–0.5 5.46 9.7
Surface tension 25 20, 30, 40 0–0.5 8.58 18.29
Figure 3.10 Liquid density of MEA -H2O-CO 2 at 30 wt% MEA
Figure 3.11 Liquid density of MEA -H2O-CO 2 at 40 wt% MEA
90095010001050110011501200
298.15 313.15 323.15 333.15 343.15 353.15 363.15Liquid Density, kg/m3
Temperature, KExp_0.10_CO2Load
Exp_0.21_CO2Load
Exp_0.32_CO2Load
Exp_0.44_CO2Load
Exp_0.56_CO2Load
Model_0.10_CO2Load
Model_0.21_CO2Load
Model_0.32_CO2Load
Model_0.44_CO2Load
Model_0.56_CO2Load
90095010001050110011501200
298.15 313.15 323.15 333.15 343.15 353.15 363.15Liquid Density, kg/m3
Temperature, KExp_0.1_CO2Load
Exp_0.21_CO2Load
Exp_0.33_CO2Load
Exp_0.45_CO2Load
Model_0.1_CO2Load
Model_0.21_CO2Load
Model_0.33_CO2Load
Model_0.45_CO2Load
56
Figure 3.12 Liquid density of MEA -H2O-CO 2 at 50 wt% MEA
Figure 3.13 Liquid density of MEA -H2O-CO 2 at 60wt% MEA
Figure 3.14 Liquid heat capacity of MEA -H2O-CO 2 at 298.15K
90095010001050110011501200Liquid Density, kg/m3
Temperature, KExp_0.1_CO2Load
Exp_0.22_CO2Load
Exp_0.34_CO2Load
Exp_0.47_CO2Load
Model_0.1_CO2Load
Model_0.22_CO2Load
Model_0.34_CO2Load
Model_0.47_CO2Load
900950100010501100115012001250Liquid Density, kg/m3
Temperature, KExp_0.1_CO2Load
Exp_0.22_CO2Load
Exp_0.34_CO2Load
Exp_0.48_CO2Load
Model_0.1_CO2Load
Model_0.22_CO2Load
Model_0.34_CO2Load
Model_0.48_CO2Load
1000150020002500300035004000
0 0.1 0.2 0.3 0.4 0.5Hear capacity, J/kg/K
CO 2loading, mol CO 2/mol MEAExp_20wt%_MEA
Exp_30wt%_MEA
Exp_40wt%_MEA
Model_20wt%_MEA
Model_30wt%_MEA
Model_40wt%_MEA
57
Figure 3.15 Liquid viscosity of MEA -H2O-CO 2 at 298.15K
Figure 3.16 Surface tension of MEA -H2O-CO 2 at 30 wt% MEA at 298.15K
0.1110100
0 0.1 0.2 0.3 0.4 0.5Liquid viscosity, mPa.S
CO2 loading, mol CO 2/mol MEAExp_20wt%_MEA Model_20wt%_MEA
Exp_30wt%_MEA Model_30wt%_MEA
Exp_40wt%_MEA Model_40wt%_MEA
0.020.030.040.050.060.070.080.09
0 0.1 0.2 0.3 0.4 0.5Surface Tension, N/m
CO2 loading, mol CO 2/mol MEAExp_20wt%_MEA
Exp_30wt%_MEA
Exp_40wt%_MEA
Model_20wt%_MEA
Model_30wt%_MEA
Model_40wt%_MEA
58
3.4 Process model development and validation at the pilot
scale
3.4.1 Introduction of the pilot plant
In this thesis, the pilot plant located at the University of Kaiserslautern (Mangalapally
and Hasse, 2011) was chosen for validating the process model . The reasons include (1)
both the absorber and the stripper use Mellapak 250Y packing, which is regarded as
appro priate structured packing type for industrial deployment (IEAGHG, 2012) ; (2) the
experimental data are c omprehensive and well presented in their publications (Notz et
al., 2012) , which helps the validation more comprehensive and could be compared with
other studies (Agbonghae et al., 2014) . The equipment features and the ranges of key
operation variables are summarized in Table 3.12. More details about this pilot plant
refer to the publication by Notz et al. (2012) .
Table 3.12 Main speci fications of the pilot plant
Flue gas source Natural gas burner
Flue gas flow rate (kg/h) 30–100
CO 2 concentration in the flue gas (mol %) 3–14
Solvent flow rate (kg/h) 50–350
CO 2 loading in lean solvent (mol CO 2/mol MEA) 0.1–0.32
Temperature of cooling water (°C) 5–10
Absorber Diameter (m) 0.125
Height of packing (m) 4.2
Packing type Structured packing Mellapak 250 Y
Operating pressure (bar) Atmospheric pressure
Operating temperature (°C) 40–70
Stripper Diameter (m) 0.125
Height of packing (m) 2.52
Packing type Structured packing Mellapak 250 Y
Operating pressure (bar) 1–2.5
Operating temperature (°C) 100–130
59
3.4.2 Process model development
3.4.2.1 Model flowsheet and process description
Figure 3.17 shows a closed -loop process flowsheet of the model developed in Aspen
Plus®. The flue gas leaving the power plant goes to a gas blower to increase its pressure
slightly above atmospheric pressure, to balance the pressure losses in the downstream
processes. Before entering the absorber, the flue gas has to be cooled down to between
40–50 °C in order to improve the absorption efficiency (Kvamsdal et al., 2011b) . The
cooling system consists of direct contact cooler (DCC) wit h a spray of water at 25 °C
and with a packing bed also with Mel lapak 250Y. The flue gas then enters the absorber,
in which MEA aqueous solvent reacts with CO 2. The scrubbed flue gas is emitted to the
atmosphere and the CO 2-rich solvent is discharged from the bottom of the absorber and
enters the stripper. The CO 2-rich solvent is regenerated inside the stripper with heat
input to the reboiler. The regenerated solvent is cooled and recirculated to the absorber
for reuse.
Figure 3.17 Process flowsheet in Aspen Plus®
3.4.2.2 Kinetics -controlled reactions
In Section 3.3, the equilibrium reactions of MEA -H2O-CO 2 mixture were described
during the thermodynamic modelling . In the rate -based model, the reaction of
dissociation of CO 2 and reaction of carbonate formation should be considered to be
kinetics -controlled reactions (Zhang et al., 2009) , presented as below:
BLOWER
H2OSPLITH2OCOOL
H2OPUMPREBOLIERCOOLER
MUMIXERPUMPLEAN
PUMPRICHINTHEX
WAT
CONDSPLFLUE1
FLUE2
WATER1
WATER2H2OOUT1
WATER3WATER4CO2OUT
LEANINPUREGAS
RICHOUT1
STEAMIN DESUPSTMREBDUTY
Q
CONDENS1 LEANOUT3
RICHOUT5RICHOUT6
LEANOUT1
LEANOUT2
H2OMU MEAMU RICHOUT2FLUE3
S1CONDENS2
S8 ABSORBERSTRIPPERDCC
60
R2*: Dissociation of CO 2
(3.11)
(3.12)
R5*: Carbonate formation
(3.13)
(3.14)
Power law expressions were used for the kinetic -controlled reactions. The kinetic s in
Equations (3.15) (Zhang et al., 2009) for Reactions R2* and R5* are in Table 3.13.
(3.15)
where is the reaction rate for reaction , is the pre -exponential factor, is the
system temperature in K, is the temperature factor, is the activation energy, is the
gas constant, is the concentration of species , and is the reaction order of
component in reaction . and for the reactions were calculated using
experimental data shown in Table 3.13.
Table 3.13 Parameters k and E in Equation (3.15) (Zhang and Chen, 2013)
Related Species Reaction direction (kmol/m3.s) (kJ/mol)
MEACOO- Forward 3.02E+10 41.2
Reverse (absorber) 5.52E+23 69.05
Reverse (stripper) 6.56E+27 95.24
HCO3- Forward 1.33E+17 55.38
Reverse 6.63E+16 107.24
61
3.4.2.3 Rate -based mass transfer
The absorber and the stripper were modelled based on two -film theory (Whitman, 1962) ,
which is used to describe the mass transfer of components between the gas phase and
the liquid phase. According to two-file theory, vapour film and liquid film with a phase
equilibrium interface are assumed between the bulk gas and bulk liquid phase. Chemical
reactions are assumed to occur in the liquid film only.
For the Ra teSep model in Aspen Plus®, Zhang et al (2009) had very detailed discussions
about correlations and settings. In this study, the flow model “VPlug” was chosen to
model the bulk properties with reasonable accuracy whilst “Countercurrent” model
sometimes causes o scillations in the temperature profile although it is most closely
approximates for the real situation (Razi et al., 2013b) . It was also pointed out that the
discretization points of the liquid film need to be over 10 to achieve accuracy , otherwise
it results in an over -prediction of the rate of mass transfer.
For the correlations related with mass transfer, Razi et al. (2013b) validated 12
correlat ion combinations with the experimental data from CESAR Pilot Data and the
results show that Billet and Schultes (1993) is one of accurate correlations provided by
Aspen Plus®. The parameters and correlations related with mass transfer used in this
study can be seen in Table 3.14. Here Fortran subroutine was used to implement
correlation of Tsa i et al. (2011) for liquid holdup calculation.
Table 3.14 Parameters and c orrelations selection for mass transfer in RateSep model
Parameters Correlations
Flow model VPlug (Razi et al., 2013b)
Film discretization points 20 (Razi et al., 2013b)
Mass transfer coefficients Billet and Schultes (1993)
Interfacial area Tsai e t al. (2011)
Liquid holdup Billet and Schultes (1993)
Heat transfer coefficient Chilton and Colburn (1934)
Pressure drop Sulzer correlation
62
3.4.3 Model v alidation
For the PCC process, the key operational parameters affecting the performance are CO 2
concentration in the flue gas, MEA concentration in solvents , lean loading and L/G ratio.
Thus, four sets of experiments from Notz et al. (2012) were chosen for the model
validation purpose. These include (1) experiment 1 –6 with different CO 2 concentration s
in the flue gas es; (2) experiment 24 –27 with different MEA concentration s at two
different CO 2 concentration s in the flue gas es; (3) experiment 28 –33 with different
solvent flow rat es at the high CO 2 concentration s in flue gas es; (4) experiment 34 –29
with different solvent flow rate s at the low CO 2 concentration s in flue gas es. Model
validations were carried out based on the same feed conditions and targeted the CO 2
loading in lean solvent (lean loading) by varying the reboiler duty of the stripper. Then
CO 2 loading in rich solvent (rich loading), CO 2 capture level and the stripper reboiler
duty could be com pared between the experimenta l data and model predictions .
Table 3.15 shows the overall validation results. MAPEs of the model predictions for the
CO 2 capture level, the stripper reboiler duty, and the rich CO 2 loading, when compared
with the experimental data from Notz et al. (2012) , are 1.78, 1.54 and 7.49%,
respectively. The MAPE s of the rich loading and the CO 2 capture level could be
acceptable. The specific duty was calculated from the reboiler duty (see Equation (2.5)).
However, in the experiment s, reboiler duty was affected by the heat loss from the
equipment and pipelines , which could not be measured directly. Although the values of
specific duty in the publication (Notz et al. , 2012) were corrected , the deviations of
themselves could not be evaluated , which may be the reason for high APE s for the
validation results of the specific duty.
63
Table 3.15 Validation results of model predictions against experimental data
Case Flue gas
flow rate
(kg/hr) CO 2
content
(mol/mol) L/G
(kg/kg) MEA
content
(kg/kg) Lean
Loading
(mol/mol) Rich loading(mol
CO 2/mol MEA) CO 2 capture level
(%) Specific duty
(GJ/ton CO 2)
Exp. Exp. Exp. Exp. Exp. Exp. Mod el APE (%) Exp. Mod el APE (%) Exp. Mod el APE (%)
1 72.0 0.085 2.8 0.275 0.265 0.386 0.379 1.81 75.91 74.02 2.49 5.01 5.24 4.59
2 72.4 0.165 2.8 0.284 0.308 0.464 0.458 1.29 51.32 51.23 0.18 3.98 4.25 6.78
3 72.1 0.055 2.8 0.287 0.230 0.308 0.313 1.62 84.93 86.89 2.31 7.18 8.25 14.90
4 71.8 0.088 2.8 0.278 0.268 0.397 0.392 1.26 76.45 78.41 2.56 5.05 5.45 7.92
5 71.8 0.130 2.8 0.284 0.306 0.446 0.446 0.00 60.67 61.48 1.34 4.19 4.43 5.73
6 72.1 0.198 2.8 0.286 0.317 0.464 0.471 1.51 43.67 44.43 1.74 3.85 4.01 4.16
24 71.8 0.085 2.8 0.221 0.251 0.392 0.399 1.79 74.63 75.24 0.82 5.11 5.36 4.89
25 71.8 0.085 2.8 0.104 0.166 0.435 0.440 1.15 68.61 68.92 0.45 5.46 5.72 4.76
26 72.8 0.164 2.7 0.217 0.288 0.474 0.475 0.21 49.29 50.32 2.09 4.13 4.51 9.20
27 72.4 0.165 2.8 0.104 0.169 0.501 0.500 0.20 42.13 44.01 4.46 4.77 5.11 7.13
28 75.6 0.164 2.0 0.298 0.266 0.470 0.477 1.49 53.42 52.30 2.10 3.68 4.05 10.05
29 76.0 0.163 2.6 0.297 0.306 0.465 0.472 1.51 53.65 53.52 0.24 3.92 4.21 7.40
30 75.1 0.159 3.3 0.264 0.316 0.459 0.463 0.87 55.91 56.50 1.06 4.38 4.62 5.48
31 75.7 0.159 3.6 0.267 0.338 0.454 0.462 1.76 55.57 56.41 1.51 4.30 4.29 0.23
32 76.6 0.156 3.9 0.259 0.335 0.449 0.460 2.45 55.39 56.12 1.32 4.57 4.56 0.22
33 77.1 0.157 4.5 0.256 0.360 0.441 0.468 6.12 54.59 55.71 2.05 4.35 4.2 3.45
34 70.3 0.083 1.1 0.300 0.146 0.417 0.425 1.92 75.87 77.42 2.04 4.85 5.58 15.05
35 70.1 0.085 1.4 0.291 0.208 0.411 0.421 2.43 76.57 76.98 0.54 4.27 4.30 0.70
36 71.1 0.083 2.1 0.274 0.252 0.393 0.401 2.04 75.98 74.57 1.86 4.68 4.99 6.62
37 71.3 0.083 2.8 0.273 0.298 0.398 0.409 2.76 74.51 75.22 0.95 5.11 4.49 12.13
38 71.3 0.085 3.5 0.276 0.308 0.385 0.401 4.16 74.69 76.01 1.77 5.40 4.53 16.11
39 71.5 0.084 3.8 0.271 0.319 0.400 0.403 0.75 74.78 74.70 0.11 5.23 4.33 17.21
MAPE (%) 1.78 1.54 7.49
64
The validations were also conducted to compare the temperature profiles and the CO 2
composition profiles in side the absorber and the stripper based on experiment A1, A2 and
A3 (Notz et al., 2012) . Figure 3.18 shows that the model predictions are in very good
agreement with the experimental data . One statement is that the total packing height is
2.25m inside stripper, the 3m position of temperature profile and liquid phase CO 2
concentration profile is Figure 3.18 (b) and Figure 3.18 (d) is the reboiler. The
comparison results show model predictions are in very good agreement s with the
experimental data.
Note: Exp represents experimental data; Model represents model prediction; CO 2
concentrations in flue gases are 8.5mol% for A1, 16.5mol% for A2 and 5.5mol% for A3
respectively.
Figure 3.18 Validation results between model predictions and experimental data, (a)
temperatur e profile of the absorber, (b) temperature profile of the stripper, (c) CO 2
composition profile inside the absorber, (d) CO 2 composition profile inside the absorber
40455055606570
0 1 2 3 4 5Absorber temperature profile (0C)
Height of the packing (m)(a)
Exp_A1
Exp_A2
Exp_A3
Model_A1
Model_A2
Model_A3
100105110115120125
0 1 2 3 4Stripper temperature profile (0C)
Height of the packing (m)(b)
Exp_A1
Exp_A2
Exp_A3
Model_A1
Model_A2
Model_A3
0.040.050.060.070.080.090.1
0 1 2 3 4 5Absorber CO 2mass fraction in liquid
(g/g)
Height of the packing (m)(c)
Exp_A1
Exp_A2
Exp_A3
Model_A1
Model_A2
Model_A3
0.040.050.060.070.080.090.1
0 1 2 3 4Stripper CO 2mass fraction in liquid (g/g)
Height of the packing (m)(d)
Exp_A1
Exp_A2
Exp_A3
Model_A1
Model_A2
Model_A3
65
3.5 Model s cale-up
To match the capacity requirement of handling the flue gas from a 453 MW e NGCC power
plant, the model of CO 2 capture process at pilot scale has been scaled up based on chemical
engineering principles about estimating of column diameter and pressure drop (Towler and
Sinnott , 2012) .
The process conditions of the flue gas from the NGCC power plant and other requirements
can be found in Table 3.16.
Table 3.16 Boundary conditions of solvent -based PCC process
Description Value
Flue gas flow rate (kg/s) 660.5 4
Flue gas CO 2 content ( mol %) 4.50
Flue gas temperature (oC) 40
Solvent MEA content (wt%) 35
Capture level (%) 90
Columns flooding (%) 65
As initial input s to the process model at the industrial scale in Aspen Plus®, first -guess
diameter s are required for both the absorber and the stripper. The column diameter can be
calculated from the maximum flooding vapour. In this study, a generalised pressure drop
correlation (GPDC) figure (see Figure 3.19) is used to estimate the maxi mum flooding
vapour . The abscissa and ordinate are presented in Equation (3.16) and Equation (3.17)
(Towler and Sinnott, 2012) respectively.
(3.16)
(3.17)
In Equation (3.16), is a flow parameter. For the absorber, the liquid feed is the lean
solvent. Its flow rate can be estimated by Equation (3.18) (Agbonghae et al., 2014) .
(3.18)
66
where is the mass flow rate of the lean solution, is the mass flow rate of the flue
gas, is the mass fraction of CO 2 in the flue gas, is required CO 2 capture
level, is the molar weight of MEA, and are the CO 2 loading in rich solvent
and lean solvent respectively , is the MEA concentration in solvent.
From Equation (3.18), (vapour mass flow rate per unit cross -sectional area) is calculated,
and then the total cross -sectional area can be obtained given the flue gas flow rate. In this
equation is a load para meter looked up from Figure 3.19, according to the value of
and specified pressure drop. is a packing factor.
Figure 3.19 Generalized pressure drop correlation (Stichlmair and Fair, 1998)
In order to achieve good liquid and gas distribution and to avoid flooding inside packing beds ,
a pressure drop of 15–50 mmH 2O per meter packing for absorber and stripper was
recommended (Towler and Sinnott, 2012) . In this study , a maximum pressure drop per unit
height of 20.83 mmH 2O (R.F., 1987) was used considering the forming of MEA solvent
(Agbonghae et al., 2014) . It should be noticed that the design of the column internals such as
gas\liquid distributors and re -distributors is crucial to ensure good gas and liquid distribution
inside the absorber and regenerator in such large diameters.
The first-guess diameter s of the absorber and the stripper can be calculated using the above
method. Starting from this, these parameters will be improved in the development of the
67
closed -loop CO 2 absorption model in Aspen Plus®. In order to directly use the detailed
equipment costs in benchmark report of IEAGHG (2012) in Chapter s 4, 5 and 7 for cost
evaluation and optimisation , the design features of the e quipment of the PCC process in this
study were set to be consistent with Scenario 3 (NGCC integrated with PCC without EGR) in
IEAGHG report . In this study, a s one key operational variable, lean loading is set at 0. 280
mol CO 2/mol MEA as an initial input for th e base case by examining the experimental data
(Notz et al., 2012) . The overall process parameters of the capture plant is shown in Table 3.17
and Table 3.18.
Table 3.17 Design parameters of the a bsorber and the stripper at the base case
Description Absorber Stripper
Cross sectional area (m2) 387.5 0 50.27
Equivalent Column diameter (m) 22.20 8.00
Packing Type Mellapa k 250Y Mellapa k 250Y
Total Packing height (m) 20.00 20.00
Column pressure (bar) 1.00 2.00
Column pressure drop (bar) 0.069 0.014
Table 3.18 Overall performance of PCC process at the base case
Description Value
CO 2 captured (kg/s) 41.04
L/G ratio (kg/kg) 1.22
Lean solvent temperature ( ℃) 40.00
Lean solvent flow rate (kg/s) 807.84
Lean loading (mol CO 2/ mol MEA) 0.280
Rich loading (mol CO 2/ mol MEA) 0.461
Lean Solvent MEA content (wt%) 35
Reboiler duty (MW th) 195.37
Specific duty (GJ/ton CO 2) 4.76
Reboiler temperature ( ℃) 120.16
68
3.6 Concluding remarks
This chapter presented the preparation of the process model. As the base of process model
development, d ifferent correlation s of the thermodynamic model were examined and
validated with the experimental data of CO 2 solubility in aqueous MEA solutions in wi de
ranges of pressure, temperature and composition conditions. At the same time, the
correlations combination in this study was compared with other three published studies
(Austgen et al., 1989; Liu et al., 1999; Zhang et al., 2011) . The results show the better
prediction performance of this study. To improve prediction accuracy of liquid mixture
density, Han and Eime r (2012) model was used by coding Fortran subroutine. Then several
key physical properties, such as liquid density, liquid heat capacity, liquid viscosity and
liquid surface tension of MEA -H2O-CO 2 system were validated with the experimental data.
A steady state rate-based process model was developed in Aspen Plus® at the pilot plant scale
referring to the pilot plant in the University of Kaiserslautern (Mangalapally and Hasse,
2011) . For kinetics -controlled rea ctions, different values were set for kinetics of the reverse
carbonate formation reactions happening in the absorber and the stripper respectively, which
improves the accuracy of the process model. Another improvement work is that the
correlation of effec tive gas liquid interfacial area was updated to Tsai et al. (2011) by coding
Fortran subroutine. The process model was then validated with series of comprehensive pilot
plant data, in terms of its absorption efficiency and thermal performance of the integrated
system. The comparison results show that model predictions are in very good agreement with
the experimental data from pilot plant , which ensure that the process model has good
accuracy for the optimisation studies in next chapters.
69
Chapter 4: Optimal Design of Solvent -based PCC
Process
In this Chapter, the cost model of PCC process is developed first. The cost breakdowns
including CAPEX, fixed OPEX and variable OPEX were analyse d. The cost model was
developed in Fortran subroutine and was dynamically linked with Aspen Plus®. Therefore it
is equivalent as a new model. Optimisation method is then explained. Cost of CO 2 avoided
(CCA) is formulated as the objective function. The key design parameters and operational
variables have been analyse d to get their reasonable variation range. The optimisation is
conducted and the performance of optimal case was compared with the base case. In order to
get comprehensive understanding, case studies we re carried out about the impact of
variation s of the key variables.
4.1 Development of cost model
4.1.1 Cost brea kdown
For operating an industrial process plant, the total cost includes capital expenditure (CAPEX)
and operational expenditure (OPEX). OPEX can be split into fixed OPEX (FOPEX) and
variable OPEX (VOPEX) (IEAGHG, 2012) . For a carbon capture plant, the costs could be
detailed as follows : (1) CAPEX includes equipment material and installation, labour cost,
engineering and management cost and other costs happened during the project contracture
and commissioning, (2) FOPEX includes overhead cost, operating and maintenance cost
(O&M) and oth er costs fixed for the plant no matter running at partial or full load or
shutdown, and (3) VOPEX mainly includes energy and utilities costs and solvent make -up
cost. It is noticed that in this chapter VOPEX does not include the emission cost of CO 2
discha rged into atmosphere. In Chapter 8, it will be involved when the optimal operation was
analyse d for the power plants integrated with whole CCS chain.
To harmonize the results for comparison with future new studies, the follow ing assumptions
were made: (1) all costs are corrected to €2015 using the harmonised consumer price index
(HICP) in Europe zone (Inflation, 2015) , (2) the captured CO 2 mixture has no economic
value, and ( 3) cooling water is sourced from a nearby body of water at the cost of pumping
and operation of a cooling tower.
70
4.1.2 CAPEX
4.1.2.1 Equipment type and material
Because of the corrosivity of the solvent at operating conditions of carbon capture plant , the
material sele ction is important to ensure the integrity of the plant design. Table 4.1 lists the
type and material selection of main equipment based on IEAGHG reports (IEAGHG, 2010;
IEAGHG, 2012) , which is the base for the cost estimation of equipment.
