Integrated Gasification Combined Cycle Power Plants

Integrated Gasification Combined Cycle
Power Plants
with Focus on Low Emission Gas Turbine Technology
by
Mohammad Mansouri Majoumerd
Thesis submitted in partial fulfillment of
the requirements for the degree of
PHILOSOPHIAE DOCTOR
(PhD)
Faculty of Science and Technology
Department of Petroleum Engineering
2014
University of Stavanger
N-4036 Stavanger
NORWAY
www.uis.no
©2014 Mohammad Mansouri Majoumerd
All rights reserved
ISBN:
ISSN:
PhD Thesis UiS, No.
Abstract
It is foreseen that global demand for electricity will increase continuously, mainly due to
population growth and improved living standards, worldwide. At the same time, the
climate change issue due to increasing GHG emissions, more specifically from the heat
and power sector, has become one of the most important global concerns. Thus there has
been a genuine demand for the delivery of innovative solutions to provide electricity in a
more sustainable way. Several pathways, which have significant potential for GHG
emissions mitigation, while providing electric power, have been introduced and
investigated during recent years. The improvement of energy efficiency and the
deployment of carbon capture and storage (CCS) in fossil-based power plants are amongst
the options to stabilize the atmospheric levels of greenhouse gas emissions, while
enabling the continued use of fossil fuels. In this regard, the integrated gasification
combined cycle (IGCC) power plant is one of the most promising power generation
technologies. Environmentally benign use of coal as primary fuel, use of highly reliable
gas turbine (GT) technology, possibilities for poly-generation of different products and for
pre-combustion CO2 capture are important features of this technology.
In 2009, the European Union co-financed the “Low Emission Gas Turbine Technology for
Hydrogen-rich Syngas (H2-IGCC)” project to achieve its targets of higher energy
efficiency and lower GHG emission levels along with greater security of energy supply.
This project aimed at providing technical solutions for using undiluted hydrogen-rich
syngas in gas turbines for IGCC application with CO2 capture.
As part of the H2-IGCC project, this PhD thesis presents investigations into the
deployment of pre-combustion CO2 capture in IGCC power plants aiming at providing
practical and realistic system integration solutions. The emphasis has been on the gas
turbine block to enable the combustion of undiluted hydrogen-rich syngas, a requirement
iii
iv
Abstract
of future IGCC technology with CO2 capture. For this purpose, different state-of-the-art
technologies for various sub-systems of the IGCC plant were proposed and the most
practical options based on the feedback from industrial partners within the H2-IGCC
consortium were selected for further thermodynamic analyses. The outcomes of these
analyses together with technical constraints related to the proposed cycle configurations
were used by other working groups as boundary conditions for the development of a gas
turbine technology optimized for undiluted H2-rich syngas. Moreover, a techno-economic
tool has been generated, which enabled economic assessments of the IGCC plant with
CO2 capture and its main fossil-based competitors, using realistic cost and performance
data confirmed by important players in the European power market.
During the implementation of this thesis, it is demonstrated that the combustion of
undiluted H2-rich syngas and the meeting of fuel flexibility targets are not possible using
the existing GT technology. Accordingly, necessary modifications were proposed and
implemented to provide an optimized GT technology suitable for the combustion of
undiluted H2-rich syngas. It is also found that investigated fossil-based power plants have
similar cost levels. The marginal difference in the cost of electricity for different plants
was within the level of uncertainties in the assessment of investment costs. Therefore,
other main drivers, apart from the cost of electricity, can affect the selection of a power
generation technology such as operational flexibility and potential for future technological
improvements.
Keywords: CO2 capture and storage, gas turbine, H2-rich syngas, IGCC, power
generation, system integration, techno-economy
Acknowledgements
This research was co-financed by the European Union’s Seventh Framework Programme
for Research and Development. Financial support from the European Commission,
Directorate-General for Energy is gratefully acknowledged.
I would like to express my sincere gratitude to Professor Mohsen Assadi for his excellent
supervision and encouragement throughout my PhD program. I also appreciate his efforts
in providing me with the unique opportunity of being involved in two European-funded
projects, the H2-IGCC project and the European North Sea Energy Alliance (ENSEA)
project. I would like to thank Peter Breuhaus at the International Research Institute of
Stavanger (IRIS) for sharing his long-term experience and profound knowledge within the
field of gas turbine technology.
Thanks to Nikolett Sipöcz, Sudipta De, and Homam Nikpey for sharing their knowledge
and the collaborative works resulting in several journal and conference papers.
I would like to thank all the project partners in the H2-IGCC project whose contributions
have been used in this work. In this regard, my special thanks go to Han Raas at
Nuon/Vattenfall for performing gasification simulations and for sharing knowledge about
operational aspects of IGCC power plants. I am also thankful to Chris Lappee at
Nuon/Vattenfall, Stuart James, James Bowers, Karel Dvorak and Adam Al-Azki at E.ON
UK for sharing their industrial perspectives and providing technical and economic inputs.
Inputs from our partners in other sub-projects of the H2-IGCC project such as
combustion, materials and turbo-machinery in addition to the perfect coordination
activities, which were carried out by European Turbine Network (ETN) are also
appreciated.
v
vi
Acknowledgements
Furthermore, my sincere thanks must go to my former colleagues at the University of
Stavanger (UiS) and my friends in Stavanger for their motivation and encouragement. I
would also like to express my gratitude to the new energy, risk management and well
construction group at IRIS.
Finally, and perhaps most importantly, I would like to take this opportunity to express my
deep gratitude to my family for their endless love and support.
Mohammad Mansouri Majoumerd,
Stavanger, Norway
Table of Contents
Abstract.............................................................................................................................. iii
Acknowledgements ............................................................................................................. v
Table of Contents...............................................................................................................vii
Nomenclature...................................................................................................................... xi
List of appended papers .................................................................................................... xxi
Additional papers not included .......................................................................................xxiii
1. Introduction ..................................................................................................................... 1
1.1. Background information ........................................................................................... 1
1.2. Objectives ................................................................................................................. 3
1.3. Limitations ................................................................................................................ 4
1.4. Methodology ............................................................................................................. 5
1.5. Outline of the thesis .................................................................................................. 6
2. Technical background ...................................................................................................... 7
2.1. Growing energy demand ........................................................................................... 7
2.2. Climate change ......................................................................................................... 8
2.2.1. Greenhouse gas emissions ................................................................................. 9
2.2.2. Climate change and the power sector .............................................................. 11
2.3. Mitigation policies .................................................................................................. 12
2.3.1. Carbon capture and storage .............................................................................. 13
vii
viii
Table of Contents
2.3.2. The European Union climate strategy .............................................................. 14
2.4. Various capture technologies in the power sector................................................... 15
2.4.1. Post-combustion CO2 capture .......................................................................... 17
2.4.2. Pre-combustion capture ................................................................................... 19
2.4.3. Oxy-fuel combustion ....................................................................................... 21
3. Coal-based power plants ................................................................................................ 25
3.1. Why coal-based power plants? ............................................................................... 26
3.2. Coal-fired power generation ................................................................................... 27
3.3. IGCC power plant’s components ............................................................................ 30
3.3.1. Air separation................................................................................................... 30
3.3.1.1. Cryogenic ASU and power island integration options .............................. 32
3.3.1.2. Other ASU technologies ........................................................................... 35
3.3.2. Gasification ...................................................................................................... 35
3.3.2.1. Entrained-flow gasifiers ............................................................................ 36
3.3.2.2. Gasification performance .......................................................................... 38
3.3.2.2.1. Coal quality ........................................................................................ 39
3.3.2.2.2. Cold gas efficiency ............................................................................ 40
3.3.3. Syngas cleaning and conversion ...................................................................... 41
3.3.3.1. Syngas cleaning ........................................................................................ 41
3.3.3.2. Water-gas shift reaction ............................................................................ 42
3.3.3.3. Acid gas removal ...................................................................................... 44
3.3.3.4. Sulfur recovery unit .................................................................................. 47
3.3.3.5. Advanced syngas cleaning and conversion ............................................... 48
3.3.4. CO2 compression and dehydration ................................................................... 50
3.3.5. Gas turbine ....................................................................................................... 52
Table of Contents
ix
3.3.5.1. Combustion process .................................................................................. 52
3.3.5.2. Turbo-machinery ...................................................................................... 54
3.3.5.3. Materials ................................................................................................... 56
3.3.5.4. Commercial syngas-fueled gas turbine ..................................................... 56
3.3.5.5. Advanced hydrogen turbine technology ................................................... 57
3.3.6. Bottoming cycle ............................................................................................... 58
3.4. Current IGCC power plants status .......................................................................... 58
4. H2-IGCC power plant.................................................................................................... 61
4.1. H2-IGCC project .................................................................................................... 61
4.2. System integration .................................................................................................. 64
4.2.1. Cryogenic air separation unit ........................................................................... 65
4.2.2. Gasification ...................................................................................................... 66
4.2.3. Syngas conversion ........................................................................................... 66
4.2.4. Acid gas removal ............................................................................................. 67
4.2.5. Gas turbine ....................................................................................................... 68
4.3. System performance analysis .................................................................................. 70
4.3.1. Software tools .................................................................................................. 71
4.3.2. Boundary conditions ........................................................................................ 74
4.3.2.1. Ambient conditions ................................................................................... 74
4.3.2.2. Feedstock properties ................................................................................. 74
4.3.2.3. Gas turbine boundaries and performance .................................................. 76
5. Economic evaluation ..................................................................................................... 81
5.1. Cost estimating methodology ................................................................................. 81
5.1.1. Costing scope ................................................................................................... 83
5.1.2. Capital costs ..................................................................................................... 84
x
Table of Contents
5.1.2.1. Step-count costing method ........................................................................ 84
5.1.2.2. Capacity adjustment .................................................................................. 86
5.1.2.3. Price fluctuations ...................................................................................... 86
5.1.2.4. Currency exchange ................................................................................... 87
5.1.3. Operation and maintenance (O&M) costs ....................................................... 87
5.1.4. Fuel cost ........................................................................................................... 87
5.1.5. CO2 cost measures ........................................................................................... 88
5.2. Uncertainty in the economic results ........................................................................ 89
6. Concluding remarks ....................................................................................................... 93
6.1. Conclusions ............................................................................................................ 93
6.2. Scientific contributions ........................................................................................... 96
6.3. Suggestions for further research ............................................................................. 98
7. Summary of appended papers ...................................................................................... 101
Bibliography .................................................................................................................... 109
Paper I .............................................................................................................................. 121
Paper II ............................................................................................................................ 133
Paper III ........................................................................................................................... 147
Paper IV ........................................................................................................................... 161
Paper V ............................................................................................................................ 175
Paper VI ........................................................................................................................... 187
Nomenclature
Abbreviations
AGR
acid gas removal
Al2O3
aluminum (III) oxide or alumina
Ar
argon
AR4
4th assessment report of Intergovernmental Panel on Climate Change
ASU
air separation unit
BFW
boiler feed water
BUA
bottom-up approach
CAESAR
CArbon-free Electricity by SEWGS: Advanced materials, Reactor-, and
process design
CaO
calcium oxide
CAPEX
capital expenditure
CCS
carbon capture and storage
CEPCI
Chemical Engineering Plant Cost Index
CHP
combined heat and power
xi
xii
Nomenclature
CH4
methane
CI
cost index
CLC
chemical looping combustion
CMD
coal milling and drying
CO
carbon monoxide
Co
cobalt
COE
cost of electricity
COS
carbonyl sulfide
COT
combustor outlet temperature
CO2
carbon dioxide
Cr
chromium
Cu
copper
DCF
discounted cash flow
DGAN
diluent gaseous nitrogen
DLN
dry low NOx
DOE
Department of Energy
EBTF
European Benchmarking Task Force
EOS
equation-of-state
EPCC
engineering, procurement and construction costs
ETS
emissions trading system
Nomenclature
xiii
EU
European Union
E-GasTM
ConocoPhillips gasifier
FBC
fluidized bed combustion
Fe
iron
FeO
ferrous or iron oxide
FGD
flue gas desulfurization
FP7
Seventh Framework Programme
GAN
gaseous nitrogen
GDP
gross domestic product
GE
General Electric
GHG
greenhouse gas
GOX
gaseous oxygen
GT
gas turbine
HHV
higher heating value
HP
high pressure
HRSG
heat recovery steam generator
HSE
health, safety and environment
HT
high temperature
HTS
high temperature shift
H2
hydrogen
xiv
Nomenclature
H2O
water
H2S
hydrogen sulfide
H2-IGCC
Low Emission Gas Turbine Technology for Hydrogen-rich Syngas
project
IC
indirect costs
IEA
International Energy Agency
IGCC
integrated gasification combined cycle
IGV
inlet guide vanes
IL
ionic liquid
IP
intermediate pressure
IPCC
Intergovernmental Panel on Climate Change
IR
index ratio
ITM
ion transport membrane
KP
Kyoto Protocol
LHV
lower heating value
LP
low pressure
LT
low temperature
LTS
low temperature shift
MAC
main air compressor
MEA
monoethanolamine
MgO
magnesium oxide
Nomenclature
xv
MHI
Mitsubishi Heavy Industry
Mo
molybdenum
M&S
Marshall and Swift cost index
NETL
National Energy Technology Laboratory
NG
natural gas
NGCC
natural gas combined cycle
NGV
nozzle guide vane
NH3
ammonia
NOx
nitrogen oxides
NPV
net present value
N2
nitrogen
N2O
nitrous oxide
OC
owner costs
OECD
Organization for Economic Co-operation and Development
OEM
original equipment manufacturer
OPEX
operational expenditure
O&M
operation and maintenance
O2
oxygen
PC
pulverized coal
PCC
pulverized coal combustion
xvi
Nomenclature
PC-SAFT
perturbed-chain statistical associating fluid theory
PGAN
pressurized gaseous nitrogen
PM
particulate matter
PR
Peng-Robinson
RE
renewable energy
R&D
research and development
SC
supercritical
SCGP
Shell Coal Gasification Process
SCOT
Shell Claus off-gas treating
SCPC
supercritical pulverized coal
SCR
selective catalytic reduction
SEWGS
sorption-enhanced water-gas shift
SFG
Siemens Fuel Gasification
SiO2
silicon dioxide
SOA
state-of-the-art
SOx
sulfur oxides
SO2
sulfur dioxide
SR
Schwarzentruber and Renon
SRU
sulfur recovery unit
ST
steam turbine
Nomenclature
xvii
SWGS
sour water-gas shift
TBC
thermal barrier coating
TDPC
total direct plant cost
TEC
total equipment costs
TEG
tri-ethylene glycol
TGT
tail gas treating
TIT
turbine inlet temperature
TOT
turbine outlet temperature
TPC
total plant costs
UHC
unburned hydrocarbons
UNFCCC
United Nations Framework Convention on Climate Change
USC
ultra-supercritical
USCPC
ultra-supercritical pulverized coal
U.S. DOE
United States Department of Energy
VIGV
variable inlet guide vanes
WGS
water-gas shift
Zn
zinc
Latin
∆ℎ
enthalpy change (kJ kg-1)
𝐴
area (m2)
xviii
Nomenclature
𝐶
cost (€)
𝑐𝑝
specific heat transfer coefficient at constant pressure (kJ mol-1 K-1)
𝐼𝑗
installation costs of a component (sub-system) (€)
𝐿𝐻𝑉
lower heating value (kJ kg-1 or kJ m-3)
𝑚̇
mass flow rate (kg s-1)
𝑝
pressure (bar)
𝑄̇
volumetric flow rate (m3 s-1)
𝑅
gas constant (kJ kg-1 K-1)
𝑆
scaling parameter
𝑇
temperature (K)
𝑊
work (kW)
𝑥̅
mean value of a parameter
Greek letters
𝛽
pressure ratio (-)
𝛾
isentropic exponent (-)
𝜅
constant
𝜂
efficiency (%)
𝑓
cost scaling exponent (-)
Subscripts
Nomenclature
xix
𝑎𝑢𝑥
auxiliary
𝑐
compressor
𝑐𝑎𝑝𝑡𝑢𝑟𝑒
power plant with carbon dioxide capture
𝑐𝑎𝑝𝑡𝑢𝑟𝑒𝑑
carbon dioxide captured at a power plant
𝑐𝑔
cold gas
𝑐𝑖
coal input
𝑒
expander
𝑒𝑙
electrical
𝑖
inlet
𝑖𝑠
isentropic
𝑗
sub-system (or component)
𝑚
mechanical
𝑜
outlet
𝑜𝑏𝑦
original base year
𝑝
pumping
𝑝𝑟𝑜𝑐, 𝑐
process contingencies
𝑝𝑟𝑜𝑗, 𝑐
project contingencies
𝑟𝑒𝑓
reference
𝑠
specific
𝑠𝑔
syngas
xx
Nomenclature
𝑠𝑡
steam turbine
𝑡ℎ
thermal
𝑢𝑏𝑦
updated base year
Units
Gt
Giga tonnes
Hz
Hertz
kWh
kilowatt hour
Mt
million tonnes
Mtoe
million tonnes of oil equivalent
MW
Megawatt
MWe
Megawatt electricity
MWh
Megawatt hour
Pa.s
Pascal second
ppm
part per million
ppmvd
part per million volumetric dry
tC
tonne of carbon
TJ
Terajoule
TWh
Terawatt hour
wt%
weight percentage
List of appended papers
Paper I
Sipöcz, N., Mansouri, M., Breuhaus, P., Assadi, M., Development of H2-rich
syngas fuelled GT for future IGCC power plants – Establishment of a
baseline, Presented at ASME Turbo Expo 2011, GT2011-45701,
Vancouver, Canada, June 2011.
Paper II
Mansouri Majoumerd, M., De, S., Assadi, M., Breuhaus, P., An EU initiative
for future generation of IGCC power plants using hydrogen-rich syngas:
Simulation results for the baseline configuration. Applied Energy, 2012, 99:
p. 280-290.
Paper III
Mansouri Majoumerd, M., Raas, H., De, S., Assadi, M., Estimation of
performance variation of future generation IGCC with coal quality and
gasification process – Simulation results of EU H2-IGCC project. Applied
Energy, 2013, 113: p. 452-462.
Paper IV
Mansouri Majoumerd, M., Assadi, M., Fuel change effects on the gas
turbine performance in IGCC application, Presented at 13th International
Conference on Clean Energy (ICEE-2014), Istanbul, Turkey, June 2014.
Paper V
Mansouri Majoumerd, M., Assadi, M., Breuhaus, P., Techno-economic
evaluation of an IGCC power plant with carbon capture, Presented at
ASME Turbo Expo 2013, GT2013-95486, San Antonio, Texas, USA, June
2013.
Paper VI
Mansouri Majoumerd, M., Assadi, M., Techno-economic assessment of
fossil fuel power plants with CO2 capture ‒ Results of EU H2-IGCC project.
International Journal of Hydrogen Energy, 2014, 39: p. 16771-16784.
xxi
Additional papers not included
Paper VII
Mansouri Majoumerd, M., Breuhaus, P., Smrekar, J., Assadi, M., Basilicata,
C., Mazzoni, S., Chennaoui, L., Cerri, G., Impact of fuel flexibility needs on
a selected GT performance in IGCC application, Presented at ASME Turbo
Expo 2012, GT2012-68862, Copenhagen, Denmark, June 2012.
Paper VIII
Nikpey Somehsaraei, H., Mansouri Majoumerd, M., Breuhaus, P., Assadi,
M., Performance analysis of a biogas-fueled micro gas turbine using a
validated thermodynamic model. Applied Thermal Engineering, 2014, 66: p.
181-190.
Paper IX
Nikpey, H., Mansouri Majoumerd, M., Assadi, M., Breuhaus, P.,
Thermodynamic analysis of innovative micro gas turbine cycles, Presented
at ASME Turbo Expo 2014, GT2014-26917, Düsseldorf, Germany, June
2014.
Paper X
Mansouri Majoumerd, M., Nikpey Somehsaraei, H., Assadi, M., Breuhaus,
P., Micro gas turbine configurations with carbon capture – Performance
assessment using a validated thermodynamic model. Applied Thermal
Engineering, 2014, 73: p. 170-182.
Paper XI
Mansouri Majoumerd, M., Assadi, M., Breuhaus, P., H2-IGCC system
integration and techno-economic analysis, Presented at 7th International Gas
Turbine Conference (IGTC-14), Brussels, Belgium, October 2014.
Paper XII
Assadi, M., Mansouri Majoumerd, Jana, K., De, S., Intelligent biogas
fuelled distributed energy conversion technologies: Overview of a pilot
study in Norway, Accepted for presentation at ASME 2014 Gas Turbine
India Conference, GTINDIA2014-8231, New Delhi, India December 2014.
xxiii
1. Introduction
This work is based on a research project co-financed by the European Union (EU)
Seventh Framework Programme for Research and Development. The aim of this project
was to provide and demonstrate technical solutions, which allow the use of state-of-the-art
highly efficient and reliable gas turbines (GT) in the next generation of integrated
gasification combined cycle (IGCC) plants with carbon dioxide (CO2) capture.
1.1. Background information
The rapid growth of the world’s population coupled with the improved living standards
has led to an increasing demand on energy, worldwide. The global economic situation
during the past few years has offset this increase to some extent by the reduction in
industrial activities. However, some factors including global population, living standards
and global economy are foreseen to be the main drivers of the energy demand increase in
coming years. Though there will be a significant improvement in energy efficiency, the
projected primary energy use in the year 2040 will be approximately 35% higher
compared to the 2010 level. Electricity generation will represent the largest driver of
energy demand by 2040 and is expected to account for about half the increase in global
demand for energy [1].
Climate change due to anthropogenic greenhouse gas (GHG) emissions is identified as the
greatest threat to mankind [2]. The major source of these GHGs is CO2 emissions, and the
heat and power sector is identified as the largest contributor to these emissions. CO2
emissions from the heat and electricity supply sector were about 42% of total global CO 2
emissions from fossil fuels in the year 2011 [3]. The present challenge for the power
1
2
Introduction
sector is to meet the ever-increasing demand for electricity and simultaneously mitigate
the greenhouse gas emissions, principally CO2.
Several sustainable solutions have been developed and introduced to cover the additional
demand and to mitigate CO2 emissions in recent years. One possible option is to replace
fossil-based power plants by renewable energy (RE) sources. In 2008, renewable energies
contributed approximately 19% of global electricity demand, and most scenarios foresee a
higher projected share in coming years [4]. Renewable sources are ultimately the most
important option for the future. However, apart from the long timescale for the complete
transformation from fossil-based fuels, RE sources suffer from the fluctuating nature or
variability of production [5]. Given all these aspects, the development of suitable
technology for large-scale power generation using fossil fuels during this transition is
urgently needed.
The other solution to mitigate CO2 emissions is the enhanced use of lower carbon
intensity fuels in power plants instead of e.g. oil and coal, which have higher carbon
contents. In this context, the use of natural gas (NG) for power generation has been
increasing in some countries, mostly due to the fact that it is more environmentally
friendly. However, coal is still the most widely used fossil fuel for large-scale power
generation [5], although the CO2 emissions from coal power plants are almost two times
higher than those from NG-fueled power plants. The International Energy Agency’s (IEA)
New Policies Scenario foresees a 25% increase in coal consumption in the year 2035
compared to the 2009 level. This increase will be 65% based on the Current Policies
Scenario [5]. The security of energy supply due to wider geographical distribution of coal
reserves and availability of abundant resources has promoted coal-fueled power
generation technologies. In addition, factors such as safe storage, easy transportation over
a long distance, less volatile pricing status motivate higher use of coal. Amongst available
reliable technologies for electric power generation, coal plants are still dominating the
market, mainly due to better economic attributes [6].
Carbon capture and storage (CCS) is also one of the key players for decarbonizing the
heat and power supply according to the European Energy Roadmap 2050 [7]. The
deployment of CCS in coal-fired power generation will maintain coal consumption at a
certain level among other fossil fuels under more restricted emissions regulations in
future.
Nomenclature
3
By virtue of the aforementioned aspects, there has been a demand for the development of
a reliable, environmentally-friendly, coal-based technology with the deployment of CO2
capture and storage.
1.2. Objectives
The vision of this PhD thesis, as part of the European H2-IGCC project, was to enable the
application of the state-of-the-art (SOA) gas turbine technology in the next generation of
IGCC plants (i.e. with CO2 capture system) with the flexibility to operate on undiluted H2rich syngas. Figure 1.1 illustrates the structure of the current thesis.
Figure 1.1. The structure of the current thesis
The overall objective of this thesis was, therefore, to provide a detailed system analysis
that generates realistic techno-economic performance indicators for future IGCC plants
with the deployment of CO2 capture. In this regard, special efforts were dedicated to:

Establish and improve a thermodynamic model in order to evaluate the
thermodynamic performance indicators of the IGCC plant with pre-combustion
CO2 capture unit.
4
Introduction






Investigate the consequences of burning H2-rich syngas in IGCC power plants at
both the system level and the GT level.
Investigate the fuel flexibility target by burning non-capture clean syngas instead
of H2-rich syngas and natural gas in the GT.
Update and improve an available GT model using new designed GT
characteristics to cope with the needs of the project and to allow future
adjustments based on the feedback from the project’s partners.
Identify different alternatives for component integration to reach higher plant
efficiency and investigate various configurations of the system in terms of the
plant’s operability.
Develop a tool for economic evaluation and assess capital expenditures
(CAPEX) and operational expenditures (OPEX) for the IGCC plants.
Compare the technical and economic performance indicators of the IGCC plants
with other fossil-based power plants, i.e. advanced supercritical pulverized coal
(SCPC) and natural gas combined cycle (NGCC) plants.
1.3. Limitations
The main focus of this research work was on system modeling and analysis for low
carbon IGCC power plants with special emphasis on the gas turbine technology.
Accordingly, different plant’s components were integrated to establish the selected IGCC
plant with CO2 capture unit. Then, the system was thermodynamically modeled and
analyzed along with continuous modifications of the gas turbine to enable the combustion
of undiluted hydrogen-rich syngas. In addition, a part of the activities was related to
techno-economic assessments of the selected IGCC plant and its fossil-based competitive
technologies. The techno-economic studies performed were exclusively reviewed by some
European utility providers, and economic figures were compared with realistic cost data
provided by project partners. It should be highlighted that the lack of large-scale IGCC
plants with CO2 capture system increases the level of uncertainty in both technical and
economic indicators. However, realistic performance and cost data used in simulations
and techno-economic assessment reflects the current development level of IGCC plants.
Moreover, some of the major alternative plant’s components contributing to efficiency
improvement have been identified and are presented here.
Nomenclature
5
This work does not address the issues revolving around the transport and storage of
captured CO2 and only focuses on the CO2 capture part. Transient (and dynamic)
simulation of the investigated cycles is also outside the scope of this thesis. While
practical heat integration was used for the selected system, thermodynamic optimization is
excluded from the present work.
1.4. Methodology
A literature review was performed to provide a process description as well as performance
data for different components of the IGCC plant based on state-of-the-art technologies.
The collected information, together with data generated during the implementation of this
work, was used for the modeling of the entire IGCC system. The thermodynamic models
described in this thesis were developed using different software tools. The simulation of
power block was performed using the heat and mass balance program, IPSEpro. For this
purpose, existing component models in IPSEpro, as well as certain models developed
during the H2-IGCC project, were used. In addition, the Enssim software, developed by a
member of the H2-IGCC project, was used to simulate and analyze the gasification block.
Simulation and modeling of the gas cleaning process of the IGCC plant were carried out
using ASPEN Plus. After establishing the baseline IGCC plant with CO2 capture, the
performance of different components and the layout of the plant were continuously
modified using the feedback from operators of similar plants.
In order to determine state-of-the-art methodologies for cost estimation in the power
sector, a literature review was performed. A Microsoft Excel-based tool has been
developed for the techno-economic comparison of different power generation
technologies, using performance data obtained from simulations and available cost data in
open literature. A techno-economic comparison between three fossil-based power
systems, i.e. IGCC, SCPC and NGCC plants, was conducted using the developed tool.
Moreover, the developed tool enabled the effects of the variation of different parameters
on the economic indicators to be investigated.
6
Introduction
1.5. Outline of the thesis
The present thesis is a summary of six scientific papers, preceded by an introduction to
the work that provides supplementary information to that presented in the papers.
Chapter 1 provides an overview of the present thesis by means of a brief background to
current energy related issues and an explanation of the objectives and limitations of this
thesis. Chapter 2 presents an extended background to the climate change concern,
different GHG mitigation options and the main CO2 capture methods. Chapter 3 gives a
brief introduction to coal power plants and concisely describes different sub-systems of an
IGCC power plant as well as different challenges for the integration of such an energy
conversion system. Chapter 4 contains the selected IGCC plant configuration, challenges
related to the use of undiluted hydrogen-rich syngas and a general description of heat and
mass balance tools used for the investigations performed during the course of this thesis.
Chapter 5 describes the economic methodology selected for the techno-economic
assessments as well as calculations performed for techno-economic studies within this
PhD project. The main conclusions of this work are presented in Chapter 6, and, finally,
Chapter 7 introduces the papers included in the thesis.
2. Technical background
The provision of energy in its most modern state, i.e. electricity, has been faced with a
major concern over the past two decades, and that is climate change. Climate change due
to anthropogenic greenhouse gas emissions is considered to be one of the most serious
threats to natural ecosystems and human life in the 21st century. The aim of this section is,
therefore, to provide a clear picture about future energy mix and its link to the climate
change concern.
Accordingly, a brief review of the most important energy indicators published by various
organizations will be presented. This review will offer an approximate picture of future
energy growth. The second part of this section is dedicated to the climate change concern
and greenhouse gas emissions. An overview of different mitigation policies to stabilize
greenhouse gas concentration in the atmosphere will be given with a focus on carbon
capture and storage. Finally, the most mature carbon capture technologies will be briefly
described.
2.1. Growing energy demand
World total primary energy consumption was 12,470 Mtoe in 2012 [8]. Global population,
global economy, energy-intensity of the global economy and living standards are the main
drivers of the world’s energy demand [9]. The global population will increase more than
30% from 2013 to 2050 and reach nearly 9.6 billion [10]. This number shows another 2.4
billion energy consumers, mainly in Asia and Africa. During recent years, the global
economy has faced the worst recession since the Second World War [11]. It began with
the crisis in the United States in 2007-08 and then the EU zone faced a weak economy.
However, using stronger measures to stimulate the economy has had a positive effect and
7
8
Technical background
a marginal economic growth has been visible recently. The projected gross domestic
product (GDP), which is an indicator of economic growth, is expected to increase at an
annual average rate of 2.8-3.7% in the next two to three decades [1, 12, 13]. Most of this
growth will come from emerging economies and non-OECD (Organization for Economic
Co-operation and Development) countries. Meanwhile, the projected GDPs of China,
India, and Africa are expected to grow by an annual average rate of about 4-5% until
2040. Global economic growth will then slow gradually as the emerging economies
become mature. Together with the rapid growth of the economy, urbanization,
industrialization and increased living standards are also projected for the future. The
projected urbanization rate for 2035 is 61%, compared to 51% in 2010 [13]. The greater
part of this increase will again come from non-OECD countries, where people want to
reach higher living standards. The shift of population to cities means a greater number of
homes and higher average energy consumption compared to rural areas, although it
enables people to have access to more efficient energy use.
As mentioned, all the fundamental drivers of energy demand (except energy-intensity)
will continuously grow in the coming decades. Energy efficiency will continue to play a
major role in moderating the energy growth. The energy-intensity (energy consumption
per capita) projection shows a downward trend worldwide over the coming decades [13].
This is an indicator of more efficient utilization of energy. However, improved living
standards will outpace energy efficiency and will ultimately result in a higher demand for
energy in the future. Therefore, global primary energy consumption, with a small
variation between data from different scenarios and organizations, is projected to grow at
an average annual rate of 1.2-1.6 % over the period of 2010 to 2030 [1, 12, 14]. To
conclude this section, the need for a greater supply of energy through the development of
efficient technologies seems inevitable in the future.
2.2. Climate change
The change in the state of the climate as a result of human activities, in addition to natural
climate variability, represents a potentially serious threat facing humanity in the 21st
century. Some of the variations in climate, which have been observed during past years,
are [15]:

Increase in global average air and ocean temperatures;
Technical background




9
rise of global average sea level;
decrease in snow and ice extent;
change in hydrological systems, e.g. increased runoff and warming of lakes and
rivers; and
changes in terrestrial, marine and freshwater biological systems.
According to the Fifth Assessment Report (AR5) of Intergovernmental Panel on Climate
Change (IPCC), “Warming of the climate system is unequivocal”. Moreover, “it is
extremely likely that human influence has been the dominant cause of the observed
warming since the mid-20th century”. The global average combined land and ocean
surface temperature increased 0.85 °C from 1880 to 2012. A comparison between the
average temperature of 1850-1900 and of 2003-2012 shows a total increase of 0.78 °C
[16].
2.2.1. Greenhouse gas emissions
A change in the atmospheric concentration of greenhouse gases such as carbon dioxide,
methane (CH4), nitrous oxide (N2O) and halocarbons altered the energy balance of the
climate system and is considered as one of the main drivers of climate change. It is very
certain that the anthropogenic increase in greenhouse gas concentrations, together with
other anthropogenic forces, is responsible for more than half of the observed increase in
global average surface temperature from the mid-20th century to 2010 [16]. Figure 2.1
shows the share of different anthropogenic greenhouse gases in total emissions in 2004
[15].
CO2 (other)
3%
CO2
(deforestation
, decay of
biomass, etc.)
17 %
CH4
14 %
N2O
8%
Fluorinated
gases
1%
CO2 (fossil
fuel use)
57 %
Figure 2.1. Global anthropogenic greenhouse gas emissions in 2004 (data adopted from IPCC AR4)
10
Technical background
The global atmospheric concentration of CO2, which is widely recognized as the most
important GHG contributor to global warming, has increased from a pre-industrial value
of about 278 to 394 ppm in 2012 [15, 17]. Figure 2.2 illustrates that CO2 concentration
has increased almost 25% between 1958 and 2012 [17]. The combustion of fossil fuel,
some energy-intensive industrial processes, land use changes (mainly deforestation),
agriculture and domestic waste disposal are the most important contributors to the
growing CO2 emissions [15, 18].
400.00
Average CO2 concentration (ppm)
390.00
380.00
370.00
360.00
350.00
340.00
330.00
320.00
310.00
1955
1965
1975
1985
Year
1995
2005
2015
Figure 2.2. Annual mean atmospheric carbon dioxide concentration at Mauna Loa Observatory,
Hawaii, USA
Given the negative effects of increasing GHG emissions, the United Nations Framework
Convention on Climate Change (UNFCCC) was set up as a first international climate
treaty in 1992. The aim was to mitigate climate change due to the global temperature rise
and to cope with its inevitable impacts. A few years later, the need for stronger measures
to limit the increasing GHG emissions resulted in the adoption of the Kyoto Protocol (KP)
on December 11, 1997 in Kyoto, Japan [19]. The main objective of this international
agreement was to legally commit its parties by setting internationally binding targets and
timetables for reducing GHG emissions. The protocol came into force in 2005, and a
heavier burden has been placed on developed nations. This is mainly because such
countries are recognized as principally responsible for the current levels of GHG
emissions in the atmosphere due to their industrial activities over the past two centuries.
The average emission reduction target for the first commitment period of this protocol
Technical background
11
(2008-2012) was 5% from the 1990 levels [19]. During the second commitment period
from 2013 to 2020, the target was set to at least an 18% reduction of GHG emissions from
the 1990 levels [20].
2.2.2. Climate change and the power sector
Currently, about 37% of global primary energy is consumed by electricity generation. The
global electricity generation was 22,126 TWh in 2012 [21], with an annual average
growth rate of 2.95% from 1990 to 2012 [8]. Fossil-fuel electricity generation accounted
for 68% of the total generation and coal, the most carbon-intensive fossil fuel, was the
largest contributor to the electricity supply in 2012. Figure 2.3 shows the share of all
sources in the production of electricity in 2012 (data adopted from key world energy
statistics 2013, International Energy Agency [21]).
Other Oil
4.5 % 4.8 %
Hydro
15.8 %
Natural gas
21.9 %
Nuclear
11.7 %
Coal
41.3 %
Figure 2.3. Electricity generation from various sources in 2012
The ever-increasing world demand for electricity generation represents the largest driver
of demand for primary energy consumption. The demand for electricity is projected to
grow more rapidly than the increase in total energy consumption over the next few
decades [1, 12]. This demand will be almost 70% higher in 2035 than the current
electricity demand [22].
As mentioned earlier, CO2 is the most important greenhouse gas contributing to climate
change and the power sector is identified as the single largest sector emitting CO2.
12
Technical background
According to the IEA, CO2 emissions from the electricity and heat supply sector
constituted about 42% of total global CO2 emissions from fossil fuels in the year 2011 [3].
2.3. Mitigation policies
The successful implementation of efficient mitigating measures is extraordinarily vital to
stabilizing GHG concentration in the atmosphere. The mitigation of greenhouse gas
emissions can be achieved through a wide variety of measures and tools in different
sectors including energy, industry, agriculture, forestry, etc. The focus of this sub-section
is on the mitigation options for the energy sector, which has the highest importance in
terms of sectorial share of global GHG emissions [9]. These options can ultimately reduce
the GHG emissions per unit of energy consumption through the following actions:



Energy conservation and efficiency improvement;
transformation/replacement of carbon-intensive fossil fuels by cleaner
technologies such as switch from coal to natural gas, enhanced use of RE
sources, and enhanced utilization of nuclear energy; and
reduction of CO2 emissions using CCS while utilizing energy from fossil fuels.
Improving energy saving and efficiency is a priority within all mitigation policies [7].
Energy saving could be performed using more stringent minimum requirements for
appliances and new buildings, high renovation rates for existing buildings and the
establishment of energy savings obligations on energy utilities. The reduction of GHG
emissions due to the efficiency improvement will, however, be gradually decreased
because of the associated cost of further improvements [9]. On the contrary, less carbonintensive technologies such as RE and GHG emissions abatement through CCS will be
more attractive because of their decreasing costs as a result of technological maturity [9].
There has been a great deal of speculation on the further utilization of nuclear energy
since 2011 after the Fukushima accident at the Fukushima Daiichi plant in Japan. Soon
after this accident, a few countries such as Germany and Switzerland adopted constrained
nuclear energy scenarios, which allow the retirement of plants over their lifetime or
earlier, without commissioning any new installations. Some countries will maintain the
total deployment of nuclear energy at current levels. Nevertheless, economic
considerations as well as the security of the energy supply would result in the domination
Technical background
13
of pre-Fukushima nuclear scenarios coupled with tighter safety requirements [23] as
several new nuclear plants are currently under construction around the globe [24].
Renewable energy has considerable potential to play an important and increasing role in
achieving GHG mitigation targets and to eliminate GHG emissions from the combustion
of fossil fuels. These energy sources are undoubtedly the only option for the future.
During recent years, many RE technologies have become increasingly market
competitive, resulting in a significant increase in their global deployment [4]. Renewablebased electricity generation is expected to continue growing over the next few decades
due to high government support and declining investment costs. Under different IEA
scenarios, the share of RE sources in total electricity generation rises from 20% in 2011 to
25-48% in 2035. However, producing energy from renewable sources is still not wholly
mature and cannot meet the present demand fully in an economic and feasible way. The
estimated timescale for the complete transformation from fossil fuels to renewable
resources is not definite and is likely to be a significant time away [5].
Carbon capture and storage could constitute an important part of the mitigation portfolio
for the stabilization of atmospheric greenhouse gas concentrations over the course of the
21st century [7, 9]. The deployment of this decarbonization strategy in the power sector
will maintain continued utilization of fossil fuels and the available infrastructure, while
also limiting the anthropogenic CO2 emissions in the near future. The widespread
deployment of CCS technologies might also prevent a drastic falling of fossil fuel
consumption as a result of the higher share of RE sources and more stringent emissions
regulations in the future.
It should be clearly underlined that no single mitigation option can provide all of the
emission reductions required for the stabilization of atmospheric GHG concentrations
[25]. Thus, a portfolio containing all the aforementioned mitigation options is necessary to
provide a comprehensive package of different sustainable solutions to tackle increasing
anthropogenic GHG emissions.
2.3.1. Carbon capture and storage
Carbon capture and storage is an essential measure designed to curb global CO 2
emissions. The commercial realization of the CCS process involves three main steps a)
separation of CO2 from industrial/energy-related sources, b) transport of the
14
Technical background
predominantly captured CO2 to a storage site (using high pressure pipelines, trucks, or
vessels) commonly in a supercritical state, and finally c) long-term isolation from the
atmosphere (please note that the latter two steps are not covered in this thesis). The
separation of the CO2 emissions, so called capture, is usually regarded as the most
expensive component in the CCS chain.
The main application of CCS is most likely to be in the large point sources (due to technoeconomic aspects) such as fossil-fuel power plants, fuel processing plants as well as other
industrial plants such as iron, steel and cement production plants. The application of CCS
in power generation, industries and fuel transformation has a mitigation potential of up to
20% by 2050 according to IEA scenarios [26]. Despite the great progress achieved in the
development of highly effective capture technologies, large-scale CCS application is not
yet commercially available for the power generation sector [27]. Although considerable
global efforts were under way to develop efficient and affordable CCS technologies, some
barriers towards the widespread deployment of CCS-related technologies remain unsolved
such as:







Lack of international agreement on cutting CO2 emissions [28];
public perception and knowledge of CCS [28];
legal and regulatory aspects such as lack of regulations on CO2 quality for
transport and storage and lack of required assessments of pipelines and storage
sites [18, 28];
market and political issues such as carbon credits and uncertainty of future
carbon costs [28];
High risk for leakage and other safety aspects associated with transport and
injection of CO2 in the designated storage sites;
high capital-intensity of most CCS technologies [26]; and
lack of commercial-scale demonstration plant, high efficiency loss, technical
maturity and uncertainties for CCS application in power plants.
2.3.2. The European Union climate strategy
Taking serious actions to mitigate the dangerous effects of global warming has been one
of the European Union’s strategic priorities during the last two decades. This will ensure
more sustainable and secure energy systems. To limit the increase of the global average
Technical background
15
temperature so that it does not exceed 2 ºC higher than the pre-industrial level, the
European Council adopted three targets to meet by 2020 in relation to the 1990 level [29]:



Reduction of GHG emissions by 20%;
improving the energy efficiency by 20%; and
increasing the share of renewable energy to 20%.
The EU has also implemented several measures to reach these targets and to stimulate the
economy and job market. Measures include, but are not limited to, are establishing the 1st
international carbon market, the EU emissions trading system (ETS), assigning national
targets for domestic GHG reduction and increasing RE sources, setting up new standards
to improve energy efficiency and reducing GHG emissions in the transport sector [30, 31].
In implementing foregoing actions and strategies, Europe has made a good progress
towards its target for GHG emissions reduction. The estimation for combined emissions
from the European member countries was about 18% below the 1990 level in 2012 [31].
The share of renewable energy sources in gross final energy consumption was about 14%,
and some countries could have already achieved their 2020 targets in 2012 [32].
The European Commission has recently announced its 2030 climate and energy goals,
including: GHG emission reduction by 40% below the 1990 level, increase of renewable
energy by at least 27% and renewed ambitions for energy efficiency policies. The
European Union aims to achieve a competitive, secure and low-carbon economy, while
maintaining the affordability of energy for end-users [33].
2.4. Various capture technologies in the power sector
The previous background information showed the necessity for CO2 emissions reduction
in the power generation sector. Various CO2 separation methods have been developed and
utilized by industry for many years. However, these technologies have not been
commercially applied in the power sector through CCS application. All of the currently
available technologies for large-scale CO2 separation require both significant additional
equipment and energy input than the standard power plants without capture [34]. Progress
in many directions connected to CO2 capture technologies has been rapid, and many
innovative concepts have been developed during the last decade. Concepts like chemical
16
Technical background
looping combustion (CLC), membrane and adsorption technologies have been explored to
find more energy-efficient and less expensive approaches [35].
In spite of extensive development of the aforementioned emerging technologies, the
timescale for the deployment of each technology in the power generation sector and its
current development status differ. Due to the urgent need for successful demonstration of
capture projects, it is important to check the near-term prospect of each capture approach
to reduce the number of available options. The current section will, therefore, give an
overview of the proven technologies for commercial CO2 capture deployment in fossilbased power plants. The available approaches for this purpose are often divided into the
three following categories:



Post-combustion capture from combustion flue gas;
pre-combustion capture or de-carbonization of the fuel stream; and
oxy-fuel combustion or direct combustion of fuel with oxygen (O 2).
These three approaches are shown in Figure 2.4. These technologies can be applied to
both gas-fired and coal-fired power systems.
Post-combustion capture
N2, O2, H2O
Fuel
Power & Heat
CO2 separation
Air
CO2
Pre-combustion capture
N2, O2, H2O
Fuel
Reforming/
Gasification
Steam or
Air/O2
Syngas
Shift reaction,
gas clean-up +
CO2 separation
H2
Power & Heat
CO2 dehydration,
compression,
transport, and
storage
Air
CO2
Oxy-fuel combustion
Recycled CO2&H2O
CO2
Fuel
Power & Heat
O2
Air
N2
Air separation
Figure 2.4. Technical options for CO2 capture from fossil-based power plants
Technical background
17
Regardless of the CO2 capture type, the following common challenges need to be
addressed by further development in CO2 capture technologies:


The complexity of the power plants inevitably increases with deployment of CO2
capture.
The operability and flexibility of the power plants are negatively affected by
deployment of CO2 capture. In particular, these items need to be assessed: the
dynamic/transient behavior of the plants during start-up, shut-down and loadchanging conditions.
2.4.1. Post-combustion CO2 capture
The post-combustion CO2 separation comprises the removal of carbon dioxide from flue
gas after combustion of the fuel. The small fraction of CO2 in the flue gas, which is mixed
with other combustion products and a large fraction of nitrogen from atmospheric air,
makes capture difficult. There are four main processes which can be utilized for largescale CO2 removal from flue gases:




Absorption using re-generable liquid solvents;
cryogenic separation anti-sublimation;
membrane technology; and
adsorption using solid adsorbents.
The absorption process, by means of a re-generable chemical solvent (Figure 2.5),
typically based on a form of amine, is currently considered as the most
common/developed technique for post-combustion capture [36, 37]. Other CO2 separation
methods still need more research and development attention to achieve mature and costeffective processes.
The solvent is counter-currently being contacted by the sour gas (gas containing CO2)
from the top of the absorber column. From the absorber bottom, the CO2-rich solvent is
then transferred to a regenerator where it is stripped of the CO 2 by heat transfer (e.g. heat
release from steam). The regenerated or CO2-lean solvent is cooled via a lean/rich solvent
heat exchanger and recirculated to the top of the absorber, completing the cycle.
18
Technical background
CO2
Lean gas
Condenser
Lean pump
Lean amine
cooler
Absorber
Blower
Rich/lean
solvent HEX
Stripper
Steam
Reboiler
Rich solvent
Flue gas
Direct
contact
cooler
Treatment
Figure 2.5. The schematic configuration of the conventional absorption system
One of the most important features of the post-combustion capture is that it can be applied
to newly designed or existing fossil-fuel power plants. In addition, this capture approach
can be applied to other industries such as cement production, oil refining and
petrochemicals. Moreover, the impact of this approach on the power conversion process is
marginal [38], especially when an external heat source for regeneration is applied.
However, the capture process is less efficient due to the low concentration of CO2 in the
flue gas [38], which is typically between 3 and 15 vol% depending on the fuel type [18].
The major challenge ahead for the widespread deployment of post-combustion capture is
the relatively large parasitic load on the power plant due to the energy intensive solvent
regeneration process [38, 39]. Other secondary challenges are the high capital costs
required for the capture unit and to develop proper solvents (in the case of the absorption
method) with low degradation rate and volatility with fewer negative environmental
impacts.
Technical background
19
2.4.2. Pre-combustion capture
Carbon dioxide can also be separated prior to the combustion process by converting the
fuel to CO2 and hydrogen (H2) and removing the CO2 from the fuel gas. The following
main processes can be utilized for large-scale CO2 separation from the fuel stream:




Absorption using physical or chemical solvents or hybrid system using
physical/chemical solvent;
pressure/temperature/electric/vacuum swing adsorption;
membrane technology; and
calcium oxide carbonation.
Similar to post-combustion, the absorption process is the most preferred technology for
pre-combustion capture, more specifically using physical solvents when the pressure of
the syngas is high. The other technologies are still at an early stage of development, and
many uncertainties remain concerning the performance of these individual technologies
when integrated into the rest of the power plant [18].
One of the promising technologies that could benefit from pre-combustion capture is the
integrated gasification combined cycle. The block flow diagram of the IGCC power plant
with CO2 capture is illustrated in Figure 2.6. The synthesis gaseous product (often known
as syngas) leaving the gasifier, where partial oxidation of the fuel (e.g. coal or oil) occurs,
is mainly a mixture of H2 and carbon monoxide (CO). By the addition of steam, the CO
content of the syngas is catalytically shifted to CO2. The CO2 is finally removed from the
H2 and the hydrogen-rich syngas is used as fuel in a gas turbine. The high CO2
concentration after CO-shift reaction allows efficient de-carbonization of the fuel stream.
Therefore, pre-combustion imposes a lower energy penalty than for post-combustion with
similar size and duty [40-42].
The other technologies which have great potential for pre-combustion capture are H2
production plants using steam reforming, partial oxidation and auto-thermal reforming of
natural gas or light hydrocarbons [18]. Detailed descriptions of these technologies are
available in standard textbooks and hence are not given here.
20
Technical background
Sulfur recovery
Sour gas
Fuel
Raw
Gasification
syngas
Gas cooling/
dedusting
Acid gas
removal
Shift reaction
Steam
Air/O2 Slag
Clean syngas
Liquid CO2
N2
N2
Air
Air separation
unit
Power & Heat
H2-rich syngas
CO2 removal
CO2
CO2 compression &
dehydration
Air
Flue gas to
atmosphere
Electricity to
grid
Stack
Figure 2.6. The block flow diagram of the IGCC power plant with CO2 capture
As with CO2 captured at higher pressure level, compression energy demand and capture
unit size (and consequently costs) are lower than those for post-combustion capture [40].
This capture approach also has the following advantages:


Chemical processing of the syngas (as in IGCC plants) coupled with CO2 capture
offers a wide range of products (e.g. H2, Fischer-Tropsch fuels) and a wide range
of downstream equipment such as gas turbines and fuel cells [40].
Due to the higher pressure and lower volume of the syngas flow to be treated in
the capture unit, the size of the capture unit is smaller than for the postcombustion capture, where flues gases are treated.
Although the pre-combustion approach offers a less-expensive CO2 mitigation
technology, the incorporation of this method has major effects on the power conversion
process [38]. This drawback limits the application of pre-combustion on the existing
power plants.
It should be highlighted that the current thesis will focus on the application of postcombustion CO2 capture in IGCC plants and its effects on the techno-economic
performance of the entire plant as well as on the GT unit.
Technical background
21
2.4.3. Oxy-fuel combustion
Oxy-fuel combustion is the last promising approach described in this chapter designed to
support the separation of CO2 from fossil-based power generation. Similar to other carbon
capture approaches, oxy-fuel has yet to be commercially deployed, while it has a lower
technological readiness than the two latter capture technologies [28, 36, 39].
The nitrogen content of the air is almost 80 vol%, which dilutes the CO2 concentration in
the flue gas from the combustion process and makes the downstream CO2 capture process
costly. In the oxy-fuel combustion process, N2 is removed from the air by means of a
large-scale air separation unit (ASU) before the combustion (refer to Figure 2.7). This
process comprises a combination of high purity oxygen (typically around 95 mol%) and
recirculation of the flue gas for combustion of the fuel. The combustion product is a gas
consisting mainly of concentrated CO2 and water (H2O). Such a process (i.e. pure O2
combustion) has a combustion temperature of about 3500 °C [18]. Current materials
cannot handle such a high temperature. Oxy-fuel combustion is not feasible for currently
available gas turbines in natural gas combined cycles since their compression and
expansion systems are not suitable for CO2 as the main working fluid instead of N2 in air
[43]. Recirculation of a part of flue gas is, hence, to control the flame temperature and
consequently NOx formation in the boiler. In addition, this recirculated stream
compensates for the missing N2 flow to carry the heat through the boiler [27, 28]. This
stream is also used to feed the fuel to the boiler in the case of coal-firing plants [28]. The
recycle stream is about 60-70% of the flue gas, depending on the fuel composition [28].
The rest of the flue gas that is not recycled is then treated for undesirable components
such as particulates and sulfur removal. The clean flue gas is finally compressed, cooled
and purified from water vapor by condensation. The final product is predominantly CO 2,
which is ready for transport and storage.
22
Technical background
Flue gas to
atmosphere
Power
N2
Air
separation
unit
Air
Pulverized
coal
O2
Gas-gas HEX
Electrostatic
precipitator
Flue gas
desulfurization
Flue gas
condenser
Fly ash
Sulfur product
Water
To CO2
purification,
compression,
transport &
storage
Flue gas secondary recirculation
Coal
Coal mill
Flue gas primary recirculation
Figure 2.7. The block flow diagram of the pulverized coal plant with oxy-fuel combustion
One of the advantages of the oxy-fuel combustion is its significantly lower size of capture
unit compared to other technologies by combusting the fuel using purified oxygen [38].
Furthermore, the oxy-fuel combustion eliminates the need for conventional CO2 removal
technologies using chemical or physical absorption. Significant cost and energy savings
can, therefore, be realized. Moreover, oxy-fuel combustion offers flexibility for the
positioning of O2 injection either into the recycled stream (as pre-mixed condition) or
directly to the burner compared to the air combustion, which may help to control pollutant
emissions e.g. CO emissions [28]. However, the major challenges of oxy-fuel combustion
also revolve around a drastic change of working fluid from conventional air combustion to
a mixture of mainly CO2 and water vapor. The other technical uncertainties regarding the
commercial deployment of oxy-fuel combustion are as follows:


The need to supply high purity oxygen results in a large efficiency penalty using
energy-intensive processes such as conventional cryogenic distillation. Such a
process is not economically viable for oxy-fuel combustion [28] and can be
replaced with emerging technologies such as ion transport membrane (ITM)
technology to reduce the costs and energy consumption.
Oxy-fuel combustion has a high impact on the power plant process, which
complicates retrofitting existing plants [38]. This approach needs substantial
modifications (redesign) for the GT or conventional steam boiler technologies
(more specifically combustion system) due to the drastic change in the working
fluid [38, 44].
Technical background




23
This approach cannot be applied just to a fraction of the main stream, so-called
slipstream, similar to post- or pre-combustion capture. This causes such an
approach to be applied only to a complete power plant module [39].
The high proportions of CO2 and H2O in the flue gases (compared to air
combustion) result in higher gas emissivity (radiative heat transfer) [27]. Thus
more sophisticated and expensive materials are required to be resistant against
the higher heat transfer rate [38], and a large volume of the flue gas needs to be
recirculated to offset the higher gas emissivity. Full-scale demonstration
boilers/GTs are required to validate radiative CFD models and thereby provide
accurate predictions of heat transfer between working fluid and materials [28].
In the case of boilers, the high concentration of CO2 (which has a high solubility
in water), high level of sulfur, and chlorine species generate a corrosive media
which requires particular caution when selecting proper materials [28, 38].
The condensation of water in the presence of a substantial amount of CO2 needs
to be carefully assessed [38].
3. Coal-based power plants
The current chapter firstly includes a synopsis of energy supply by coal and CO2
emissions from coal-based power generation technologies. Different coal-based power
systems are then reviewed. An overview of different available technologies and
components which constitute an integrated gasification combined cycle power plant is
further presented and discussed. Finally, the existing IGCC plants are listed and their
specifications are reviewed. The outline of Chapter 3 of this thesis is shown in Figure 3.1.
Why coal-based power
plants?
(3.1)
Coal-fired power generation
(3.2)
Integrated gasification combined cycle (3.3)
Air separation
Gasification
(3.3.1)
(3.3.2)
Syngas cleaning
and conversion
(3.3.3)
CO2 compression
and dehydration
(3.3.4)
Current IGCC power plants
status
(3.4)
Figure 3.1. Outline of Chapter 3
25
Gas turbine
(3.3.5)
Bottoming
cycle
(3.3.6)
26
Coal-based power plants
3.1. Why coal-based power plants?
Coal showed 2.5% higher consumption in 2012 compared to 2011 and has been the
fastest-growing fossil fuel during recent years [8]. The global distribution of proved coal
reserves (total: 860,938 Mt), together with coal consumption (total: 3,730 Mtoe) and
production (total: 3,845 Mtoe) are shown in Figure 3.2 (data from Ref. [8]). The IEA’s
New Policies Scenario projected a 25% increase in coal consumption in the year 2035
compared to 2009, while the other IEA scenario, Current Policies, predicted 65% higher
coal consumption than the level of 2009 [5].
2%
14 %
29 %
31 %
12 %
1%
4%
4%
68 %
35 %
0%
0%
(b)
(a)
North America
1%
12 %
South & Centerl
America
Europe & Eurasia
14 %
0%
70 %
3%
Middle East
Africa
Asia Pacific
(c)
Figure 3.2. Global share of (a) proved coal reserves, (b) coal production, and (c) coal consumption
at the end of 2012
Coal-based power plants
27
The main reasons for the continued utilization of coal are the abundant resources of coal
(more than 100 years with current proved reserves and consumption rate), its widespread
availability and less-volatile price compared to the other fossil fuels [8, 45, 46]. Thus, coal
has emerged as the most widely used fossil fuel for large-scale power generation, though
natural gas use is also increasing, mostly in localities of availability due to the fact that it
is more environmentally friendly [5].
Coal is not only the most used fossil fuel for power generation but it is, unfortunately, the
most polluting fuel due to its heavy carbon content per unit of energy released. Carbon
content is 15.3 tC/TJ for natural gas, while it is almost double, 25.8 to 28.9, for different
types of coal [47]. Carbon dioxide emissions from coal for heat and power production
were 8.9 Gt CO2 in 2011, about 28.5% of the global anthropogenic CO 2 emissions [3].
Given the continued need to use coal as primary fuel and requirements to limit its CO 2
and conventional emissions, a genuine demand for the development of reliable and lowemission coal technology has been generated. In addition, some political actions have
been taken which are in favor of clean coal technologies. New regulations which enforce
the construction of new coal-fired power plants with CCS demonstration or CCS ready
capability are among those [27]. However, it should be mentioned that due to low costs
for CO2 allowances, there is still not enough pressure on owners to build plants with CO2
capture.
3.2. Coal-fired power generation
The major coal-based power generation technologies available today are pulverized coal
combustion (PCC), fluidized bed combustion (FBC) and IGCC. This sub-section will
briefly touch on PCC and FBC technologies, while the rest of this chapter will focus on
IGCC.
PCC technology has been dominating coal-fired electricity generation worldwide for
almost 100 years. Figure 3.3 illustrates the block flow diagram of the typical pulverized
coal-fired plant (with post-combustion CO2 capture unit). Typical operating parameters of
pulverized coal plants using a sub-critical steam cycle are 163 bar pressure and a
temperature of 538 °C for both superheat and reheat [48]. The efficiency of a steam cycle
is a function of the steam pressure, superheat and reheat temperatures, which are all
28
Coal-based power plants
dependent to the advances in materials that are selected for the boiler and turbine and
pipework connecting them [49]. In order to achieve higher technical performance and
lower emissions, supercritical and ultra-supercritical pulverized coal plants have been
developed. These plants are operated beyond the critical point of water, i.e. 221 bar and
374 °C [49]. Supercritical and ultra-supercritical technologies are more beneficial,
avoiding surface tension between liquid and gas phase and eliminating the use of a drum
for separation of water and steam in sub-critical plants. They are also better suited to
frequent load variations compared to sub-critical boilers. The typical pressure range of
SCPC is more than 245 bar, with the temperature in excess of 550 °C for both superheat
and reheat steam, and the temperature range of ultra-supercritical pulverized coal plants
(USCPC) is around 600°C or higher [48].
Although the share of SCPC and USCPC plants currently under construction or planned is
increasing, sub-critical technology has still continued to dominate coal-fired power plants.
However, the share of supercritical and ultra-supercritical plants might be increased with
stricter requirements for CO2 emissions [50].
Steam/water to and from capture plant
Feedwater
heater
system
Air
Ammonia
Oxidation air
Electrostatic
precipitator
Pulverized coal boiler
Water
Coal
Limestone
Flue
gas
Flue
gas
Pre-heated air
Coal and ash
handling
DeNOx plant
FGD
Capture unit
Bottom ash
CO2 product
Fly ash
Effluent
HP
turbine
IP
turbine
LP
turbine
Gypsum
Condenser
Flue gas
Stack
Figure 3.3. Block flow diagram of typical pulverized coal-fired plant with CO2 capture
Coal-based power plants
29
The other coal-based plant, i.e. FBC technology, contributes in niche applications, e.g. in
the combustion of low quality coals [51]. This technology offers both atmospheric and
high pressure operation [52]. A fluidized bed combustion system is generally
characterized by acceptable availability and fuel flexibility and has a good emissions
performance. Its emissions control is usually cheaper than that of PCC technology.
Although development efforts have been focused on scaling up the technology, the
capacity of this technology is far behind the conventional PCC plants [53]. Similar to
SCPC and USCPC technologies, the development of advance materials to cope with
higher pressure and temperature will improve the technical performance of this
technology [27].
Carbon dioxide emissions can be captured from both the abovementioned technologies,
i.e. PCC and FBC, using oxy-fuel and post-combustion methods. However, challenges
including the parasitic effects of CCS using post-combustion (refer to Chapter 2) require
technical progress such as achieving higher plant efficiency.
The other alternative coal-fired technology is the IGCC power system. The integrated
gasification combined cycle is currently one of the most promising technologies for the
efficient use of coal. This technology enables the conversion of coal (and other solid or
liquid fuels) into synthetic gas fuel, while still maintaining ambitious emissions targets
and high efficiency. IGCC technology benefits from its widely known environmental
credentials such as low emissions of sulfur dioxide (SO2) and nitrogen oxides (NOx) [54].
Although this technology suffers from high capital costs and is perceived to be more
complex than other technologies, e.g. pulverized coal plants, its significantly better
emissions performance is of high interest for future large-scale deployment [27, 49]. In
addition, the IGCC technology offers co-gasification of biomass1, good performance with
lower grade coals and other feedstock [55], as well as co-production of H2 and electricity
[56].
Moreover, IGCC technology is technically well suited for CO2 capture. If CCS becomes
mandatory for the next generation of fossil-based power plants, high-efficient precombustion carbon capture methods can be incorporated into the IGCC system. The
additional cost due to the capture unit will be significant but probably lower than for PCC
1
The biomass co-gasification can be utilized to achieve even a CO2-free or CO2-negative condition
when CO2 capture is integrated into the cycle.
30
Coal-based power plants
systems [57]. Regardless of the lack of demonstration activities for IGCC plants with CO2
capture, every component of this system has been commercially utilized in other
industries, such as chemical industries, petrochemical complexes, etc. The heart of the
power generation unit, i.e. gas turbine technology, suited to an IGCC system with and
without carbon capture unit, when the diluted syngas is used, is also currently available on
the market [58, 59].
3.3. IGCC power plant’s components
An IGCC power plant consists of several components which can be categorized in
different sub-systems depending on their main processes. The main sub-systems of an
IGCC plant are as follows:








Coal receiving and storage unit;
air separation unit;
coal milling, drying, and gasification;
syngas cleaning and conversion unit;
water-gas shift reaction unit (in a plant with CO2 capture unit);
acid gas removal (AGR) unit;
CO2 compression and dehydration (in a plant with CO2 capture unit); and
power island consisting of a gas turbine and a heat recovery steam generator
(HRSG), steam turbine, generator, auxiliaries, etc.
With the exception of coal receiving and storage units, process descriptions of the stateof-the-art technologies for main sub-systems together with currently potential alternative
technologies for each sub-system will be presented here.
3.3.1. Air separation
Oxygen supply to the gasifier represents a major part of the energy consumption and
capital costs of any IGCC power plant. The technology currently used for oxygen
production is the cryogenic separation of the air by distillation, a mature technology used
for over 100 years.
In a typical cryogenic air separation unit (refer to Figure 3.4), the air is initially
compressed to a pressure of about 5 bar [18]. It is then purified using multiple fixed bed
adsorption units to remove water, CO2, N2O and trace hydrocarbons. Such components
Coal-based power plants
31
could accumulate to undesirable levels in the cryogenic parts such as the reboilercondenser, causing a blockage (due to freezing of CO2 and H2O) and other safety issues
for the plant operation. The adsorption units are regenerated by either temperature or
pressure swing with a low pressure N2 stream. Then, the air is cooled to about its dew
point by heat transfer with returning products (O2 and N2) in the main heat exchanger. The
air is finally separated into oxygen, nitrogen and, optionally, argon (Ar) streams in the
separation part. The separation process depends on the relative volatility of the more
volatile components (N2 and Ar) relative to less volatile O2. A basic arrangement for the
separation part involves a double distillation column which has a reboiler-condenser
between two columns [60]. The O2 product can be withdrawn from the base of the low
pressure column (upper column in Figure 3.4) either as a liquid or a gas.
Air
Filter
MAC
Pretreatment
Low pressure
column
GOX
GAN
Expander
Product
compression
Reboiler/
condenser
Main
heat exchanger
Subcooler
High pressure
column
Figure 3.4. Schematic configuration of the cryogenic air separation unit
The main parameter controlling the power consumption of a cryogenic ASU is the main
air compressor (MAC) discharge pressure, which is inherently affected by the pressure
32
Coal-based power plants
balance and reboiler-condenser design. Consequently, numerous alternatives for the
configuration of heat exchange, distillation, compression and pumping exist to minimize
the energy consumption of an air separation unit. The second important parameter
affecting power consumption is the number of product streams and their purity. The
higher purity of the O2 product (typically higher than 97%) requires a higher number of
separation stages, which results in higher MAC discharge pressure, capital and operational
costs [60]; hence, there is a trade-off between capital cost and power consumption and the
purity of oxygen.
Generally, the main areas to reduce specific energy consumption and costs are:



Efficiency improvement by integration of ASU with other sub-systems such as
GT compressor;
development of other air separation technologies to reduce specific energy
consumption for O2 production (kWh per unit of O2 product); and
improvement of basic components of cryogenic ASU.
3.3.1.1. Cryogenic ASU and power island integration options
The power generation block of an IGCC plant could be integrated with the air separation
unit (refer to Figure 3.5) through the following ways:


Gas turbine air extraction (full or partial air integration) to supply the ASU; and
N2 supply from the ASU to the GT for dilution purposes (NOx control), for
turbine cooling, for GT power augmentation, and for increase of steam
generation in the HRSG.
Air extracted from a GT compressor can be used to partially or fully supply the
requirements of the ASU, which can be defined according to the following equation:
Air − side integration =
Air to ASU from the GT
Total Air to ASU
(Eq. 3.1)
Full GT-ASU integration means that the feed air for the ASU is completely supplied by
the gas turbine air compressor [61]. The integration between the ASU and the gas turbine
can significantly affect GT performance [62], which will be discussed later in this chapter.
Most European IGCC designs have selected full GT-ASU integration targeting maximum
overall plant efficiency [63]. However, this integration option generates some operation
problems. The main difficulty is control of the ASU when the GT operates at variable
Coal-based power plants
33
load. Increase of GT power output may result in an increased GT compressor discharge
pressure, which causes a pressure rise of the air delivered to the ASU. Consequently, the
boiling pressure and temperature of the liquids in the ASU will be elevated, meaning that
liquids in the columns will be sub-cooled. The net vapor flows will then be reduced, while
the GT combustor requires higher fuel flow for the increased power production and vice
versa. This problem may be efficiently resolved utilizing an expansion turbine before air
injection to the ASU to maintain the discharge pressure of the GT compressor similar to
that required by the ASU. The other problem with full air integration arises during start-up
of the ASU and the gasification system. Gasification needs O2 and N2 (depending on the
technology selected for gasification) to produce syngas (i.e. the GT fuel). Therefore, the
gas turbine should operate on NG or liquid fuels to supply the initial amount of air, or a
supplementary air compressor for start-up of the ASU should be considered.
Partial air integration means that only a part of the air required for the ASU is supplied
from the GT compressor and the rest is provided by a supplementary compressor. This
configuration allows the GT system to be started after the start-up of the ASU and
gasification processes. The amount of air flow to be withdrawn from the GT compressor
depends on the air flow required for the plant start-up (to be supplied by a separate
compressor) and the amount of air available from the GT (based on the GT design [64]
and the prevailing atmospheric conditions). The optimal situation is to ensure that the
overall loading of the GT expander is maximized (choked). The required thermal energy
input of the GT shows a substantial increase in fuel gas flow in the case of using diluted
syngas fuel due to its significantly lower calorific value compared to that of NG. The
additional fuel flow (compared to the NG case) possibly results in bleeding of the GT
compressor air to avoid an increase in flow rate expanding in the GT expander, which is
already maximized. Consequently, this is the available air feed which could be allocated
to the ASU [60].
34
Coal-based power plants
To gasification
O2
To atmosphere
N2
Air separation
unit
Syngas
fuel
Air
LP DGAN
Heat recovery
steam generator
Stack
HP DGAN
Air
LP DGAN
Heat
integration
LP DGAN
(and H2O)
Air
Figure 3.5. Integration options of the ASU and the power island
The zero supply of air from the GT compressor to the ASU (or non-integrated air-side
GT-ASU) is usually only optimum when higher operational flexibility, availability, and
reliability of the overall IGCC system is the main concern [65]. However, this option may
be also applied under circumstances when the air flow from the GT compressor is limited
(e.g. due to the re-allocation of the GT compressor air for cooling purposes of expander’s
parts) [60] and there is no need for dilution using N2 coming from the ASU.
In addition to air-side GT-ASU integration, high pressure (HP) diluent gaseous nitrogen
(DGAN) from the ASU can be integrated into the GT as diluent to control NOx emissions
[66]. Nitrogen could be compressed and then heated by the extracted air feed stream from
the GT in the case of partial or full air integration. It should be mentioned that, in the most
recently developed GTs, the margin allowing for extra fuel (or added N2) is limited and
depends on atmospheric conditions [60]. N2 injection could be performed directly to the
combustor or as a mixture with the syngas. This nitrogen can also contribute to increased
power output from the expander [62]. Low pressure (LP) diluent gaseous nitrogen
(DGAN) is commonly used in the ASU as a source to cool the compressed air feed stream
from the GT [67]. Low pressure DGAN together with chilled water from the ASU could
be fed to the inlet of the gas turbine compressor to reduce the bulk air temperature and
thereby increase the air mass flow rate to achieve higher GT power output. The use of
Coal-based power plants
35
nitrogen as diluent also provides the opportunity to exploit higher steam generation in the
HRSG due to the lower dew point of flue gases containing higher amounts of N2.
3.3.1.2. Other ASU technologies
The other important oxygen production technologies are adsorption, polymeric membrane
and ion transport membrane processes. Ongoing research and development will continue
to improve both the economy and the energy efficiency of these technologies. Unlike
cryogenic plants, which need approximately two hours to produce O2 and N2 from a coldcondition start-up, adsorption and membrane systems can be started and powered up to
full load within a few minutes [68]. Adsorption and polymeric membrane systems are less
complex and more passive compared to cryogenic systems. However, neither technology
could yet compete with cryogenics for large-scale O2 production, especially at high
purities. Moreover, neither technology is capable of directly producing argon [69].
Ion transport membrane technology is a breakthrough ASU technology and the most
promising alternative to cryogenic technology for the production of large quantities of
oxygen. Compressed high-temperature air (at about discharge pressure of GT compressor
and 800-900 °C) is electrochemically passed through highly selective ceramic membranes
at high flux. Oxygen on the feed side (i.e. air) is ionized on the surface of the membrane
and diffuses through the membrane as ions forming oxygen molecules on the permeate
side [60]. The primary advantage of such technology is its significant potential for capital
costs reduction compared to cryogenic systems [70]. This potential could be up to 30%
compared to cryogenics in IGCC application [71]. Furthermore, ITM offers the possibility
of providing O2 with a less adverse effect on the efficiency of the power plant than the
cryogenic system, although this performance improvement is not strong (in the range of a
few decimal points) [72]. However, similar to other non-cryogenic technologies, ITM has
shortcomings concerning the production of pure and liquid by-products [69]. Moreover,
this technology needs to be commercially developed and integrated into the IGCC system
[73].
3.3.2. Gasification
The gasification process is one of the most important parts of the IGCC system and has
gained special significance in the context of future generation IGCC plants with CO2
capture. This process is to convert coal (can also be other feedstock, e.g. biomass, liquid
36
Coal-based power plants
fuels, etc.) through sub-stoichiometric reaction with oxidant agents, either air or O2 at a
temperature exceeding 700 °C to produce a synthetic gaseous product [41]. Compared to
conventional pulverized coal combustion, gasification offers great opportunities for both
higher efficiency and improved capture of pollutants. The commercial gasification
technologies can be classified into three categories according to the flow geometry:
entrained-flow, fluidized bed, and moving bed gasification technologies [74]. In most
existing industrial plants, including IGCC power plants, the entrained-flow gasifiers have
had extensive operating experience [75]. Thus, the description of other main categories,
i.e. fluidized bed and moving bed are excluded here and can be found in references [41,
74].
3.3.2.1. Entrained-flow gasifiers
Entrained-flow gasifiers allow high operating pressures (20-80 bar) and temperatures
(1200-1600 °C). The high pressure and temperature environment of the gasifier facilitates
the gasification of the fed coal [76]. However, challenges corresponding to measurement
techniques and instrumentation due to the rigid environment, and possible problems with
slag handling and removal still need more development [77, 78].
High operating temperatures enable a favorable slagging process to remove ash and
render gasification almost tar-free. The released heat results in the melting of the ash
content and the production of molten, inert slag (eventually as the only solid waste).
Meanwhile, and under extremely hot conditions, the carbon content in the coal is
converted mainly to CO due to the reducing environment of the gasifier. This type of
gasifier typically provides a high H2/CO ratio syngas. Apart from these combustible
compounds, products typical for combustion, CO2 and H2O, are also produced. Steam or
other compounds are added to the gasifier to moderate the hot temperature of the process.
Further details such as dominant reactions within the gasifier can be found in Paper III of
this thesis.
The oxidant agent can be oxygen or air. Amongst the various gasification technologies,
oxygen-blown gasification is an attractive process for the production of high calorific
value syngas (mainly due to its high H2 content). The plant’s components (gasifier and
downstream equipment) are also much smaller than that with the air-blown technology
due to oxygen combustion in the gasifier. On the other hand, the absence of a large ASU
in air-blown gasification offers some advantages in terms of capital and operating costs
Coal-based power plants
37
and efficiency. However, the increase in the capital costs associated with a less effective
capture process (due to the removal of CO2 from a larger volume syngas diluted by N2 in
air) offsets the reduced power consumption of an air-blown system [79].
The feedstock can be fed either dry (using N2 as a conveying gas) or wet (using slurry
water) into the entrained-flow gasifier. In a dry-fed system, there is no need for water
evaporation (like for those slurry systems) in the gasifier, leading to high cold gas
efficiencies 1 compared to (single stage) slurry-fed entrained-flow gasifiers [80]. In slurryfed gasifiers, pulverized coal is mixed with water to produce a slurry feed. The typical
range of slurry (ratio of solid to whole mixture) varies from 35 to 70 wt%, depending on
the coal’s characteristics [65, 81, 82]. The slurry type of gasifier utilizes a slurry pump to
feed the slurry into the gasifier, enabling the process to have a higher operating pressure
compared to dry-fed systems. High operating pressures result in a more efficient CO2
separation due to the high partial pressure of CO2 in the syngas. In slurry-fed gasifiers
some CO and H2 burning is required to vaporize the slurry water. The syngas, therefore,
has a relatively high content of the combustion products (i.e. CO2 and H2O), which is
again suitable for the operation of downstream shift reaction and CO2 capture units. The
relatively high operating pressure of slurry-fed gasifiers compared to dry-fed gasifiers
results in a higher partial pressure of CO2 and consequently a lesser energy penalty due to
the removal process [83]. However, the ratio of hydrogen sulfide (H2S) to CO2 is higher in
dry-fed gasifiers, which improves sulfur recovery using a conventional absorption system
unit [84]. In addition, the dry-fed gasifiers show better performance when operating on
low quality fuels (with low calorific values) compared to the slurry-fed. Moreover, the
quality of produced syngas in dry-fed gasifiers is relatively constant compared to slurryfed types, even when low calorific fuel is gasified.
The aforementioned features of entrained-flow gasifiers are very desirable for large-scale
power generation. Hence, almost all the commercially useful coal gasifiers deployed for
large-scale power generation are of this type. Some of the leading companies in the power
sector have patented their gasification technologies, such as Shell Coal Gasification
Process (SCGP), General Electric (GE) gasifier (formerly Texaco), and ConocoPhillips
(E-GasTM) gasifier for O2-blown and Mitsubishi Heavy Industry (MHI) for air-blown
1
This term will be introduced later in this section.
38
Coal-based power plants
entrained-flow type. The main characteristics of commonly used oxygen-blown and airblown entrained-flow gasification technologies are shown in Table 3.1.
Table 3.1. The main characteristics of various commercial gasifiers
Specification
SCGP
GE/Texaco
E-GasTM
Flow regime
Entrained-flow
Entrained-flow
Entrained-flow
Type of ash
Slag
Slag
Slag
Oxidant
O2-blown
O2-blown
O2-blown
Dry/slurry
Dry-fed
Slurry-fed
Slurry-fed
Feed type
PC
PC
PC
Pressurization
Lock hopper
Slurry pump
Slurry pump
Number of stages
Single
Single
Double
Slag removal
Lock-hopper
Lock-hopper
Continuous
system
pressure letdown system
[41]
Flow direction
Upward flow
Downward flow
Upward flow
Boiler position
Side-fired
Top-fired
Side-fired
Quenching type
Quenching with
Full water quench, Two-stage
recycle gas and
radiant cooler, and gasification
radiant cooler
radiant/convective
coolers
Reactor type
Membrane-wall
Refractory-lined
Refractory-lined
[80]
Cold gas
78-83% [80]
69-77% [85]
71-80% [85]
efficiency
Carbon conversion Above 99% [80]
Above 96%
Above 99% [86]
Availability targets
92%[87]
88-90% [83]
92%
MHI
Entrained-flow
Slag
Air-blown
Dry-fed
PC
Lock hopper
Double
Lock-hopper
Upward flow
Side-fired
Two-stage
gasification
Membrane-wall
70-75% [85]
Above 99%
[76]
Not available
3.3.2.2. Gasification performance
Coal properties and characteristics such as ash content and reactivity are amongst the most
important parameters affecting the performance of gasifiers in IGCC application. The
effects of important coal properties on the performance of the gasification process are
briefly presented here. In addition, cold gas efficiency, which is an indicator of
gasification performance, is also introduced later.
Coal-based power plants
39
3.3.2.2.1. Coal quality
There are various types of coal, and each has specific properties. Coal is typically
classified based on the content of fixed carbon and volatile matters. Table 3.2 shows the
four classes of coal with their characteristics and thermal properties [41].
Class
Anthracite
Bituminous
Sub-bituminous
Brown coal (lignite)
Table 3.2. Coal classification
Volatile matter (wt%)
Fixed carbon (wt%)
<8
>92
8-22
78-92
22-27
73-78
27-35
65-73
HHV (MJ/kg)
36-37
32-36
28-32
26-28
An advantage of entrained-flow gasification is its fuel flexibility. This type of gasifier
allows the choice of a wide range of feedstock with different prices, including low-rank
coals with lower prices. The main specifications of low-rank coals (e.g. lignite coals) are
typically high levels of ash, moisture, sulfur, chlorine and alkali metals as well as low ash
melting point [55]. It is estimated that 53% of global coal reserves consist of average and
low-rank coals, i.e. sub-bituminous and lignite [8]. Even though an entrained-flow gasifier
can process a wide range of feedstock [65, 78], the feedstock characteristics significantly
influence the gasification performance [56, 78]. The existing gasifiers show a substantial
increase in cost combined with a drastic reduction in performance operating on low-rank
feedstock, e.g. lignite coals [83]. Nevertheless, the utilization of such types of coals can
broaden the range of suppliers and consequently improve the security of the energy supply
[55].
The main parameters for selection of coal type in IGCC plants are ash content, slag
viscosity and coal reactivity. A low ash content coal is favorable for IGCC power plants
since it produces a lower fly ash and lower bottom slag that can result in a possible
plugging of exit pipes and downstream heat exchangers [78]. The slag viscosity directly
determines the operating conditions of the gasifier. Higher slag viscosity induces the
possibility of a blockage of the slagging system and also requires a higher gasification
temperature, which decreases the lifetime of the refractory materials. In order to have a
continuous slag tapping process, a viscosity less than 25 Pa.s (250 Poise) is required [41,
88]. The viscosity of the slag tends to be high at high concentrations of Al2O3 and SiO2.
Conversely, the viscosity has a tendency to be low if the CaO, MgO and FeO contents are
high [55, 88]. The slag viscosity needs to be reduced when it is higher than the
40
Coal-based power plants
abovementioned critical value. Utilizing a fluxing agent (such as limestone) or mixing
with a coal which has a lower slag viscosity are the main solutions for lowering the
viscosity. Coal reactivity determines the amount of required oxidant agent for gasification.
The lower coal reactivity results in a higher injection of oxidant agent and consequently
lower gasifier performance. In summary, the best coal type for IGCC to reduce operation
difficulties and shutdowns appears to be one which contains low ash, has low slag
viscosity and high coal reactivity [74, 78].
The secondary parameters for coal selection are coal water and sulfur contents. Generally,
coal containing lower surface moisture would be beneficial in terms of lower drying cost
in dry-fed gasifiers and lower oxygen consumption in slurry-fed gasifiers [78]. A higher
sulfur content results in a higher loss of H2 content produced within the process, such as
H2S, and has a detrimental effect on electricity production.
3.3.2.2.2. Cold gas efficiency
One of the main parameters to determine gasifier performance is cold gas efficiency. This
parameter is an indication which shows how much of the energy input has been recovered
as chemical energy in syngas [78]. The cold gas efficiency is defined as:
ηcg =
LHVsg Q̇sg
LHVci ṁci
(Eq. 3.2)
The cold gas efficiency of a single-stage slurry-fed entrained-flow gasifier is lower than
that for dry-fed gasifiers (refer to Table 3.1). A slurry-fed gasifier requires 20-25% more
O2 for vaporization compared to a dry-fed gasifier due to the higher water content
(because of slurry) [76]. Therefore, more carbon in coal is oxidized to CO 2 in the slurryfed gasifier, which reduces the cold gas efficiency. The problem is even larger when the
coal rank is low. The higher moisture content of the coal is not useful for the slurry’s
transport properties and a large amount of water is still required for the slurry.
Consequently, the overall efficiency of the plant is reduced by an increase in the ASU size
and higher auxiliary power demand. This water content results in a higher H2/CO ratio
(details concerning the gasification’s CO-shift reaction are available in Paper III). On the
contrary, the dry-fed gasifier can handle a wide range of feedstock such as any type of
coal with a relatively lower effect on the produced syngas’ properties and cold gas
efficiency [80].
Coal-based power plants
41
3.3.3. Syngas cleaning and conversion
Raw syngas produced in a gasifier contains many impurities such as particulate matters
(PM), heavy metals, undesirable gaseous components such as acid gases, etc. It also
contains a high amount of carbon monoxide, which needs to be converted to carbon
dioxide for CO2 capture application; hence, cleaning, conditioning and conversion of
syngas is required for its efficient use in IGCC applications.
3.3.3.1. Syngas cleaning
The cleaning of the syngas produced in the gasifier is unavoidable before its combustion
in a gas turbine to protect the GT and to keep the pollutant emissions below the
environmental restriction levels [89]. The cleaning process consists of the removal of ash
and particulates, as well as control of ammonia (NH3) and heavy metals (such as mercury,
arsenic, selenium, etc.). It should be mentioned that the separation of H 2S and carbonyl
sulfide COS is excluded here and will be described later in Section 3.3.3.3 (Acid gas
removal).
Most of the coal ash is removed from the gasifier as slag in all entrained-flow gasification
technologies. The remaining ash in syngas is captured in the downstream equipment. The
clean-up configuration strongly depends on the gasification process. Ash and PM control
consists of cyclones, candle filters and a syngas scrubber in the case of the SCGP and the
E-GasTM gasifiers, while it consists of a water quench and a syngas scrubber for the GE
gasifier [90]. De-dusted syngas exiting the water wash scrubber is almost free of
chlorides, NH3, SO2 and PM. Water used for quenching purpose or scrubbing is then sent
to a sour water stripper for treatment. For mercury removal efficiency, the design target is
about 90-95% [90], although environmental targets (if available) for mercury control
differ, based on the local regulations [91, 92]. Mercury removal is typically performed via
an adsorption process. Sulfur-impregnated activated carbon is used as adsorbent and the
lifetime of the adsorption bed is up to two years. This process is performed prior to the
acid gas removal unit in power plants with and without CO2 capture.
Conventional syngas cleaning commonly consits of multi-stage pollutants’ separation.
Advanced syngas clean-up technologies are thus being developed to eliminate several
plant components for contaminants control [77]. Such clean-up processes will be
42
Coal-based power plants
concisely presented in Section 3.3.3.5 (Advanced syngas cleaning and conversion), since
they simultaneously remove H2S in addition to other pollutants and convert syngas.
3.3.3.2. Water-gas shift reaction
The produced syngas from commercial gasification technologies for IGCC application
contains high amounts of CO (25–50%) [93]. In IGCC power plants with CO2 capture, the
water-gas shift (WGS) process is the first step in converting the gasifier product into a
high hydrogen content syngas. This process is a moderate exothermic reaction, which is
used to convert CO as a component of the syngas into CO2. This is carried out by shifting
the CO with steam over a catalyst bed (Reaction 3.1).
KJ
(44mole)
CO(g) + H2 O(g) ↔
CO2 (g) + H2 (g)
(Reaction 3.1)
The reaction is equilibrium limited, implying the dependency of CO conversion on
reaction temperature, which is thermodynamically favored at low temperatures. On the
other hand, WGS reaction is kinetically favored at high temperatures (higher catalyst
activity as well as faster reaction is achieved at higher temperatures). Therefore, this
reaction is typically designed in two sequential reactors, where the first reactor (operating
at a higher temperature) converts the bulk of CO and the second reactor (operating at a
lower temperature) increases the overall CO conversion [94, 95]. However, factors such
as desired CO2 capture efficiency, sulfur emission limit (will be described later), coal
quality, gasifier design, etc. change the number of reaction stages and process design [96].
The operating temperatures of each stage are determined by catalyst type used in the
reactors, the amount of steam injected to the syngas stream and heat integration with other
components. The temperature range is between 150 and 530 °C [96, 97].
The reactor can be located either upstream (sour shift) or downstream (sweet shift) of the
acid gas removal unit (Figure 3.6). The location depends on the type of catalysts used for
the reaction. Some catalysts such as Fe, Cr or Cu-based are poisoned by a small amount of
sulfur compounds (higher than a few ppm levels) in the syngas. Hence, such catalysts
should be utilized after separation of sulfur compounds [97]. In contrast, Co or Mo-based
catalysts have the advantage that the sulfur compounds do not need to be separated from
the syngas prior to the WGS unit [98]. It has to be borne in mind that such catalysts need a
minimum level of sulfur compounds to operate actively. Shift catalysts based on
Coal-based power plants
43
molybdenum sulfide need a certain H2S concentration to stabilize the catalytic active
phase (higher than 100 ppm depending on the temperature level). The sulfur levels
required by catalysts may not be reached with low sulfur coals. Therefore, coal
characteristics are also a key element to be considered for efficient WGS design [99].
Steam
WGS
HT shift
LT shift
Syngas
High CO2 &
H2 content
Compressed CO2
Sour
Fuel
Gasification
Raw
Gas
syngas cleaning
Sweet
H2S
removal
CO2
capture
To atmosphere
O2
Air
Slag
ASU
Heat recovery steam generator
Stack
HP
Air
IP/LP
Gas turbine
Figure 3.6. Schematic configuration of the IGCC plant with sour or sweet WGS unit
In order to protect the catalytic bed from carbon deposit, to control the reaction
temperature, as well as to increase equilibrium conversion of CO to CO2, the WGS
reaction requires a large amount of steam (much larger than stoichiometric requirement)
[100]. Syngas produced in dry-fed gasifiers (e.g. SCGP and Siemens Fuel Gasification
(SFG) technologies) has lower water content and requires an injection of a considerable
amount of steam (mostly from the steam cycle) to ensure acceptable CO conversion. On
the contrary, syngas produced in slurry-fed gasifiers (e.g. GE gasifier) has higher water
content and requires lower supplementary steam injection. However, higher CO2 content
in the syngas produced in such gasifiers changes the equilibrium reaction direction to a
backward WGS reaction. Hence, it requires higher residence time for the reactor to reach
the targeted conversion of carbon monoxide.
44
Coal-based power plants
For IGCC power plants with CO2 capture, a sour WGS (SWGS) reaction may be a better
option. This helps to avoid additional cooling of the syngas required by the conventional
AGR unit (refer to the next section) upstream of the WGS unit and then reheating to the
level required for the catalyst’s activation in the WGS unit. It is also beneficial in order to
postpone the water condensation that occurs during the conventional AGR process
downstream of the WGS unit, as the WGS unit requires the existence of a considerable
amount of steam.
Despite the extensive ongoing research into finding improved catalysts [101], innovative
WGS configuration has also been investigated in order to reach a higher technical
performance of the WGS unit in IGCC application. An advanced WGS reaction
configuration equipped with syngas splitting has been utilized to feed four WGS reactors
in a staged configuration with intermediate water and synthesis gas quenches. The
potential of such a configuration for steam reduction is significant (54%), while it is
moderate for a reduction in efficiency penalty (2.7%) compared to the conventional WGS
[102]. However, the increased number of reaction units as well as the amount of water
quench should be optimized to balance the steam reduction with the higher capital costs.
These higher capital costs compared to the conventional unit are not only associated to a
higher number of reaction units but to the overall larger volume caused by a lower
thermodynamic driving force for CO conversion [100].
3.3.3.3. Acid gas removal
In IGCC plants, the sulfur content of the coal is mainly converted to H2S and COS due to
the highly reducing conditions of the gasifier [89]. Such gaseous components can produce
acidic solutions after dissolving in water and hence, are corrosive under moist conditions.
The combustion process converts H2S and COS to sulfur oxides, which are precursors of
acid rain. Their emissions to the atmosphere are, thus, limited by stringent environmental
regulations. Acid gases must, therefore, be removed from the syngas prior to the gas
turbine to avoid GT damage and to comply with legislation [103].
The removal of acid gases from the gaseous streams has been widely practiced using the
gas-liquid scrubbing process. This process consists of three solvent-based methods
including physical, chemical and hybrid (physical/chemical) solvents. Though some of the
current IGCC plants (without CO2 capture) have utilized chemical solvents (mainly
amine-based solvents), physical solvents (e.g. Selexol or Rectisol) are the most preferred
Coal-based power plants
45
choice for the IGCC application with CO2 capture [104]. The reasons are high partial
pressure of acid gases in the syngas, highly efficient sulfur removal process and moderate
operation costs offered by these solvents, and low desorption heat for solvent regeneration
[105-107]. Figure 3.7 schematically highlights the better performance of physical solvents
than chemical solvents at a higher partial pressure of acid gases (based on the data
available in [107]). As shown in Figure 3.7, the solubility of acid gases in a physical
solvent follows Henry’s law and increases linearly, unlike the chemical solvents which
plateau at a higher partial pressure of acid gases [45].
Chemical solvents
Solvent loading
Physical solvents
Partial pressure
Figure 3.7. Schematic comparison between loading characteristics of chemical (monoethanolamine
(MEA)) and physical (Selexol) solvents
Different physical solvents for absorption processes have their own advantages and
disadvantages. In this regard, the following criteria should be considered for the selection
of a proper solvent [104, 107-109]:






High loading capacity for different acid gases and high thermal stability;
low vapor pressure for minimal solvent losses and low viscosity;
non-reactive as well as non-corrosive;
high availability with a reasonable price;
low degradation rates; and
low health, safety, and environmental impacts.
46
Coal-based power plants
Amongst several physical solvents, Selexol (dimethyl ethers of polyethylene glycol), has
been extensively employed for acid gas removal. Its main advantages are high H2S
solubility, low vapor pressure, wide operating conditions, chemical stability, non-toxicity
and biodegradable material [107]. The characteristics of Selexol solvent are presented in
Table 3.3, below [107, 108].
Table 3.3. Characteristics of Selexol
Gas
Unit
Value
H2 solubility
-a
0.047
CO solubility
-a
0.10
CO2 solubility
-a
3.63
COS solubility
-a
8.46
H2S solubility
-a
32.4
Chemical formula
CH3(CH2CH20)nCH3
Density
kg/m3
1030
Molecular weight
g/mol
280
Vapor pressure
mbar
9.7e-4
Viscosity
Pa.s
5.8e-3
a
Solubility (
gas volume
Selexol volume
Remark
3≤n≤9
at 25 °C
at 25 °C
at 25 °C
) at 25 °C and 1 atm.
According to Table 3.3, solubility of both H2S and CO2 is much greater than CO and H2,
which results in a limited co-absorption of such combustible gases. Hence, Selexol offers
a good match prior to the GT in the IGCC application. Selexol’s potential for H2S
removal is greater than that of CO2, since the solubility of H2S in Selexol is about nine
times higher than that of CO2. In IGCC plants with CO2 capture, where both H2S and CO2
should be removed, the Selexol process typically takes place in two successive and
typically independent absorption-regeneration stages, in which the H2S is first removed
from the shifted syngas and consequently the CO2 is separated in the second stage of
absorption. The process is similar in principle to what was presented for CO2 capture
using an absorption process in Chapter 2 (Technical background). The syngas enters from
the bottom of the first absorption column, where the H2S is removed by a counter-current
flow of the solvent. The H2S-rich solvent is then thermally regenerated in a stripper. The
regenerated solvent is cooled, pressurized and recycled back to the top of the H2S
absorber, while acid gases are sent to a sulfur recovery unit (SRU). The H 2S-lean syngas
enters the second absorber for CO2 removal. Similar to the first stage, the CO2-rich
solvent exits the absorber bottom and then passes through a few flash drums in series.
Carbon dioxide is released from the physical solvent as a result of stepwise pressure
Coal-based power plants
47
reduction, unlike the chemical solvents which need significant thermal energy input [107].
The CO2 released from flash drums goes to the compression unit, while the clean and
CO2-lean syngas is sent to the GT. The CO2-lean solvent (after the flash drums) is also
cooled, pressurized and recycled back to the absorption column.
In order to achieve higher sulfur removal (more than 99%) from the syngas in IGCC
plants, it is necessary to add a COS hydrolysis unit to convert COS to H 2S in the case of
CO2 capture trip, according to Reaction 3.2 [104, 110, 111].
KJ
(33.6mole)
COS(g) + H2 O(g) ↔
H2 S(g) + CO2 (g)
(Reaction 3.2)
In the case of using a sour WGS reaction unit (IGCC with CO2 capture), COS hydrolysis
is directly carried out in the WGS unit, avoiding a dedicated reactor compared to the
IGCC plant without CO2 capture [112].
3.3.3.4. Sulfur recovery unit
The AGR process results in three product streams, i.e. the fuel gas to the GT, a CO 2-rich
stream and an acid gas stream. The acid gas stream from the AGR unit cannot be directly
vented to the atmosphere, according to stringent environmental regulations [104]. A sulfur
recovery unit is, therefore, required to treat the acid gas stream and recover sulfur (with
more than 99% recovery) as a by-product. The conventional SRU typically is based on the
Claus process for oxidizing H2S, obtaining elemental sulfur. The Claus process
catalytically converts H2S to elemental sulfur by the following reactions:
H2 S + 3⁄2 O2 → SO2 + H2 O
(Reaction 3.3)
2H2 S + SO2 ↔ 3⁄2 S2 + 2H2 O
(Reaction 3.4)
The Reaction (3.4), the Claus reaction, is equilibrium limited. The overall reaction is:
3H2 S + 3⁄2 O2 ↔ 3⁄2 S2 + 3H2 O
(Reaction 3.5)
The oxygen required for the Claus combustion is supplied by the ASU without any major
penalty on the overall plant efficiency [113]. Since the Claus reaction is exothermic, HP
48
Coal-based power plants
steam production usually follows the Claus furnace. Moreover, LP steam is raised in the
condenser downstream of the HP steam recovery section [104].
To reach more than 99.8% sulfur removal efficiency, the tail gas from the Claus plant
needs to be further cleaned-up, an exercise which is widely practiced using a Shell Claus
off-gas treating (SCOT) unit. The SCOT unit treats the Claus tail gas by employing a
dedicated absorption unit (typically amine-based) and recycles the resulting acid gas to
the AGR unit. The tail gas treating (TGT) can also be performed by recycling the Claus
tail gas to the AGR unit [104].
3.3.3.5. Advanced syngas cleaning and conversion
In general, advanced gas cleaning and conversion processes are under investigation to
enhance both the technical and the economic performance of IGCC plants with CO2
capture [65]. Most of the reaserch activities have focused on the development of reliable
adsorption, membrane, improved scrubbing, and hybrid technologies.
In the case of IGCC plants with CO2 capture, the syngas exiting the conventional gas
cleaning needs to be heated up to certain limits for downstream sour WGS reaction. It
requires cooling down again for H2S separation in the AGR unit, which typically operates
at near-ambient temperature. The clean syngas may be reheated before combustion in the
GT. These repeated heating and cooling processes cause the inherent energy losses and
have detrimental effects on the plant’s overall efficiency [99]. To perform the removal of
H2S and multi contaminants in fewer unit operations, and to avoid the penalties associated
with syngas cooling and heating, a warm (hot) gas cleaning process is being developed to
operate at high temperatures (250-700 °C). Such a cleaning method employs sorbents
(typically metallic type e.g. Zn-based) to remove H2S or alkali species [103]. Significant
capital cost redcution and efficiency improvement could be achieved by replacing the cold
gas clean-up systems with warm systems [77]. The use of warm gas cleaning may also
allow heat extraction from IGCC power plants that could be beneficial for combined heat
and power (CHP) application [114]. However, the low duarbility and thermal stability of
the sorbents increases the operation costs and reduces the avaialability of such processes
[42]. In addition, such technologies still require a considerable time frame to be
commercially developed for large-scale implementation in IGCC power plants [42, 99].
Coal-based power plants
49
To reduce the penalty of the capture process, advanced technologies such as membrane
separation are also being developed. Membrane technology has attracted the attention of
the research community due to its process simplicity, which can separate different gas
components through a continuous process. Different kinds of membranes can selectively
separate either H2 or CO2. The combination of warm gas clean-up technology (~480 °C)
with CO2 separation by membrane technology is projected to reduce the cost of electricity
by 14% [77]. Membrane technology can also be integrated into the system in order to
enhance WGS reaction either by the permeation of CO2 (according to Figure 3.8) or H2
[18, 115].
Catalyst particle
H 2O
High pressure side
Syngas
(CO, H2, CO2)
CO + H2O
Retentate
H2, H2O (CO, CO2)
CO2 + H2
Membrane
CO2
CO2
CO2
CO2
Permeate CO2
in sweep flow
Sweep flow
Low pressure side
Figure 3.8. Operating principle of an enhanced WGS reactor by membrane technology
For scrubbing process in the AGR unit, research activities are foccused on the
improvement of the acid gases’ loading to achieve better techno-economic performance.
Solvents that could show a good performance for acid gases at higher temperatures are
highly required in order to avoid the cooling necessary for current scrubbing agents. Some
salty compounds such as ionic liquids (ILs), which are at the early stage of development,
could be suitable alternatives for the current physical/chemical solvents for CO2
absoprtion. They can opperate at high temperatures (up to 200 °C), which is a good match
for the warm gas clean-up process [116].
Another promising concept for pre-combustion CO2 capture is the sorption-enhanced
water-gas shift (SEWGS) process. Simultaneous removal of CO2 in the WGS reactor will
enhance the conversion of carbon monoxide [94]. This system offers a higher CO2 capture
rate, higher H2 recovery in the fuel, simultaneous separation of H2S with CO2 and avoids
the cooling required by conventional AGR and CO2 capture [113]. The process comprises
of multiple sorbents beds (containing WGS catalysts), operated in parallel, that adsorb
50
Coal-based power plants
CO2 at high temperature and pressure and release it at lower pressure. The combination of
CO conversion and instantaneous CO2 removal enhances H2 production and thereby the
purity of the fuel feeding the GT combustor. A separate CO2 stream (mixed with H2S) can
be recovered from the sorbents by regenerating the bed. Regeneration is carried out by a
pressure swing, producing a low-pressure CO2-rich stream. The CO2 stream, which
contains certain amount of H2S, needs to be further treated for the removal of H2S from
CO2 for final compression and storage [113]. The COE for the IGCC plant using SEWGS
can be reduced by 4%, while the overall efficiency can be improved by 2-3% compared to
the conventional IGCC plants with solvent-scrubbing CO2 capture [117]. However,
similarly to other sorbent-based cleaning methods, practical issues such as handling and
regeneration of the solid sorbent materials need to be addressed by ongoing research [42].
3.3.4. CO2 compression and dehydration
Carbon dioxide captured at a power plant can be stored in depleted oil and gas reservoirs
and deep saline formation or utilized for enhanced oil and gas recovery [18]. The captured
CO2 can be transported by several means including ships, pipeline, railways or roads.
Amongst those, however, ships and pipelines are more cost-effective for the transportation
of substantial amounts of CO2, depending on the distance to storage sites [18]. In order to
provide an optimum condition for transportation of large amounts of CO2, it is necessary
to transform gaseous CO2 into a phase comprising less volume and more density, i.e. a
liquid, solid or supercritical state. For pipeline transportation, the suitable condition is in
the supercritical region, as shown in Figure 3.9 (data for triple and critical points are from
[118]). Being above supercritical pressure eliminates the risk for the two-phase flow
regime due to temperature variations along the pipeline [18]. Recompression stages are
commonly considered in order to keep the pressure over supercritical pressure and to
overcome the pressure drops whenever the length of CO2 pipeline is more than 150 km
[119]. For tank transportation (e.g. ship), the most economically feasible condition is to
keep the CO2 in the liquid state at about 7 bar and -50 °C [119, 120]. Irrespective of the
choice of transportation, CO2 compression has a negative impact on the plant’s technical
and economic performance. The loss of overall plant efficiency associated with CO2
compression is approximately 5 percentage points. Considering such a unit for an IGCC
plant with CO2 capture increases the capital costs and cost of electricity by approximately
10% [65].
Coal-based power plants
51
Pressure, p (bar)
Meltin
Solid
g line
Supercritical
region
Liquid
Satu
n
ratio
line
Critical point
Tc = 31.0 ºC
pc = 73.8 bar
5.2
Triple point
Vapor
-56.6
Temperature, T (ºC)
Figure 3.9. The schematic temperature-pressure diagram for CO2
There are several compression alternatives to reach the required pressure of the CO 2 for
transportation/storage [121]. In reality, a process closer to isothermal compression, such
as that which occurs in compression with intercooling, is beneficial to reduce the
compression power demand [122]. Inter-cooled compressors offer smaller sized
compression units and hence reduce costs and increase overall plant efficiency by
appropriate heat integration with other sub-systems at higher costs [122]. A further
reduction in the demand for power to pressurize the CO2 could be accomplished by
eliminating the final stage of the inter-cooled compressor using a less power-, costintensive pumping process [121]. Initial compression to the condition where the CO2 is
transformed to a liquid state is carried on by the intercooled compressor, while the pump
is utilized to reach the final pressure.
In addition to compression, CO2 needs to be treated for the removal of accompanying
water to prevent the risk of corrosion and the formation of gas hydrates in the
52
Coal-based power plants
transportation pipeline [120]. In addition to water vapor, the CO2 stream from the AGR
unit of the IGCC plants with CO2 capture contains minor species such as N2, Ar, H2, CO
and traces of H2S [107]. All these impurities have a negative impact on the compression
power demand, although these effects are marginal due to their trace existence [123].
Dehydration can be accomplished using an absorption process, vapor-liquid separator
drums or an adsorption process [122]. Maximum allowable water content of the CO2
stream is a critical factor in order to select a suitable dehydration process [120]. Often, a
dehydration unit based on glycol solvents such as tri-ethylene glycol (TEG) is considered
to absorb water from the CO2 stream for IGCC application with CO2 capture [43]. It
should be mentioned that the saturated water content first decreases by the increased
pressure of the high-purity CO2 stream then increases again at pressures above 60 bar
[107]. Thus, the optimum pressure and location of the dehydration unit to remove water
content (i.e. the lowest saturated water content) is about 60 bar at 25 °C [124].
3.3.5. Gas turbine
Due to the continuous need for coal utilization in power generation, the development of
reliable, environmentally friendly and cost-competitive gas turbine technologies for
hydrogen-rich syngas combustion is highly essential. The performance of the GT varies
with changes in the properties of the fuel gas [65]. The behavior of the gas turbine
changes with the transformation from NG (as conventional fuel for the GT industry) to a
H2-rich syngas, which is a typical fuel in IGCC plants with CO2 capture. The current
section presents various operational challenges and effects of using syngas instead of NG
on the existing gas turbine.
3.3.5.1. Combustion process
The state-of-the-art combustion technology for NG operation is dry low NOx (DLN) premixed burners. Such burners principally work at lean condition by forcing more air than
stoichiometric in the primary combustion zone, which results in moderate flame
temperatures [125]. Unfortunately, the available pre-mixed technology could not comply
with flammability limits for H2-rich fuels, which are much larger than those for natural
gas [126]. Moreover, high hydrogen content syngas has higher adiabatic flame
temperature, higher flame speed, and higher flashback potential compared to NG,
complicating the use of a combustor that is designed using NG design criteria [60, 127].
The SOA combustion technology for burning H2-rich syngas (25-40 vol%) is the diffusion
Coal-based power plants
53
flame burner [125]. Note that this value could be much higher in an IGCC plant with CO 2
capture, depending on the performance of upstream operation units. In general, diffusion
flame burners produce considerably more NOx than the pre-mixed combustors for NG
combustion and this is exacerbated when burning high hydrogen content fuels which is
typical for IGCC power plants with CO2 capture [27, 127].
Stoichiometric adiabatic flame temperature is a representative indicator for NOx
formation in diffusion flame combustors [126]. In order to lower the flame temperature
(down to about 2300 K [125]) and, consequently, to minimize NOx formation (25-45
ppmvd @ 15% O2 [125]), hydrogen-rich syngas is normally mixed with a diluent gas such
as nitrogen [128]. There are several methods to control NOx emissions from diffusion
flame burners of gas turbines including:


Saturation with water [126], steam or N2 injection [60], and combination of
saturation, steam and N2 injection [128]; and
use of selective catalytic reduction (SCR) in the bottoming cycle [126].
Irrespective of the method selected for controlling NOx emissions, all strategies work on
the basis of lowering the adiabatic flame temperature [126]. Nitrogen dilution is generally
recognized as the most efficient method due to its availability in conventional IGCC
plants [60]. It results in reduction of water consumption, though it increases auxiliary
power requirements (for N2 compression) [66]. Steam injection or syngas saturation with
water causes a higher convective heat transfer coefficient between combustion products
(hot stream) and the expander blade materials due to the change in hot stream composition
compared to an undiluted case. Accordingly, the blade metal temperature will increase at
a given turbine inlet temperature (TIT), geometry and cooling flow, which results in faster
life consumption of the blades [125]. In order to keep the same blade metal temperature,
TIT is commonly de-rated (reduced) [129] or new geometry and design should be adopted
to increase cooling flows. In the case of using SCR in the HRSG unit, a highly efficient
upstream AGR process (COS+H2S < 20 ppmv) needs to be employed to prevent
ammonium sulfate from fouling due to the presence of SOx in the flue gas [104]. In
addition to the higher costs induced by the SCR technology, a larger HRSG is required for
such a process [125].
As mentioned earlier, N2 is typically used as diluent in conventional IGCC plants with
cryogenic ASU. In this regard, the presence of a large amount of N 2 in the syngas derived
54
Coal-based power plants
from an air-blown IGCC plant is advantageous to minimize N2 dilution [126]. The coal
feeding system into the gasifier also has some effects on the NOx formation. The required
steam for NOx control in IGCC plants with slurry-fed gasifiers (e.g. GE and E-GasTM) is
considerably lower than that for dry-fed gasifiers due to the greater water content of the
syngas in slurry systems [64].
3.3.5.2. Turbo-machinery
Syngas property significantly affects GT operation. Syngas combustion results in different
product composition and thermo-physical properties (such as heat transfer coefficients)
compared to NG combustion. The fuel change (from NG to syngas) leads to [125]:



Different expansion line (enthalpy drop) in the expander;
change in the hot gas flow rate at the expander inlet; and
different cooling flow required for expander blades.
As a consequence, there a new compressor design might be needed to match the new
turbine characteristics, or compressor re-design might be considered to provide different
cooling flows.
The isentropic enthalpy drop in the expander for a H2-rich syngas is higher than for NG.
This drop will be significantly increased by steam dilution or water saturation to the
minimum required level for NOx emissions. If the working fluid is assumed to be an ideal
gas, the isentropic enthalpy drop can be evaluated using the following equation:
𝑇
∆ℎ𝑖𝑠 = ∫𝑇 𝑖 𝑐𝑝 (𝑇)𝑑𝑇 = 𝑐̅𝑝 (𝑇𝑖 − 𝑇𝑜,𝑖𝑠 )
𝑜,𝑖𝑠
(Eq. 3.3)
According to Equation 3.3, isentropic enthalpy drop is a function of average 𝑐𝑝 and
temperature drop through the expansion. The 𝑐̅𝑝 is enhanced by fuel transformation from
NG to H2-rich syngas and increases even more in the case of steam or water dilution. The
temperature drop is influenced by the change in isentropic exponent (𝛾) according to
isentropic 𝑝 − 𝑇 relation. Assuming constant expander inlet pressure, inlet temperature
and outlet pressure, the temperature drop reduces by an increase of 𝑐̅𝑝 and simultaneous
reduction in 𝛾. Consequently, turbine outlet temperature (TOT) is also increased, which
threatens the lifetime of the expander blades. TOT increase is intensified in the case of
water saturation or steam injection. In the case of N2 dilution, the change in isentropic
Coal-based power plants
55
enthalpy drop from NG combustion remains almost constant with an increase in N2
dilution as the hot gas already contains a considerable amount of N2 from the combustion
air [125].
The heating value of the produced syngas in IGCC plants is lower than that of NG, a
commonly used fuel for GT design [62]. This results in a higher flow rate of the fuel gas
to the GT to reach the same order of TIT and thereby a similar efficiency level at a given
compressed air flow [130]. The flow rate of the hot gas stream into the expander further
increases by the dilution process (in the case of SOA combustion technology, i.e.
diffusion flame burners). This increase is exacerbated in the case of N2 dilution, which
requires more injection to lower the adiabatic flame temperature and thereby NOx
formation due to its lower 𝑐𝑝 value compared to steam [125]. Irrespective of dilution type
and use of undiluted syngas, increased fuel flow rate:



Affects the compressor/expander matching [126];
induces higher back pressure to the compressor [125]; and
reduces available surge margin [130].
An increase in pressure ratio adversely affects the expander blade wall temperature. The
mass flow of the cooling stream increases by pressure ratio due to the higher cooling flow
density. The enhanced density results in higher convective heat transfer coefficients for
both fluid and outer blade wall. The higher heat transfer coefficient of the cooling stream
could not compensate for the higher outer blade wall heat transfer coefficient, which
increases the blade wall temperature beyond its admissible level [125]. The lifetime of the
turbine materials will then be shorter than normal operation (NG operation). The cooling
flow temperature also rises by the higher pressure ratio, which results in less effective
cooling [130]. In the case of using an existing GT, which is originally designed for NG,
the following options are available to improve GT operation for syngas fuel.