Table 4.1 Equipment type and material selection of PCC process
Category Name of equipment Type Material
Separation
equipment DCC Rectangular tank Concrete with epoxy
lining
DCC packing Mellapa k 250Y SS316
Absorber Rectangular column Concrete with
polyproylene lining
Absorber packing Mellapak 250Y SS316L
Stripper Vertical cylinder SS316L
Stripper packing Mellapak 250Y SS316L
Heat
exchanger Stripper reboiler Vertical shell & tube
thermosyphon SS316L
Stripper condenser Shell and tubes SS 304
Cross h eat exchanger Plate and frame SS316L
Lean cooler Plate and frame SS304
DCC water cooler Plate and frame CS
Pressure
Change Flue gas blower Axial CS
Rich solvent pump Centrifuge SS316L
Lean solvent pump Centrifuge SS304
DCC water pump Centrifuge CS
Compression
train Compressor Multi -stage Integrally
geared Type Cr Ni alloy
casing/impeller
Knock out drum Vertical tank SS304
Inter -stage cooler Shell and tubes SS304
71
4.1.2.2 Direct cost of equipment
The accuracy of the equipment cost estimation depends on the available design details at
different project phase. The major equipment costs of PCC process in IEAGHG report were
estimated by contacting the vendors after detailed engineering design (IEAGHG, 2012) . The
method could be regarded as Class 2 detailed estimates whose accuracy could be in the range
from -15% to 20% (Feng and Rangaiah, 2011) . The direct material costs and other fixed costs
in the report could be trusted although the process simulation may not be accurate enough. In
this study, the base case was set up with same equipment design features and process
boundary conditions of Scenario 3 (NGCC integrated with PCC without EGR) in I EAGHG
report. Thus, the direct material costs can be derived from IEAGHG report. For the absorber
and the stripper, the cost s of the packing with internals can be directly calculated by the
volume s of the packing beds. For the other cases, the direct mater ial cost could be calculated
based on their reference value in the base case and the specific scaling factor for different
types of equipment, by Equation (4.1) (Mores et al., 2012) .
(4.1)
where is the value of selected scaling factor related to equipment capacity, is cost index
for different year and area, is the specific factor and the value is 1.0 for structured packing
inside the columns and 0.6 for other equipment according to six -tenths rule (Sweeting, 1997) .
is the direct material cost of the base case. Table 4.2 lists the main investment items
considered including the construction material of each one of the m. The flue gas cooling
system includes flue gas blower, DCC, DCC pump and DCC cooler.
4.1.2.3 Annualized CAPEX
For a PCC plant, the major equipment costs include the costs of the absorbers, strippers,
pumps and all other plant items which are listed in Table 4.2. The other direct and indirect
costs were estimated using factors of the overall major equipment cost. Those factors are
listed in Table 4.3.
72
Table 4.2 Direct material costs and the scaling factor of equipment
Name of equipment X, Selected scaling factor , Base cost* (€@2011 )
Flue gas cooling system flue gas flow rate 8,768,110
Absorber column shell shell surface area 995,908
Stripper column shell shell surface area 18,095,040
Unit price of packing volume of packing 4,565 (€/m3) *
Stripper reboiler heat duty 25,539,000
Stripper condenser heat duty 9,287,000
Inter Heat Exchanger heat duty 1,963,000
Lean cooler heat duty 557,000
Rich solvent pump electricity consumption 51,000
Lean solvent pump electricity consumption 51,000
Compression train electricity consumption 8,256,245
*The values were derived from IEAGHG (2012)
Table 4.3 Factors for total project cost calculation
Description Percentage of major
equipment cost * Direct material Major equipment 1
Piping 0.1500
Control and instrumentation 0.0200
Electrical 0.0400
Catalysts and other chemicals 0.0085
Civil/steelwork/buildings 0.280 3 Labour
only
contracts Mechanical 0.109 7
Electrical/instrumentation 0.036 6
Scaffolding/lagging/rigging 0.030 5 Other cost Engineering service/construction management 0.045 7
Commissioning 0.010 1
Soft costs contractor (inc contingency & profit) 0.5026
Soft costs owner 0.201 1
CAPEX 2.434 9
73
*The values were derived from IEAGHG (2012)
The annualized CAPEX is the total CAPEX multiplying by capital recovery factor (CRF)
(McCollum and Ogden, 2006) , which is calculated by Equation (4.2) (Mores et al., 2012) .
(4.2)
where is the economic life of plant and is the interest rate. It is assumed a project life of
25 years and 12% of interest rate (McCoy and Rubin, 2008) .
4.1.3 Fixed OPEX
Fixed OPEX (FOPEX) includes long term service agreement costs, overhead cost, operating
and maintenance cost (O&M) and other costs fixed for the plant no matter if it is running at
partial or full load or shutdown. FOPEX can be simply calculated by Equation (4.3)
(4.3)
4.1.4 Variable OPEX
For operating a carbon capture process integrated with a power plant, the power plant could
supply electricity and lower pressure steam to the capture plant . Other utilities could also be
provided from the power plant accessory facilities. However, in this chapter, the study scope
only includes the PCC process with CO 2 compression. Each utility cost will be calculated by
multiplying the market unit price with it s amou nt obtained from the simulation results.
Furthermore, for heat input required for solvent regeneration, the low pressure steam
consumption is converted into equivalent power electricity consumption. The utility unit
prices can be seen in Table 4.4 with the costs given in Euro. VOPEX includes the cost of
power electricity consumption for pumps/blower/compressor, the cost of power electrici ty for
solvent regeneration, the cost of cooling utilities and the cost of MEA solvent make -up. The
water make -up is neglected because Kvamsdal et al. (2010) proven the water in a solvent –
based PCC process coul d be in a neutral balance without make -up.
74
Table 4.4 Key economic evaluation cost inputs
Description Unit Value Source
Electricity price €/kW 0.0775 1st quarter of 2012 of APEA
Cooling water price €/m3 0.0317 1st quarter of 2012 of APEA
MEA price €/t 1,452 Alibaba (2016)
Operating hours hr/year 8,000
Project economic life year 25
Interest rate /year 0.12
4.1.5 The costs of the base case
The case of PCC process after scale -up in Section 3.5 (process parameters can be seen in
Tables 3.16 – 3.18) was defined as t he base case in this c hapter. With all the basic cost s and
relevant correlations in above sections, the costs of the base case (process parameters can be
seen in Table s 3.16 – 3.18) were calculated. Table 4.5 shows the costs of the base case of
PCC standalone. In the base case, the annualized CAPEX, FOPEX and VOPEX account for
38.46%, 8.87% and 52.67% of the total annual cost respectively. For the variable OPEX,
power electricity cost is the biggest part and solvent make -up cost is the second largest part.
The CCA is 86.85 €/ton CO 2.
Table 4.5 Costs of the base case
Description Base case
CO 2 captured rate (ton/year) 1,179,064
CAPEX (M€) 302.85
Annualized CAPEX (M€/year) 39.37
Fixed OPEX (M€/year) 9.09
Variable
OPEX Power electricity (M€/year) 49.72
Cooling water (M€/year) 1.46
Solvent make -up cost (M€/year) 2.75
Total annual cost (M€/year) 102.3 8
CCA (€/ton CO 2) 86.85
75
4.2 Optimisation methodology
4.2.1 Sequential quadratic programming (SQP)
The SQP method has been one of the most successful general methods for solving large -scale
nonlinear constrained optimization problems (Boggs and Tolle, 2000) . A typical optimisation
model consists of an objective function supplemented with equality and inequality constraints.
This optimisation problem can be formulated as follow s:
Minimize (4.4)
Subject to the process constrains and operation constrains:
(4.5)
(4.6)
where is the objective function ; is the vector of the coefficients in the objective function
and constrains; is the vector of the design variables ( e.g. diameters and packing heights of
the absorber and stripper, the operating pressure and operatin g temperature of the towers).
And is the vector of operational variables (i.e. , capture level, , lean loading,
, solvent and flue gas ratio and , reboiler duty).
The Lagrangian for this problem is:
(4.7)
where λ and σ are Lagrange multipliers; T denotes the vector transpose .
The SQP method converges fast with a few iterations but it needs numerical derivatives for
all decision and tear variables at each iteration. At an iterate , a basic SQP algorithm defines
an appropriate search direction as a solution to the quadratic programming (QP)
subproblem, in which a quadratic objective function is minimize d subject to inequality or
equality constraints.
76
Minimize:
(4.8)
Subject to:
(4.9)
(4.10)
where , are the gradients.
The SQP method used in Aspen Plus® has a novel feature which is that tear streams can be
partially converged using Wegstein for each optimization iteration (AspenTech, 2008a) . Then
the solving can start with only a single po int and does not need to calculat e the derivative s
(Wegstein, 19 58), which normally stabilizes convergence s and reduce s the total number of
iterations.
4.2.2 Objective function
For techno -economic evaluation or cost optimisation of a power plant integrated with carbon
capture process, different economic indexes have been used in different studies , including (a)
total annual operating profits; (b) total annualized cost; (c) levelised cost of electricity
(LCOE); (d) cost of CO 2 avoided. In this Chapter, the study scope only includes PCC process
and compression train so that the cost of CO 2 avoided (CCA) was formulated to be the
objective function of the optimisation .
CCA was calculated through dividing total annual cost by annual numbers of CO 2 captured as
in Equation (4. 11). The total annual cost is a sum of annualized CAPAX, FOPEX and
VOPEX as in Equation (4. 12).
(4.11)
(4.12)
(4.13)
77
4.2.3 Optimisation constraints
4.2.3.1 Equality constraints
Equality constraints relate d to the mass balances, reactions and phase balance were embedded
in the first principle process model built in Aspen Plus® described in Chapter 3.
4.2.3.2 Inequality constraints
The consideration of the first constrain t is that current perspective studies predict 90%
capture level from the fossil -power plants is required to reach the target of CO 2 emission
control (IPCC, 2005; DECC, 2013) . For operating the column, the constrain ts abo ut the
flooding and pressure drop are strict considering MEA solvent easily cause fo aming inside
packing beds (Agbonghae et al., 2014) at this context.
(4.14)
(4.15)
(4.16)
4.2.4 Optimisation variables
4.2.4.1 Key design variables
4.2.4.1.1 Diameter of the absorber and the stripper
Previously, a maximum column diameter of 12.6 m for carbon capture process was suggested
by Chapel et al. (1999) . In recent years, with different column internal technologies
developed by different equipment manufacturers (Carbon Capture Journal, 2013; Sulzer,
2014; Koch -Glitsch, 2014) , the upper limit of column diameter is increasing. For Fluor's CO 2
capture demonstration plant using Econamine FG PlusSM Technology, Reddy et al. (2013)
reported that a maximum diameter of 18.0 m was used as the criterion for deciding the
numbers of the column requi red. It is also noticed that concrete rectangular tower with
appropriate lining rather than cylindrical metal material tower could be used for the absorber
(IEAGHG, 2012; SASKPOWER, 2015) in a power plant to get a better economic profile
because the operating pressure of the absorber is near the atmosphere pressure. In line with
the IEAGHG report, the absorber is a rectangular column with a size of 15.5m x 25m
(equival ent to a cylindrical column with 22.2 m diameter) and the stripper is a cylind rical
column with 8.0 m diameter. However, for model input format in Aspen Plus®, the diameter
78
of the absorber was given by calculating same from the cross -section area of the rectangular
column. This could also give a generic sizing of the absorber for comparison with other
publications . Then the variation ranges of the diameters of the absorber and the stripper in the
optimisation are presented as:
(4.17)
(4.18)
4.2.4.1.2 Packing height of the absorber and the stripper
In the pilot plant at the University of Texas at Austin (Dugas, 2006) , the packing heights of
absorber and stripper are 6.1m and CO 2 capture level could be over 95% in some scenarios
(Zhang et al., 2009) . Notz et al (2012) report ed the capture level could reach 90% and lean
loading could reach 0.1 mol CO 2/mol MEA with the packing height of 4.2 m for the absorber
and of 2.25 m for the stripper in the pilot plant at the University of Kaiserslautern
(Mangalapally and Hasse, 2011) . In the studies (IEAGHG, 2012; Sipöcz and Tobiesen, 2012;
Kvamsdal et al., 2010; Mores et al., 2014; Biliyok and Yeung, 2013; Agbonghae et al., 2014)
on industrial scale PCC process , the packing height of the columns varies from 6 m to 30 m.
In this optimisation , the variation ranges of the packing height of the absorber and the stripper
are presented as:
(4.19)
(4.20)
4.2.4.2 Operating pressure and temperature
Operating pressure and temperature of the columns
Generally , low operating temperature and high operating pressure is benefi cial in chemical
absorption efficiency (Sinnott and Towler, 2009) . However, the operating pressure of the
absorber of PCC process is normally set at near atmosphere pressure because it is very costly
to compress the flue gas with a huge volume tric flow rate ( 599.195 m3/s in the base case in
this study).
79
For the stripper, Abu-Zahra et al. (2007b) investigated that increasing the operating pressure
from 90 kPa to 210 kPa le ads to an 8.5% r eduction in heat requirement in the reboiler and
lower electricity consumption of the CO 2 compression train. One limitation of the trade -off is
that MEA solvent degradation increases sharply if the temperature increases over 1 25 °C
(Davis and Roche lle, 2009; Rochelle, 2012) . Corres pondingly, the operating pressure for
stripper is then set up to 2.0 bar.
For both the absorber and the stripper, t he operating temperature is a dependant variable once
the operating pressure is specified. It is affected by the feeding conditions and the reaction
heat releas ed or heat input for solvent generation.
Temperature of flue gas
Kvamsdal et al . (2011b) found the specific duty decreases from 2.87 to 2.71 GJ/ton CO 2
when flue gas temperature changes from 50 °C to 30 °C. So the benefit of low flue gas
temperature is not significant , especially considering the extra cooling cost.
Temperature of lean solvent
In the study by Abu -Zahra (2007b) , when the lean solvent temperature decreased from 50 °C
to 25 °C, the specific duty decreased from 4.15 to 3.75 GJ/ton CO 2. However, more cooling
energy is required i n the lean solvent cooler.
Then the variation ranges of the operating pressure of the columns and the temperature of the
feeding streams in the optimisation are presented as below:
(4.21)
(4.22)
°C (4.23)
°C (4.24)
4.2.4.3 Key operational variables
MEA concentration
In the study of Abu -Zahra et al (2007b) , it was found that thermal energy requirement
decrease s by about 5–8% when MEA concentration in the lean solvent increase from 20 wt%
80
to 40 wt%. However, higher MEA concentration leads to pronounced corrosive effects to the
equipment and cause higher solvent degradation loss (Davis and Rochelle, 2009; Rochelle,
2012) .
Lean Loading
Lean loading has lar ge impact for operating the absorber in terms of both the absorption
efficiency and hydraulic performance. Abu -Zahra et al . (2007b) examined the lean loading
with a range of 0.18 –0.38 mol CO 2/mol MEA while Agbonghae et al. (2014) investigated the
range of 0.1 0–0.30 mol CO 2/mol MEA . Lean loading can be controlle d by adjusting the heat
input to the reboiler o f the stripper.
Liquid/Gas ratio
In the contribution of Agbonghae et al. (2014) , the range of L/G (kg/kg) ratio is from 0.7 0 to
2.75 for gas fired power plant and is from 2.0 0 to 5.5 0 for coal -fired power plant. It is notice d
that the interactions between these key operational variables are complex and nonlinear. For a
certain capture task (fixed flue gas flow rate and capture level requirement), amongst MEA
concentration, lean loading and L/G ratio, when two of them are specified, the other one is
then independent. The variation ranges of these three key operational variables in this
optimisation are presented as below:
(4.25)
(4.26)
(4.27)
4.3 Optimisation result s
Table 4.6 shows optimisation results compared with the base case. The CCA of the optimal
case decreased by 18.7% compared with the base case . The main contribution is the saving
from the CAPEX. Both the diameter and packing height of the columns in the optimal case
are less than the base case. VOPEX in the optimal case is also lower than the base case as the
81
reboiler duty is less with optimal lean loading although higher L/G ratio means high
operating cost for the solvent circulations.
Table 4.6 Comparison of the optimal case and the base case
Input conditions
Parameter Values
Flue gas flow rate (kg/s) 660.05
CO 2 concentration in flue gas (mol%) 4.50
CO 2 capture level (%) 90.00
MEA concentration in solvents (wt%) 35.00
Technical performance comparison
Parameter Base case Optimal case
Lean loading (mol CO 2/mol MEA) 0.280 0.294
L/G (kg/kg) 1.22 1.37
Reboiler duty (MW th) 195.37 166.2 1
Speci fic duty (GJ/ton CO 2) 4.76 4.05
Absorber column diameter (m) 22.20 18.93
Absorber column packing height (m) 20.00 10.43
Stripper column diameter (m) 8.00 7.72
Stripper column packing height (m) 20.00 7.67
Economic performance comparison
Parameter Base case Optimal case
ACAPEX (M€/year ) 39.37 29.79
FOPEX (M€/year ) 9.09 6.41
VOPEX (M€/year ) 53.93 47.51
TAC (M€/year ) 102.3 8 81.71
CCA (€/ton CO 2) 86.85 69.13
4.4 Optimisation s in respon se to variations of key variable s
As can be seen in Table 4.6, the optimisation result is a series of the value s of each parameter
in the optimal case and i t could not reflect the impact of the variables on the objective
function and the interactions be tween different variables. In order t o obtain more
comprehensive understanding of the influence of important process parameters, systematic
82
case studies about variation of MEA concentration in solvents, CO 2 concentration in flue gas
and flow rate of flue g as are carried out in this section
In the case studies, the selected variables were specified with discrete values to see that how
the objective function and related variables change, driven by the optimisation s. It should be
pointed out that , in all the figures in this section , each point on lines is concerned with a
steady state solution for one discrete optimisation running . The lines present the trends of the
changes instead of continual changes of the variables.
4.4.1 Variation of MEA concentration in solvent
The MEA con centration in solvent affects both the CO 2 physical solubility and chemical
reactions, thus it influence s on the energy requirement of solvent regeneration. In the study of
Abu-Zahra et al (2007b) , it was found that thermal energy requirement decrease about 5–8%
when MEA concentration in the lean solvent increase from 20 wt% to 40 wt%. However,
higher MEA concentration leads to pronounced corrosive effects to the equipment and cause
higher solvent degradation loss (Davis and Rochelle, 2009; Rochelle, 2 012). In this case
study, the MEA concentration in solvent was specified at 20, 25, 30, 35, 40 wt% respectively .
The results could be seen in Table 4.7 and Figure 4.1.
With increasing MEA concentration , the CO 2 solubility increases because there are more
MEA molecule s available to react with CO 2 molecules. The CO 2 loading in rich solvent is
also increasin g (see Figure 4.1(a)).The results show the required L/G ratio decreases because
of increasing of absorption capacity of the solvent (the difference of CO 2 loading between
rich and lean solvent ).
Lower L/G ratio (Figure 4.1(b)) results in column cross sectional area decreases by 17.74%
for the absorber a nd by 29.49 % for the stripper (Figure 4.1(d)). The reason is that the flow
rate of the flue gas entering the absorber remains same in the case study. With higher MEA
concentration in solvent, the optimal packing height of the absorbe r slightly decreas es (Figure
4.1(e)) because of lower solvent flow rate. T he optimal packing height of the stripper
increases significantly in Case M5 because the absorption capacity of solvent increases
largely which means the required solvent regeneration degree increase s in the stripper.
The e conomic results (see Figure 4.1 (f)) shows CCA decreases 14.54%, from 80.34 €/ton
CO 2 to 68.6 6 €/ton CO 2. The cost breakdown shows ACAPEX decreases by 15.88 % and
VOPEX decrease by 13.47 %. It is also noticed that the CCA in Case M5 is just slightly
83
lower than in Case M4 (see Table 4.7). However at 40 wt% concentration, MEA degradation
in the system will cause significant solvent los s.
Table 4.7 Optimisation results with variation of MEA concentration
Case tag M1 M2 M3 M4 M5
MEA concentration (wt%) 20 25 30 35 40
Flow rate of flue gas (kg/s) 660.05 660.05 660.05 660.05 660.05
CO 2 concentration in flue gas (mol%) 4.50 4.50 4.50 4.50 4.50
CO 2 capture level (%) 90 90 90 90 90
CO 2 captured (kg/s) 41.04 41.04 41.04 41.04 41.04
Lean Loading (mol CO 2/ mol MEA) 0.324 0.317 0.299 0.294 0.310
Rich Loading (mol CO 2/ mol MEA) 0.428 0.439 0.456 0.462 0.477
Flow rate of lean solvent (kg/s) 2856.68 1933.54 1274.52 905.60 894.17
L/G ratio (kg/kg) 4.32 2.93 1.93 1.37 1.35
Reboiler duty (MW th) 177.90 179.17 172.66 166.21 157.13
Specific duty (GJ/ton CO 2) 4.33 4.37 4.21 4.05 3.83
Diameter of absorber (m) 20.65 20.18 19.44 18.93 18.63
Packing height of absorber (m) 11.89 11.38 11.21 10.42 10.84
Diameter of stripper (m) 9.17 8.43 7.54 7.72 7.70
Packing height of stripper (m) 3.23 4.44 6.26 7.68 12.67
ACAPEX (M€/year) 34.26 32.29 29.72 27.79 28.82
FOPEX (M€/year) 7.91 7.45 6.86 6.41 6.65
VOPEX (M€/year) 52.79 51.52 49.27 47.51 45.68
TAC (M€/year) 94.96 91.26 85.84 81.71 81.15
CCA (€/ton CO 2) 80.34 77.21 72.62 69.13 68.66
84
Figure 4.1 Optimisation results with variation of MEA concentration in solvent
0102030405060
4045505560657075808590
20% 25% 30% 35% 40%
ACAPEX and VOPEX (M €/year)CCA (€/ton CO 2)
MEA mass concentration in solvent(f) Economic performance
CCA
ACAPEX
VOPEX0.250.30.350.40.450.5
20% 25% 30% 35% 40%CO 2 Loading (mol CO 2/mol MEA)
MEA mass concentration in solvent(a) CO 2loading in solvent
Lean Loading
Rich Loading
3.8003.9004.0004.1004.2004.3004.400
155160165170175180185
20% 25% 30% 35% 40%
Optimal Specific Duty (GJ/tonCO 2)Optimal reboiler duty (MW th)
MEA mass concentration in solvent(c) Thermal performance
Reboiler duty
Special duty
02468101214
20% 25% 30% 35% 40%Optimal packing height (m)
MEA mass concentration in solvent(e) Packing height
PH_ABS
PH_STR0.001.002.003.004.005.00
050010001500200025003000
20% 25% 30% 35% 40%
L/G ratio (kg/kg)Lean solvent flow rate (kg/s)
MEA mass concentration in solvent(b) Solvent
Solvent flow rate
L/G Ratio
0510152025
20% 25% 30% 35% 40%Optimal column diamter(m)
MEA mass concentration in solvent(d) Column diamter
DI_ABS
DI_STR
85
4.4.2 Variation of CO 2 concentration in flue gas
The CO 2 concentration differs in the flue gases produced by different type of emitters . For
example, CO 2 concentration in flue gas is around 4.4 mol % for a NGCC power plant without
EGR (Biliyok et al., 2013) while it is around 13.5 mol % for coal -fired power plant
(Agbonghae et al., 2014) . The CO 2 concentration in flue gases from refinery and cement are
around 20 –33 mol% (IPCC, 2005) . The changes of CO 2 concentration in flue gases do
significantly affect key equipment design features as well as the economic range of the key
operational variables because it not only changes the required capacity of the capture plant
but also impact s the absorption efficiency. In this case study, the CO 2 concentration was
specified at 4. 5, 7.5, 13.5, 20, 30 mol% respectively . The optimisation results could be seen
in Table 4.8 and Figure 4.2.
With the increase of the CO 2 concentration but fixing the capture lev el, the amount of CO 2
captured increase s from 41.04 kg/s to 243.85 kg/s (see Table 4.8). The optimal CO 2 rich
loadings ( Figure 4.2(a)) have been pushed high towards the saturated loading (could be
roughly estimated with the CO 2 solubility data in Figure 3.2–Figure 3.4) for CO 2
concentration s from 7.5 mol% to 30 mol%. At the same time, optimal CO 2 lean loading
gradually increases which means the solvent regeneration degree becomes low. This helps to
keep the increase of reboiler duty (Figure 4.2(c)) not such sharp.
The solvent flow rate increases sharply to meet the capture capacity , resulting in big
increasing of L/G ratio from 1.37 kg/kg to 9.60 kg/kg (Figure 4.2(b)). The optimal diameter s
of the columns also increase significantly (Figure 4.2( d)). The optimal packing height (Figure
4.2(e)) of the absorber does not show clear change trend. The optimal packing height of the
stripper decrease s slightly because lean loading increase s which means lower solvent
regeneration degree .
From Case C1 to Case C5, ACAPEX increases by 192.39 % and VOPEX increases by
358.8 3%, which indicate that the operating cost is more sensitive to the change of CO 2
concentration. However, CCA decreases by 34.5%, from 69.13 €/ton CO 2 to 45.28 €/ton CO 2,
because the capture d CO 2 increases 494.16% . It reflects that high CO 2 concentration in the
flue gas benefits low CCA, which is consistent with the result of EGR study in Chapter 5.
86
Table 4.8 Optimisation results with variation of CO 2 concentration in flue gas
Case tag C1 C2 C3 C4 C5
CO 2 concentration in flue gas ( mol% ) 4.40 7.50 13.50 20.00 30.00
MEA concentration (wt%) 35 35 35 35 35
Flow rate of flue gas (kg/s) 660.05 660.05 660.05 660.05 660.05
CO 2 capture level (%) 90.00 90.00 90.00 90.00 90.00
CO 2 captured (kg/s) 41.04 68.64 119.54 174.89 243.85
Lean Loading (mol CO 2/ mol MEA) 0.294 0.298 0.308 0.315 0.331
Rich Loading (mol CO 2/ mol MEA) 0.462 0.476 0.474 0.474 0.475
Flow rate of lean solvent (kg/s) 905.60 1542.67 2973.22 4565.87 6338.98
L/G ration (kg/kg) 1.37 2.34 4.50 6.91 9.60
Reboiler duty (MW th) 166.21 265.84 463.44 676.94 943.83
Specific duty (GJ/ton CO 2) 4.05 3.87 3.88 3.87 3.87
Diameter of absorber (m) 18.93 19.71 21.54 23.22 24.58
Packing height of absorber (m) 10.42 12.50 11.54 11.19 10.60
Diameter of stripper (m) 7.72 10.30 12.91 15.55 18.09
Packing height of stripper (m) 7.68 6.50 6.29 5.93 5.77
ACAPEX (M€/year) 27.79 36.63 50.30 64.89 81.26
FOPEX (M€/year) 6.41 8.45 11.61 14.97 18.75
VOPEX (M€/year) 47.51 69.87 113.56 160.54 217.98
TAC (M€/year) 81.71 114.95 175.46 240.40 317.99
CCA (€/ton CO 2) 69.13 58.15 50.97 47.73 45.28
87
Figure 4.2 Optimisation results with variation of CO 2 concentration in flue gas
050100150200250
4045505560657075
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%
ACAPEX and VOPEX (M €/year)CCA (€/ton CO 2)
CO 2mole concentration in flue gas(f) Economic performance
CCA
ACAPEX
VOPEX0.250.30.350.40.450.5
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%CO 2 Loading (mol CO 2/mol MEA)
CO 2mole concentrtion in flue gas(a) CO 2loading in solvent
Lean Loading
Rich Loading
3.83.944.14.24.34.4
01002003004005006007008009001000
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%
Optimal Specific Duty (GJ/tonCO 2)Optimal reboiler duty (MW th)
CO 2mole concentration in flue gas(c) Thermal performance
Reboiler duty
Special duty
051015202530
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%Optimal column diamter (m)
CO 2mole concentrtion in flue gas(d) Column diamter
DI_ABS
DI_STR024681012
01000200030004000500060007000
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%
L/G ratio (kg/kg)Lean solvent flow rate (kg/s)
CO 2mole concentrtion in flue gas(b) Solvent
Solvent flow rate
L/G Ratio
02468101214
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%Optimal packing height (m)
CO 2mole concentrtion in flue gas(e) Height of packing
PH_ABS
PH_STR
88
4.4.3 Variation of flue gas flow rate
Another major change is the flow rate of the flue gas, which reflects different sizes of NGCC
power plants, or if a part of flue gas is designed to bypass the capture process (Mores et al.,
2014) or if operating at part load condition . In this case study, the flow rate of flue gas was
specified at 50, 75, 100, 125, 150 % of its value in the base case in Table 4.9. The
optimisation results could be found in Table 4.9 and Figure 4.3.
The flow rate change of flu e gas ( with same CO 2 concentration in the flue gas and same
capture level) just means simple scale -up or scale -down. With only technical considerations,
it is exp ected that the equipment size should change in proportion with the flow rate change
of flue gas and the optimal values for key operational variables should be kept same.
However the optimisation results give different answers. The optimal CO 2 rich loading keeps
high towards its saturated loading. The optimal CO 2 lean loading (Figure 4.3 (a)) is stable at
around 0.294 in Case F2, F3 and F4 with flue g as flow rate change range of 75 –125%. But it
rises to 0.315 in Case F1 (50% of flow rate of flue gas) and down to 0.275 in Case F5 (150%
of flow rate of flue gas). The optimal reboiler duty is roughly in proportion with the flue gas
flow rate which results in a relatively stable specific duty at range of 3.99–4.04 GJ/ton CO 2
(Figure 4.3(c)).
It is easy to understand that the diameters of both the absorber and the stripper increase with
the increase of flue gas flow rate (Figure 4.3(d)). The optimal packing heights (Figure 4.3(e))
of the absorber are relatively stable in a range of 10.43–11.92 m while the optimal packing
heights of the stripper significantly increase from 5.44 m to 9.425 m because of higher
solvent regeneration degree (difference between lean loading and rich loadin g).
The economic results ( Figure 4.3(f)) shows that from Case F1 to Case F5, CCA decreases by
26.97%, from 86.538 €/ton CO 2 to 63.2 €/ton CO 2, although ACAPEX increases by 119.13 %
and VOPEX increas es by 137.11 % because the capture CO 2 increases by 200%. The result s
reflect scales effect here. It indicates that, if the capacity meets the requirement, a single train
design would be more cost -effec tive than a multi -train design for a capture plant.
89
Table 4.9 Optimisation results with variation of flue gas flow rate
Case tag F1 F2 F3 F4 F5
Flue gas flow rate ( % of the base case) 50 75 100 125 150
CO 2 concentration in flue gas (mol%) 4.50 4.50 4.50 4.50 4.50
MEA concentration (wt%) 35 35 35 35 35
CO 2 capture level (%) 90.00 90.00 90.00 90.00 90.00
CO 2 captured (kg/s) 20.52 30.78 41.04 51.30 61.56
Lean Loading (mol CO 2/ mol MEA) 0.315 0.294 0.294 0.293 0.275
Rich Loading (mol CO 2/ mol MEA) 0.463 0.465 0.462 0.464 0.467
Flow rate of lean solvent (kg/s) 542.41 701.25 905.60 1257.0
6 1337.1
0
L/G ration ( kg/kg) 1.64 1.42 1.37 1.52 1.35
Reboiler duty (MW th) 82.92 123.70 166.21 207.37 245.67
Specific duty (GJ/ton CO 2) 4.04 4.02 4.05 4.04 3.99
Diameter of absorber (m) 13.37 16.17 18.93 20.65 22.75
Packing height of absorber (m) 11.29 11.51 10.42 11.10 11.92
Diameter of stripper (m) 5.44 6.67 7.72 7.92 9.42
Packing height of stripper (m) 5.35 6.84 7.68 8.87 9.48
ACAPEX (M€/year) 18.74 23.41 27.79 31.92 36.97
FOPEX (M€/year) 4.32 5.40 6.41 7.37 8.53
VOPEX (M€/year) 28.07 37.48 47.51 57.83 66.56
TAC (M€/year) 51.14 66.29 87.71 97.11 112.05
CCA (€/ton CO 2) 86.54 74.78 69.13 65.73 63.20
90
Figure 4.3 Optimisation results with variation of flue gas flow rate
010203040506070
405060708090
50% 75% 100% 125% 150%
ACAPEX and VOPEX (M €/year)CCA (€/ton CO 2)
Flue gas flow rate change percent(f) Economic performance
CCA
ACAPEX
VOPEX0.250.30.350.40.450.5
50% 75% 100% 125% 150%CO 2 Loading (mol CO 2/mol MEA)
Flue gas flow rate change percent(a) CO 2loading in solvent
Lean Loading
Rich Loading
3.83.944.14.24.34.4
050100150200250300
50% 75% 100% 125% 150%
Optimal Specific Duty (GJ/tonCO 2)Optimal reboiler duty (MW th)
Flue gas flow rate change percent(c) Thermal performance
Reboiler duty
Special duty
0510152025
50% 75% 100% 125% 150%Optimal column diamter (m)
Flue gas flow rate change percent(d) Column diamter
DI_ABS
DI_STR00.511.522.5
400600800100012001400
50% 75% 100% 125% 150%
L/G ratio (kg/kg)Lean solvent flow rate (kg/s)
Flue gas flow rate change percent(b) Solvent
Solvent flow rate
L/G Ratio
02468101214
50% 75% 100% 125% 150%Optimal packing height (m)
Flue gas flow rate change percent(e) Height of packing
PH_ABS
PH_STR
91
4.5 Concluding remarks
In this chapter, the cost model of solvent -based PCC process was developed based on the
major equipment costs provided by vendors after detailed engineering design in IEAGHG
(2012 ). The uncertainty of this method could be in the range of from −15% to 20%, instead of
other correlation -based methods with uncertainty in the range of from −30% to 50%. The cost
model was then integrated into the process model by coding Fortran subroutine in Aspen
Plus®. Using this model, the optimisation studies were carried out for the carbon capture
process a nd on the impact s of variation s of the key variables.
Compared to the base case, the optimal case has small er diameter and lower packing height
for both the absorber and the stripper . This leads to significant saving s in CAPEX. There are
also slight savings of VOPEX because of optimal values obtained for key operational
variables such as lean loading and L/G ratio . As a result , the cost of CO 2 avoided of the
optimal case is 69.13€/ton CO 2, which is about 18.4% lower than the base case of 8 6.83€/ton
CO 2.
Findings from case studies on cost optimisation in respon se to variations in several key
variables include :
In the optimisations in all cases, t he rich solvent reaches its upper loading limit in all
the cases at the temperature, pressure and composition conditions.
The range of optimal lean loading in these three case studies is 0.275 –0.331 mol
CO 2/mol MEA compared with 0.132–0.234 mol CO 2/mol MEA from the literature .
The optimal packing height of the stripper significantly depends on the solvent
regeneration degree (difference between CO 2 loading in lean solvent and rich solvent ).
The reduction of CCA is more significant when MEA concentration in solvent s
increases from 20 wt% to 30 wt% than increas ing from 35 wt% to 40 wt%.
For scale -up of the optimal design , scale effect impacts not only economic terms but
also process variable s. New optimisation could be carried out for each single case to
obtain optimal values of both the equipment sizes and key operational variables. It
should be aware that optimal values of process variables may drift off recommended
ranges by experimental studies. In this case, the optimisation model should be
carefully checked to see whether it includes relevant process constraints discovered
by experimental studies .
92
Chapter 5: Integration of NGCC Power Plant and
Solvent -based PCC Process and CO 2
Compression Train
This chapter aims to explore the integration between the NGCC power plant and the PCC
process and the CO 2 compression train. The steady state model of a 453MW e NGCC power
plant was developed, including the gas turbine, HRSG and steam turbine. The general
interfaces between NGCC and PCC were discussed . Exhaust gas recirculation (EGR) was
investigated. The supersonic shock wave compress or was adopted for the CO 2 compression
and its heat integration options were explored .
5.1 NGCC power plant model
5.1.1 Modelling of gas turbine
In the model of a 453MW e of NGCC power plant, GE PG9371FB gas turbine from General
Electric was employed. The modelling of gas turbine in Aspen Plus® was performed by
combining three process sections including air compressor, combustion reactor and gas
expander (Ong'iro et al., 1995) . The compressor and expander sections were simulated as
isentropic compressors or turbines using Compr block in Aspen Plus®, in which, isentropic
efficiency and mechanical efficiency could be specified to improve the prediction accuracy
(Canepa et al., 2013) . The combustor section was simulated with an RGibbs reactor block
(Ong'iro et al., 1995) . The vent oxygen is controlled to a certain value to ensure complete
(equilibrium) combustion. It calculates the equilibriums by the Gib bs free energy
minimization thus the complicated calculations of reaction stoichiometry and kinetics are
avoided with only required inputs of the temperature and the pressure of the reactor. PR -BM
(Peng -Robinson equation of state with Boston -Mathias modifi cations) was used for the
property calculations for this gas turbine (Canepa et al., 2013) .
One key factor affecting the accuracy of modelling of the gas turbine is the modelling of the
turbine cooling. This relatively large cooling flow (approximatel y 20% of the inlet air flow)
has two negative effects: firstly it reduces the temperature of the gas expanding through the
turbine, and therefore its power output, secondly it adds losses connected with the mixing of
the cooling air with the turbine workin g fluid. Here a simple and generic method proposed by
Jonsson et al. (2005) was used, compared with very detailed and complex modelling in some
specific software package s such as Thermoflex®. In th is model, a part of fresh air is split from
93
the exit of the air compressor and bypass es the combustor to mix with the hot combustion gas
before entering the gas expander (see Figure 5.1).
Figure 5.1 Schematic of gas turbine developed in Aspen Plus®
The mass flow rate of this bypass air for turbine blade cooling is calculated by Equation
(5.1) (Jonsson et al., 2005) .
(5.1)
where and are tunable parameters; is the mass flow rate of the hot combustion gas;
represents the maximum blade surface temperature , and is the average specific heat
capacity of bypass air and hot gas between combustion exit and gas expander.
This pressure drop acros s the hot gas -cooling air mixer is calculated using by Equation
(5.2) (Jonsson et al., 2005) , which is derived from momentum balance .
(5.2)
where, is a tunable parameter; is the pressure of hot gas entering the gas
expander.
During the thermal performance analysis of gas turbine, these three model parameters (b, K
and s) can be tuned to represent different type of the gas turbines. The main model inputs and
LOSSCOOLMIXERAIRSPLIT
AIRCOMPCOMBUST
EXPENDERHOTGAS2
HOTGAS3AIRBYPASHOTGAS1
AIR1AIR2
AIRIN
HOTGAS4 NGGAS
94
the results of the model in this study are summarized in Table 5.1, while model performance
and validation results are shown in Table 5.2.
Table 5.1 GT modelling assumptions and tuning parameters
Process conditions
Ambient temperature (°C) 15
Atmospheric pressure (bar) 1.00
Air mass flow rate (ton/h) 2,365
Compressor pressure ratio 18.3
Compressor isentropic efficiency 0.915
Expander isentropic efficiency 0.915
Combustor exit temperature (°C) 1,425
Tb (°C) 860
Tuned parameter s
S 1
B 0.1668
K 1.1927
Table 5.2 Model validation with manufactory product data sheet
This study GE Product Data
(GE Power, 2016)
Cooling mass flow rate (kg/hr) 28,700 –
Cooling loss (bar) −2.45 –
Compressor discharge pressure ( bar) 18.3 18.3
Exhaust temperature (°C) 641.8 642.0
Net power output (MW e) 299.1 299.0
Net Efficiency (%, LHV) 38.7 38.7
Exhaust Energy (MJ/h) 1,704.3 1681.0
5.1.2 Model development for NGCC power plant
Another part of electricity is generated from the steam cycle. PR -BM method is used for the
gas cycle and STEAMNBS (AspenTech, 2012b) property method is used for steam cycle.
95
Figure 5.2 show the flow diagram of the model. At the ambient conditions (ambient
temperature is 9 °C and ambient pressure is 1.01 bar), fresh air is compressed and mixed with
natur al gas before entering the combustion chamber . The hot gas leaves the combustion
cham ber at a temperature of 1 ,427 °C. The hot gas expands in the gas turbine and
consequently generate s a part of the total electricity output of the NGCC power plant .
Exhaust gas es from the gas turbine is used to generate steam in the HRSG . Steam s from
differ ent stages of the HRSG go to the high pressure steam turbine (HP -ST), the intermediate
pressure steam turbine (IP -ST) and the low pressure steam turbine (LP -ST) to generate
another part of the total electricity output of the NGCC power plant . The main mode l
parameters are given in Table 5.3.
Figure 5.2 The flowsheet of NGCC power plant standalone
One feature of this model is that the steam conditions of HRSG are higher in both
temperature and pressure than what is currently obtainable . The pressure and temperature of
high pressure steam are 170 bar and 600 °C compared to about 120 bar and 556°C in existing
cases . The pressure and temperature of intermediate p ressure steam are 40 bar and 600 °C
compared to about 30 bar and 550°C in existing cases . An explanation is described in the
referred benchmark report (IEAGHG, 2012) as the original equipment manufacturers are
already considering similar steam conditions and it is considered that utilizing these
conditions in NGCC power plant will be proven and typical by 20 20.
96
Table 5.3 Model parameters of NGCC power plant
Parameters Value
Natural gas
composition CH 4 (vol%) 89
C2H6 (vol%) 7
C3–C5 (vol%) 1.11
CO 2 (vol%) 2
N2 (vol%) 0.89
Steam turbine Steam inlet of HP turbine (bar/°C) 172.6/601.7
Steam inlet of IP turbine (bar/°C) 41.5/601
Steam inlet of LP turbine (bar/°C) 5.8/293.1
HP/IP/LP turbine efficiencies (%) 92/94/90
Minimum
temperature
approach of
HRSG Steam and gas (°C) 25
Gas and boiling liquid (°C) 10
Liquid and gas (°C) 10
Approach of economizer (°C) 5
Condenser pressure and temperature (bar/°C) 0.039/29.0
5.1.3 Model validation
For large scale NGCC power plant simulations, operational or experimental data for model
validation purpose is not available. In this study, the simulation results using Aspen Plus®
were compared with the simulation results using another software package, GT Pro®, in order
to make a brief validation. Table 5.4 shows the compariso n results of Aspen Plus® and GT
Pro®
, which appear to be in good agreement.
97
Table 5.4 Comparison of the simulation results for model validation
Input
Fuel flow rate (kg/s) 16.62
Air flow rate(kg/s) 656.94
Validation results
IEAGHG, (2012) This study
Temperature of flue gas to HRSG (°C) 638.4 638.4
Flow rate of flue gas to HRSG (kg/s) 114.97 114.97
HP turbine inlet pressure, temperature (bar/°C) 172.5/601.7 172.6/601.7
IP turbine inlet pressure, temperature (bar/°C) 41.4/601.5 41.5/601.0
LP turbine inlet pressure, temperature (bar/°C) 5.81/293.3 5.8/293.1
Condenser pressure and temperature (mbar/°C) 0.04/29.2 0.039/29.0
Gas turbine power output (MW e) 295.2 4 295.03
Steam turbine power output (MW e) 171.78 170.71
Net plant power output (MW e) 455.15 453.872
Net plant efficiency (%,LHV) 58.87 58.74
5.2 Integration of NGCC with PCC process and CO 2 compression
5.2.1 General interfaces of the integration
When a n NGCC power plant is designed or r etrofitted with a PCC and compression process es,
some structural modifications are required for basic interfaces . These include : (1) connecting
flue gas from exit of HRSG of the power plant to the PCC process, (2) extracti ng low
pressure steam from the steam cycle of the power plant to provide heat for solvent
regeneration in the PCC process, (3) connecting steam condensate outlet in the PCC process
to the steam cycle of power plant , and (4) electrical power connection from the power plant
to service electrical power consumption in the PCC and CO 2 compression processes .
The process flow diagram can be seen in Figure 5.3. Flue gas leaves the HRSG at a
temperature of around 80 °C and enters a gas conditioning unit which consists of a direct
contact column (DCC), a water recirculating pump and a blower. The flu e gas is then cooled
down to 40 –50 °C (Kvamsdal et al., 2011b) by a spray of water at 25 °C, in order to improve
the absorption efficiency and to reduce solvent evaporation losses in the absorber (Wang et
98
al., 2011) . At the same time, a part of water is removed from the flue gas due to the
condensation. The flue gas is pressurized by the blower before it feeds into the absorber.
Figure 5.3 The flowsheet of NGCC power plant with EGR integrated with PCC process and
compression
Heat input for the solvent regeneration process is provided by extracting low pressure steam
from the steam cycle of the NGCC power plant into the stripper reboiler of the PCC process.
The flow rate of the steam extraction is decided by the operating conditions of PCC process
and has a large impact on the output of the power plant. Considering that high temperature
would result in thermal d egradation of the solvent in the reboiler and the stripper, normally,
the temperature of the reboiler is maintained between 110 °C to 130 °C at an operating
pressure of 1.6 –2.0 bar. There are three potential configurations , clutched turbine, throttled
turbine and floating crossover pressure, in the NGCC power plant process for steam
extraction (Kang et al., 2011) . In this study, the steam is extracted off from the floating IP/LP
crossover, the most feasible solutions for steam extraction (Lucquiaud and Gibbins, 2009) , at
5.8 bar and 303 °C. Before the steam enters the reboiler, it is cooled down just above
saturation temperature with a spray of the condensate circu lated from the reboiler, which
helps to reduce the requirement of steam to be extracted from the power plant. After heat
HRSG
Gas Turbine
Steam TurbineCompression Pretreat Capture
99
exchange inside the reboiler, the steam is cooled down to condensate, which then is returned
to the deaerator in the HRSG of the power plant for cycling.
5.2.2 EGR technology
One disadvantage of using PCC process to NGCC power plant is that the CO 2 concentration
in flue gas is as low as 3 –4 mol% whilst it is 11 –13 mol% for a coal fired power plant,
resulting in lower absorption efficiency and larger equipment size in PCC capture plant
(Jonshagen et al., 2011; Biliyok et al., 2013) . EGR is regarded as an effective solution. The
underpinning of EGR is that the O 2 concentration in the flue gas leaving from the HRSG is
still high (11.41 mol% in this study). Even though EGR is applied, a relatively high oxygen
concentration in the combustion air can be ensured with an appropriate recirculation ratio .
However, in order to ensure the combustion efficiency in the burner, the minimum oxygen
concentration in combustion air should be 16 – 18 mol % (Ulfsnes et al., 2003; Canepa et al.,
2013) .
In this study, EGR ratio of 0.38 is selected to ensure the minimum oxygen content of 16
mol% . Table 5.5 presents the comparison results of the integration without EGR and with
EGR . With EGR, 4.73 kg/s less steam is extracted for solvent regeneration. So the net power
generated from the gas turbine section decreases by 0.39 MW e but steam turbines section
generates more by 4.13 MW e. At the same time, significant equipment size reductions are
achieved for both absorber and stripper in the case with EGR. In this case, the flow rate of the
flue gas feed reduces by 38% which results in 37.39% and 9.36% reduction of the cross –
section area of the absorber and stripper respectively. The reason for the di fference between
these values is that the required cross -section area of a column is decided by both gas phase
and liquid phase loadings inside the column. Although the flow rate of flue gas reduces 38%,
the flow rate of lean solvent only decreases by 8.1% because higher liquid gas ratio (L/G
ratio) is needed for higher CO 2 concentration in the flue gas (it increases from 4.5 mole% to
7.32 mol%). The economic results shows CCA in the case with EGR decreases 5.12%, from
69.713 €/ton CO 2 to 66.142 €/ton CO 2 while CAPEX decreases by 6.81% and VOPEX
decrease by 2.93%.
100
Table 5.5 Optimal design of the integration without EGR and with EGR
Category Parameter without EGR with EGR
Power plant
performance Flow rate of fresh air intake (kg/s) 656.94 407.45
O2 concentration in combustion air (mol %) 20.74 16.0
O2 concentration in gas turbine vent gas
(mol %) 11.4 6.45
Gas turbine power output (MW e) 295.03 294.64
Steam turbine power output (MW e) 113.56 117.69
Steam extracted for reboiler (kg/s) 76.39 71.06
Flow rate of flue gas to PCC (kg/s) 660.5 4 408.75
CO 2 concentration in flue gas (mol %) 4.5 7.32
PCC
technical
performance CO 2 captured (kg/s) 41.04 40.94
Lean loading (mol CO 2 /mol MEA) 0.294 0.299
Rich loading (mol CO 2 /mol MEA) 0.462 0.472
Flow rate of lean solvent ( kg/s) 905.60 987.65
L/G ( kg/kg) 1.37 1.495
Reboiler duty (kW) 166.2 6 164.003
Specific duty (GJ/ton CO 2) 4.05 4.01
Absorber pressure loss (bar) 0.048 0.036
Absorber diameter (m) 18.93 16.13
Packing height of absorber (m) 10.43 10.37
Stripper pressure loss (bar) 0.012 0.011
Stripper diameter (m) 7.72 7.35
Packing height of stripper (m) 7.68 6.87
PCC
economic
performance CAPEX (M€) 213.7 8 199.21
FOPEX (M€/year ) 6.41 5.89
VOPEX (M€/year ) 47.51 46.11
TAC (M€/year ) 81.71 77.99
CCA (€/ton CO 2) 69.13 66.14
101
5.3 Heat integration options based on supersonic shock wave
compression
5.3.1 CO 2 compression technology
After the CO 2 leaves the capture plant, it will be transfered for geologic sequestration. In the
transport section of CCS, pipelines are the preferred method for onshore and offshore
transport of large volumes of CO 2. The dense phase is regarded as the most energy -efficient
condition due to its high density and low viscosity. Consequently, current operating practice
for CO 2 pipelines is to maintain the pressure well above the critical pressure. Considering the
pressure drop along the length of the pipeline and the i mpact of the elevation change and
impurities, the entry pressure of the CO 2 pipel ine network is as high as 110 –150 bar. Thus a
compression train is required to pressurize the CO 2 stream from PCC captured plant to reach
a so high entry pressure.
One limita tion of conventional compressor is that the pressure ratio per stage is normal ly less
than 3, otherwise the efficiency would decrease drastically as the temperature rises with the
pressure during the adiabatic compression process . Thus 6 – 16 stage compres sor is normally
required . Witkowski et al. (2013) performed a thermodynamic evalua tion of various CO 2
compression configurations with 6 – 12 stages compress or based on conventional centrifugal
compressor and integrally geared compressor. In Section 6.4 in this study , a comprehensive
techno -economic evaluation was conducted for different configuration of compression train.
The optimal option was selected to get a minimum annual cost including annualized capital
cost, operating and maintenance cost and energy cost. The optimal configuration of the
compression train compromises 6 stages integrally geared follow ing pumping and
intercoolers with an exit temperature of 20 oC. The mu lti-stage compression means a great
capital investment in terms of the equipment material cost, const ruction and installation cost .
With the a im to address the challenge of the high investment cost, s upersonic shock wave
compression technology was developed by RAMGEN Power System (Lawlor, 2009) for CO 2
compression. The shock wave compression only needs two stages of compression ( compared
with 6 to 16 stages for the conventional multi -stage approach), and the potential capital cost
saving for compression chain is up to 50% (Ciferno et al., 2009) in addition to reduced
footprint requirement. The discharge temperature of compressed CO 2 is as high as 246–
285oC (Witkowski et al., 201 3) due to higher pressure ratio of each stage, providing an
opportunity for compression heat integration .
102
In this study, s upersonic shock wave compression technology was adop ted for the CO 2
compression. T he compression model was developed also in Aspen Plus® and was validated
with published data from RAMGEN Power System (Lawlor, 2009) . After that, the inlet and
outlet pressure of the compression train were modified to adopt for the boundary conditions
in this study. The model parameters are seen in Table 5.6. In terms of reliability , there is no
report about the comparison between this new compression technology and conventional
compressors . However, it should be aware that reliability of compr essors is an important
performance for real industrial applications.
Table 5.6 Process boundary conditions and parameters of CO 2 compression
Parameters Value
Flow rate of CO 2 stream (kg/s) 41.04
Inlet pressure (bar) 1.9
Inlet temperature (oC) 20
Outlet pressure (bar) 136
Stage number 2
Exit temperature of stage 1(oC) 214.5
Exit temperature of stage 2 (oC) 230.5
pressure ratio per stage 8.65
Isentropic efficiency (%) 85
Intercooler exit temperature (oC) 20
Last stage exit temperature (oC) 20
Pressure drop of intercooler (%) 3
Power consumption (MW e) 14.8
5.3.2 Heat integration case setups
Table 5.6 shows that the exit temperature of the CO 2 stream is as high as 214.5 –230.5 °C for
the supersonic shock wave compressors . At this temperature, the compression heat could be
recovered by integrating the pressurized streams with low temperature streams in the NGCC
power plant and the PCC process. T wo options could be justified as below.
1) The compression heat is integrated into the steam generation cycle of HRSG to
generate more steam.
103
In the NGCC power plant, the steam coming out of the LP -ST is cooled down to condensate
with a temperature of 29 .0 °C at a pressure of 0.0 39 bar before it is pressurized to a high
pressure by a pump. Then the subcooled water enters the economizer section of HRSG, in
which, it is heated to around 158 °C by the hot flue gas in normal case . Applying this heat
integration opt ion, this subcooled water could be lined to the compression train first as a
refrigerant of the intercoolers. With this additional heat recovered from compression process,
more LP steam generation is expected to go to the LP -ST to generate more electricity .
2) The compression heat is integrated into the stripper reboiler of the PCC process for
solvent regeneration.
In the PCC process, the operating temperature range of the stripper reboiler is from 110°C to
125°C, which is much lower than the exit temperature of each stage of the compressor. So the
compression heat could be transferred to provide heat to the reboiler. However, the reboiler
duty is so high that the compression exhaust heat cannot satisfy the reboiler duty requirement .
Thus, the steam from the po wer plant is still required at the same time using a multiple shell
kettle reboiler (Shah and Sekulic, 2003) .
In previous sections, different process integration options were discussed when NGCC power
plant is integrated with a PCC process and the CO 2 compression. A case study was conducted
for the evaluation of power consumptions and heat requirement of different options for
comparison purpose s. For the case setup, five scenarios were summarized as below:
1) Reference case : NGCC power plant without integration with PCC and compression
2) Case 1: NGCC power plant without EGR integrated with PCC and compression
without co mpression heat integration
3) Case 2: NGCC power plant with EGR integrated with PCC and compression without
compression heat integration
4) Case 3: NGCC power plant with EGR integrated with PCC and compression with
compression heat integration into the steam cyc le of HRSG
5) Case 4: NGCC power plant with EGR integrated with PCC and compression with
compression heat integration into the reboiler of the stripper
104
5.3.3 Results and discussion
Table 5.7 shows the results of energy and electricity consumptions of each case. By
comparing the reference case (NGCC standalone) and Case 1, a to tal 9.58 %-points net power
efficiency decrease is obse rved when the NGCC power plant integrated with the PCC process.
This obvious reduction is caused by three main factors: 1) the steam through the LP -ST
decreases hugely to lead to a power output reduction because of steam extraction, which
contributes 7.40% -points net efficiency decrease; 2) the electricity consumption of CO 2
compression contributes 1.92% -points net efficiency decrease; 3) auxiliary electricity
consumption of the blower and solvent circulation pumpers accounts for 0.55% -points net
efficiency decrease.
The results of Case 2 shows EGR help to achieve a 0.77% efficiency improvement compared
with Case 1. The reason is dissected as follow s. Firstly, the specific reboiler duty decreases to
4.31 MJ th/kg CO 2 from 4.54 MJ th/kg C O2. The absorption efficiency is improved because of
increase in the CO 2 concentration in the flue gas (from 4.5 mol% to 7.32 mol% in Table 5.5 ),
which leads to a higher rich solvent loading and then a lower recirculating solvent flow rate .
The above results in a lower reboiler duty for the solvent regeneration . Secondly, the power
consumption of the PCC process reduces to 2.0 4 MW e from 4.24 MW e. With EGR at a ratio
of 0.38 , the flue gas flow rate decreases significantly, which causes a great reduction of the
power consumption of the blower at the upstream of the absorber . Meanwhile , the simulation
results show the discharge pressure of the blower also decreases because of the decrease of
the whole tower pressure drop of the absorber (see Table 5.5 ).
Applying compression heat integration s into the main process of NGCC and PCC , Case 3 and
Case 4 improves the net LHV efficiency of the power plant to 50.25% and 50.47%
respectively. In Case 3, the subcooled water from the feed water pump of the HRSG is lined
to the compression train and is heated to around 65°C before entering the economizer of the
HRSG . One limitation of this option is that the temperature of the water leaving the
economizer should be lower than its boiling temperature, otherwise there would be vapo ur
phase exiting in its downstream pump. In Case 4, the temperature of the stream from
compression train is 135 °C , which is still higher than expected recoverable temperature of
90 °C , after it exchanges heat the stripper reboiler . Thus more efficiency im provement could
be achieve d by combining other low-temperature heat recover technology .
105
Table 5.7 Performance comparison results of different cases
Description Reference Case 1 Case 2 Case 3 Case 4
Major process components NGCC NGCC +PCC NGCC +PCC NGCC +PCC NGCC +PCC
The application of EGR without EGR without EGR with EGR with EGR with EGR
Compression heat integration without without Without with HRSG With reboiler
Gas turbine power output (MW e) 295.03 295.03 294.64 294.64 294.64
Steam turbine power output (MW e) 170.71 113.56 117.69 120.14 121.85
Power island power consumption (MW e) 11.69 9.7 9.7 9.7 9.7
CO 2 compression power consumption ( MW e) – 14.8 14.8 14.8 14.8
Power consumption in PCC (MW e) – 4.24 2.04 2.04 2.04
Stripper reboiler duty (MW th) – 186.8 176.2 176.2 176.2
Steam extracted for reboiler (kg/s) – 76.39 71.06 71.06 65.50
CO 2 captured (kg/s) – 41.11 40.92 40.92 40.92
Specific reboiler duty (M Jth/kg CO 2) – 4.54 4.31 4.31 4.31
Net plant power output (MW e) 453.87 379.85 385.80 388.2 5 389.9 6
Net plant efficiency (%, lower heating value) 58.74 49.16 49.93 50.25 50.47
Efficiency decrease (% -points) compared with
reference case – 9.58 8.81 8.49 8.27
Efficiency increase (%-points ) compared with
Case 1 – – 0.77 1.09 1.31
106
As a comparison of these two options, Marchioro Ystad et al. (2013) reported that employing
a CO 2 Rankine cycle with an additional turbine improves the thermal efficiency by 1.63% –
points. In this study, there is no major capital cost required for the integration options in both
Case 3 and Case 4. These efficiency improvements are meaningful especially considering
great amount of the total electricity output from the gas-fired power plants . Taking the total
number of gas -fired electricity consumption in EU (2016b) in 2015 as the calculation base ,
the annual saving could be around 100 M€, assuming that this heat integration is applied to
even only 5% of the gas-fired power plant s in Europe .
5.4 Conclu ding remarks
This chapter presents the investigation on thermal performances of different the integration
options of a 453MW e NGCC equipping with a PCC process and a CO 2 compression train.
The process models of each process were developed us ing Aspen Plus® and were validated
with published data or experimental data . The effect of EGR to the performance of the
integration was investigated first. Significant saving s was achieved with the contributions
from both the CAPEX and VOPEX. The CCA of the case with EGR decreases 5.12%
compared with the case without EGR, from 69.13 €/ton CO 2 to 66.14 €/ton CO 2.
Integrated with the PCC process and the compression train, t he thermal efficiency (LHV) of
the NGCC power plant deceases from 58.74% to 49.16% . This reduction includes 7.40% –
points decrease due to steam extraction, 0.55 %-points reduction due to PCC power
consumption and 1.92 %-points reduction due to compression train power consumption. With
the application of EGR in the NGCC power plant at a recirculation ratio of 0.38, the net
efficiency increases 0.77% -points while the cross -section area s of the absorber and stripper in
the carbon capture process reduce d by 37.39% and 9.36% respectively . The compression heat
integration option s have been analyse d by applying supersonic shock wave compression
technology . Compression heat integration into the steam cycle of HRSG and stripper reboiler
achieves 0.32%-points and 0.54%-points net efficiency improvement separately without
major capital investment required. The study indicate s that EGR technology, supersonic
shock wave compression technology and compression heat integration s could be future
direction s for commercial PCC deployment in NGCC power plants.
107
Chapter 6: Optimal Design of CO 2 Transport
Pipeline Network
This chapter presented the study on optimal design of the pipeline network planned in the
Humber region of the UK. Steady state process simulation models of the CO 2 transport
pipeline network were developed using Aspen HYSYS®. The simulation models were
integrated with Aspen Process Economic Analyser® (APEA). Techno -economic evaluations
for different options were conducted for the CO 2 compression train and the trunk pip elines
respectively. The evaluation results were compared with other published cost models.
Optimal options of compression train and trunk pipelines were applied for the whole pipeline
network . The overall cost of CO 2 transport pipeline network was analyse d and compared
between the base case and the optimal case.
6.1 Pipeline network system
In the UK, the Humber region offers good opportunities for CCS deployment as it is not only
the biggest CO 2 emission area in the UK, but also the area with easy reach to CO 2 offshore
storage sites in the North Sea (Lazic et al., 2013) . There are two advanced proposals for CCS
power st ation developments that utilise the trunk pipelines: The Don Valley Power Project
(DVPP) and the White Rose CCS Project (even though cancelled very recently) . DVPP will
use pre -combustion carbon capture technology at a new -build integrated gasification
combined cycle (IGCC) power plant of 920 MW e gross output (CCSA, 2014) . The White
Rose CCS project is a demonstration project of an oxy -fuel power plant of 450 MW e gross
output (CPL, 2013) .
Figure 6.1 shows the proposed route corridor of the pipeline network. CO 2 captured from
DVPP will be transported in gaseous phase at a maximum allowable operating pressure
(MAOP) of 35 bar, and would then be boosted to dense phase by a compressor near the multi –
junction site, before joining the dense phase CO 2-rich stream from the White Rose CCS plant.
The combined CO 2-rich stream will then be transported in dense phase via an onshore trunk
pipeline with a MAOP of 136 bar. The onshore pipelines are buried under ground 1. 2 m. A
booster pumping station located near the coast will boost the pressure of the CO 2-rich stream
before it is transported in the offshore trunk pipeline with a MAOP of 186 bar to a saline
aquifer storage site more than 1 km beneath the bed of the North Sea. Table 6.1 presents the
key parameters of the pipelines. The material of pipelines is carbon steel and the size of
pipelines follows ANSI standard.
108
Figure 6.1 The pipeline sketch for the Humber case study
Table 6.1 Parameters of the pipelines
Emitter Flow rate
range Collecting pipelines Onshore trunk
pipeline Offshore trunk
pipeline
Length Internal
diameter Length Internal
diameter Length Internal
diameter
Mt/a km mm km mm km mm
Don Valley 0.91–6.27 15 738.2
71 571.8 91 559.2
White Rose 0.61–2.65 5 295.5
An entry specification for the CO 2-rich stream is needed to define the acceptable range of
composition, taking into account safety, impact on pipeline integrity and hydraulic efficiency
(Race et al., 201 2). In this case study, the entry specifi cation was defined to be 96 mol % CO 2
and a mixture of nitrogen, oxygen, hydrogen, argon and methane w ith hydrogen limited to
2.0 mol% and oxygen limited to 10 ppmv.
6.2 Process model development and economics evaluat ion
methodology
For a real CO 2 pipeline network project described in Section 6.1, the techno -economic
evaluation should be more detailed and realistic. In this study, process simulation models
were developed in Aspen HYSYS®. Then the simulation results were exported into APEA for
economic evaluation.
109
6.2.1 Process simulation model development for the base case
6.2.1.1 Physical property method
The cubic equation of state (EOS) has been widely used to calculate the physical properties of
CO 2 for pipeline transport modelling (Li and Yan, 2009) . Peng -Robinson EOS (Peng and
Robinson, 1976) is most frequently used. More complex EOS such as Lee Kesler (Lee and
Kesler, 1975) , SAFT (Wertheim, 1984; Wertheim, 1986) , Span and Wagner (Span and
Wagner, 1996) and GERG (Kunz and Wagner, 2012) were used in recent studies. There is no
consensus in the literature regarding the best EOS for the design of CO 2 pipelines.
(Diamantonis et al., 2013) compared the results of several EOS with experimental data and
found that PR EOS is of reasonable accuracy, even when compared with more advanced
EOS, when binary interaction parameters are used. In this study, PR EOS has been selected
considering both the accuracy and the simplicity.
In this study, PR EOS with calibrated binary in teraction parameters has been used considering
both the accuracy and the simplicity. Table 6.2 lists the calibrated binary interaction
parameters for PR-EOS used in this study. The A PEs between the calculations of PR EOS
and the experimental data were listed for corresponding values. For calibration of of
CO 2-H2, there is no good agreement among the available experimental data and there is no
liquid volume experimental data .
One weakness of PR EOS reported by E.ON’s report (E.ON, 2010) is that it is very accurate
in the near -critical region. This study focuses on the techno -economic evaluations based on
steady state simulations. For the trunk pipeline transport section, CO 2-rich stream is in the
subcooled liquid phase. The temperature range is from 4oC to 20oC and the pressure range is
from 101 bar to 150 bar, which is far away from the critical region of the CO 2. In this T/P
range, the deviation of pure CO 2 density is from -4.8% to 0.1% for the calculations of PR
compared to the calculations of SW according to the comparison results from E.ON’s report.
110
Table 6.2 APE between experimental data and PR -EOS for corresponding values
Binary Bubble pressure Liquid volume
Reference Temp.
(K) Pressure
(MPa) APE
(%) Temp.
(K) Pressure
(MPa) APE
(%)
CO 2- N2 −0.007 220–301 1.4–16.7 3.73 209–
320 1.4–16.7 1.54 Li and Yan (2009)
Diamantonis et al. (2013)
CO 2- Ar 0.141 288 7.5–9.8 2.32 288 2.4–14.5 1.83 Diamantonis et al. (2013)
CO 2- H2 0.147 290.2 5.0–20.0 5.6% – – – Foster et al. (2010)
6.2.1.2 Assumptions, constraints and inputs
The maximum entry flow rates from both the White Rose plant and the Don Valley plant and
the highest ambient temperature were chosen as the base case. This is considered as the worst
case scenario with respect to the energy requirement since it would require the highest entry
pressure and the greatest boosting pressure at the pump station. Figure 6.2 shows the
flowsheet of pipeline network developed in this study.
Figure 6.2 The flowsheet of pipeline network in Aspen HYSYS®
The assumptions made for the pipeline network model are as follow s: (1) the pressure drops
across valves and other fittings are negligible; (2) the adiabatic efficiencies of compressors
and pumps used in this model are fixed at 75%.
The pressure settings of key sections are based on two operational constraints: (1) the entry
pressure (i.e. the outlet pressure of the compressor at each capture plant) should be high
enough to maintain a minimum pipeline operating pressure of 101barg to avoid two phase
flow in the common pipeline; (2) a constant injection pressure of 126 bar is specified to
satisfy the injection rate. In reality, the required injection pressure at the offshore storage site
will rise over the lifetime of the operation wit h injection pressures below 126 bar being
111
sufficient in the initial phase. The input and boundary conditions for the base case are
specified in Table 6.3.
Table 6.3 Input and boundary conditions of the base case
unit White Rose Don Valley
Capture technology – oxy-fuel IGCC
Composition of CO 2 -rich stream mol% 96%CO 2, 2%N 2,
2%Ar 96%CO 2, 2%N 2,
2%H 2
Flow rate t/h 334.596 791.667
Suction pressure of compression bar 1 1
Suction temp. of compression oC 20 20
Number of compression stages – 5 4
Exit pressure of compression bar 112.5 35
Exit temp. of compression oC 20 20
Number of mid -compressor stages – – 2
Exit pressure of mid -compression bar – 121.23
Trunk pipelines entry temperature oC 20 20.0
Differential pressure of pump station bar 43
Offshore platform arrival pressure bar 126
6.2.1.3 Model validation
For large scale CO 2 pipeline network simulations, operational or experimental data for model
validation purpose is not available as the projects considered are currently only in the
planning stage. In this study, the results of the Aspen HYSYS® base case model were
compared with the results from another software package, PIPE -FLO® from industrial
collaborator National Grid. Table 6.4 shows the results of Aspen HYSYS® and PIPE -FLO®
,
which appear to be in good agreement.
112
Table 6.4 Comparison of the simulation results
Entry
pressure at
White Rose Entry
pressure at
Don Valley DP of mid –
booster for
Don Valley DP of pump
station Arrival
pressure
bar bar bar bar bar
Aspen
HYSYS® 120.5 35 86.92 43 126
PIPEFLO® 120.2 35 86.70 42.4 126
APE (%) 0.25 – 0.25 1.40 –
6.2.2 Economic evaluation methodology
Economic evaluations were conducted using APEA V8.0 using data from the 1st quarter of
2012. APEA becomes an industry -standard tool known to be far more accurate than
correlation -based economic approaches and is used for engineering design of many projects .
APEA includes design procedures and costs data for hundreds of types of materials of
projects. A bottom -up approach is used in APEA. When the simulation models are exported
into APEA, the unit operations are mapped and sized according to relevant design codes.
Then the cost was estimated for single piece of equipment.
The total cost includes capital expenditure (CAPEX) and operational expenditure (OPEX).
OPEX can be split into fixed OPEX (operating and maintenance (O&M) cost) and variable
OPEX (mainly the energy and util ities cost). In this study, for a clearer comparison, the
annual cost and the levelise d cost (per ton of CO 2) were used. The total annual cost was split
into annualized capital investment cost (capital return factor is 0.15), annual O&M cost and
annual ene rgy and utilities cost. In consistency with that, the levelise d cost was split into
levelise d capital cost, levelise d O&M cost and levelise d energy and utilities cost.
To harmonize results for comparison with other studies, the follow ing assumptions are ma de:
1) the project begins in January 2012; 2) all costs are corrected to €2012 using the average
inflation index; 3) the captured CO 2 mixture has neither economic value nor disposal cost; 4)
cooling water is sourced from a nearby body of water at the cost of pumping and operation of
a cooling tower. Other important cost inputs are provided in Table 6.5, with the costs given in
Euro.
113
Figure 6.3 The work flow of the techno -economic evaluation
Table 6.5 Economic evaluation cost inputs
Description Unit Value
Electricity price €/kW 0.0775
Cooling water price €/m3 0.0317
The price of refrigerant -Freon 12 €/t 0.17
Carbon steel price €/kg 500
Interest rate % 15
Contingency % 5
Project economic life a 25
Diameter calculation
in different
correlation methodsInput information
Simulations on trunk
pipelines
Economic evaluations
on trunk pipelines
Comparison with the
literature and analysis
Simulation for whole
pipeline networkCompression
technology analysis
Simulation on
compression train
Economics evaluation
on compression train
Comparison with the
literature and analysis Comparison of
compression options
to select optimal
optionComparison of
pipeline options to
select optimal option
Economics evaluation
for whole pipeline
network
114
For a given base case, the simulation model of whole pipeline network was developed first to
check its accuracy and to gain basic data of streams and processes. The evaluations for
different technical options go forward for compression train and trunk pipe lines respectively,
in order to confirm whether the designs of base case are optimal. Otherwise, optimal options
would be applied for an optimal case. Finally, the overall costs of the whole pipeline network
in the base case and the optimal case are summar ized and compared. Figure 6.3 shows the
work flow of the techno -economic evaluation in this study.
6.3 Techno -economic evaluation of CO 2 compression
Variou s types of compression configurations for CO 2 pipeline transport were found in
literature. In the study of Zhang et al. (2006) , 5 stage centrifugal compression was applied for
pressurization power consumption analysis. McCollum and Ogden (2006) evaluated the
energy cost, CAPEX and O&M cost of the compression achieved with 5 s tage centrifugal
compression follow ed by pumping. Witkowski et al. (2 013) performed a thermodynamic
evaluation of various CO 2 compression configurations and only the power requirements of
those options were compared. In this section, the case studies about the compression train at
the White Rose plant was conducted to ge t optimal configuration. The results were compared
with other published studies in literature.
6.3.1 Compression configuration options
After the conditioning process, the CO 2 captured in the White Rose plant will be pressurized
from 1bar to 136 bar for dense ph ase transport by a compression train. Four compression
configurations were selected ( see Table 6.6) for techno -economic evaluation and compared
with the base case. For options C3 and C4, the CO 2 mixture is initially pressurized to
supercritical pressure (80 bar considering the impurities c ontent in this study) and then further
pressurized to the final exit pressure 136 bar by pumping. The difference between option C3
and C4 is the exit temperature of the intercoolers.
115
Table 6.6 Compression technology options and their process definition
Option Unit Base Case C1 C2 C3 C4
Description Centrifugal 5 stages
with 4 intercoolers Centrifugal 16
stages 4 intercoolers 8 stages centrifugal
geared with 7
intercoolers 6 stages integrally
geared with 5
intercoolers to 20 oC
+pumping 6 stages integrally
geared with 5
intercoolers to 38 oC
+pumping
Capacity t/h 334.60 334.60 334.60 334.60 334.60
Suction pressure bar 1 1 1 1 1
Suction temp. oC 20 20 20 20 20
Pumping suction
pressure bar – – – 80.0 80.0
Pumping suction temp. oC – – – 20 20
Exit pressure bar 136.0 136.0 136.0 136.0 136.0
Stage – 5 16 8 6 6
Isentropic efficiency % 75 75 75 75 75
Interstage cooler exit
temperature oC 20 38 38 20 38
Last stage exit temp. oC 20 20 20 20 20
116
6.3.2 Results and analysis
The comparison of energy and utilities requirement for the five compression configurations
can be seen in Table 6.7. Option C3 has the lowest annual energy and utilities cost. The
intercooling performance is one of the key factors related with the energy consumption of the
compressor. Option C2 has less co mpressor stages but more intercoolers than option C1.
Option C2 has 12.78% annual saving of energy and utilities cost , compared to option C1 .
Compared to option C4, the lower intercooler exit temperature in option C3 results in 3.10%
energy saving , resulti ng in 4.59% saving in annual energy and utilities cost. Op tion C3 has
5.44% energy saving but only 2.07% annual energy cost saving compared with option C2.
The reason is that , in option C3 , a suction temperature of 20 oC is specified to cool down the
CO 2 mixture with suction pressure at 80 bar, to avoid any gas formation for the pumping,
which cause higher refrigerant cost.
Table 6.7 Energy and utilities requirements of compression technologies
Cases Energy
requirements
(kWh) Cooling duty
(m3/h) Refrigerant
(t/h) Energy and
utilities cost
(M€/a)
Base case 34546 – 1257 23.13
C1 39921 2540 656 26.29
C2 34832 2977 423 22.93
C3 31921 – 1197 21.42
C4 32972 2304 592 22.45
Figure 6.4 shows the comparison of levelise d cost in breakdown of these five compression
technologies. The range of total levelise d costs is from 11.81 €/ton CO 2 to 14.99 €/t on CO 2.
Energy and utilities cost is the biggest part with a proportion of 65.6 –71.3%. Option C3 has
the lowest total levelised cost of 11.81 €/t on CO 2 although levelise d capital cost of option C3
is 0.25 €/ ton CO 2 higher than the base case. The reason is that lower pressure ratio of each
stage compression benefits a big saving of energy and utilities consumption. Compared with
the base case, option C3 has an annual saving of 1.13 M€.
117
Figure 6.4 Comparison of levelise d costs of different compression options
6.3.3 Comparison with other studies in the literature
There is a little published literature about the cost estimate of CO 2 compression. IEAGHG
(2002) proposed an equation for the calculation of capital cost of compression based on the
power required. Ogden et al. (2004) developed a correlation summarizing the data from
Carbon Capture Project (CCP). The annual O&M cost was calculated by applying a factor of
0.04 to the total capital cost . Wong (2005) reported that the typical levelise d cost of CO 2
compression varies from 5.5 € to 7.4 € per ton of CO 2 with an estimated capital cost of 4.12
M€ per 3000HP in average. The method for O&M calculation was not mentio ned in the
paper. McCollum and Ogden (2006) studied the cost of the compression train with 5 -stage
compression follow ed by pumping and the O&M factor is also 0.04.
For the energy and utilities cost, it is generally accepted that it can be accurately calculated
based on the consumption data of process simulation results. So it was not included in the
comparison. Figure 6.5 shows the comparison of levelise d capital cost and O&M cost of
different cost models used in IEAGHG (2002) , McCollum and Ogden (2006) and this study.
The method used by McCollum and Ogden (2006) failed to distinguish the costs of different
options as a flow -based equation was applied for the capital cost calculation. The comparison
shows the O&M cost in this study is much higher than in other two methods.
0.002.004.006.008.0010.0012.0014.0016.00
Base case C1 C2 C3 C4Levelized costs ( €/t-CO2)
OptionsLevelized Energy and utilities cost
Levelized O&M cost
Levelized capital cost
118
Figure 6.5 The co mparison of levelise d cost of different cost model
6.4 Techno -economic evaluation of trunk pipelines
In Section 6.1, the diameters of the onshore and offshore trunk pipelines were selected with a
velocity -based equation for the base case. In this section, different published pipeline
diameter models were used for diameter calculation and different results were obtained.
Steady -state simulations were conducted to do a rating calculation for different diameters in
order to compare process performance and econ omic evaluation. The model used in this
section only includes the trunk pipelines and the booster pump station. The same entry
conditions were used for each simulation model ( see Table 6.8). The results of different
models were compared and the optimal diameter was chosen for the optimal design.
Table 6.8 Input and boundary conditions
Condition unit Value
Composition of CO 2 mixture stream mole% 96%CO 2, 2%N 2, 1.41%H 2,
0.59%Ar
Flow rate t/h 1126.263
Entry pressure bar 136
Entry temperature oC 20
Minimum arrival pressure at offshore
platform bar 126
012345
Base
caseC1 C2 C3 C4Levelized cost ( €/t-CO 2)
McCollum and Ogden (2006)012345
Base
caseC1 C2 C3 C4Levelized cost ( €/t-CO 2)
IEA GHG (2002)012345
Base
caseC1 C2 C3 C4Levelized cost ( €/t-CO 2)
This studyLevelized O&M cost
Levelized capital cost
119
6.4.1 Calculation of pipeline diameter
The diameter is a key factor for both technical and economical assessments in designing a
pipeline system. For a given CO 2 pipeline transport task, several published models can be
used to calculate the diameter of the pipelines. Table 6.9 shows an overview about the
equations of several models. The velocity based equation is often used to do an initial
estimation by setting input velocity in an experienced economical r ange. The (extensive)
hydraulic equation is only capable for the fluid transport. McCoy and Rubin model can be
used for both gaseous and liquid phase transport because it integrates the equation of state of
real gas with the energy conservation and hydraul ic equations.
For the diameter calculation, the parameters of the CO 2 mixture stream were obtained from
the process simulation results and are substitute d into each equation. Table 6.10 presents the
results of calculated diameter of trunk pipelines. For the velocity based method, 1.0m/s, 1.5
m/s and 2.0m/s were selected for t he diameter calculation. The results show the velocity range
of other three methods is from 1.3 to1.8 m/s, which is close to the most effective velocity
range of 1.5 to 2.0 m/s (Pershad et al., 2010) . As only standard size pipeline diameters (ANSI
standard) are specifie d in APEA, the calculated diameters were rounded off to the nearest
whole number. With a diameter of 20 inches, the exit pressure of the onshore trunk pipeline is
below 101 bar, which does not meet the operational constraint. The diameters of 22, 24 and
28 inches were then selected for the next techno -economic evaluations.
120
Table 6.9 Overview of the different diameter calculation methods in literature
Name Formula Abbreviations Limitation Source
Velocity
based
equation
=diameter (m), =mass flow
(kg/s), =velocity (m/s), =density
(kg/m3) Wildenborg et al.
(2004),
Element Energy
(2010),
Chandel et al. (2010)
Hydraulic
equation
=Fanning friction factor, =length
(m), =overall pressure drop (Pa) Re < 2000 or
Re>4000 Heddle et al. (2003) ,
Van den Broek et al.
(2010)
Extensive
hydraulic
equation
=Manning friction factor,
=height diffirence (m), =gravity
constant (9.81m/s2) originally
developed for
open channel
flow Piessens et al. (2008)
McCoy and
Rubin
model
=Average fluid compressibility,
=Gas constant (8.31Pa M3/mol K),
=average fluid temperature (K),
=molecular weight of flow
(kg/kmol), =Pressure at inlet (Pa),
=Pressure at outlet ( Pa), =
Average pressure in the pipeline=
Re < 2000 or
Re>4000 McCoy and Rubin
(2008),
Gao et al. (2011)
121 Table 6.10 The calculation results of different diameter models
Item Calculated
diameter
(m) Velocity
(m/s) Selected diameter
in APEA
(inch)
Velocity based equation 0.699 1.0 28
0.5713 1.5 24
0.4948 2 20
Hydraulic equation 0.5262 1.77 22
Extensive hydraulic equation 0.6173 1.29 24
McCoy and Rubin model 0.5672 1.52 22
6.4.2 Results and analysis
The selected diameters were used as the inputs in steady state models in order to simulate the
hydraulic performance of the pipeline. The results of each simulation were exported into
APEA to do the economic evaluations. Table 6.11 shows hydraulic results and power
requirement of each simulation. Higher velocity results in a greater pressure drop of the CO 2-
rich stream in the onshore and offshore trunk pipelines. Higher boosting pressure of the pump
station is then needed to compensate the pressure loss to maintain a constant arrive pressure at
the offshore storage platform.
Table 6.11 Technical performance of trunk pipeline s system in different diameters
Pipeline
diameter
(inch) Actual
initial
velocity
(m/s) Pressure drop
of onshore
pipeline
(bar) Pressure drop
of offshore
pipeline
(bar) Boosting
pressure of
pump station
(bar) Energy
required of
pump station
(kWh)
28 1.08 5.9 10.0 5.9 301.5
24 1.49 13.5 20.6 24.1 1243
22 1.81 22.1 32.2 44.3 2305
Figure 6.6 illustrates the comparison of the levelise d cost in breakdown of three options with
different diameters. The comparison shows that the saving of capital cost is much bigger than
the penalty of energy cost when the diameter of the pipelines decreases from 28 inches to 24
inches and then to 22 inches. The option with 22 -inch diameter has the lowest total levelise d
122 cost of 7.59 €/t on CO 2. Compared with the option of 24 -inch diameter in the base case, the
option with 22 -inch diameter has an annual saving of 7.34 M€.
Figure 6.6 Annual cost comparison for different diameters of the pipelines
6.4.3 Comparison with other studies in the literature
There are some models for cost evaluation of the CO 2 pipelines. Some of cost evaluation
methods do not include a cost assessment of the booster pump. None of them can make an
economic evaluation of the pipelines integrated with the energy cost of booster pump statio n.
For the comparison, the capital costs of trunk pipelines were calculated respectively by
different methods developed by IEAGHG (2002) , McCollum and Ogden (2006) , Piessens et
al. (2008) and Van den Broek et al. (2010) . Figure 6.7 shows a large range of the capital cost
per kilometre of pipeline for different cost models. The total capital cost calculated in this
study and Piessens et al. (2008) is much higher than those calculated wit h the other models.
One main reason is that the method used by Piessens et al. (2008) is a weight -based model
while the other methods are mainly based on the historical cost data of natural gas pipelines.
Thos e correlation models, except for the weight -based models, do not consider the adaptation
for the higher operation pressure of CO 2 pipeline transport. Normally, higher design pressure
requires higher wall thickness of the pipelines, which results in a signi ficant increase of the
material cost.
0.002.004.006.008.0010.0012.0014.00
28 in. 24 in. 22 in.Levelized costs ( €/t-CO 2)
Options in different pipeline diamterLevelized energy cost
Levelized O&M cost
Levelized capital cost
123
Figure 6.7 Comparison of capital cost of different cost models
6.5 Overall cost of CO 2 transportation pipeline network
6.5.1 Comparison of the base case and the optimal case
In Sections 6.3 and 6.4, techno -economic evaluations were conducted for the compressors
and trunk pipelines respectively. The options, which have the lowest annual costs, were used
to optimise the design of the pipeline network in this study. For the compressi on train at the
White Rose plant, 6 -stage compression followed pumping and 5 intercoolers with 20 oC exit
temperature is the optimal option. The compression train for the Don Valley plant applied the
similar configuration but it includes two parts, 5 -stage compression and 1 -stage compression
as the CO 2 mixture will be transported in the gaseous phase at a pressure of 35 bar first and
then boosted to dense phase at a pressure of 136 bar before entering the trunk pipelines. For
the trunk onshore and offshore pipelines, 22 -inch diameter is the optimal option. The overall
cost of the base case (conceptual design provided by Nati onal Grid) and the optimal case
were compared as shown in Figure 6.8. The total capital cost was split into the costs of trunk
pipeline and collecting system for a better comparison. The collecting system includes the
collecting pipelines and compression trains.
00.511.522.533.544.5
22 in. 24 in. 28 in.Capital cost (M €/km)
Pipeline diamterThis study
IEA GHG, 2002
MaCollum and Ogden, 2006
McCoy and Rubin, 2008
Piessense et al., 2008
Van Den Broek et al., 2010
124
Figure 6.8 Comparison of annual costs of base case and optimal case
The comparison shows the annual O&M cost is almost same for two cases as there are similar
processes. The annual energy and utilities cost is also very close to each other. Table 6.7 in
Section 6.3 shows a significant compression energy saving for the opti mal case compared to
the base case. However the smaller diameter of trunk pipelines in optimal case increased the
frictional pressure drop along the pipelines. This requires higher boosting pressure and
therefore, higher energy consumption of the booster p ump. The annual capital cost of trunk
pipelines of the optimal case is obviously lower than the base case. Smaller diameter of
pipelines has the advantage of incurring lower material cost and the construction cost may
also be lower. Compared to the base ca se, optimal case has an annual total saving of 22.7 M€.
It should be noticed that, having a larger diameter trunk pipelines, the base case provides the
opportunities to transport extra CO 2 mixture from additional electricity generation capture
plants or in dustrial capture plants in the future.
6.5.2 Comparison with other studies in the literature
Public data are scarce for a cost comparison about the whole pipeline network for CO 2
transport. Most published studies present the costs evaluation for the pipelines wi thout
including a cost assessment of the compression train. The few studies that carried out an
evaluation of the compression train failed to link it to the whole pipeline network system. In
the study of Roussanaly et al. (2013) , the economic evaluations were conducted to compare
0.0050.00100.00150.00200.00250.00
Base case Optimal caseAnnual cost (M €/a) Annual capital cost of trunk
pipelines
Annual capital cost of
collecting system
Annual energy and utilities
cost
Annual O&M cost
125 different options for the COCATE project. The cost evaluation of the onshore pipelines
option presented a typical pip eline network comprising a collecting pipeline system
(including compression) around 40 km long and an onshore trunk pipelines around 620 km
long.
The levelise d costs per ton of CO 2 were summed up in each of the studies as shown in Figure
6.9. The levelise d energy and utilities cost is close for these two studies. The levelise d capital
cost of trunk pipelines for the COCATE project is about 5.5 €/ton CO 2 , much lower than 8.1
€/ton CO 2 of the optimal case in this study, despite the fact that length of the COCATE
pipeline is 620 km while the length of pipeline used in this study is 162 km. The evaluations
of COCATE project used a spe cific pipeline cost model based on pipeline data of several
published cost models. The reason for low capital costs predicted by most of the published
models was analyse d in Section 6.4. The levelise d capital cost of collecting system in
COCATE project is only 0.2 €/ton CO 2. The details of the evaluation method used for the
collecting pipeline system in the COCATE project were not reported in the paper.
Figure 6.9 Comparison of levelise d cost of the optimal case and COCATE project
6.6 Concluding remarks
The aim of this chapter is to conduct simulation -based techno -economic evaluations for the
optimal design of the CO 2 transport pipeline network. A detailed s teady state model was
0.005.0010.0015.0020.0025.00
Optimal case COCATE projectLevelized cost ( €/t-CO 2) Levelized capital cost of
trunk pipelines
Levelized capital cost of
collecting system
Levelized energy and
utilities cost
Levelized O&M cost
126 developed, including CO 2 mixture streams from two emitters, the compression train, the
onshore and offshore trunk pipelines and the booster pump station. The simulation results
were exported in to APEA for conducting the techno -economic evaluations. The optimal
options with the lowest annual cost for compression train and trunk pipelines were selected
after a comparative study of the different economic evaluation results. The overall costs of
base case and optimal case were also compared. The optimal case has an annual total saving
of 22.7 M€. For the optimal case, levelised energy and utilities cost is 7.62 €/ ton CO 2,
levelised capital cost of trunk pipeline is about 8.11 €/ ton CO 2 and levelise d capital cost of
collecting system is 2.62 €/ ton CO 2. The cost evaluation results of the compression train,
trunk pipeline and whole pipeline network were compared with the cost evaluation results in
the literature respectively to gain more insights , as fol lows.
For CO 2 compression, the lower intercooler exit temperature (20 oC vs. 38 oC in this
study) and lower pressure ratio per stage leads to lower energy and utilities consumption of
compression train.
The correlation cost models for CO 2 compression train cannot give good cost
predictions for some different configuration options. The O&M factor of 0.04 in those models
is very small comparing with the result of this study.
The pipeline diameter models in the literature are generally relia ble. Among these
models, the hydraulic equation method gives the most accurate predictions. The initial
velocity of CO 2 mixture is around 1.7m/s in the optimal case in this study.
A large range of capital cost was obtained after applying different publishe d cost
models for the trunk pipelines. Most of the pipeline cost models in the literature predicted a
much lower capital cost and the weight -based model in the study of Piessens et al. (2008) has
the best predict ion compared with the results in this study.
Simulation -based techno -economics evaluation method offers a powerful tool for
optimal designs for the projects, especially for the decision making support about the detailed
technical options selection.
127 Chapter 7: Optimal operation under different
market conditions based on whole CCS chain
consideration
This chapter aims to explore the optimal operation under different market conditions for an
assumed existing NGCC power plant integrated with whole CCS chain includi ng solvent –
based PCC process, CO 2 compression, CO 2 transport and storage. Two major question s were
answered: (1) what is the optimal carbon capture degree under different market situations ? (2)
what are the optimal values of key operational variables for an optimal capture level ?
Compared with the optimisation model for solvent -based PCC process in Chapter 4, the
model in this chapter was updated. Firstly, the objective function of optimisation is changed
to levelise d cost of electricity (LCOE), which could direct ly reflect the changes of electricity
cost. Secondly, specific values were considered for two key operational variables, capture
level ( CL) and lean load ing, although they are continuous in real process. Economic
evaluation was carried out for the base case of the integrated system including CO 2 emission
penalty cost and CO 2 T&S cost. The optimal operations were investigated for the carbon
capture level under different carbon price (IPCC, 2007) , fuel price and CO 2 T&S price.
7.1 Optimisation methodology update
The optimisation algorithm could also be re presented by Equation s (4.4–4.10) in Chapter 4.
But several updates have been made to reflect the natures of optimal operation of the whole
integrated system in this chapter, compared with optimal design for PPC process only in
Chapter 4. The optimal operation study was conducted for an assumed existing plant whose
design features such as the equipment sizes sho uld be fixed during the study. Details are
presented in the follow ing sections.
7.1.1 Objective functi on
In Chapter 4, several potential objective functions were discussed and CCA was finally
chosen. However, in the optimisation studies in this chapter, LCOE was formulated to be the
objective function of the optimisation . LCOE is one indicator normally used to directly
present the electricity cost in the context of whole CCS chain consideration. LCOE was
calculated by dividing total annual cost by annual net power output in Equation ( 7.1). The
total annual cost is a sum o f annualized CAPEX , FOPEX and VOPEX as in Equation ( 7.2).
128
(7.1)
(7.2)
(7.3)
In this chapter, the integrated system in the study scope includes the NGCC power plant,
PCC process and compression process. Correspondingly, CAPEX and FOPEX in this chapter
present the cost for the above processes. At the same time , the electricity is supplied by the
power plant and low pressure steam is also extracted from the power plant. Consequently, the
costs of power electricity and LP steam are replaced by the cost of fuel. So V OPEX in this
chapter includes the fuel cost, cooling utilities cost, solvent make -up cost, carbon emission
cost and CO 2 T&S cost, as presented in Equation ( 7.3). It would be noticed that this study
focuses on the optimal operation of NGCC power plant with the carbon capture process. Its
CAPEX and FOPEX are assumed to be fixed neglecting the tax and labour cost changes.
Only VOPEX was considered to vary in response to different market conditions .
7.1.2 CO 2 emission cost
In order to achieve the target of global climate control, carbon credit (EU, 2010) was set to
drive the actions of reducing CO 2 emission. Under this policy, there is a CO 2 emission cost
for plant operators if CO 2 emission is over the cap. The first part is the cost for buying carbon
credits thought the The EU Emissions Trading System (ETS) at one floating carbon price
determined by the market . If the CO 2 emission is still over the allowance, a noncompliance
cost will be charged as a penalty at a much higher carbon price. Current carbon price in
Europe is around €7/ton CO 2 (EEEAG, 2015) . However f uture carbon price will increase
with the time and could be highly uncertain (USDO E, 2010a) .
7.1.3 CO 2 T&S cost
CO 2 transport and storage are two important sections of whole CCS chain and are also cost –
intensive processes. Collecting CO 2 mixture from several emitters into trunk pipelines for
geologic storage is more cost -effective than the use of separate pipelines (IPCC, 2005;
Chandel et al., 2010) . Other companies may operate CO 2 transport and storage infrastructure
129 and charge the emitters for the CO 2 stream entering the network. One example is that
National Grid plc will construct and operate the CO 2 transport pipelines and the permanent
CO 2 undersea storage facilities at a North Se a site in the Yorkshire and Humber CCS Project
in the UK (National Grid, 2014) .
The previous predictions of the costs of CO 2 transport and storage are in a wide range with
high uncertainties. For th e pipeline transport cost, IPCC predicted to be 9.9–14.9 €/ton CO 2
(IPCC, 2005) . The study in Chapter 6 also estimated that transport cost is around €17/ton
CO 2. For the CO 2 storage cost, IPCC predicted it to be 0 –7.9 €/ton CO 2 for onshore storage
and 6–30.8 €/ton CO 2 for ocean storage (IPCC, 2005) . Department of Energy and Climate
Change (DECC) (2013) reported that the transport and storag e cost accounts for a big part of
the increment of LCOE. Under FID 2013, 2020 and 2028 CCS technology scenarios, the CO 2
T&S cost is 49.7, 19.2 and 4.5 €/MWh, which correspond to equivalent prices of 102.5, 39.54
and 9.32 €/ton CO 2.
7.1.4 Equality constraints
In this study, equality constraints such as the mass balances, reactions and phase balance were
formulated in the first principle process models built in Aspen Plus® described in Chapter 3
and Chapter 5 . For this optimal operation of an assumed existing plant, the design variables
such as diameters and packing heights of the absorber and the stripper would not change. In
this study, the values of key design variables can be seen in the tables for the base case in
Chapter 3 and Chapter 5 .
7.1.5 Inequality const raints
The inequality constraints are imposed in the form of upper bounds for product flow rates for
different cases. Those inequality constraints for controlled operational variables in this study
are listed in Equation s (7.4–7.7) considering the flexible operation range of packing towers
and other equipment.
(7.4)
(7.5)
130 (kg/kg) (7.6)
(7.7)
For the optimisation of such a large scale model for rate -based first principle PCC process
integrated with the NGCC power plant , high computational requirements and convergence
problems often occur although commercial software package AspenPlus® was used.
Compromising on those challenges, specific values were considered for two key operational
variables although they are continuous in real process. Their value sets were presented in
Equation ( 7.8) and ( 7.9) respectively.
(7.8)
(7.9)
7.2 Techno -economic evaluation of the base case
In this section, the technical performance was evaluated according to the process simulation
results. Then the cost of whole chain for capturing carbon from NGCC power plant was
evaluated for the base case by combining calculation results and the literatur e data, in order to
give a basis for the optimal operation study in Section 7.5.
The base case was set up based on the PCC process described in Chapter 5 with 90% carbon
capture level for the NGCC power plant with EGR. For the economic evaluation, CAPEX
and fixed OPEX were referred to published benchmark report (IEAGHG, 2012) . Variable
OPEX was summarized fr om each subcost calculated based on the simulation results from
process model. To harmonize results for comparison with other studies, follow ing
assumptions were made: 1) all costs are corrected to €2015 using the harmonised consumer
price index (HICP) in Europe zone; 2) the captured CO 2 mixture has no economic value; 3)
cooling water is sourced from a nearby body of water at the cost of pumping and operation of
a cooling tower. Other important cost inputs are provided in Table 7.1 with the costs given in
Euro.
131 Table 7.1 Key economic evaluation cost inputs
Description Value Reference
Carbon price (€/kg) 7.0 EEEAG (2015)
NG price (€/GJ) 6.58 Ycharts (2015)
MEA solvent price (€/ton) 1452 Alibaba (2016)
CO 2 T&S cost (€/ton) 39.54 DECC (2013)
Project economic life (yea r) 25
Table 7.2 shows the comparison of the results between the reference case of NGCC
standalone and the base case of NGCC integrated with whole CCS chain . In the base case, the
annualized CAPEX of PCC proc ess is close to the annualized CAPEX of NGCC power plant
and the variable OPEX accounts for 65% of the total annual cost. For the variable OPEX of
NGCC standalone, the fuel cost is the biggest part and carbon emission cost is the second
largest part. Howev er when NGCC is integrated with PCC process, the fixed OPEX increases
significantly because of new expense items such as CO 2 T&S cost and MEA solvent make -up
cost.
Table 7.2 Cost comparison
Description Unit NGCC
standalone Base case of NGCC
integrated with CCS
ACAPEX of NGCC* M€/year 44.23 41.26
ACAPEX of PCC M€/year – 23.91
FOPEX of NGCC* M€/year 8.34 7.78
FOPEX of PCC M€/year – 5.89
VOPEX Fuel cost M€/year 160.42 160.42
Carbon emission M€/year 9.70 0.970
CO 2 T&S cost M€/year – 51.03
Solvent make -up cost M€/year – 3.00
Refrigerant cost M€/year – 0.67
TAC M€/year 222.6 9 294.93
LCOE €/MWh 56.00 87.26
Note: * the cost refers to a benchmark report from IEAGHG (2012) .
132 7.3 Optimal operation
The economic evaluation of the base case in Section 7.2 shows high capital cost as well as
wide ranging operating cost occurring for carbon capture from the NGCC power plant. In this
section, optimisation study was carried to find the optimal carbon capture degree under
different market situations and the optimal values of key operational variables for this
optimal capture level , for an assumed existing NGCC power integrated with whole CCS
chain.
7.3.1 Optimal capture level under different carbon price
The economic performances with regard to LCOE were examin ed under different carbon
prices of €7, €50, €100 and €150 per ton of CO 2 in this study.
The results were summarized in Fig ures 7.1–7.4. Under low carbon price of €7/ ton CO 2
(Figure 7.1) , LCOE gets the minimum value of €78.28 /MWh with 60% CL at an optimal lean
loading of 0.26 mol CO 2/mol MEA. Figure 7.1 also shows LCOE increase obviously with
higher CL no matter what the lean loading would be. The trend indicates that the carbon
emission cost cannot justify the high operating cost of the PCC process under l ow carbon
price. The optimal operation in terms of minimum LCOE is to vent the flue gas to the
atmosphere through bypassing the PCC process. With higher carbon price of €50/ton CO 2,
the differences of LCOE of different CLs become smaller as indicated in Figure 7.2 . For the
scenario of carbon price of €100/ ton CO 2, the values of LCOE distribute in a very narrow
range ( Figure 7.3) which means the carbon emission penalty cost can just justify the extra
VOPEX for carbon capture. With high carbon price of €150/ ton CO 2, the optimal value of
LCOE of 90% CL and 95% is very close at a lean loading of 0.26 –0.28 mol CO 2/mol MEA
whilst LCOE is around €94.35 /MWh (Figure 7.4) .
133
Figure 7.1 LCOE of different capture level with carbon price of 7 €/ton CO 2
Figure 7.2 LCOE of different capture levels with carbon price of 50 €/ton CO 2
75.0080.0085.0090.0095.00100.00105.00110.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
75.0080.0085.0090.0095.00100.00105.00110.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
134
Figure 7.3 LCOE of different capture levels with carbon price of 100 €/ton CO 2
Figure 7.4 LCOE of different capture levels with carbon price of 150 €/ton CO 2
75.0080.0085.0090.0095.00100.00105.00110.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
75.0080.0085.0090.0095.00100.00105.00110.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
135 The optimal values for key operational variables at diffe rent capture levels were displayed in
Figure 7.5 . The economic range of the lean loading was found to be 0.26 –0.3 mol CO 2/mol
MEA for the capture level in a range from 60% to 95%. It is noticed that this result is
different with the optimal values such as 0.158 mol CO 2/ mol MEA in the study of Mores et
al. (2014) and 0.2 mol CO 2/ mol MEA in the study of Agbonghae et al. (2014) . The reason is
that the studies implemented optimisation studies for both design and operation. In this
situation, lower lean loading required smaller L/G rat io (kg/kg) which results in a reduction
of the required crossing section al area of the absorber. However the diameter of the absorber
was fixed in this study for optimal operation. Here, one practice could be obtained for the
optimal operating of an existi ng PCC plant, which is that increasing the lean loading until
reaching the maximum capacity of the absorber could reduce reboiler duty to achieve a lower
energy cost.
Figure 7.5 Optimal lean loading and L/G ratio for different capture levels
The trend of L/G ratio is different from the lean loading. Except for being relevant with the
difference between lean loading and rich loading, L/G ratio relies more on the rate of CO 2
captured. As shown in Figure 7.5 , L/G ratio increases as more solvent is required for
absorb ing more CO 2 at higher capture level. It is also noticed that the required L/G ratio for a
same capture level varies for different CO 2 concentration in the flue gas. The range of L/G
ratio in mass is from 0.5 to 1.5 for a NGCC power without EGR (4.04 mol% CO 2 content in
the flue gas) and it is from 1.2 to 2.2 for a NGCC with EGR (7.32 mol% CO 2 content in the
flue gas) in this study. As a comparison, it is from 2.0 to 5.0 for a subcritical coal -fired power
plant with PCC process (13.5 mol% CO 2 content in flue gas) (Agbonghae et al ., 2014) .
136
Figure 7.6 Optimal reboiler duty and specific duty for different capture levels
The speci fic duty was calculated from the reboiler duty divided by the mass flow rate of CO 2
captured. The range of specific duty is from 3.25 to 4.35 GJ/ton CO 2 for PCC process for gas –
fired power plant in previous studies (Agbonghae et al., 2014; Canepa and Wang, 2015;
Mores et al., 2014; Sipöcz and Tobiesen, 2012) . Figure 7.6 presented the speci fic duty is from
4.05 to 4.32 GJ/ton CO 2 while the reboiler duty increases greatly when the capture level
increase from 60% to 95%.
Figure 7.7 Thermal efficiency of the NGCC with PCC at different capture levels
Figure 7.7 gives the trend of thermal efficiency of the NGCC with PCC at different capture
levels, which is easy to be justified because more steam would be extracted from the steam
system of the NGCC power plant for providing heat to the stripper reboiler of the PCC
process at the higher capture levels.
100120140160180200220
60% 70% 80% 85% 90% 95%44.054.14.154.24.254.34.35Specific duty
(GJ/ton CO 2)
Specific duty
Reboiler duty
Reboiler duty
(MW th)
137 7.3.2 The effect of NG price
In Section 7.2, the economic evaluation results show fuel cost is the largest part of variable
OPEX and is a huge expense even compared with annualized CAPEX. It is realized that the
uncertain NG price would have big impact to decide the optimal operation strategy.
Figure 7.8 LCOE of different capture level with carbon price of €100/ton CO 2 and NG price
of €2/GJ
Figure 7.9 LCOE of different capture level with carbon price of €100/ton CO 2 and NG price
of €6.58/GJ
55.0056.0057.0058.0059.0060.0061.0062.0063.0064.0065.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
90.0091.0092.0093.0094.0095.0096.0097.0098.0099.00100.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
138
Figure 7.10 LCOE of different capture level with carbon price of €100/ton CO 2 and N G price
of €12/GJ
Figures 7.8–7.10 shows the results of the optimal capture level under different fuel prices
with fixed carbon price of €100/ton CO 2. At the scenario of low NG price at €2/GJ ( Figure
7.8), the higher capture level shows a low LCOE because the CO 2 emission penalty can
easily justify the fuel cost. The situation reverses when NG price rises up to €12/GJ (Figure
7.10). Thus a carbon price higher than €100/ton CO 2 is required to draw the balance back for
carbon capture.
Figure 7.11 presents the required carbon price for driving the capture level to 90% in
response to the changes of fuel price. The result shows a range of LCOE is 57.27–131.2
€/MWh when the NG price rises from €2/GJ to €12/GJ. For the based case in Section 7.2
with 90% capture level, the required carbon price is around €101.50 /ton CO 2 with a LCOE of
€92.09 /MWh.
125.00127.00129.00131.00133.00135.00137.00139.00141.00143.00145.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
139
Figure 7.11 Required carbon price for driving 90% capture level in response to different fuel
prices
7.3.3 The effect of CO 2 T&S price
The change of the CO 2 T&S price may affect the optimal operation decision largely. In this
section, the optimisations were carried out on three different CO 2 T&S equivalent prices of
102.5, 39.54 and 9.32 €/ton CO 2.
Figure 7.12 LCOE of different capture level with carbon price of €100/ton CO 2 and T&S
price of €9.32/ton CO 2
5060708090100110120130140
2 4 6 8 10 1260708090100110120130140
LCOE (€/MWh)
Fuel Price ( €/GJ)Required carbon price for 90% CL
(€/ton CO2)
Carbon price
LCOE
75.0077.0079.0081.0083.0085.0087.0089.0091.0093.0095.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
140
Figure 7.13 LCOE of different capture level with carbon price of €100/ton CO 2 and T&S
price of €39.54/ton CO 2
Figure 7.14 LCOE of different capture level with carbon price of €100/ton CO 2 and T&S
price of €102.5/ton C O2
The results were displayed in Figure s 7.12–7.14. With low CO 2 T&S price of €9.32/ton CO 2,
the optimal capture level is 90 –95% compared to 80–90% at the intermediate price of
€39.54/ton CO 2. At the high CO 2 T&S price of is €102.5/ton CO 2, the high cost of carbon
capture cannot be justified ( see Figure 7.14 ) and a carbon price higher than €100/ton CO 2 is
90.0091.0092.0093.0094.0095.0096.0097.0098.0099.00100.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
100.00103.00106.00109.00112.00115.00118.00121.00124.00127.00130.00
0.2 0.24 0.26 0.28 0.3 0.32 0.36LCOE (EUR/MWh)
Lean Loading (mol CO2/mol MEA)95% CL
90% CL
85% CL
80% CL
70% CL
60% CL
141 needed to provide driving force for carbon capture. Otherwise bypassing PCC process is the
optimal choice .
Figure 7.15 presents the required carbon price for driving the capture level to 90% in
response to changes of CO 2 T&S price. The result shows a range of LCOE is 73.99–117.33
€/MWh when the CO 2 T&S price rises from 0 to €100/ton CO 2. When the CO 2 T&S cost is 0,
the carbon price is required to be €51.5/ton CO 2 for 90% capture level €54/ton CO 2 for the
case without considering the CO 2 compression, transport and storage in the study by Mac
Dowell and Shah (2013). Comparing the results of Figure s 7.11 and 7.15, it is noticed that
CO 2 T&S price has a lower sensitivity than fuel price to LCOE at 90% capture level.
Figure 7.15 Carbon price for driving 90% capture level in response to different CO 2 T&S
price
7.4 Concluding remarks
In this thesis , the optimal operation of NGCC power plant integrated with PCC process was
investigated under different market conditions such as different carbon price, fuel price and
CO 2 T&S cost. The objective function to be minimized in the optimisation is the LCOE
obtained by dividing total annual cost by annual net power output. The economic estimate
was carried out for the reference case of NGCC standalone and the base case of NGCC
integrated with CCS chain . With the deployment of carbon capture at 90% captur e level,
LCOE increase s from €56.00 /MWh to €87.26 /MWh. It is also found that fuel cost, carbon
emission cost and CO 2 T&S cost are major parts of VOPEX.
60708090100110120130140
010 20 30 40 50 60 70 80 90100406080100120140160180200
LCOE (€/MWh)
CO2T&S Price ( €/ton CO2)Required carbon price for 90% CL
(€/ton CO2)
Carbon price
LCOE
142 For an assumed existing 453 M We NGCC power plant with CCS whole system, current low
carbon price of €7/ton CO 2 is not able to drive power generators to run carbon capture
process because the carbon emission cost cannot is much lower than the operating cost of
CCS based on the results from this study . During UNFCCC Paris Conference in 2015, EU
stated a cut of at least 40% of greenhouse emission in its 2030 climate & energy framework
and a further cut of 80% by 2050 (EU, 2016a) . It could be predicted that the carbon price
will increase when the cap of the total amount of CO 2 emission is reduced over time to
achieve the GHG emission controlling target (EU, 2010) . Predicted by this study, c arbon
price needs to be up to around €100–120/ton CO 2 to dri ve carbon capt ure level to 90% , which
is similar with the penalty carbon price for noncompliance emission of €100/ton CO 2 in the
phase 2 of EU ETS from 2008 to 2012 (EU, 2016c) . From this study, it is also found that the
required carbon price would be affected by the fuel price and CO 2 T&S price.
In a summary, th is study indicates the coactions of carbon price, fuel price and CO 2 T&S
price will significantly affect the decision making about the optimal capture level for
operating carbon capture process for a NGCC power plant.
143 Chapter 8: Conclusions and recommendations for
future research
8.1 Conclusions
This thesis presents the studies on optimal design and operation of MEA -based PCC process
and the integrated system with NGCC power plant through process modelling, simulation and
optimisation, aiming to reduce the cost of PCC process commercial deployment. As the base
of optimisation studies, both process model and cost model were developed with high
accuracy.
In C hapter 3, the process model was developed and validated at different stages . The
validation show ed prediction s of thermod ynamic model developed in this study is better than
other three correlation combinations (Austgen et al., 1989; Liu et al., 1999; Zhang et al.,
2011) . It also indicates that the uncertainties of those correlations should be carefully
considered when they are used in process modelling and simulation. In order to improve the
accuracy of the model predictions, two correla tions were updated by coding Fortran
subroutines, including Han and Eimer (2012) model for liquid mixture density and Tsai et al.
(2011) effective gas liquid interfacial area . For kinetics -controlled reactions, different values
were set for kinetics of the reverse carbonate formation reactions happening in the absorber
and the str ipper respectively, which improves the accuracy of the process model. The process
model was then validated with comprehensive pilot plant experimental data, in terms of the
absorption efficiency and thermal performance of the integrated system. The compari son
results show that model predictions are in very good agreement s with the experimental data,
which ensure s that the process model has good accuracy for the optimisation studies in the
following chapters.
In Chapter 4, the cost model was developed in Fortran and then integrated into the process
model by coding Fortran subroutine in Aspen Plus®. Using this newly developed model, the
optimisation studies were carried out for the PCC process and on the impacts of variation s of
the key variables. The optim isation results show the cost of CO 2 avoided in the optimal case
is 69.13 €/ton CO 2, which is about 18.4% lower than the base case which have a value of
86.85 €/ton CO 2. Findings from case studies on cost optimisation in response to variations in
several k ey variables include:
144 The optimal rich loadings should be saturated CO 2 loadings under the temperature,
pressure and composition conditions.
The range of optimal lean loading in these three case studies is 0.275 –0.331 mol
CO 2/mol MEA.
The optimal packing height of the stripper significantly depends on the solvent
regeneration degree.
The reduction of CCA is more significant when MEA concentration in solvent s
increases from 20 wt% to 30 wt% than increas ing from 35 wt% to 40 wt%.
For scale -up of the optimal design , size effect impacts not only economic terms but
also process parameters. New optimisation should be carried out for each single case
to obtain optimal values of both the equipment sizes and key operational variables .
Chapter 5 presented the investigation on thermal performances of different integration
options of a 453MW e NGCC equipped with a PCC process and CO 2 compression train.
Employing EGR achieved s ignificant saving with the contributions from both the CAPEX
and VOPEX. The CCA of the case with EGR decrease d from 69.13 €/ton CO 2 to 66.14 €/ton
CO 2 about 5 .12% compared with the case without EGR . The thermal efficiency (LHV) of the
NGCC power plant deceases from 58.74% to 49.16% when i ntegrated with the PCC process
and the compression train. This reduction includes 7.40% -points decrease due to steam
extraction, 0.55% -points reduction due to PCC power consumption and 1.92% -points
reduction due to compression train power consumption. With the application of EGR in the
NGCC pow er plant at a recirculation ratio of 0.38, the net thermal efficiency increases
0.77% -points while the cross -section area s of the absorber and stripper in the carbon capture
process reduced by 37.39% and 9.36% respectively .
The compression heat integration option s have been analysed by applying supersonic shock
wave compression technology . Compression heat integration into the steam cycle of HRSG
and stripper reboiler achieves 0.32% -points and 0.54% -points net efficiency improvement
separately without major capital investment required. The study indicate s that EGR
technology, supersonic shock wave compression technology and compression heat
integration s could be future direction s for commercial PCC deployment in NGCC power
plants.
In Chapter 6, a detailed steady state model was developed for transport system comprising
CO 2 mixture streams from two emitters, the compression train, the onshore and offshore trunk
145 pipelines and the booster pump station. The overall costs of the base case and the optimal case
were also compared. The optimal case has an annual total saving of 22.7 M€. The cost
evaluation results of the compression train, trunk pipeline and whole pipeline network were
compared with the cost evaluation results in open literature respe ctively to gain more insights.
For CO 2 compression, lower intercooler exit temperature (20 oC vs. 38 oC in this
study) and lower pressure ratio per stage leads to lower energy and utilities consumption of
compression train.
The correlation cost models for CO 2 compression train do not give good cost
predictions for some configuration options. The O&M factor of 0.04 in those models is very
small compared to the result of this study.
The pipeline diameter models in literature are generally reliable. Among th ese models,
the hydraulic equation method gives the most accurate prediction. The initial velocity of CO 2
mixture is around 1.7m/s in the optimal case in this study.
Large range of capital cost was obtained after applying different published cost
models fo r the trunk pipelines. Most of the pipeline cost models in literature predicted much
lower capital cost and the weight -based model in the study of Piessens et al. (2008) has the
best predi ction compar ed to the results in this study.
Simulation -based techno -economic evaluation method offers a powerful tool for
optimal designs for the projects, especially for the decision making support on the detailed
technical options selection.
In Chapter 7, the optimal operation of NGCC power plant integrated with whole CCS chain
was investigated under different market conditions such as carbon price, fuel price and CO 2
T&S cost. The objective function to be minimized in the optimisation is the LCOE obtained
by dividing total annual cost by annual net power output. The economic estimate was carried
out for the reference case of NGCC standalone and the base case of NGCC integrated w ith
CCS chain. With the deployment of carbon capture at 90% capture level, LCOE increases
from €56.00 /MWh to €87.26 /MWh. It is also found that fuel cost, carbon emission cost and
CO 2 T&S cost are major parts of VOPEX.
For an assumed existing 453 MW e NGCC p ower plant with CCS whole system, current low
carbon price of €7/ton CO 2 is not able to drive power generators to run carbon capture
146 process because the carbon emission penalty do not justify the operating cost of CCS. Carbon
price needs to increase to around €100/ton CO 2 to justify the cost of carbon capture and needs
to increase to around €120/ton CO 2 to drive carbon capture level to 90%. An economic range
of lean loading is 0.26 –0.3 mol CO 2/mol MEA for the capture levels from 60% to 95%. The
fuel price and the CO 2 T&S price have great impact on optimal operation results and the cost
of electricity. In a summary, this study indicates the coactions of carbon price, fuel price and
CO 2 T&S price will significantly affect the decision making about the o ptimal capture level
for operating carbon capture process for a n NGCC power plant.
8.2 Recommendations for future research
During the thermodynamic modelling in Chapter 3 , it is found that, even for MEA , the most
commonly used solvent, the experimental data are not available to cover the full range of
temperature, pressure and composition conditions. Thus the correlations regressed from this
data for some typical published models have some uncertaint ies. Thus two recommendations
are (1) new experimental measurements of VLE and physical properties of MEA -H2O-CO 2 to
cover the full range of temperature, pressure and composition conditions, (2) the correlations
in thermodynamic model and physical property calculation needs to be improved with new
experimental data.
One challenge of the industrial deployment of solvent -based PCC process is the large sizes of
the equipment. But there is a big gap in terms of the equipment sizes between the pilot plants
and industrial scale plants. For example, the diameter of the absorber is around 20 m at the
industrial scale and is only 0.427 m at the pilot plant. This gap would cause big uncertainty
for the scale -up study of the process . As a result, comprehensive running data of existing
demonstration plants (e.g. Boundary Dam CCS, Canada (SASKPOWER, 2015) ) should be
share d with the research communities to promote studies in this field . It is also important that
the rese arche rs at industrial scale should seek more collaboration with relevant stakeholders
including power plant operators, key equipment suppliers such as column internals design
companies and compressors manufactories.
In terms of the whole CCS chain, the costs of CO 2 transport and storage in the publications
are found to be normally underestimated . One of the reasons is that most estimation models
are empirical correlations developed on historical cost data of natural gas pipeline projects
and do not refle ct the process conditions of different CO 2 pipeline projects. Furthermore,
current cost models (including this study) did not consider the costs related with safety issues,
147 such as monitoring and protection of facilities namely CO 2 pipelines located in dense
population areas, the CO 2 geological storage sites etc. Therefore, more extensive studies in
this field are required.
148 Appendix A: Publications from this thesis
Peer reviewed journal papers:
Luo, X., Wang, M., Oko, E., Okezue, C., 2014. Simulation –based techno –economic
evaluation for optimal design of CO 2 transport pipeline network. Applied Energy , 132, 610 –
620.
Luo, X., Wang, M., Chen, J., 2015. Heat integration of natural gas combined cycle power
plant integrated with post –combustion CO 2 capture and compression. Fuel, 151, 110 –117.
Luo, X. , Wang, M., 2016. Optimal operation of MEA -based post -combustion carbon capture
for natural gas combined cycle power plants under different market conditions. Inter national
Journal of Greenhouse Gas Control, 48, Part 2 , 312-320.
Peer reviewed conference papers:
Luo, X., Mistry, K., Okezue, C., Wang, M., Cooper, R., Oko, E., Field, J., 2014. Process
Simulation and Analysis for CO 2 Transport Pipeline Design and Operati on – Case Study for
the Humber Region in the UK, Computer Aided Chemical Engineering , pp. 1633 –1638.
Luo, X., Wang, M., 2015, Optimal operation of MEA –based post –combustion carbon capture
process for gas –fired CCGT power plants, In 7th International Exergy , Energy and
Environment Symposium (IEEES7): University of Valenciennes et du Hainaut –Cambrésis –
ENSIAME – Valenciennes – FRANCE . April 27 –30, 2015.
Peer reviewed book chapter:
Luo, X., Wang, M.,. Process Simulation and Integration of Natural Gas Combined Cycle
Power Plant Integrated with Chemical Absorption Carbon Capture and Compression, In: The
Water–Food–Energy Nexus: Processes, Technologies and Challenges , Edited by I.M.
Mujtaba, R. Srinivasan and N. O. Elbashir, In Press.
149 Reference
Abu-Zahra, M. R. M., Niederer, J. P. M., Feron, P. H. M. & Versteeg, G. F. (2007a) CO 2
capture from power plants: Part II. A parametric study of the economical performance
based on mono -ethanolami ne. International Journal of Greenhouse Gas Control ,
1(2), 135 -142.
Abu-Zahra, M. R. M., Schneiders, L. H. J., Niederer, J. P. M., Feron, P. H. M. & Versteeg, G.
F. (2007b) CO 2 capture from power plants Part I. A parametric study of the technical
performan ce based on monoethanolamine. International Journal of Greenhouse Gas
Control , 1(1), 37 -46.
Agbonghae, E. O., Hughes, K. J., Ingham, D. B., Ma, L. & Pourkashanian, M. (2014)
Optimal Process Design of Commercial -Scale Amine -Based CO 2 Capture Plants.
Industrial & Engineering Chemistry Research , 53(38), 14815 -14829.
Alibaba. (2016) RE: Monoethanolamine (MEA) price. Quotation to Luo, X.
Amrollahi, Z., Ystad, P. A. M., Ertesvåg, I. S. & Bolland, O. (2012) Optimized process
configurations of post -combustio n CO 2 capture for natural -gas-fired power plant –
Power plant efficiency analysis. International Journal of Greenhouse Gas Control ,
8(0), 1 -11.
Aresta, M., Dibenedetto, A. & Angelini, A. (2013) The changing paradigm in CO 2 utilization.
Journal of CO 2 Utilization , 3-4, 65 -73.
Arias, A. M., Mores, P. L., Scenna, N. J. & Mussati, S. F. (2016) Optimal design and
sensitivity analysis of post -combustion CO 2 capture process by chemical absorption
with amines. Journal of Cleaner Production , 115, 315 -331.
Aronu, U. E., Gondal, S., Hessen, E. T., Haug -Warberg, T., Hartono, A., Hoff, K. A. &
Svendsen, H. F. (2011) Solubility of CO 2 in 15, 30, 45 and 60 mass% MEA from 40
to 120°C and model representation using the extended UNIQUAC framework.
Chemical Engineering Science , 66(24), 6393 -6406.
Aspelund, A. & Jordal, K. (2007) Gas conditioning —The interface between CO 2 capture and
transport. International Journal of Greenhouse Gas Control , 1(3), 343 -354.
150 AspenTech. (2008a) Aspen Plus User Guide 10.2 . Burlington, MA: Aspen Technology, Lnc.
AspenTech. (2008b) Rate-based Model of the CO 2 Capture Process by MEA using Aspen
Plus. Burlington, MA Aspen Technology, Inc.
AspenTech. (2012a) Aspen Icarus Reference Guide: Icarus Evaluation Engine (IEE) V8.0 .
Burlington, USA: Aspen Tech nology, Inc.
AspenTech. (2012b) Aspen Physical Property System: Physical Property Methods .
Burlington, USA: Aspen Technology, Inc.
Austgen, D. M., Rochelle, G. T., Peng, X. & Chen, C. C. (1989) Model of vapor -liquid
equilibria for aqueous acid gas -alkanola mine systems using the electrolyte -NRTL
equation. Industrial & Engineering Chemistry Research , 28(7), 1060 -1073.
Bassily, A. M. (2007) Modeling, numerical optimization, and irreversibility reduction of a
triple -pressure reheat combined cycle. Energy , 32(5) , 778 -794.
Bates, R. G. & Pinching, G. D. (1951) Acidic dissociation constant and related
thermodynamic quantities for monoethanolammonium ion in water from 0°C to 50°C.
J. Res. Natl. Bur. Stand. , 46(5), 349 -352.
BBC. (2015) UK's coal plants to be phased o ut within 10 years [Online]. Available:
http://www.bbc.co.uk/news/business -34851718 [Accessed 6th June 2016].
BIGCO2. (2007) CO 2 Management Technologies for Future Power Generation [Online].
2007–2011: BIGCO2 Research & Development Platform. Available:
http://www.bigco2.com [Accessed 6th June 2016].
Biliyok, C., Canepa, R., Wang, M. & Yeung, H. (2013) Techno -Economic Analysis of a
Natural Gas Combined Cycle Power P lant with CO 2 Capture. In: Andrzej, K. & Ilkka,
T. (eds.) Computer Aided Chemical Engineering. Elsevier.
Biliyok, C. & Yeung, H. (2013) Evaluation of natural gas combined cycle power plant for
post-combustion CO 2 capture integration. International Journal of Greenhouse Gas
Control , 19(0), 396 -405.
Billet, R. & Schultes, M. (1993) Predicting mass transfer in packed columns. Chemical
Engineering and Technology , 16(1), 1 -9.
151 Blasing, T. J. (2013) Current Greenhouse Gas Concentrations [Online]. CDIAC Web site.
Available: http://cdiac.ornl.gov/pns/current_ghg.html [Accessed 6th June 2016].
Boggs, P. T. & Tolle, J. W. (2000) Sequential quadratic programming for large -scale
nonlinear optimization. Journal of C omputational and Applied Mathematics , 124(1–
2), 123 -137.
Bollas, G. M., Chen, C. C. & Barton, P. I. (2008) Refined electrolyte -NRTL model: Activity
coefficient expressions for application to multi -electrolyte systems. AIChE Journal ,
54(6), 1608 -1624.
Boot -Handford, M. E., Abanades, J. C., Anthony, E. J., Blunt, M. J., Brandani, S., Mac
Dowell, N., Fernandez, J. R., Ferrari, M. -C., Gross, R., Hallett, J. P., Haszeldine, R.
S., Heptonstall, P., Lyngfelt, A., Makuch, Z., Mangano, E., Porter, R. T. J.,
Pourkash anian, M., Rochelle, G. T., Shah, N., Yao, J. G. & Fennell, P. S. (2014)
Carbon capture and storage update. Energy & Environmental Science , 7(1), 130 -189.
Böttinger, W., Maiwald, M. & Hasse, H. (2008) Online NMR spectroscopic study of species
distribution in MEA -H2O-CO 2 and DEA -H2O-CO 2. Fluid Phase Equilibria , 263(2),
131-143.
BP. (2014) BP Statistical Review of World Energy 2014 [Online]. Available:
http://www.bp.com/content/dam/bp -country/de_de/PDFs/brochures/BP -statistical –
review -of-world -energy -2014 -full-report.pdf [Accessed 6th June 2016].
Bravo, J. L., Rocha, J. A. & Fair, J. R. (1985) Mass transfer in gauze packings. Hydrocarbon
Processing , 64(1), 91 -95.
Brelvi, S. W. & O'Connell, J. P. (1972) Corresponding states correlations for liquid
compressibility and partial molal volumes of gases at infinite dilution in liquids.
AIChE Journal , 18(6), 1239 -1243.
Buhre, B., Elliott, L., Sheng, C., Gupta, R. & Wall, T. (2005) Oxy -fuel combustion
technology for coal -fired power generation. Progress in energy and combustion
science , 31(4), 283 -307.
Buizert, C., Gkinis, V., Severinghaus, J. P., He, F., Lecavalier, B. S ., Kindler, P.,
Leuenberger, M., Carlson, A. E., Vinther, B., Masson -Delmotte, V., White, J. W. C.,
152 Liu, Z., Otto -Bliesner, B. & Brook, E. J. (2014) Greenland temperature response to
climate forcing during the last deglaciation. Science , 345(6201), 1177 -1180.
Canepa, R. & Wang, M. (2015) Techno -economic analysis of a CO 2 capture plant integrated
with a commercial scale combined cycle gas turbine (CCGT) power plant. Applied
Thermal Engineering , 74(0), 10 -19.
Canepa, R., Wang, M., Biliyok, C. & Satta, A. (201 3) Thermodynamic analysis of combined
cycle gas turbine power plant with post -combustion CO 2 capture and exhaust gas
recirculation. Proceedings of the Institution of Mechanical Engineers, Part E: Journal
of Process Mechanical Engineering , 227(2), 89 -105.
Carbon Capture Journal. (2013) Internals for carbon capture plants [Online]. Available:
http://www.carboncapturejournal.com/news/internals -for-carbon -capture –
plants/3349.aspx?Category=featured [Accessed 6th June 2016].
CASTOR. (2004) EU -Project CO 2, from Capture to Storage (CASTOR): integrated research
project under the 6th frame work programme of the European commission, . contract
nr. ses6 -ct-2004 -502586: project duration from 01.02.2004 to 31.01.2008.
CCP. (2000) CO 2 Capture Project (CCP): CCP1, CCP2, and CCP3. project duration from
2000 to 2013.
CCSA. (2014) Don Valley Power Plant and CCS Project [Online]. Available:
http://www.ccsassociation.org/about -us/our -members/sargas -power/ [Accessed 6th
June 2016].
CESAR. (2008) EU -Project CO 2 Enhanced Separation and Recovery (CESAR): integrated
research project under th e 7th frame work programme of the European commission.
project duration from 01.02.2008 to 31.01.2011: contract nr. 213569.
Chaczykowski, M. & Osiadacz, A. J. (2012) Dynamic simulation of pipelines containing
dense phase/supercritical CO 2-rich mixtures for carbon capture and storage.
International Journal of Greenhouse Gas Control , 9, 446 -456.
Chandel, M. K., Pratson, L. F. & Williams, E. (2010) Potential economies of scale in CO 2
transport through use of a trunk pipeline. Energy Conversion and Management ,
51(12), 2825 -2834.
153 Chapel, D. G., Mariz, C. L. & Ernest, J. (Year) Recovery of CO 2 from flue gases: commercial
trends. In: Canadian society of chemical engineers annual meeting, 1999.
Chapoy, A., Burgass, R., Tohidi, B., Michael Austell, J. & Eickhoff, C. (2011) Effect of
common impurities on the phase behavior of carbon -dioxide -rich systems:
Minimizing the risk of hydrate formation and two -phase flow. SPE Journal , 16(4),
921-930.
Chen, C. -C. & Evans, L. B. (1986) A local composition model for the excess G ibbs energy of
aqueous electrolyte systems. AIChE Journal , 32(3), 444 -454.
Chen, C. C., Boston, J. F., Evans, L. B. & Britt, H. I. (1982) Local composition model for
excess Gibbs energy of electrolyte systems. Part I: Single solvent, single completely
dissociated electrolyte systems. AIChE Journal , 28(4), 588 -596.
Chen, C. C., Britt, H. I., Boston, J. F. & Evans, L. B. (1979) Extension and application of the
pitzer equation for vapor ‐liquid equilibrium of aqueous electrolyte systems with
molecular solutes. AIChE Journal , 25(5), 82 0-831.
Cheng, S., Meisen, A. & Chakma, A. (1996) Predict amine solution properties accurately.
Hydrocarbon Processing , 75(2), 81 -84.
Chilton, T. H. & Colburn, A. P. (1934) Mass transfer (absorption) coefficients prediction
from data on heat transfer and fluid friction. Industrial & engineering chemistry ,
26(11), 1183 -1187.
Ciferno, J. P., Fout, T. E., Jones, A. P. & Murphy, J. T. (2009) Capturing carbon from
existing coal -fired power plants. Chemical Engineering Progress , 105(4), 33 -41.
CO2CRC. (2003) The Cooperative Research Centre for Greenhouse Gas Technologies
(CO2CRC). project duration from 2003 to 2015.
Cohen, S. M., Rochelle, G. T. & Webber, M. E. (2012) Optimizing post -combustion CO 2
capture in response to volatile electricity prices. International Journal of Greenhouse
Gas Control , 8(0), 180 -195.
CPL. (2013) About the White Rose CCS Project [Online]. Capture Power Limited. Available:
http://www.whiteroseccs.co.uk/about -white -rose [Accessed 6th June 2016].
154 Dahowski, R. T., Li, X., Davidson, C., Wei, N. & Dooley, J. (2009) Regional opportunities
for carbon dioxide capture and storage in China . Richland, WA, USA: Pacific
Northw est National Laboratory.
Danckwerts, P. (1951) Significance of liquid -film coefficients in gas absorption. Industrial &
Engineering Chemistry , 43(6), 1460 -1467.
Dang, H. & Rochelle, G. T. (2003) CO 2 absorption rate and solubility in
monoethanolamine/pipera zine/water. Separation Science and Technology , 38(2), 337 –
357.
Davis, J. & Rochelle, G. (2009) Thermal degradation of monoethanolamine at stripper
conditions. Energy Procedia , 1(1), 327 -333.
DECC. (2013) CCS Cost Reduction Task Force: final report [Online] . Available:
https://www.gov.uk/government/publications/ccs -cost-reduction -task-force -final-
report [Accessed 6th June 2016].
Diamantonis, N. I., Boulougouris, G. C., Mansoor, E., Tsangaris, D. M. & Economou, I. G.
(2013) Evaluation of Cubic, SAFT, and PC -SAFT Equations of State for the Vapor –
Liquid Equilibrium Modeling of CO2 Mixtures with Other Gases. Industrial &
Engineering C hemistry Research , 52(10), 3933 -3942.
Dillon, D., Panesar, R., Wall, R., Allam, R., White, V., Gibbins, J. & Haines, M. (Year) Oxy –
combustion processes for CO 2 capture from advanced supercritical PF and NGCC
power plant. In: 7 th International Conference on Greenhouse Gas Technologies,
Vancouver, Canada, 2004.
Dugas, R. E. (2006) Pilot plant study of carbon dioxide capture by aqueous
monoethanolamine. MSE Thesis, University of Texas at Austin .
E.ON. (2010) Equation of State Prediction of Carbon Dioxide Pro perties [Online].
Available:
http://webarchive.nationalarchives.gov.uk/20111209170 139/https:/www.decc.gov.uk/
en/content/cms/emissions/ccs/demo_prog/feed/e_on_feed_/transport/transport.aspx
[Accessed 6th June 2016].
155 Easterling, D. R. & Wehner, M. F. (2009) Is the climate warming or cooling? [Online].
Available: http://dx.doi.org/10.1029/2009GL037810 [Accessed 6th June 2016].
Edgar, T. F., Himmelblau, D. M. & Lasdon, L. S. (2001) Optimization of Chemical
Processes : McGraw -Hill.
Edwards, T., Maurer, G. , Newman, J. & Prausnitz, J. (1978) V apor‐liquid equilibria in
multicomponent aqueous solutions of volatile weak electrolytes. AIChE Journal ,
24(6), 966 -976.
EEEAG. (2015) EU Emission Allowances | Secondary Market [Online]. European Energy
Exchange AG. Available: https://www.eex.com/en/market -data/emission –
allowances/spot -market/european -emission -allowances#!/2016/06/06 [Accessed 6th
June 2016].
EIA. (2016) How much carbon dioxide is produced per kilowatthour when generating
electricity with fossil fuels? [Online]. Available:
https://www.eia.gov/tools/faqs/faq.cfm?id=74&t=11 [Accessed 6th June 20 16].
EU. (2010) EU ETS Handbook [Online]. Available:
http://ec.europa.eu/clima/publications/docs/ets_handbook_en.pdf [Accessed 6th June
2016].
EU. (2016a) Climate strategies & ta rgets [Online]. EU. Available:
http://ec.europa.eu/clima/policies/strategies/index_en.htm [Accessed 6th June 2016].
EU. (2016b) Electricity production and supply statistics [Online]. Available:
http://ec.europa.eu/eurostat/statistics –
explained/index.php/Electricity_production_and_suppl y_statistics#Further_Eurostat_i
nformation [Accessed 6th June 2016].
EU. (2016c) Phases 1 and 2 (2005 -2012) of The EU Emissions Trading System (ETS)
[Online]. Available: http://ec.eur opa.eu/clima/policies/ets/pre2013/index_en.htm
[Accessed 6th June 2016].
Faber, R., Köpcke, M., Biede, O., Knudsen, J. N. & Andersen, J. (2011) Open -loop step
responses for the MEA post -combustion capture process: Experimental results from
the Esbjerg pilo t plant. Energy Procedia , 4(0), 1427 -1434.
156 Fakouri Baygi, S. & Pahlavanzadeh, H. (2015) Application of the perturbed chain -SAFT
equation of state for modeling CO 2 solubility in aqueous monoethanolamine
solutions. Chemical Engineering Research and Design , 93, 789 -799.
Feng, Y. & Rangaiah, G. P. (2011) Evaluating capital cost estimation programs. Chemical
Engineering , 118(8), 22 -29.
Foster NR, B. K., Dehghani F. . (2010) Modeling of Phase Equilibria for Binary and Ternary
Mixture of Carbon Dioxide, Hydrogen and Methanol [Online]. Available:
http://webarchive.nationalarchives.gov.uk/201112 09170139/https:/www.decc.gov.uk/
en/content/cms/emissions/ccs/demo_prog/feed/e_on_feed_/transport/transport.aspx
[Accessed 6th June 2016].
Gao, L., Fang, M., Li, H. & Hetland, J. (2011) Cost analysis of CO 2 transportation: Case
study in China. Energy Proced ia, 4, 5974 -5981.
GCCSI. (2011) Economic assessment of carbon capture and storage technologies [Online].
GCCSI. Available: http://www.globalccsinstitute.com/publications/economic –
assessment -carbon -capture -and-storage -technologies -2011 -update [Accessed 2014 –
02-26].
GE Power. (2016) 9F.05 Gas Turbine (50 Hz). GE Power.
GerbilNow. (2012) Why choose between solar and coal when you can use both? [Online].
Available: http://blog.gerbilnow.com/2012/11/solar -coal-power.html [Accessed 6th
June 2016].
Gibbins, J., Crane, R., Lambropoulos, D., Booth, C., Roberts, C. & Lord, M. (2004)
Maximising the effectiveness of post -combustion CO 2 capture systems. Proceedings
of GHGT , 7, 5 -9.
Godoy, E., Benz, S. J. & Scenna, N. J. (2011) A strategy for the economic optimization of
combined cycle gas turbine power plants by taking ad vantage of useful
thermodynamic relationships. Applied Thermal Engineering , 31(5), 852 -871.
Godoy, E., Scenna, N. J. & Benz, S. J. (2010) Families of optimal thermodynamic solutions
for combined cycle gas turbine (CCGT) power plants. Applied Thermal Engine ering ,
30(6-7), 569 -576.
157 Gross, J. & Sadowski, G. (2001) Perturbed -chain SAFT: An equation of state based on a
perturbation theory for chain molecules. Industrial and Engineering Chemistry
Research , 40(4), 1244 -1260.
Gross, J. & Sadowski, G. (2002) Applica tion of the Perturbed -Chain SAFT Equation of State
to Associating Systems. Industrial & Engineering Chemistry Research , 41(22), 5510 –
5515.
Han, J., Jin, J., Eimer, D. A. & Melaaen, M. C. (2012) Density of Water (1) +
Monoethanolamine (2) + CO 2 (3) from (29 8.15 to 413.15) K and Surface Tension of
Water (1) + Monoethanolamine (2) from (303.15 to 333.15) K. Journal of Chemical
& Engineering Data , 57(4), 1095 -1103.
Harriott, P. (1989) Effect of liquid maldistribution on the performance of packed stripping
columns. Environmental Science & Technology , 23(3), 309 -314.
Harris, F., Kurnia, K. A., Mutalib, M. I. A. & Thanapalan, M. (2009) Solubilities of carbon
dioxide and densities of aqueous sodium glycinate solutions before and after CO
absorption. Journal of Chemical and Engineering Data , 54(1), 144 -147.
Hartmann, D., Klein Tank, A., Rusicucci, M., Alexander, L., Broenniman, B., Charabi, Y.,
Dentener, F., Dlugokencky, E ., Easterling, D. & Kaplan, A. (2013) Observations:
Atmosphere and Surface. In: Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change . Cambridge, Uni ted Kingdom and New
York, NY, USA: Cambridge University Press.
Hartono, A., Mba, E. O. & Svendsen, H. F. (2014) Physical Properties of Partially CO 2
Loaded Aqueous Monoethanolamine (MEA). Journal of Chemical & Engineering
Data , 59(6), 1808 -1816.
Heddle, G. , Herzog, H. & Klett, M. (2003) The economics of CO 2 storage. The economics of
CO 2 storage.
Hegerl, G., Zwiers, F., Braconnot, P., Gillet, N., Luo, Y., Marengo, J., Nicholls, N., Penner, J.
& Stott, P. (2007) Understanding and attributing climate change. In: Solomon, S.,
Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M. & Miller, H.
158 L. (eds.) Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change Cambridge, United Kingdom and New York, NY, USA: Cambridge
University Press.
Hessen, E. T., Haug -Warberg, T. & Svendsen, H. F. (2010) The refined e -NRTL model
applied to CO 2-H2O-alkanolamine systems. Chemical Engineering Science , 65(11),
3638-3648.
Higbie, R. (1935) The rate of absorption of a pure gas into a still liquid during short periods
of exposure. Transactions of the A.I.Ch.E., , (31), pp. 365 –389.
Hilliard, M. D. (2008) A Predictive Thermodynamic Model for an Aqueous Blend of
Potassi um Carbonate, Piperazine, and Monoethanolamine for Carbon Dioxide
Capture from Flue Gas. PhD, University of Texas at Austin.
Hoek, P. J. (1983) Large and small scale liquid maldistribution in a packed column. TU
Delft, Delft University of Technology.
IEA. (2008) Energy Technology Perspectives 2008 – In support of the G8 Plan of Action .
Paris: International Energy Agency.
IEA. (2010) Energy Technology Perspectives 2010 . Paris: International Energy Agency.
IEA. (2012) Energy Technology Perspectives 2012 . Pari s: International Energy Agency.
IEA. (2015) CO 2 emissions from fuel combustion -highlights 2015 [Online]. Paris: IEA.
Available: https://www.iea.org/publications/freepublications/publication/co2 –
emissions -from -fuel-combustion -highlights -2015.html [Accessed 6th June 2016].
IEAGHG. (2002) Pipeline transmission of CO 2 and energy. Transmission study report. IEA
GHG.
IEAGHG. (2010) Corrosion and Materials Selection in CCS Systems . Orchard, UK:
International Energy Agency.
IEAGHG. (2012) CO 2 Capture at Gas Fired Power Plants. International Energy Agency.
159 Inflation. (2015) Historic harmonised inflation Europe – HICP inflation [Online]. Available:
http://www.inflation.eu/inflation -rates/europe/historic -inflation/hicp -inflation –
europe.aspx [Accessed 6th June 2016].
IPCC. (20 05) Carbon Dioxide Capture and Storage . UK: Cambridge University Press.
IPCC. (2007) Mitigation of climate change [Online]. IPCC. Available:
http://www.ipcc.ch/publi cations_and_data/publications_and_data_reports.shtml
[Accessed 6th June 2016].
IPCC. (2015) Climate Change 2014: Mitigation of Climate Change (Vol. 3). UK: Cambridge
University Press.
Jonshagen, K., Sipöcz, N. & Genrup, M. (2011) A Novel Approach of Retrof itting a
Combined Cycle with Post Combustion CO 2 Capture. Journal of Engineering for Gas
Turbines and Power , 133((1)), 011703.
Jonsson, M., Bolland, O., Bücker, D. & Rost, M. (2005) Gas turbine cooling model for
evaluation of novel cycles. Proceedings of E COS , 20-22.
Jordal, K., Ystad, P. A. M., Anantharaman, R., Chikukwa, A. & Bolland, O. (2012) Design –
point and part -load considerations for natural gas combined cycle plants with post
combustion capture. International Journal of Greenhouse Gas Control , 11(0 ), 271 –
282.
Jou, F. -Y., Mather, A. E. & Otto, F. D. (1995) Solubility of CO 2 in a 30 mass percent
monoethanolamine solution. Canadian Journal of Chemical Engineering , 73(1), 140 –
147.
Kang, C. A., Brandt, A. R. & Durlofsky, L. J. (2011) Optimal operation of an integrated
energy system including fossil fuel power generation, CO 2 capture and wind. Energy ,
36(12), 6806 -6820.
Kanniche, M., Gros -Bonnivard, R., Jaud, P., Valle -Marcos, J., Amann, J. -M. & Bouallou, C.
(2010) Pre -combustion, post -combustion and oxy -combustion in thermal power plant
for CO 2 capture. Applied Thermal Engineering , 30(1), 53 -62.
160 Karl, T. R., Arguez, A., Huang, B., Lawrimore, J. H., McMahon, J. R., Menne, M. J.,
Peterson, T. C., Vose, R. S. & Zhang, H. -M. (2015) Possible artifacts of data biases in
the recent global surface warming hiatus. Science , 348(6242), 1469 -1472.
Kenig, E. Y., Schneider, R. & Górak, A. (2001) Reactive absorption: Optimal process design
via optimal modelling. Chemical En gineering Science , 56(2), 343 -350.
Kim, Y. (2007) Equation of state for carbon dioxide. Journal of Mechanical Science and
Technology , 21(5), 799 -803.
King, C. (1966) Turbulent liquid phase mass transfer at free gas -liquid interface. Industrial &
Engineerin g Chemistry Fundamentals , 5(1), 1 -8.
Knoope, M. M. J., Ramírez, A. & Faaij, A. P. C. (2013) A state -of-the-art review of techno –
economic models predicting the costs of CO 2 pipeline transport. International Journal
of Greenhouse Gas Control , 16, 241 -270.
Koch-Glitsch. (2014) Carbon capture [Online]. Available: http://www.koch –
glitsch.com/masstransfer/pages/carbon -capture.aspx [Accessed 6th June 2016].
Koch -Glitsch. (2015) Struct ured Pakcing [Online]. Available: http://www.koch –
glitsch.com/Document%20Library/KGSP.pdf [Accessed 6th June 2016].
Koornneef, J., Ramirez, A., van Harmelen, T., van Horssen, A., Turkenb urg, W. & Faaij, A.
(2010) The impact of CO 2 capture in the power and heat sector on the emission of
SO 2, NO x, particulate matter, volatile organic compounds and NH 3 in the European
Union. Atmospheric Environment , 44(11), 1369 -1385.
Kunz, O. & Wagner, W. (2012) The GERG -2008 Wide -Range Equation of State for Natural
Gases and Other Mixtures: An Expansion of GERG -2004. Journal of Chemical &
Engineering Data , 57(11), 3032 -3091.
Kvamme, B., Kuznetsova, T., Jensen, B., Stensholt, S., Bauman, J., Sjo blom, S. & Nes
Lervik, K. (2014) Consequences of CO 2 solubility for hydrate formation from carbon
dioxide containing water and other impurities. Physical Chemistry Chemical Physics ,
16(18), 8623 -8638.
161 Kvamsdal, H. M., Chikukwa, A., Hillestad, M., Zakeri, A . & Einbu, A. (2011a) A
comparison of different parameter correlation models and the validation of an MEA –
based absorber model. Energy Procedia , 4(0), 1526 -1533.
Kvamsdal, H. M., Haugen, G. & Svendsen, H. F. (2011b) Flue -gas cooling in post –
combustion capt ure plants. Chemical Engineering Research and Design , 89(9), 1544 –
1552.
Kvamsdal, H. M., Hetland, J., Haugen, G., Svendsen, H. F., Major, F., Kårstad, V. &
Tjellander, G. (2010) Maintaining a neutral water balance in a 450 MWe NGCC -CCS
power system with po st-combustion carbon dioxide capture aimed at offshore
operation. International Journal of Greenhouse Gas Control , 4(4), 613 -622.
Kvamsdal, H. M., Jordal, K. & Bolland, O. (2007) A quantitative comparison of gas turbine
cycles with CO 2 capture. Energy , 32(1), 10 -24.
Kvamsdal, H. M. & Rochelle, G. T. (2008) Effects of the Temperature Bulge in CO2
Absorption from Flue Gas by Aqueous Monoethanolamine. Industrial & Engineering
Chemistry Research , 47(3), 867 -875.
Lawal, A., Wang, M., Stephenson, P. & Obi, O. (20 12) Demonstrating full -scale post –
combustion CO 2 capture for coal -fired power plants through dynamic modelling and
simulation. Fuel, 101(0), 115 -128.
Lawal, A., Wang, M., Stephenson, P. & Yeung, H. (2009) Dynamic modelling of CO 2
absorption for post combus tion capture in coal -fired power plants. Fuel, 88(12), 2455 –
2462.
Lawlor, S. (2009) CO 2 Compression Using Supersonic Shock Wave Technology [Online].
Ramgen Power Systems. Available:
http://www.netl.doe.gov/technologies/coalpower/ewr/co2/co2compression/supersonic.
html [Accessed 6th June 2016].
Lazic, T., Oko, E. & Wang, M. (2013) Case study on CO 2 transport pipeline network design
for Humber region in the UK. Proceedings of the Institution of Mechanical Engineers,
Part E: Journal of Process Mechanical Engineering .
162 Lee, B. I. & Kesler, M. G. (1975) A generalized thermodynamic correlation based on three –
parameter corresponding states. AIChE Journal , 21(3), 510 -527.
Lee, J. I., Otto, F. D. & Mather, A. E. (1974) The solubility of H 2S and CO 2 in aqueous
monoethanolamine solutions. Can. J. Chem. Eng. , 52(6), 803 -805.
Lee, J. I., Otto, F. D. & Mather, A. E. (1976) Equilibrium between carbon dioxide and
aqueous monoeth anolamine solutions. J Appl Chem Biotechnol , 26(10), 541 -549.
Lepaumier, H., Grimstvedt, A., Vernstad, K., Zahlsen, K. & Svendsen, H. F. (2011)
Degradation of MMEA at absorber and stripper conditions. Chemical Engineering
Science , 66(15), 3491 -3498.
Li, H. , Jakobsen, J. P., Wilhelmsen, Ø. & Yan, J. (2011) PVTxy properties of CO 2 mixtures
relevant for CO 2 capture, transport and storage: Review of available experimental data
and theoretical models. Applied Energy , 88(11), 3567 -3579.
Li, H. & Yan, J. (Year) Im pact of Impurities in CO 2-Fluids on CO 2 Transport Process. In:
ASME Turbo Expo 2006: Power for Land, Sea, and Air, May 8 –11 2006 Barcelona,
Spain. pp. 367 -375.
Li, H. & Yan, J. (2009) Evaluating cubic equations of state for calculation of vapor –liquid
equilibrium of CO 2 and CO 2-mixtures for CO 2 capture and storage processes. Applied
Energy , 86(6), 826 -836.
Li, H., Yan, J., Yan, J. & Anheden, M. (2009) Impurity impacts on the purification process in
oxy-fuel combustion based CO 2 capture and storage system. Applied Energy , 86(2),
202-213.
Liljemark, S., Arvidsson, K., Mc Cann, M. T. P., Tummescheit, H. & Velut, S. (2011)
Dynamic simulation of a carbon dioxide transfer pipeline for analysis of normal
operation and failure modes. Energy Procedia , 4, 3040 -3047.
Lin, Y. & Franzke, C. L. (2015) Scale -dependency of the global mean surface temperature
trend and its implication for the recent hiatus of global warming. Scientific reports , 5.
163 Littel, R. J., Versteeg, G. F. & Van Swaaij, W. P. M. (1992) Kinetics of CO 2 with primary
and secondary amines in aqueous solutions -II. Influence of temperature on zwitterion
formation and deprotonation rates. Chemical Engineering Science , 47(8), 2037 -2045.
Liu, Y., Zhang, L. & Watanasiri, S. (1999) Representing Vapor−Liquid Equilib rium for an
Aqueous MEA−CO 2 System Using the Electrolyte Nonrandom -Two-Liquid Model.
Industrial & Engineering Chemistry Research , 38(5), 2080 -2090.
Lucquiaud, M. & Gibbins, J. (2009) Retrofitting CO 2 capture ready fossil plants with post –
combustion capture . Part 1: Requirements for supercritical pulverized coal plants
using solvent -based flue gas scrubbing. Proceedings of the Institution of Mechanical
Engineers, Part A: Journal of Power and Energy , 223(3), 213 -226.
Mac Dowell, N., Llovell, F., Adjiman, C. S ., Jackson, G. & Galindo, A. (2009) Modeling the
Fluid Phase Behavior of Carbon Dioxide in Aqueous Solutions of Monoethanolamine
Using Transferable Parameters with the SAFT -VR Approach. Industrial &
Engineering Chemistry Research , 49(4), 1883 -1899.
Mac Dow ell, N., Pereira, F. E., Llovell, F., Blas, F. J., Adjiman, C. S., Jackson, G. &
Galindo, A. (2011) Transferable SAFT -VR Models for the Calculation of the Fluid
Phase Equilibria in Reactive Mixtures of Carbon Dioxide, Water, and n -Alkylamines
in the Contex t of Carbon Capture. The Journal of Physical Chemistry B , 115(25),
8155 -8168.
Mac Dowell, N. & Shah, N. (2013) Identification of the cost -optimal degree of CO 2 capture:
An optimisation study using dynamic process models. International Journal of
Greenhouse Gas Control , 13(0), 44 -58.
Mahgerefteh, H., Denton, G. & Rykov, Y. (Year) Pressurised CO 2 pipeline rupture. In, 2008.
869-879.
Mangalapally, H. P. & Hasse, H. (2011) Pilot plant study of post -combustion carbon dioxide
capture by reactive absorption: Metho dology, comparison of different structured
packings, and comprehensive results for monoethanolamine. Chemical Engineering
Research and Design , 89(8), 1216 -1228.
164 Marchioro Ystad, P. A., Lakew, A. A. & Bolland, O. (2013) Integration of low -temperature
transc ritical CO 2 Rankine cycle in natural gas -fired combined cycle (NGCC) with
post-combustion CO 2 capture. International Journal of Greenhouse Gas Control ,
12(0), 213 -219.
Mason, J. W. & Dodge, B. F. (1936) Equilibrium absorption of carbon dioxide by solutions
of the ethanolamines. Trans. Am. Inst. Chem. Eng , 32, 27 -48.
McCollum, D. L. & Ogden, J. M. (2006) Techno -economic models for carbon dioxide
compression, transport, and storage & Correlations for estimating carbon dioxide
density and viscosity [Online]. A vailable:
http://www.canadiancleanpowercoalition.com/pdf/CTS1%20 -%202006_UCD -ITS-
RR-06-14[1].pdf [Accessed 6th 2016 2016].
McCoy, S. & Rubin, E. (2008) An engineering -economic model of pipeline transport of CO 2
with application to carbon capture and storage. International Journal of Greenhouse
Gas Control , 2(2), 219 -229.
Merkel, T. C., Lin, H., Wei, X. & Baker, R. (2010) Power plant post -combustion carbo n
dioxide capture: an opportunity for membranes. Journal of membrane science , 359(1),
126-139.
Mores, P., Rodríguez, N., Scenna, N. & Mussati, S. (2012) CO 2 capture in power plants:
Minimization of the investment and operating cost of the post -combustion p rocess
using MEA aqueous solution. International Journal of Greenhouse Gas Control ,
10(0), 148 -163.
Mores, P. L., Godoy, E., Mussati, S. F. & Scenna, N. J. (2014) A NGCC power plant with a
CO 2 post-combustion capture option. Optimal economics for different
generation/capture goals. Chemical Engineering Research and Design , 92(7), 1329 –
1353.
NASA. (2016) Carbon Dioxide [Online]. NASA. Available: http://climate.nasa.gov/vital –
signs/carbon -dioxi de/ [Accessed 6th June 2016].
National Grid. (2014) Yorkshire and Humber CCS Project [Online]. Available:
http://www.ccshumber.co.uk/ [Accessed 6th June 2016].
165 NERC. (2014) Storing carbon dioxide beneath the Moray Firth explored [Online]. Available:
http://www.bgs.ac.uk/research/highlights/2010/scottishCCS.html [Accessed 6th June
2016].
Nimtz, M., Kl att, M., Wiese, B., Kühn, M. & Joachim Krautz, H. (2010) Modelling of the
CO 2 process – and transport chain in CCS systems —Examination of transport and
storage processes. Chemie der Erde – Geochemistry , 70, 185 -192.
NOAA. (2015) The Recent Global Surface Wa rming Hiatus [Online]. Available:
https://www.ncdc.noaa.gov/news/recent -global -surface -warming -hiatus [Accessed 6th
June 2016].
Notz, R., Mangalapally, H. P. & Hasse, H. (201 2) Post combustion CO 2 capture by reactive
absorption: Pilot plant description and results of systematic studies with MEA.
International Journal of Greenhouse Gas Control , 6, 84 -112.
Ogden, J. M., Yang, C., Johnson, N., Ni, J. & Johnson, J. (2004) Conceptu al design of
optimized fossil energy systems with capture and sequestration of carbon dioxide
[Online]. Available: http://www.osti.gov/scitech/servlets/purl/829538 [Accessed 6th
June 2016].
Ong'iro, A. O., Ugursal, V. I., Al Taweel, A. M. & Blamire, D. K. (1995) Simulation of
combined cycle power plants using the ASPEN PLUS shell. Heat Recovery Systems
and CHP , 15(2), 105 -113.
Oyenekan, B. A. & Rochelle, G. T. (2007) Alternative stripper configu rations for CO 2
capture by aqueous amines. AIChE Journal , 53(12), 3144 -3154.
Pachauri, R. K., Allen, M., Barros, V., Broome, J., Cramer, W., Christ, R., Church, J., Clarke,
L., Dahe, Q. & Dasgupta, P. (2014) Climate Change 2014: Synthesis Report.
Contribut ion of Working Groups I, II and III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change [Online]. Available:
http://www.ipcc.ch/report/ar5/syr/ [Accessed 6th June 2016].
Panahi, M. & Skogestad, S. (2011) Economically efficient operation of CO 2 capturing
process part I: Self -optimizing procedure for selecting the best controlled variables.
Chemical Engineering and Processing: Process Intensification , 50(3), 247 -253.
166 Peng, D. Y. & R obinson, D. B. (1976) A new two -constant equation of state. Industrial and
Engineering Chemistry Fundamentals , 15(1), 59 -64.
Pershad, H., Harland, K., Stewart, A. & Slater, S. (2010) CO 2 pipeline infrastructure: An
analysis of global challenges and opportu nities [Online]. Cambridge: Element
Energy. [Accessed.
Piessens, K., Laenen, B., Nijs, W., Mathieu, P., Baele, J. M., Hendriks, C., Bertrand, E.,
Bierkens, J., Brandsma, R., Broothaers, M., De Visser, E., Dreesen, R., Hildenbrand,
D., Lagrou, D., Vendegin ste, V. & Welkenhuysen, K. (2008) Policy Support System
for Carbon Capture and Storage.
POST. (2009) Postnote 335 – carbon capture and storage . London, United Kingdom: The
Parliamentary Office of Science and Technology.
R.F., S. (1987) Random packings and packed towers: Design and applications (Second ed.).
Houston: Gulf Publishing Company.
Race, J. M., Wetenhall, B., Seevam, P. N. & Downie, M. J. (2012) Towards a CO 2 pipeline
specification: defining tolerance limits for impurities. Journal of Pipeline
Engi neering , 11(3), 173 -190.
Rao, A. B. & Rubin, E. S. (2002) A technical, economic, and environmental assessment of
amine -based CO2 capture technology for power plant greenhouse gas control.
Environmental Science and Technology , 36(20), 4467 -4475.
Rao, A. B. & Rubin, E. S. (2005) Identifying Cost -Effective CO 2 Control Levels for Amine –
Based CO 2 Capture Systems. Industrial & Engineering Chemistry Research , 45(8),
2421 -2429.
Razi, N., Bolland, O. & Svendsen, H. (2012) Review of design correlations for CO 2
absorption into MEA using structured packings. International Journal of Greenhouse
Gas Control , 9, 193 -219.
Razi, N., Svendsen, H. F. & Bolland, O. (2013a) Cost and energy sensitivity analysis of
absorber design in CO 2 capture with MEA. International Jour nal of Greenhouse Gas
Control , 19, 331 -339.
167 Razi, N., Svendsen, H. F. & Bolland, O. (2013b) Validation of mass transfer correlations for
CO 2 absorption with MEA using pilot data. International Journal of Greenhouse Gas
Control , 19(0), 478 -491.
Reddy, S., S cherffius, J. R., Yonkoski, J., Radgen, P. & Rode, H. (2013) Initial Results from
Fluor's CO 2 Capture Demonstration Plant Using Econamine FG PlusSM Technology
at E.ON Kraftwerke's Wilhelmshaven Power Plant. Energy Procedia , 37, 6216 -6225.
Rochelle, G. T. ( 2009) Amine Scrubbing for CO 2 Capture. Science , 325(5948), 1652 -1654.
Rochelle, G. T. (2012) Thermal degradation of amines for CO 2 capture. Current Opinion in
Chemical Engineering , 1(2), 183 -190.
Roussanaly, S., Bureau -Cauchois, G. & Husebye, J. (2013) Cos ts benchmark of CO 2
transport technologies for a group of various size industries. International Journal of
Greenhouse Gas Control , 12(0), 341 -350.
Rumpf, B. & Maurer, G. (1992) Solubilities of hydrogen cyanide and sulfur dioxide in water
at temperatures f rom 293.15 to 413.15 K and pressures up to 2.5 MPa. Fluid Phase
Equilibria , 81(C), 241 -260.
Samanta, A., Zhao, A., Shimizu, G. K., Sarkar, P. & Gupta, R. (2011) Post -combustion CO 2
capture using solid sorbents: a review. Industrial & Engineering Chemistry Research ,
51(4), 1438 -1463.
SASKPOWER. (2015) The world's first post -combustion coal -fired CCS facility [Online].
Available: http://saskpowerccs.com/ [Accessed 6th June 2016].
SCCS. (2016) CCS-the full chain [Online]. Scottish Carbon Capture and Storage Research
Group. Available: http://www.sccs.org.uk/education -and-training/downloads
[Accessed 6th June 2016].
Schach, M. -O., Schneider, R., Schramm, H . & Repke, J. -U. (2013) Control Structure Design
for CO2 -Absorption Processes with Large Operating Ranges. Energy Technology ,
1(4), 233 -244.
Seevam, P. N., Race, J. M., Downie, M. J. & Hopkins, P. (2008) Transporting the Next
Generation of CO 2 for Carbon, Capture and Storage: The Impact of Impurities on
168 Supercritical CO 2 Pipelines. 2008 7th International Pipeline Conference. Calgary,
Alberta, Canada: International Petroleum Technology Institute and the Pipeline
Division.
Shah, R. K. & Sekulic, D. P. (2003) Fundamentals of heat exchanger design : John Wiley &
Sons.
SIEMENS. (2016) Siemens sets new performance and efficiency world record at Düsseldorf
power plant [Online]. Available:
http://www.siemens.com/press/en/pressrelease/index.php?content%5B%5D=Corp&se
arch=Irsching&date -1-dd=26&date -1-mm=04&date -1=2015&date -2-dd=26&date -2-
mm=04&date -2=2016&intern=1 [Accessed 6th June 2016].
Sime, L. C. (2014) Greenland deglaciation puzzles. Science , 345(6201), 1116 -1117.
Sinnott, R. & Towler, G. (2009) Chemical engineering design (5th ed.). Oxford: Butterworth
heinemann.
Sipöcz, N. & Tobiesen, F. A. (2012) Natural gas combined cycle power plants with CO 2
capture – Opportunities to reduce cost. International Journal of Greenhouse Gas
Control , 7(0), 98 -106.
Soave, G. (1972) Equilibrium constants for modified Redlich -Kwong equation of state.
Chem. Eng. Sci. , 27, 1196 -1203.
Song, J. H., Yoon, J. H., Lee, H. & Lee, K. H. (1996) Solubility of carbon dioxide in
monoethanolamine + ethylene glycol + water and monoet hanolamine + poly(ethylene
glycol) + water. Journal of Chemical and Engineering Data , 41(3), 497 -499.
Song, Y. & Chen, C. C. (2009) Symmetric electrolyte nonrandom two -liquid activity
coefficient model. Industrial and Engineering Chemistry Research , 48(16) , 7788 –
7797.
Span, R. & Wagner, W. (1996) A new equation of state for carbon dioxide covering the fluid
region from the triple -point temperature to 1100 K at pressures up to 800 MPa.
Journal of Physical and Chemical Reference Data , 25(6), 1509 -1596.
169 Stichl mair, J. & Fair, J. R. (1998) Distillation: principles and practices : Vch
Verlagsgesellschaft Mbh.
Sulzer. (2014) Putting a chill on global warming [Online]. Available:
http://www.sulzer.com/en/ –
/media/Documents/Cross_Division/STR/2008/2008_3_4_Sulzer_e.pdf [Accessed 6th
June 2016].
Sulzer. (2015) Structured Packings_Energy -efficient, innovative and profitable [Online].
Available: http://www.sulzer.com/cs/ –
/media/Documents/ProductsAndServices/Separation_Technology/Structured_Packing
s/Brochures/Structured_Pa ckings.pdf [Accessed 6th June 2016].
Svensson, R., Odenberger, M., Johnsson, F. & Strömberg, L. (2004) Transportation systems
for CO 2––application to carbon capture and storage. Energy Conversion and
Management , 45(15 -16), 2343 -2353.
Sweeting, J. (1997) Project Cost Estimating: Principles and Practice : IChemE.
Szulczewski, M. L., MacMinn, C. W., Herzog, H. J. & Juanes, R. (2012) Lifetime of carbon
capture and storage as a climate -change mitigation technology. Proceedings of the
National Academy of Sciences , 109(14), 5185 -5189.
Tobiesen, F. A., Svendsen, H. F. & Juliussen, O. (2007) Experimental validation of a
rigorous absorber model for CO 2 postcombustion capture. AIChE Journal , 53(4), 846 –
865.
Tong, D., Trusler, J. P. M., Maitland, G. C., Gibbins, J. & Fen nell, P. S. (2012) Solubility of
carbon dioxide in aqueous solution of monoethanolamine or 2 -amino -2-methyl -1-
propanol: Experimental measurements and modelling. International Journal of
Greenhouse Gas Control , 6, 37 -47.
Towler, G. & Sinnott, R. K. (2012) Chemical engineering design: principles, practice and
economics of plant and process design (2nd ed.). Oxford: Butterworth -Heinemann
Tsai, R. E., Seibert, A. F., Eldridge, R. B. & Rochelle, G. T. (2011) A dimensionless model
for predicting the mass -transfe r area of structured packing. AIChE Journal , 57(5),
1173 -1184.
170 Ulfsnes, R., Karlsen, G., Jordal, K., Bolland, O. & Kvamsdal, H. M. (2003) Investigation of
physical properties of CO 2/H2O-mixtures for use in semi -closed O 2/CO 2 gas turbine
cycle with CO 2-capture. Department of Energy and Process Eng., Norwegian Univ. of
Science and Technology 9.
UNFCCC. (2015) Historic Paris Agreement on Climate Change [Online]. Available:
http://newsroom.u nfccc.int/unfccc -newsroom/finale -cop21/ [Accessed 6th June 2016].
USDOE. (2010a) Energy Market and Economic Impacts of the American Power Act of 2010.
Washington, DC, USA.
USDOE. (2010b) Interagency Task Force on Carbon Capture and Storage . Washington,
DC, USA.
Valdés, M. & Rapún, J. L. (2001) Optimization of heat recovery steam generators for
combined cycle gas turbine power plants. Applied Thermal Engineering , 21(11),
1149 -1159.
Van den Broek, M., Brederode, E., Ramírez, A., Kramers, L., Van der Kuip, M ., Wildenborg,
T., Turkenburg, W. & Faaij, A. (2010) Designing a cost -effective CO 2 storage
infrastructure using a GIS based linear optimization energy model. Environmental
Modelling & Software , 25(12), 1754 -1768.
Vargas, J. V. C. & Bejan, A. (2000) Thermo dynamic optimization of the match between two
streams with phase change. Energy , 25(1), 15 -33.
Wagner, M., von Harbou, I., Kim, J., Ermatchkova, I., Maurer, G. & Hasse, H. (2013)
Solubility of Carbon Dioxide in Aqueous Solutions of Monoethanolamine in the Low
and High Gas Loading Regions. Journal of Chemical & Engineering Data , 58(4),
883-895.
Wang, M., Lawal, A., Stephenson, P., Sidders, J. & Ramshaw, C. (2011) P ost-combustion
CO 2 capture with chemical absorption: A state -of-the-art review. Chemical
Engineering Research and Design , 89(9), 1609 -1624.
Wegstein, J. H. (1958) Accelerating convergence of iterative processes. Commun. ACM , 1(6),
9-13.
171 Weiland, R. H. (199 6) Carbon dioxide removal by mixtures of amines: Final report : Gas
Research Institute.
Weiland, R. H., Dingman, J. C. & Cronin, D. B. (1997) Heat capacity of aqueous
monoethanolamine, diethanolamine, N -methyldiethanolamine, and N –
methyldiethanolamine -based blends with carbon dioxide. Journal of Chemical and
Engineering Data , 42(5), 1004 -1006.
Weiland, R. H., Dingman, J. C., Cronin, D. B. & Browning, G. J. (1998) Density and
viscosity of some partially carbonated aqueous alkanolamine solutions and their
blends. Journal of Chemical and Engineering Data , 43(3), 378 -382.
Wertheim, M. S. (1984) Fluids with highly directional attractive forces. I. Statistical
thermodynamics. Journal of Statistical Physics , 35(1 -2), 19 -34.
Wertheim, M. S. (1986) Fluids with highly directional attractive forces. IV. Equilibrium
polymerization. Journal of Statistical Physics , 42(3 -4), 477 -492.
Whitman, W. G. (1962) The two film theory of gas absorption. International Journal of Heat
and Mass T ransfer , 5(5), 429 -433.
Wildenborg, T., Holloway, S., Hendriks, C., Kreft, E., Lokhorst, A., Brook, M., Brandsma,
R., Egberts, P. & Larsen, M. (2004) Cost curves for CO 2 storage. Part 2: European
sector.
Witkowski, A. & Majkut, M. (2012) The impact of CO 2 compression systems on the
compressor power required for a pulverized coal -fired power plant in post -combustion
carbon dioxide sequestration. Archive of Mechanical Engineering , 59(3), 343 -360.
Witkowski, A., Rusin, A., Majkut, M., Rulik, S. & Stolecka, K. (2013) Comprehensive
analysis of pipeline transportation systems for CO 2 sequestration. Thermodynamics
and safety problems. Energy Conversion and Management , 76, 665 -673.
Wong, S. (2005) Module 4: CO 2 compression and transportation to storage reservoir
[Online]. Asia -Pacific Economic Cooperation (APEC). Available:
https://hub.globalccsinstitute.com/publications/building -capacity -co2-capture -and-
storage -apec-region -training -manual -policy -makers -and-practitioners/module -4-co2-
compression -and-transportation -storage -site [Accessed 6th June 2016] .
172 Xu, Q. & Rochelle, G. (Year) Total pressure and CO 2 solubility at high temperature in
aqueous amines. In: Energy Procedia, 2011. 117 -124.
Yan, Y. & Chen, C. -C. (2010) Thermodynamic modeling of CO 2 solubility in aqueous
solutions of NaCl and Na 2SO 4. The Journal of Supercritical Fluids , 55(2), 623 -634.
Ycharts. (2015) European Union Natural Gas Import Price [Online]. Available:
http://ycharts.com/indicators/europe_natural_gas_price [Acces sed 6th June 2016].
Ying, J. & Eimer, D. A. (2012) Measurements and correlations of diffusivities of nitrous
oxide and carbon dioxide in monoethanolamine+ water by laminar liquid jet.
Industrial & Engineering Chemistry Research , 51(50), 16517 -16524.
Zhang, Y. & Chen, C. -C. (2013) Modeling CO 2 Absorption and Desorption by Aqueous
Monoethanolamine Solution with Aspen Rate -based Model. Energy Procedia , 37(0),
1584 -1596.
Zhang, Y. & Chen, C. C. (2011) Thermodynamic modeling for CO 2 absorption in aqueous
MDEA so lution with electrolyte NRTL model. Industrial and Engineering Chemistry
Research , 50(1), 163 -175.
Zhang, Y., Chen, H., Chen, C. -C., Plaza, J. M., Dugas, R. & Rochelle, G. T. (2009) Rate –
Based Process Modeling Study of CO 2 Capture with Aqueous Monoethanola mine
Solution. Industrial & Engineering Chemistry Research , 48(20), 9233 -9246.
Zhang, Y., Que, H. & Chen, C. -C. (2011) Thermodynamic modeling for CO 2 absorption in
aqueous MEA solution with electrolyte NRTL model. Fluid Phase Equilibria , 311,
67-75.
Zhang, Z. X., Wang, G. X., Massarotto, P. & Rudolph, V. (2006) Optimization of pipeline
transport for CO 2 sequestration. Energy Conversion and Management , 47(6), 702 –
715.
Copyright Notice
© Licențiada.org respectă drepturile de proprietate intelectuală și așteaptă ca toți utilizatorii să facă același lucru. Dacă consideri că un conținut de pe site încalcă drepturile tale de autor, te rugăm să trimiți o notificare DMCA.
Acest articol: FACULTY OF SCIENCES AND ENGINEERING SCHOOL OF ENGINEERING PhD Thesis PROCESS MODELLING, SIMULATION AND OPTIMISATION OF NATURAL GAS COMBINED CYCLE… [628922] (ID: 628922)
Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.