Allowing the pressure to increase up to the minimum surge margin level or
adding one or more high pressure (rear) stages to the compressor in order to
increase the available surge margin [130] at the price of higher costs [127].
Keeping the existing compressor design while widening the swallowing capacity
of the expander without increasing the pressure ratio, although it results in
higher blade wall temperature [130].
Partially closing the compressor variable inlet guide vanes (VIGV), which can
compensate for part of the higher hot gas flow rate [125].
56
Coal-based power plants


De-rating the TIT which has a negative effect on the performance of the GT.
The reduction of TIT can marginally diminish the problem of higher pressure
ratio and consequently surge margin [130].
Air bleeding from the GT compressor to the ASU to maintain the pressure ratio
increase [62]. As mentioned earlier, this option results in higher overall plant
efficiency but reduces the operational flexibility, more specifically during
transient conditions [63]. Note that the higher degree of air-side integration leads
to improved surge margin and reduced pressure ratio at a given TIT and amount
of diluent and vice versa [130]. Effects of lower integration degree on the blade
wall temperature are similar to that presented for higher hot gas flow and
pressure ratio and hence, are not repeated again.
3.3.5.3. Materials
The combustion of H2-rich syngas results in higher heat transfer to the hot section
materials. The other major material concern, applicable to any coal-derived syngas, is the
need to protect the gas turbine from the corrosive effects of sulfur compounds [131],
alkali metal salts [132], and fly ash deposition [133]. The existence of such compounds
coupled with high temperature media can boost hot corrosion of metallic alloys [132] and
results in extensive thermal barrier coating (TBC) spallation [133].
Advanced turbine aerodynamic and cooling schemes are, therefore, required to maintain
the lifetime of the hot path at existing gas turbines. Otherwise, advanced high temperature
low conductivity TBC materials and superalloys need to be incorporated into the expander
hot path for the combustion of syngas [127].
3.3.5.4. Commercial syngas-fueled gas turbine
The original equipment manufacturers (OEMs) simultaneously improve their GT
technology for NG operation and perform required developments to enable syngas
operation and thus the integration of GTs into IGCC plants [63]. There are many
commercially available heavy-duty GT models originally designed for NG operation with
different modifications (e.g. different burner design) for syngas application. Amongst
those are Siemens (E and F frames), GE (EB and FB models), and MHI (e.g. M701DA)
[58, 59, 63, 134]. Nearly all these models use diluted syngas (using water saturation, N2,
or steam injection), and all are equipped with diffusion flame burners. The general
specifications of different commercially available 50 Hz F-frame gas turbine models
Coal-based power plants
57
suitable for syngas operation are shown in Table 3.4. Please note that available data are
based on NG operation.
Table 3.4. Main specifications of two commercial 50 Hz gas turbine models for syngas operation
Parameter
Unit
9FB a [135]
SGT5-4000F [136]
OEM
General Electric
Siemens
TIT
°C
1360 [126]
1265 b
COT
°C
1454 [126]
1500 b
TOT
°C
623
577
19.7 [137]
18.2
Pressure ratio (𝛽)
Exhaust flow rate
kg/s
745
692
Gross power output
MW
338
292
%
40.0+
39.8
Gross efficicency (𝜂𝑔𝑟𝑜𝑠𝑠 )
Power output (combined cycle) c
MW
510
423 [138]
Heat rate c
kJ/kWh
5894
6164 [138]
%
60.0+
58.4 [138]
Net thermal plant efficiency (𝜂𝑡ℎ ) d
Number of compresssion stages
14
15 e
Number of expansion stages
4
4
a 2011 model.
b According to personal comminucation.
c Value represents 1×1 combined cycle configuration.
d Value is LHV basis and represents 1×1 configuration.
e The modified SGT5-4000F has an additional compressor rear stage in order to accommodate
higher back-pressure caused by diluted syngas in the case of zero or low air-side integration
with ASU [63].
3.3.5.5. Advanced hydrogen turbine technology
One of the most important on-going research and development (R&D) programs is the
development of fuel flexible hydrogen turbine technology, which has great potential for
the further development of IGCC plants through improved thermal efficiency [127]. Most
recently, many R&D activities have focused on the development of low NOx gas turbine
technology for undiluted hydrogen-rich syngas operation [73]. A report by the U.S.
Department of Energy (DOE) confirms that significantly higher overall efficiency (1.3
HHV%) and lower specific plant costs (9%) can be achieved by the deployment of such
technology in IGCC plants with CO2 capture [72]. In addition, novel aerodynamic
designs, advanced cooling schemes, advanced TBC systems and superalloys are under
development to further enhance GT performance [127]. The use of advanced gas turbine
technologies operating at elevated firing temperatures and pressure ratios such as H-class
machines will also significantly improve the overall efficiency of IGCC plants with CO 2
58
Coal-based power plants
capture [65]. Utilizing such GT technologies (60+% efficiency for NG operation) will
tackle the increasing GHG emissions from another area, i.e. energy efficiency [64, 128].
3.3.6. Bottoming cycle
The gas turbine exhaust gas is used to produce steam in an HRSG for electricity
generation in steam turbines. The process is quite similar to that of the conventional
bottoming cycle in natural gas combined cycles, with few changes in terms of thermal
energy inputs and outputs. The bottoming cycles in existing IGCC plants are typically
based on a triple-pressure level design with reheat consisting of an HRSG, steam turbine
(ST), condenser, and associated auxiliary pumps. The HRSG section includes
economizers, evaporators, super-heaters, and re-heaters for the three pressure levels. The
HP steam turbine inlet conditions in the existing IGCC units are about 100-120 bar and
520-540 °C with 520-540 °C reheat inlet temperature for the intermediate pressure (IP)
stage [139].
The steam turbine power output changes significantly in IGCC plants compared to that in
conventional combined cycles due to the many interactions between HRSG and other subsystems. The primary thermal energy input to the HRSG is from the GT exhaust gas,
which enters at about 560-590 °C [139]. The other energy inputs are HP, IP, or LP steams
generated in the gasifier, syngas coolers and WGS unit, depending on the heat integration
scheme between different process units. The HRSG supplies the high and intermediate
pressure boiler feed water (BFW) used in the gasifier, syngas cooler’s water-gas shift
process, the IP steam used in the WGS and the LP steam used in the AGR unit (i.e. for the
solvent regeneration). In IGCC plants with capture, the ST power output decreases due to
the extraction of steam for the WGS reaction unit. Consequently, the WGS unit is
typically viewed as a burden on the steam cycle [96]. In addition to steam production and
consumption, the operation of the steam cycle is dependent on ambient temperature which
imposes a vacuum at the condenser and, therefore, controls the performance of the steam
turbine [43].
3.4. Current IGCC power plants status
Seven IGCC power plants have been operated on coal as main feedstock (refer to Table
3.5), but none has been equipped with a CO2 capture unit [27, 49, 79]. Although
Coal-based power plants
59
operational experience from the current plants has proven the viability of IGCC
technology, the demonstration of a full-scale IGCC plant with integrated CO2 capture unit
is highly essential. Some of the European IGCC plants, even without CO2 capture, are
simply not economically viable under current electricity market conditions. Buggenum
IGCC plant has recently been decommissioned, and Puertollano plant is at risk of closure.
Therefore, further IGCC deployment is tightly connected with its competitiveness against
other power generation technologies. Important areas which have tremendous effects on
the economy of this technology, that should therefore be improved or developed, include
[27]:
 Higher plant availability and reliability for all types of coals;
 cheaper solutions e.g. quench gasification for low grade coals;
 development of highly efficient multi-pollutant gas clean-up systems; and
 development of gas turbine technology burning hydrogen-rich fuels.
In addition to the coal IGCC plants mentioned in Table 3.5, many IGCC plants with and
without CO2 capture have been in the evaluation, planning, and construction phases,
especially in the USA and China [140, 141]. However, only a few, such as Kemper
County (Mississippi, USA) and GreenGen (China), show significant progress due to the
sustainability of government incentives and supports [142, 143].
60
Name
Location
Coal-based power plants
Table 3.5. Specification of the operating coal-based IGCC power plants
Buggenum
Wabash River Vresova
Polk
Puertollano Nakoso
[27, 49]
[27, 49]
[27]
County [27, [27, 49]
[27, 49]
49]
The
Indiana, USA
Czech
Florida,
Spain
Japan
Netherlands
Republic
USA
Edwardsport
[144, 145]
Indiana,
USA
Start
1994
1995
1996
1996
1998
2007
2013
Gasifier
SCGP
E-GasTM
Lurgi (26
fixed-bed
gasifiers)
GE
Prenflo
MHI
GE
Feed
Dry
Slurry
Dry
Slurry
Dry
Dry
Slurry
Oxidant
O2
O2
O2
O2
O2
Air
O2
GT
Siemens
V94.2
(SGT52000E)
GE 7FA
GE 9E
GE 7FA
Siemens
V94.3
(SGT54000F)
MHI
M701DA
GE 7FB
Air-side
integration
Full
Zero
Zero
Zero
Full
Partial
Partial
Diluent
(NOx
control)
Saturation,
N2 dilution
Saturation
(>20 vol%)
[139]
Steam
injection
N2 dilution,
saturation
[66]
Saturation,
N2 dilution
Without
dilution,
with SCR
N2 dilution
Net output
(MWe)
253
262
350
250
300
250
618
40.0
Not
available
36.7
42
42.0 a
~ 44.0 a
[146]
Efficiency 43.0
(LHV%)
a estimated value.
4. H2-IGCC power plant
The current chapter briefly presents the main objectives of the European co-financed Low
Emission Gas Turbine Technology for Hydrogen-rich Syngas (H2-IGCC) project, as well
as its multiple research areas together with its main outcomes. Furthermore, the selected
IGCC configuration with CO2 capture consisting of several sub-systems is briefly
described. This configuration represents a realistic and practical integration of various
state-of-the-art technologies for different components of the plant. Limitations which have
arisen from the selection of each technology and operating mode compared to other
alternatives are also presented. As the focus of the H2-IGCC project was on the
development of a combustor for burning undiluted high H2 content syngas fuels in gas
turbine technology, various challenges tackled by different sub-project groups are briefly
summarized. Finally, the methodology for the performance analysis of the selected IGCC
plant with CO2 capture is described. Different software tools used for thermodynamic
modeling, together with reasons for the selection of each tool as well as boundary
conditions of the entire cycle and the gas turbine, are then presented.
4.1. H2-IGCC project
As mentioned previously in Chapter 3, current GT technology (preliminary developed for
natural gas and used) in IGCC application with CO2 capture suffers from several
challenges. Amongst those are the wide variation of fuel composition compared to NG,
increased hot gas flow to the expander, increased heat transfer between the hot gas and
expander materials, high NOx emissions from diffusion flame burners, and high dilution
rate to control NOx emissions.
61
62
H2-IGCC power plant
In November 2009, the H2-IGCC project was started, aiming at the knowledge
development necessary to overcome the abovementioned drawbacks, while burning H2rich fuels. The overall objective was to provide and demonstrate technical solutions which
allow the use of SOA highly efficient, reliable GTs in the next generation of IGCC plants
after introducing CO2 capture. The goal was to enable the combustion of undiluted H2rich syngas with low NOx emissions and also to allow for high fuel flexibility by enabling
the burning of back-up fuels (e.g. NG) with limited adverse effects to reliability and
availability. Twenty-four partners including academia and manufacturers, as well as plant
operators from ten European countries, worked together to achieve the abovementioned
goals. The project was divided into four major research areas, namely, combustion,
materials, turbo-machinery and system analysis. Figure 4.11 shows different research
areas with their overlaps on a schematic configuration of the IGCC plant with CO2
capture.
Compressed CO2
Coal
O2
Air
SWGS,
H2S removal,
CO2 capture
Raw
Gas
syngas cleaning
Gasification
Slag
ASU
C
b
om
us
n
tio
Materials
To atmosphere
Heat recovery steam generator
Turbo-machinery
Stack
HP
IP/LP
System analysis
Figure 4.1. The structure of the H2-IGCC project
1
The GT image is courtesy of Siemens SGT5-4000F gas turbine.
H2-IGCC power plant
63
Different technical sub-projects (research areas) had various main objectives, as detailed
below [147]:




Combustion group aimed to develop and demonstrate a safe and low emission
pre-mixed combustion technology for undiluted H2-rich syngas.
Materials group aimed to develop and demonstrate improved materials with
advanced coatings able to protect base materials of the blades and combustor
against the potentially more aggressive temperature and composition of the
exhaust gas.
Turbo-machinery group aimed to provide required design for the
compressor/expander aerodynamics and cooling schemes to cope with changed
fluid properties of the hot gas.
System analysis group aimed to evaluate optimum IGCC plant configurations
and to establish guidelines for optimized full-scale integration. Moreover, a
detailed systems analysis needed to be performed to generate realistic technoeconomic results for IGCC plants with pre-combustion carbon capture.
As the main target of the project was to develop and demonstrate a reliable and low
emission combustion technology, great efforts have been dedicated to the GT block. Note
that the successful implementation of this project could only be realized by intensive
collaboration between project partners due to the cross-disciplinary nature of the project’s
tasks and objectives. The interactions between the system analysis group and other project
groups (please refer to Figure 4.1) can be summarized as follows:





Syngas composition and mass flow, total mass flow (cold/hot path), turbine inlet
temperature and pressure from the system analysis group to the combustion
group, necessary for successful experimental campaigns and burner designs;
total mass flow (cold/hot path), turbine inlet temperature and pressure from the
system analysis group to the materials group, required for the selection of
materials and coatings as well as testing of the blades;
syngas composition and mass flow, total mass flow (cold/hot path), turbine inlet
temperature and pressure from the system analysis group to the turbo-machinery
group, necessary for modifying the GT designs (compressor and expander
aerodynamics);
flue gas composition and nitrogen demand (if any) for dilution from the
combustion group to the system analysis group, which influenced the steam
cycle calculations;
flue gas temperature and mass flow from the turbo-machinery group to the
system analysis group, which influenced performance analysis of the steam
cycle; and
64
H2-IGCC power plant

gas turbine characteristics (expander and compressor maps, cooling flows, TIT,
combustor outlet temperature) from the turbo-machinery group to the system
analysis group, required for thermodynamic modeling of the GT.
The importance of an appropriate sub-system selection and integration, as well as overall
system analysis, should be clearly highlighted. It should be underlined that every decision
made at the entire system level had some impacts on the component targeted by the H2IGCC project (i.e. the gas turbine) and vice versa. Amongst those decisions, but not
limited to them, are:



Employing the current SOA technologies1 with respect to both the gas turbine
and also the entire IGCC system;
selecting a proper degree of integration between GT and ASU to achieve higher
flexibility, availability, and operability of the plant; and
defining fuel flexibility targets considering both planned and sudden changes in
fuel composition due to trip of the carbon capture unit or a failure upstream of
the gas turbine.
4.2. System integration
The thermodynamic performance calculations required the establishment of a reference
IGCC plant with carbon capture. This was performed based on a comprehensive review of
the present IGCC technology as well as openly available data. The results of
thermodynamic simulations of the baseline case have been published in Paper I.
The configuration (and input data/settings) of the plant has then been improved,
incorporating more realistic performance data reflecting industrial experiences from the
operation of similar plants during the project’s development (for details see Paper II). The
major sub-systems and the way in which they were integrated into the cycle are presented
here. Further information can be found in Papers I-III. In addition, the current section
briefly reviews the pros and cons of such sub-systems in terms of the operability and
thermodynamic performance of the overall IGCC plant.
1
Most of improved technologies for different sub-systems (e.g. ITM, hot gas clean-up, etc.) are not
likely to be commercially available in the time frame for plants discussed within the H2-IGCC
project, i.e. 2020.
H2-IGCC power plant
65
4.2.1. Cryogenic air separation unit
The ASU is the most power-demanding auxiliary unit in the IGCC plant, and the level of
its integration to other sub-systems has to be properly analyzed with respect to costs,
efficiency, operational flexibility and plant availability.
Operating experience from Buggenum IGCC plant by an industrial partner of the H2IGCC project (NUON/Vattenfall) confirmed that a fully-integrated GT-ASU solution
adversely affects the availability of the plant. Therefore, a stand-alone ASU was
considered for the H2-IGCC project. The advantage obtained by selecting no air-side
integration between the GT and the ASU is also driven by higher plant operability.
However, it should be noted that the overall plant efficiency increases with the degree of
integration due to the higher isentropic efficiency of the GT compressor [113]. Lower
efficiency of the non-integrated GT-ASU case could be balanced with the selection of an
inter-cooled MAC to achieve similar overall equivalent compression work compared to a
fully-integrated GT-ASU case. Furthermore, as there is no need for injection of diluent
gaseous nitrogen into the GT for the dry-low NOx combustion, heat integration between
the GT compressor bleed air and DGAN from the ASU is not an option in order to
enhance the overall plant efficiency. Using undiluted syngas in the GT has also resulted in
the selection of a low pressure ASU, where O2 and N2 are produced at near atmospheric
pressure, as there is no need to reduce the compression work required for injection of
DGAN into the GT.
Due to the high competitiveness within the oxygen production industry, compressors’
characteristics, performance data, and detailed cost of cryogenic air separation plants are
difficult to obtain. Consequently, the performance data for main air, pure gaseous
nitrogen, and gaseous oxygen compressors, as well as for a number of inter-cooling stages
for all ASU compressors, have been gathered from H2-IGCC industrial partners based on
currently available technologies (presented in Paper II). The purity of the final product of
the ASU (95 mol% for O2) reflects an economic choice, maintaining the balance between
higher capital expenditure and higher efficiency loss. Moreover, as the combustion
process in the GT uses air as the oxidant agent, additional N2 and Ar in the syngas
produced by the gasifier does not make any major difference compared to when higher
purity O2 is produced by the ASU.
66
H2-IGCC power plant
4.2.2. Gasification
The gasification technology is based on the Shell Coal Gasification Process. Such a
technology was selected due to its highest cold gas efficiency and its operating pressure
level. A key parameter governing the overall plant pressure is the operating pressure of
the GT combustor. The pressure prior to the combustion chamber was fixed at about 30
bar to overcome the pressure loss over the fuel valves for pre-mixing of fuel and air in the
combustor. The pressure of the gasification block was then calculated and fixed at 45 bar,
considering all pressure losses from the gasifier to the combustion chamber and
eliminating any supplementary syngas compression. The conventional Shell gasifier had a
slightly lower pressure (~ 42 bar) at the time the gasification technology was selected (in
2010). However, the fast pace of technology improvement could result in higher operating
pressures (e.g. 45 bar) of dry-fed gasifiers in the period of 2015-2020, when the results of
the H2-IGCC project could be commercially demonstrated. The adoption of SCGP
technology was also justified by the availability of a validated gasification model
provided by a member of the consortium, Nuon/Vattenfall, who operated the Buggenum
IGCC plant. An assessment of the impact and behavior of various gasification
technologies fed by different coal types and qualities on the overall technical performance
of the IGCC plant with CO2 capture was investigated and has been presented in Paper III.
A relatively cheaper slurry-fed gasification technology could be an appropriate substitute
for the SCGP technology, more specifically when high quality coal reserves are available.
Nevertheless, dry-fed gasifiers can offer more stable performance, even when fed by low
quality coals or biomass.
4.2.3. Syngas conversion
A sour water-gas shift unit has been selected for modeling the IGCC cycle with CO2
capture. This type of shift reaction helps to avoid the additional cooling of the syngas
required by conventional AGR units and then reheating to the level required for the
catalyst’s activation in the SWGS unit. It is also beneficial in order to postpone the water
condensation, which occurs during the conventional AGR process downstream of the
SWGS unit, as the SWGS unit requires the existence of a considerable amount of steam. It
should be noted that the dry-feed characteristics of the Shell gasifier required the injection
of a considerable amount of steam, which adversely affects the steam turbine power
H2-IGCC power plant
67
output. The selection of sour shift reaction was also motivated by industrial partners of the
project.
In order to increase the lifetime of the shift catalyst by eliminating the carbon deposition,
a large amount of steam injection, indicated by a high steam to CO ratio (i.e. 2.4 molar
basis), has been considered.
4.2.4. Acid gas removal
A double-stage physical absorption system using Selexol was selected for the H2S and
CO2 removal from the shifted syngas. The heat required for the regeneration of the solvent
from the acid gas has been provided from the low quality heat, which must be rejected
downstream of the SWGS before the conventional Selexol unit.
The CO2 capture target for simulations has been set to be 90%, as it was found to be an
optimal capture efficiency for IGCC power plants [65]. In order to reach this target, coabsorption of CO2 with H2S should be minimized. This has been achieved by use of the
pre-loaded solvent in the AGR unit. However, using the pre-loaded solvent may have an
adverse effect on the H2S recovery through the decrease in temperature rise within the
absorber column [108]. Furthermore, the solubility data presented in Table 3.3 (see
Chapter 3) may be differentiated due to the higher interactions between polar compounds
such as CO2 and H2S in the pre-loaded solvent [104].
In the absorption process, a part of the combustible constituent of the syngas, i.e. H2 and
CO, are also co-absorbed by the rich solvent. Lowering the pressure to separate CO2 or
H2S from the rich solvent will result in a loss of the combustible gases. Therefore,
products of the first flash drums after H2S and CO2 absorbers are compressed and recycled
to the absorbers to minimize the CO and H2 slips. The high CO2/H2S ratio in the syngas
from the SWGS unit together with a requirement for 90% CO2 capture efficiency resulted
in the production of an unsuitable acid gas stream to the Claus plant due to increased coabsorption of CO2. Therefore, an acid gas enrichment unit was considered in order to
reach a higher H2S content (> 35 mol%) of the acid gas stream.
H2S removed in the AGR section is sent to the sulfur recovery unit, which has not been
modeled in this work. However, the oxygen required for the Claus plant has been
considered for the calculation of the capital costs of the ASU. Furthermore, net steam
68
H2-IGCC power plant
required for the SRU has been assumed to be zero, as the heat required to keep the sulfur
molten and to regenerate the SCOT solvent is balanced by the steam raised by H2S
combustion in the Claus plant, according to [113].
4.2.5. Gas turbine
The baseline GT design has been selected considering the best available gas turbine
technologies. Accordingly, a Siemens SGT5-4000F/Ansaldo Energia V94.3A gas turbine
was chosen, as the manufacturers are partners of the H2-IGCC project. Suitable values for
relevant parameters (e.g. for pressure ratio, gross power, etc.) have been selected, taking
the present SOA technology and the OEMs available data into consideration.
As mentioned in Section 4.2.2 (gasification), one of the most important interactions
between the overall IGCC system and the GT is the required fuel pressure at the GT fuel
valves. The inlet pressure of the fuel upstream of the combustor is dictated by the
compressor outlet pressure. Higher inlet pressure to the fuel valves compared to
compressor outlet pressure should be considered to compensate for a certain pressure loss
between the fuel valves and nozzles. The flame temperature and thereby NOx emissions
are principally controlled by premixing the fuel and air in dry low NOx burners. Such
burners are equipped with a certain number of swirlers to stabilize the flame and to create
the necessary turbulent conditions. This eventually results in higher pressure loss in premixed burners compared to diffusion flame burners. Therefore, a high pressure loss (~10
bar) through the fuel injection system has been considered for the H2-IGCC project.
Syngas can be preheated (up to 200-300 °C) in the IGCC plant prior to the GT
combustion to increase overall plant efficiency, exploiting available waste heat. The
selection of preheating temperature is a compromise between the thermodynamic benefits
at higher temperatures and the operational risks for handling hydrogen-rich, high
temperature syngas (compared to NG) as well as higher fuel system costs [126]. However,
lowering the risk for auto-ignition of H2-rich syngas, this alternative was considered
neither in simulation tasks nor in experimental tests within the H2-IGCC project.
As previously mentioned, for IGCC plants with CO2 capture, either syngas dilution (with
N2 or steam) or syngas saturation (with water) is often considered to control the NOx
emissions from the diffusion flame burners. However, as the goal of the project was to
develop a pre-mixed combustor for the combustion of “undiluted” hydrogen-rich syngas,
H2-IGCC power plant
69
this strategy was not applicable. Once H2-rich syngas was considered as the GT fuel in the
existing GT (i.e. SGT5-4000F/Ansaldo Energia V94.3A) designed for NG operation, the
operating parameters and performance of the GT deviated from the original design.
Therefore, a full off-design analysis was performed in order to realistically simulate those
changes. An existing compressor model was improved using a characteristics map
provided within the H2-IGCC project. The turbine off-design operation was modeled
considering a constant swallowing capacity at choking condition, which is a reasonable
assumption for heavy duty gas turbines:
Swallowing capacity = Constant =
𝑚̇𝑖 √𝑇𝑖
𝜅 𝐴𝑖 𝑝𝑖
(Eq. 4.1)
where,
𝛾
𝜅=√ (
2
𝑅 𝛾+1
)
𝛾+1
𝛾−1
(Eq. 4.2)
The sizing of the entire IGCC plant is governed by the gas turbine as it requires a specific
amount of fuel depending on the fuel composition. The operating condition of the GT has
been determined by matching the operating characteristics of the compressor and the
expander. Thus, if the gas flow rate, e.g. due to the change of syngas composition, varies
at the expander inlet, the operating condition of the GT adapts to this change. This could
result in a change of pressure ratio, even at similar firing temperatures.
When using H2-rich syngas, gas turbine power output increased due to the higher hot gas
flow expanding in the turbine at a certain TIT compared to the NG operation. It should be
highlighted that, in the case of using slurry-fed gasifiers (or water saturation in diffusion
flame burners), the potential for enhanced power output is higher. This is because of the
higher enthalpy drop through the expander due to the higher H2O content in the syngas
and consequently in the flue gas, according to [126].
As shown in Eq. 4.1, the syngas flow rate at the expander inlet is proportional to the
square root of the temperature. In the GT designed for NG in IGCC application, once the
fuel flow rate is increased due to the change in upstream operations (e.g. slip of CO2
capture unit) or transformation of fuel gas, the compressor stability and expander hot gas
path could be affected. Different alternatives to solve various problems incurred by the
70
H2-IGCC power plant
introduction of H2-rich syngas instead of NG are reviewed in Paper IV. As mentioned
previously in Chapter 3, to maintain the GT operation’s stability and safety, TIT could be
de-rated (refer to Eq. 4.1) at the expense of lower GT efficiency.
The addition of one or more high pressure stages to the end of the compressor can resolve
the problem of reduced surge margin due to the higher mass flow of the H2-rich fuel
compared to NG [125]. The turbo-machinery group of the H2-IGCC project has
investigated this option, and the results of their calculation showed that the stable
operation of the compressor could be maintained by just adding one rear stage. The other
strategy adopted by the H2-IGCC project was to modify the turbine, i.e. re-staggering or
opening up the expander nozzle guide vanes (NGVs) in order to increase the swallowing
capacity of the expander. This strategy reflects the fact that industry prefers modifications
to the expander side as it has fewer stages and requires less effort compared to the
compressor. Nevertheless, extensive modifications to the expander should be avoided as
they will be costly and are unlikely to be accepted by the industry. Hence, only
modifications to the first stator of the turbine were followed as the main alternative by the
project.
Note that the modifications could be minor in the case of using an integrated GT-ASU
(air-side). The proper degree of integration could result in just a change of cooling scheme
and no modifications to the expander/compressor designs, as pointed out in [130].
Therefore, the opportunity to keep the NG-designed GT for operation on H2-rich syngas
has been lost by selecting a non-integrated GT-ASU to achieve simpler operation of the
plant, higher reliability and the possibility to run the GT only on NG. The zero integration
necessitated modifications to both expander hot gas path and cooling scheme to keep the
blade wall temperature under its prescribed level provided by the materials group (895 °C
for the 1st expander stator).
4.3. System performance analysis
One of the most important criteria for supporting any decision for investment in a
technology (here, the IGCC technology) is to analyze its performance both technically and
economically. The methodology for technical performance analysis is presented here,
while the method for economic evaluation will be given in the next chapter.
H2-IGCC power plant
71
For the field of power generation, thermodynamic analyses by means of computer-aided
tools have become the most widely used practice. In this regard, thermodynamic
simulations by heat and mass balance programs are cost-effective and fast. In order to
obtain realistic performance indicators, different heat and mass balance programs have
been utilized by this project; these are briefly presented here.
4.3.1. Software tools
The entire IGCC power plant with CO2 capture (and also the NGCC with CO2 capture for
the techno-economic assessments) was modeled by simulation of several sub-systems
mentioned in Chapter 3. Each sub-system’s model and embedded characteristics represent
commercially available technologies, as each major component/sub-system of the IGCC
plant has been broadly utilized in industrial and power generation applications. Access to
the experienced utility owners and operators of similar plants in the H2-IGCC project
provided realistic performance characteristics for the relevant components.
As the main focus of the project was on the gas turbine, IPSEpro software tool, a
commercial heat and mass balance program by SimTech [148], was initially selected for
the modeling of the entire plant as well as the turbo-machinery parts. This choice was
made to reduce the number of software tools and thereby data exchanges between them. It
should be noted that most commercial heat and mass balance programs provide a limited
number of component models and relevant details. Furthermore, the necessary
modifications to the component models are often difficult as access to the source code of
the models and their underlying assumptions is restricted. The main feature of the
IPSEpro software is its component-by-component approach. This capability enables the
modeling of virtually any type of power plant by the integration of basic modules such as
expander, compressor, combustor, steam section, heat exchanger, etc. In addition, the
performance of each component (e.g. gas turbine, HRSG, steam turbine, and pumping
units) can be effectively predicted at their design and off-design points by means of
embedded component characteristics. Though the calculation process does not include the
dimensional design of any components, it is accurate enough to estimate system level
performance of the power block. Moreover, different parameters can be calibrated to
reproduce the performance of advanced gas turbines as realistically as possible. However,
this software suffers from some limitations, including upper limit of the operating
pressure of gaseous streams, as well as lack of enough chemical elements generated
72
H2-IGCC power plant
during coal gasification. Moreover, simulation results for the acid gas removal showed a
considerable difference compared to the results from Aspen Plus, as well as results from
an industrial partner’s simulator (ProMax). This was justified by the differentiated
solubility data of Selexol solvent, which was only based on a certain operating
temperature and pressure and for non-, pre-loaded solvent in IPSEpro.
The mentioned limitations, as well as the specific capabilities of different tools to model
certain sub-systems, resulted in the use of a combination of different software tools,
including IPSEpro, for simulation tasks. In addition to IPSEpro, two main software tools
have been employed to establish the thermodynamic models of the power plant system
and thereby to analyze the thermodynamic performance in this thesis as follows:


Enssim: simulation tool developed by Enssim Software [149]; and
Aspen Plus: commercial process engineering software by Aspen Tech [150].
This approach has been selected to obtain reliable results and to utilize the possibility of
incorporating detailed component characteristics into relevant sub-system models. Even
though different software tools have been used for the simulation of different sub-systems,
proper matching between those tools enables simulation of the entire plant. Data exchange
between software tools was performed manually to find the optimal match, which was a
time-consuming process. A combination of the following simulation tools was used to
model the IGCC and NGCC power plants as follows:




Detailed modeling of the gasification block including various processes, e.g. coal
milling and drying (CMD), gasification, raw syngas cooling and scrubbing, was
performed using the Enssim software tool. Selection of this software was
justified by the fact that a validated gasification model against real plant
operational data was provided by Nuon/Vattenfall that could simulate the process
with a high level of accuracy. The validation results for the Shell gasifier are
available in Paper III. It should be noted that the interface between simulations
performed by the author of this thesis and the Enssim software was only at the
level of data and information exchange to modify the existing gasification model
and to simulate the entire IGCC system.
The air separation unit was modeled using Aspen Plus. The Peng-Robinson (PR)
equation-of-state (EOS) was selected as the properties’ method.
The sour water-gas shift reaction was modeled in Aspen Plus using PR EOS.
The acid gas removal unit was modeled in Aspen Plus. Two different equationsof-state, i.e. Peng-Robinson and perturbed-chain statistical associating fluid
H2-IGCC power plant




73
theory (PC-SAFT), were used for simulation. However, based on a
benchmarking study with one of the industrial partners, the simulation using PCSAFT equation-of-state was selected.
For IGCC plant without CO2 capture, the COS hydrolysis unit and H2S removal
(i.e. AGR unit) was modeled in Aspen Plus, using PR EOS and PC-SAFT EOS,
respectively.
The compression of captured CO2 and dehydration of CO2 stream were modeled
in Aspen Plus, using PR EOS and Schwarzentruber and Renon (SR polar)
equation-of-state, respectively.
The power block, including the GT, and the triple-pressure steam cycle were
modeled in IPSEpro.
The NGCC including the gas turbine, the triple-pressure steam cycle, and the
amine plant for CO2 removal were modeled using IPSEpro software.
Enssim
Gasification
Aspen
O2 & N2
ASU
IPSEpro
H2-rich
syngas
Gas turbine
Steam
Coal milling
and drying
SWGS
Raw syngas
cooling
H2S removal
and
CO2 capture
Syngas
scrubbing
Water
HRSG
Water
Raw
syngas
Steam
turbine
CO2
compression
and
dehydration
Steam
Figure 4.2. Schematic figure of the interface and parameter exchange between different software
tools
As shown in Figure 4.2, the software tools had various interactions with each other,
including the amount of O2 and high pressure N2 from the ASU to the gasifier;
intermediate and high pressure BFW from the HRSG to the gasifier; the high,
74
H2-IGCC power plant
intermediate, and low pressure BFW and IP steam from the HRSG to the SWGS; the
composition and operating parameters of the produced syngas from the gasification block
to the SWGS; the composition and the operating parameters of the syngas from the gas
cleaning unit to the GT; the required syngas flow by the GT to the upstream units; and
different BFW flows and steam flows to the HRSG.
The calculation of the syngas fuel composition was performed by Aspen Plus software. It
was then manually transferred into the IPSEpro gas turbine model. The input parameters
to IPSEpro include the composition of the syngas, any inputs or bleeds of steam or hot
water. Once the fuel flow was determined by the IPSEpro GT model, the backward
calculations were performed to update the coal flow, ASU duties, auxiliary compression
and pumping power demands, etc. Heat integration was finally performed between the
Aspen Plus and IPSEpro models, where heating and cooling streams was required. Given
the final values for heating and cooling inputs, the calculation of the steam turbine power
output and HRSG duties was carried out using the IPSEpro HRSG model.
4.3.2. Boundary conditions
In this section the basic assumptions for thermodynamic calculations are presented,
including the ambient conditions, characteristics of the fuels, and boundary conditions of
the gas turbine as the main focus of this project.
4.3.2.1. Ambient conditions
For gas turbine modeling within the H2-IGCC project (and this thesis), ISO standard was
used as a standard choice in the power industry, as shown in Table 4.1.
4.3.2.2. Feedstock properties
The design feedstock for simulation of the IGCC power plant is a mixture of various trade
coals on the world market (mainly Russia, but also USA, Colombia and South Africa).
The composition and thermal properties of the design coal (bituminous coal) and the
natural gas (used in NGCC simulation) are listed in Table 4.2.
H2-IGCC power plant
75
Table 4.1. Ambient conditions and air composition
Parameter/ component
Unit Value
Ambient air pressure
bar
1.013
Ambient air temperature
°C
15
Relative humidity
%
60
Air composition
N2
wt% 75.10
O2
wt% 23.01
Ar
wt% 1.21
H2O
wt% 0.63
CO2
wt% 0.05
Table 4.2. Composition and thermal properties of bituminous coal and natural gas
Fuel type
Parameter/ component
Unit
Value
Proximate analysis (dry basis)
Moisture
wt%
10
Ash
wt%
12.50
Volatile matter
wt%
27.00
Fixed carbon
wt%
50.50
LHV
MJ/kg 25.10
HHV
MJ/kg 26.20
Ultimate analysis (as received)
C
wt%
64.10
Coal
H
wt%
5.02
N
wt%
0.70
O
wt%
16.09
S
wt%
1.50
Cl
wt%
0.09
Main ash composition
SiO2
wt%
55.00
Al2O3
wt%
24.00
Fe2O3
wt%
5.50
CaO
wt%
4.50
CH4
wt%
95.53
C3H8
wt%
4.02
Natural gas
CO2
wt%
0.40
N2
wt%
0.05
LHV
MJ/kg 49.70
76
H2-IGCC power plant
4.3.2.3. Gas turbine boundaries and performance
Different assumptions made for thermodynamic modeling of various sub-systems of the
selected IGCC cycle with CO2 capture have been presented in Papers I-III and hence are
not given here. However, the general assumptions for the thermodynamic modeling of the
gas turbine designed for H2-rich syngas operation, which have not been presented in the
previous papers, are listed in the following Table 4.3.
a
Table 4.3. Technical assumptions for the modeling of the gas turbine
Parameter
Unit
Value
Compressor
Air flow at the compressor inlet
kg/s
685.2
Air flow at the compressor outlet
kg/s
497.0
Pressure ratio
18.2
st
Cooling flow 1 stator
kg/s
45.7
Cooling flow 1st rotor
kg/s
42.8
Cooling flow 2nd stator
kg/s
31.5
Cooling flow 2nd rotor
kg/s
24.5
Cooling flow 3rd stator
kg/s
12.8
Cooling flow 3rd rotor
kg/s
17.2
Low pressure cooling flow a
kg/s
13.7
Compressor isentropic efficiency
%
89.0
Mechanical efficiency
%
88.7
Expander
Combustor outlet temperature
ᵒC
1500
Turbine inlet temperature
ᵒC
1265
Expander isentropic efficiency
%
92.9
Expander total inlet pressure
bar
17.9
Expander static outlet pressure
bar
1.1
Mechanical efficiency
%
88.7
Gas turbine
Exhaust flow rate
kg/s
709
Exhaust temperature
ᵒC
574
This cooling flow shows a part of the cooling flow which does not go
through the expander and was assumed for cooling of the shaft and
bearings.
H2-IGCC power plant
77
The compressor characteristics map, which relates the compressor mass flow, pressure
ratio, and isentropic efficiency, has been implemented in the compressor model. The
compression power demand was then calculated based on the operating points on the
compressor map. Figure 4.3 shows the generic characteristics map used for modeling the
GT compressor.
22
Pressure ratio [-]
18
IGV
100%
IGV 90%
16
IGV 80%
14
IGV 70%
12
IGV 60%
10
Isentropic efficiency (%)
95%
20
94%
93%
92%
91%
IGV 50%
90%
8
85
90
95 100 105 110 115 120
Corrected mass flow [-]
(a)
8
10
12 14 16 18
Pressure ratio [-]
20
22
(b)
Figure 4.3. Generic compressor characteristics maps, (a) pressure ratio versus corrected mass flow
and (b) isentropic efficiency versus pressure ratio, for different IGV positions
The targeted lumped surface temperature provided by the materials group of the H2-IGCC
project is presented in Table 4.4. It should be mentioned that the equations for the
calculation of the metal temperature have not been incorporated into the GT model.
Hence, the provided data are presented here just to give an overview of the temperature
figures at the expander side, where only the first three stages are cooled with the cooling
flows.
78
H2-IGCC power plant
Table 4.4. Targeted surface lumped temperatures
Parameter
Unit
Value
1st stator
°C
895
1st rotor
°C
879
2nd stator
°C
820
2nd rotor
°C
807
3rd stator
°C
787
3rd rotor
°C
757
4th stator
°C
772
4th rotor
°C
771
The temperature increase for cooling air between extraction and injection due to the heat
loss from the combustion chamber has been set to be zero for the HP cooling flow (for 1st
stator and rotor) and 20 °C for the cooling flows to the 2nd and 3rd stators and rotors, as
shown in Figure 4.4.
20 ºC temperature increase
Fuel
VIGV
1-5
6-9
10-13
Compressor
14-15
S1
No temperature
increase
R1
S2
R2
S3
R3
S4
R4
Expander
Exhaust gas
Air
Shaft cooling
Figure 4.4. Temperature increase for the cooling flows
The calculation of system performance was started using a change of NG to H2-rich
syngas, given the fact that the gas turbine model was able to calculate off-design behavior.
Nevertheless, during an extensive iterative process within the H2-IGCC project, the model
was improved to represent a gas turbine designed for operating on undiluted H2-rich
syngas. During the evolutionary calculation process, off-design operations were
considered to be limited only to the GT and not to the HRSG and steam turbine. This
H2-IGCC power plant
79
could be justified as the gas turbine is extremely sensitive to its design, while the HRSG
and the steam turbine are more flexible and can be adapted to different operating
conditions.
During the modeling and experimental activities in the H2-IGCC project, the following
limits have been considered for continuous operation of the gas turbine and not during
start-up or shut-down.




SO2 emissions from the gas turbine were considered to be less than 10 ppmvd at
15% O2. This resulted in 99.9% removal of the sulfur content by the acid gas
removal unit.
NOx emissions from the gas turbine were considered to be less than 25 ppmvd at
15% O2.
CO emissions from the gas turbine were considered to be less than 10 ppmvd at
15% O2.
Unburned hydrocarbons (UHC) from the gas turbine have been considered to be
less than 10 ppmvd at 15% O2.
Upstream
sub-systems
Waux
Fuel
Gas Turbine
Steam
Turbine
Wp
Wc
We
Wst
Ambient air
Figure 4.5. The boundary for efficiency calculation of the entire IGCC plant
Figure 4.5 shows the boundary for efficiency of the whole IGCC plant, which is
calculated by the following Equation 4.3, considering the mechanical losses, generator
loss, and all auxiliaries.
𝜂𝑛𝑒𝑡 =
(𝑊𝑒 +𝑊𝑐 ) 𝜂𝑚 𝜂𝑒𝑙 +𝑊𝑠𝑡 𝜂𝑚 𝜂𝑒𝑙 +𝑊𝑝,𝐻𝑅𝑆𝐺+ 𝑊𝑎𝑢𝑥
𝑚̇𝑐𝑖 . 𝐿𝐻𝑉𝑐𝑖
× 100
(Eq. 4.3)
5. Economic evaluation
Widespread utilization of any power generation technology depends heavily on its
economic viability in addition to its technical benefits. The demonstration of a new power
plant’s competitive position, compared to other potential technologies, is therefore
essential to attract market attention.
In this regard, a comprehensive cost estimating methodology was adopted and adjusted to
reality, based on the feedback from industrial partners within the H2-IGCC project. In
addition, a techno-economic comparative study was performed to highlight the economic
feasibility as well as the advantages/disadvantages of the IGCC plant compared to other
competing fossil-fuel power plants. One important goal of this chapter is to provide a brief
description of different steps in order to perform the techno-economic evaluation of the
selected energy conversion systems.
5.1. Cost estimating methodology
A complete analysis of any electricity generating system is carried out by an evaluation of
current and future projected costs as well as its performance characteristics. Technoeconomic assessments play an important role in determining the competitiveness of a
selected technology against existing/reference technologies by evaluation of CAPEX and
OPEX in addition to the technical indicators. Such assessments are crucial to investigate
whether and under what circumstances investment in the selected technology is
economically viable. The economic evaluation consists of different stages. Estimations of
capital costs, operation and maintenance (O&M) costs, and fuel costs are necessary to
calculate the cost of electricity (COE).
81
82
Economic evaluation
Economic assessments are not definite and rely on the underlying assumptions as well as
on the choice of selected parameters. There are significant differences in the cost
estimating methods and basis of the calculations employed by various authors and
organizations performing economic assessments of fossil-fuel power plants with CO2
capture [151]. These inconsistencies complicate a fair comparison between the COEs for
different fossil-fuel power plants using different CO2 capture options from various
publishing sources. However, a cost comparison between different alternative systems
based on the same sort of assumptions and methodology is valid even in the presence of
uncertainty in absolute costs of the plant’s components.
Various publicly available reports by different organizations presented their recommended
approaches for cost estimation of power plants [73, 90, 152, 153]. Amongst these reports,
two publicly available reports have been initially selected as sources for equipment cost
data and reference cost estimating methodologies. These two reports are from the
European Benchmarking Task Force (EBTF) under the EU-FP7 CAESAR project [153]
and the National Energy Technology Laboratory (NETL) of the U.S. Department of
Energy (DOE) [90, 154]. The benefit of choosing these two studies from different
organizations was to highlight the effects of different costing methodologies (e.g. different
cost layers and assumptions) and various electricity market conditions (i.e. European and
American types) on the projected capital investments. The methodology used for Paper V
is based on the same methodology used in [90], while the methodology used for Paper VI
is based on what is presented in [153]. Although these reports share many common
features, the final cost estimating method selected for this work is based on the study
provided by the EBTF report [153]. This selection can be justified by the enhancement of
the exploitation of the results achieved during implementation of other European-funded
projects.
A set of assumptions has then been made in order to evaluate the economic indicators of
the selected cycle, i.e. the H2-IGCC plant with CO2 capture, on a consistent basis. The
economic viability of the selected cycle has been measured through the cost of electricity.
This cost indicator is a standard metric used in the assessment of project economics,
which represents the revenue per unit of electricity that must be met to reach breakeven
over the lifetime of a plant. In other words, it is the selling price of electricity that
generates a zero profit. For this purpose, the net present value (NPV) or discounted cash
flow (DCF) computations have been carried out in order to place expenditures that occur
Economic evaluation
83
in different time periods on a common value basis. In addition to cost estimation for the
selected H2-IGCC plant, other alternative fossil-fuel power plants, i.e. a super-critical
pulverized coal plant and a natural gas combined cycle, have been techno-economically
evaluated, and the results have been published in Papers V and VI. The main purpose of
these articles was to compare the technical and economic performance of the selected
power plants. Special emphasis was placed on constructing a set of realistic parameters,
ensuring that the comparison is performed in a consistent and fair way. In Paper VI, in
addition to COE, different aforementioned plants were economically compared using the
cost of CO2 avoided, which will be described later in this chapter. Moreover, economic
sensitivity analyses of the selected plants were investigated, considering the realistic
variation of the most uncertain parameters.
5.1.1. Costing scope
The performed techno-economic studies focus on the commercial installation of each
plant (or nth-of-a-kind technology) and do not cover the costs for the demonstration plants.
The following general considerations have also been taken into account in the technoeconomic studies performed:





The assessments carried out for this project were based on the reference years of
2012 and 2013.
The power plant boundary was defined as the total power plant facility within the
“fence line”. Moreover, site-specific considerations were not taken into account,
and cost estimations were based on a complete power plant on a generic
greenfield site.
Coal receiving and water supply systems were within the battery limit.
Costs associated with CO2 transport, storage, and monitoring were included in
the reported cost of electricity for Paper V, while only the CO2 compression cost
was included in Paper VI.
All performance data are based on nominal base-load operation under clean and
new conditions
Due to large uncertainties in the available cost data for some cost elements, they were
excluded from the assessments. Hence, any labor incentives; costs associated with plant’s
decommissioning; costs associated with the transmission networks, handling distribution
84
Economic evaluation
network and administration of supply; as well as all taxes (with the exception of property
taxes) were excluded from the assessments.
5.1.2. Capital costs
The following sub-sections firstly present the method and equations used for overall
capital costs assessment and then the equations used for costing any component/subsystem of the selected IGCC plant.
5.1.2.1. Step-count costing method
The capital cost assessment for the IGCC plant was based on a bottom-up approach
(BUA). The BUA is the step-count exponential costing method using dominant
parameters or a combination of parameters derived from the mass and energy balance
simulation. The capital costs levels is illustrated in Figure 5.1, showing that there are three
main levels, i.e. total direct plant costs (TDPC), engineering, procurement and
construction costs (EPCC), and total plant costs (TPC).
Figure 5.1. Capital costs levels and their elements for a bottom-up costing approach
The following equations show the calculation method for total equipment costs (TEC) and
TDPC, respectively. Table 5.1 lists various sub-systems and plant components which
were systematically grouped according to their processes in the IGCC plant.
Economic evaluation
85
TEC = ∑𝑛𝑗=1 𝐶𝑗
(Eq. 5.1)
TDPC = ∑𝑛𝑗=1 𝐶𝑗 + ∑𝑛𝑗=1 𝐼𝑗
(Eq. 5.2)
Table 5.1. Major plant components of the IGCC plant with CO2 capture
#
Plant components
1
Coal handling
2
Gasifier
3
Gas turbine
4
Steam turbine
5
Heat recovery steam generator
6
Low temperature heat recovery
7
Cooling
8
Air separation unit
9
Ash handling
10
Acid gas removal
11
Gas cleaning
12
Water treatment
13
Sour water-gas shift
14
Claus burner
15
SELEXOL plant
16
CO2 compression
The engineering, procurement and construction costs were calculated using the following
equation:
EPCC = TDPC + IC
(Eq. 5.3)
The indirect costs (IC) were considered for the integration of the individual modules into
the entire plant, such as costs for piping/valves, civil works, instrumentations, and
electrical installations. The indirect costs can be simplified as a fixed percentage of the
TDPC. An example of the simplified indirect costs with relevant assumptions is shown in
Table 5.2.
The total plant costs are the sum of EPCC, owner costs (OC) and contingencies (process
and project contingencies), which is shown in the following equation:
TPC = EPCC + OC + Cproc,c + Cproj,c
(Eq. 5.4)
86
Economic evaluation
Table 5.2. Breakdown of the indirect costs
Indirect costs
% of TDPC
Yard improvement
1.5
Service facilities
2.0
Engineering/consultancy costs 4.5
Building
4.0
Miscellaneous
2.0
Total indirect costs
14.0
5.1.2.2. Capacity adjustment
The capital costs for plant components could be found in the open literature. However,
these data could not be used unless they were made consistent by using correction of size
and the reference year. Calculation of the equipment cost for a certain plant, based on
utilization of the cost data for different component sizes, could be performed using the
following equation:
𝑆
C𝑗 = C𝑗,𝑟𝑒𝑓 ( 𝑗⁄𝑆
𝑗,𝑟𝑒𝑓
)
𝑓
(Eq. 5.5)
The term (𝑓), cost scaling exponent, incorporates economies of scale in the equation and
indicates that the percentage change in cost is smaller than the percentage change in size
for each major component. The typical values of the scaling exponent for power utilities
vary between 0.6-0.7 [155].
5.1.2.3. Price fluctuations
The economic evaluation, based on the cost data found in the literature, should consider
the economic ups and downs (market fluctuations) from the date of the original cost data
to the current time. The cost adjustments are necessary since equipment cost estimates
correspond to a specific time. All the cost data used in the economic evaluation need to be
brought to the same reference year to reflect the market conditions for that specific year.
The adjustment for price fluctuations of equipment, materials, and labor over time could
be performed using a suitable cost index (CI) such as the Chemical Engineering Plant
Cost Index (CEPCI), the Marshall and Swift (M&S) cost index, etc. The cost index ratio
(IR) for a component is achieved by using the following equation:
𝐼𝑅 =
𝐶𝐼𝑢𝑏𝑦
𝐶𝐼𝑜𝑏𝑦
(Eq. 5.6)
Economic evaluation
87
The updated cost for a component from its original base year could be then adjusted using
the following equation which is derived from Eq. 5.5 and Eq. 5.6.
𝑆
C𝑗 = C𝑗,𝑟𝑒𝑓 ( 𝑗⁄𝑆
𝑗,𝑟𝑒𝑓
𝑓
) . 𝐼𝑅
(Eq. 5.7)
5.1.2.4. Currency exchange
In the economic calculations carried out, all figures extracted from the literature given in
different currencies (e.g. US$ or €) were recalculated to the desired currency using the
universal currency conversion XE rates [156].
5.1.3. Operation and maintenance (O&M) costs
The operations and maintenance costs are the costs associated with operating and
maintaining the power plants over their expected lifetimes. These costs usually include:





Operating labor;
maintenance (materials and labor);
consumables;
waste disposal and management; and
co-product/by-product credits (negative costs for any co/by-products sold can be
considered).
The abovementioned costs are classified in two categories: the fixed O&M costs, which
are independent of the plant output (products) such as labor cost, overheads, insurance,
and property taxes; the O&M costs that vary proportionally to the plant output are
variable costs. These costs include consumables (such as water, chemicals, solvent, and
catalysts) and waste disposal.
5.1.4. Fuel cost
The fuel cost, similar to variable O&M costs, is dependent on the plant output. Although
the coal cost, based on mine-mouth coal prices, has been stable over recent years, the
market price shows significant variation. The price for the bituminous coal was, therefore,
based on the market price. The fuel quantity, for the fuel cost calculation, was taken from
88
Economic evaluation
simulation results, and the corresponding cost was determined on the basis of yearly
consumption.
5.1.5. CO2 cost measures
A variety of measures are used in the literature to report the cost of CO 2 capture and
storage systems for power plants. The most common measures include the cost of CO 2
avoided and cost of CO2 captured [18].
The cost of CO2 avoided compares a plant with carbon capture with a reference plant
without capture and quantifies the cost of avoiding CO2 emissions for the provision of
electricity, which is defined as:
𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑂2 𝑎𝑣𝑜𝑖𝑑𝑒𝑑 [€/𝑡𝐶𝑂2 ] =
𝐶𝑂𝐸𝑐𝑎𝑝𝑡𝑢𝑟𝑒 −𝐶𝑂𝐸𝑟𝑒𝑓 [€/𝑀𝑊ℎ]
𝐶𝑂2 𝑠,𝑟𝑒𝑓 −𝐶𝑂2 𝑠,𝑐𝑎𝑝𝑡𝑢𝑟𝑒 [𝑡𝐶𝑂2/𝑀𝑊ℎ]
(Eq. 5.8)
where 𝐶𝑂2 𝑠 is tonne of CO2 emissions to the atmosphere per MWh (based on the net
capacity of each power plant), and the subscripts “𝑐𝑎𝑝𝑡𝑢𝑟𝑒” and “𝑟𝑒𝑓” refer to plants
with capture and without capture (or reference plant), respectively. It should be
highlighted that the cost of CO2 avoided can be more comprehensive, incorporating the
costs associated with CO2 capture, transport and storage rather than only considering the
capture part. However, the boundary conditions for this study did not include transport
and storage steps, as these areas are different research fields that could not be covered by
the H2-IGCC project.
As shown in Eq. 5.8, calculation of the cost of CO2 avoided requires the definition of a
reference plant. This could be an identical/similar plant of the same type as the plant with
CO2 capture or a different plant type. The choice of an identical/similar reference plant is
typically made to quantify the cost of CO2 avoidance for a particular technology. Such a
choice is also made assuming that the investigated technology has a similar chance to be
built in future under a no-carbon-constraint scenario [157].
Another important cost measure is the cost of CO2 captured for a particular capture
technology in a specific type of power plant [18]. This measure is to quantify only the cost
of capturing CO2 and the economic viability of a CO2 capture system could be evaluated
Economic evaluation
89
using this measure compared to the CO2 market price as an industrial commodity [42].
The cost of CO2 captured for a power plant is defined as:
𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑂2 𝑐𝑎𝑝𝑡𝑢𝑟𝑒𝑑 [€/𝑡𝐶𝑂2 ] =
𝐶𝑂𝐸𝑐𝑎𝑝𝑡𝑢𝑟𝑒 −𝐶𝑂𝐸𝑟𝑒𝑓 [€/𝑀𝑊ℎ]
𝐶𝑂2 𝑠,𝑐𝑎𝑝𝑡𝑢𝑟𝑒𝑑 [𝑡𝐶𝑂2 /𝑀𝑊ℎ]
(Eq. 5.9)
where the subscript “𝑐𝑎𝑝𝑡𝑢𝑟𝑒𝑑” shows the total mass of CO2 captured per net MWh for
the power plant with capture. It should be noted that, in this case, the reference plant is the
same type as the plant with capture unit.
The cost of CO2 captured is always lower than the cost of CO2 avoided, mainly because
the efficiency penalty caused by the CO2 capture unit means that more CO2 is captured
than avoided per net MWh generated (see also Figure 5.2). The values illustrated in
Figure 5.2 are based on the selected IGCC cycle with and without capture unit in Paper
VI.
CO2 emitted
CO2 captured
60
700
50
600
500
40
400
300
€/t CO2
CO2 avoided
CO2 produced (kg/MWh)
800
30
20
200
10
100
0
0
Reference plant
Plant with capture
Cost of CO2 captured
Cost of CO2 avoided
Figure 5.2. The relationship between the CO2 emitted, avoided and captured (left) and the cost of
CO2 captured and avoided (right)
5.2. Uncertainty in the economic results
Generally, some degree of uncertainty is expected in economic and technical performance
data for any technology. Additional uncertainties are commonly encountered in executing
a project which results in an increase in cost [158]. Uncertainty reflects lack of
knowledge/experience about the precise value(s) of one or more parameters affecting the
90
Economic evaluation
economic (or technical) performance of a technology [157]. In any case, the most mature
technologies show the smallest range of uncertainty compared to what is demonstrated by
the new technologies.
The IGCC technology is a complex energy conversion system. Moreover, operating
experience with IGCC power plants is limited compared to e.g. NGCC and SCPC plants.
In addition, currently there is no pre-combustion carbon capture system operating on a
commercial scale. As a consequence, there are substantial uncertainties associated with
cost data and technical performance for any economic assessment related to IGCC plants
with CO2 capture [65]. The most important uncertainty factors or sources of uncertainties
in the economic assessments carried out are summarized below:




The current and expected heat and electricity market conditions can have a major
impact on the capital costs of the plants, as well as financial assumptions such as
discount rate. With the current market condition for fossil-fuel power plants,
which is considered a volatile market, a low capacity factor can be considered
due to the increased share of renewable energy sources; a high discount rate may
be applied as investors try to gain a return on their investments as fast as they
can. In addition, assumptions about market prices for e.g. chemicals, catalysts,
etc. are uncertain.
Different technical assumptions such as process design assumptions and
parameters used for simulation such as equipment sizing parameters,
requirements for catalysts, chemicals and consumables are also sources of
uncertainty in the economic results.
As no existing full-scale carbon capture plant has been integrated into a power
plant on a commercial scale, any estimates have been made from scaling up from
prototypes or detailed bottom-up engineering estimates. Therefore, there is a
high degree of uncertainty in the cost data of the CO2 capture systems, including
capital costs and O&M costs, apart from technical performance such as the
additional energy consumption required for the capture unit.
There is uncertainty as to how the state of technology of all CO2 capture systems
(including pre-combustion) will be developed in the future, even though it is
expected that the costs of capture technologies will decrease in future [34].
However, this cost reduction is strongly connected to experience gained by more
demonstration plants and incremental technological improvements.
Economic evaluation

91
The IGCC technology is not currently a widely deployed technology so the cost
of IGCC plant itself (even without CO2 capture unit) is somewhat uncertain.
There is also the possibility that substantially cheaper technologies may become
commercially available (e.g. ITM for O2 production in IGCC cycles).
Given all these sources for uncertainties in economic (as well as technical) results,
performing a sensitivity analysis is a way to examine the effects of uncertainties (or
variability) in key parameters on the economic results. Therefore, such analyses were
carried out in order to disclose the effect of a plant’s capacity factor (or load factor) and
fuel price on the economic attributes of the selected IGCC plant.
6. Concluding remarks
The ever-increasing demand for electricity has been faced with a global concern, i.e.
increasing worldwide GHG emissions. Several potential pathways to mitigate these
emissions have been investigated during recent years. The most important ones, having
substantial impacts, are increase in the renewable energy share in the power mix, increase
in energy efficiency, and carbon capture and storage.
As one of the leading stakeholders, the European Union set a 20% reduction in GHG
emissions (compared to 1990 level) by 2020 and has included CCS in the portfolio of
technologies to meet this target. Accordingly, many R&D projects were financed by the
Directorate-General for Energy (European Commission) under the Sixth and Seventh
Framework Programmes including the Low Emission Gas Turbine Technology for
Hydrogen-rich Syngas (H2-IGCC) project in 2009. As mentioned earlier, this PhD study
has been carried out as a part of the research activities under the framework of the H2IGCC project. The following sub-chapters will summarize the main findings/conclusions
of this work and the scientific contributions of this study as well as offering some
suggestions for further investigations which can be accomplished by future research
activities.
6.1. Conclusions
The two important driving forces for defining, obtaining financial support for, and
implementing the H2-IGCC project were the continuing need to use coal as primary fuel
for the security of the energy supply and the requirements to curb CO2 emissions. The
electricity supply must be secured by utilizing various environmentally-friendly
technologies in every modern society. Undoubtedly, IGCC plants can contribute to the
93
94
Concluding remarks
security of the electricity supply under stringent emission regulations. However, it should
be clearly underlined that electricity must be supplied at an affordable cost so that the
global competitiveness of countries/regions is not affected in negative way. The technoeconomic results presented by this study showed that the three fossil-fuel power
generation alternatives without CO2 capture perform quite similarly with respect to the
cost of electricity. However, IGCC and SCPC are advantageous among plants without
capture based on underlying assumptions. The marginal difference in the cost of
electricity was within the level of uncertainties in the assessment of investment costs.
Therefore, other main drivers, apart from the cost of electricity, affect the selection of a
power generation technology including:






operational flexibility and availability;
compatibility with grid requirements assuming much higher share of renewable
energy sources in future energy mix and the risks for underperformance;
compatibility with utilities’ experience;
availability and diversity of equipment and technology suppliers;
various aspects relevant to health, safety and environment (HSE); and
potential for future improvements.
Given these criteria, opportunities for substantial economically attractive investments in
IGCC plants without CO2 capture remain questionable. Under current electricity market
conditions, new investments even in “standard” fossil-fuel power plants, i.e. pulverized
coal and NG simple and combined cycles (without CO2 capture), are foreseen to be
limited in Europe. This is mainly due to the increasing share of renewable energy sources
in the European power mix, which has had a tremendous impact on operating strategies
and the profitability of fossil-fuel power plants. Anyhow, intermittent RE requires
reliable-, fast balancing/backup power plants as well as large storage capacity. Hence, gas
turbine cycles fueled with NG, which are much faster, are superior compared to coalbased plants (e.g. IGCC and SCPC). However, the reduction in the coal price in Europe,
mainly due to coal import from the USA as well as inexpensive costs for carbon
emissions, has recently resulted in the increased use of old coal plants (which have
already repaid their investment costs) in this region, compared to costly, high-efficiency
and low emission NGCC plants.
Concluding remarks
95
The limited or non-existent tendency of European power market might change from
standard fossil-fuel plants to IGCC plants in a carbon-constrained future when CCS
technologies will play an important role in the mitigation of GHG emissions. In this
context, the value of CO2 credits should be established with certainty, and appropriate
regulations on the required CO2 quality, storage access, monitoring of the storage sites,
etc. should be introduced. Even then, the IGCC technology with CO2 capture should
prove its competitive position against other low-carbon emissions technologies with
respect to issues like economic viability, operability, availability and reliability at a high
share of renewable energies in the global power mix. Assuming all these issues will be
resolved by policy changes and technological improvements, the findings of the current
thesis indicated that higher carbon prices should be set for the economic benefits of cycles
with capture compared to their reference plants (i.e. without capture).
Given all the aforementioned uncertainties and challenges facing the application of IGCC
technology, the precautionary principle suggests that doing nothing is not the best choice.
Indeed, it is quite clear that investigations such as those presented by this work are highly
necessary, especially when political and market conditions are being changed to force new
fossil plants to be built only with CO2 capture. Perhaps under these assumptions, and
given the need to keep the electricity supply versatile (energy diversity), IGCC application
will still be limited in some localities with abundant coal reserves such as China and the
USA.
In addition to aspects related to the aforementioned general view, such as current market
condition, economic viability, and the risk of CCS deployment, technological aspects of
the H2-IGCC project also need to be presented. Theses aspects include the application of
pre-mixed combustion for H2-rich syngas, knowledge built within the field of system
analysis, and techno-economic assessment.
It is questionable whether and under what condition the technology proposed in this thesis
for the combustion of H2-rich syngas can be employed in gas turbines specifically for
IGCC application in the near future. However, rapid changes in the global structure of
heat and electricity supply and demand have a tremendous impact on the application of
combustion technology developed within the H2-IGCC project. Likewise, via the
application of power-to-gas to reach the maximum utilization of renewable energy
sources, the hydrogen produced needs to be stored, perhaps in the existing NG pipelines.
96
Concluding remarks
This will assist the existing infrastructure (e.g. pipelines) to be maintained and the carbon
footprint to be reduced. Knowledge built in this project will then enable the use of premixed combustion technology for high H2 content natural gas in the gas turbines, which is
amongst the options to balance/back-up renewable energies.
With respect to the system analysis performed for this project, the knowledge built will be
helpful in taking a holistic approach to analyzing any other energy conversion systems. In
addition, the level of detail in every component of the system was appropriate to provide
the necessary boundary conditions and data for combustor design, gas turbine design and
techno-economic assessment. However, it should be highlighted that different types of
optimizations are still required for successful utilization of the IGCC technology, such as
heat integration, cost-benefit optimization, desired level of integration, etc.
The cost estimate presented in this work clearly confirmed the considerable negative
impact of applying CO2 capture systems in different power plants, as the total investment
and cost of electricity are much higher compared to the same plant type without CO 2
capture. Accordingly, in order to make power plants with CO2 capture economically
attractive, the cost of emitting CO2 must be much higher than the current cost for CO2
allowances. Finally, it should be noted that the economic calculations performed here are
relevant based on data available today and underlying assumptions. Such uncertainties
should be kept in mind when interpreting the outcomes of this thesis.
6.2. Scientific contributions
The research work presented in this thesis places emphasis on the development of
technical solutions to allow the use of highly efficient gas turbine technology in the IGCC
plants with CO2 capture suitable for combusting undiluted H2-rich syngas. The two major
contributions of this study are:
1.
System analysis and integration: system analysis using detailed validated models
provides highly valuable contributions concerning low cost, reliable results prior
to any piloting and demonstration activities. In this regard, system integration
alternatives with a high degree of complexity in both the IGCC plant and the
integrated pre-combustion carbon capture were evaluated. This has shed light on
the pros and cons of various alternatives, paving the way for future
Concluding remarks
2.
97
implementation of the most efficient and practical system integration
alternatives.
Techno-economic model and assessments: the available techno-economic
approaches for the power plant were thoroughly reviewed, and the most suitable
method was selected. Accordingly, correction/adjustment of these methodologies
was carried out. Realistic cost and performance data supported by the industrial
partners of the project were then used to establish a solid base for a comparative
techno-economic study. The tool developed in the Microsoft Excel environment
provided the opportunity to update and modify any underlying assumptions and
enabled the economic evaluation of the IGCC plant with carbon capture as well
as its main competitors with a good level of accuracy.
The following list presents the other secondary contributions, which enhance the current
knowledge in this field:
i.
An undiluted H2-rich syngas was used for the gas turbine modeling and
simulation, and the plant’s configuration was established and modified compared
to what is available in the literature.
ii. A non-integrated ASU-GT was selected to provide more availability and
flexibility to the operation of the IGCC plant. The plant’s overall performance
data could be marginally better in more integrated layouts but at the expense of
additional costs as well as less availability.
iii. Different gasification technologies have been investigated and integrated into the
selected IGCC configuration in order to explore the most appropriate option for
the application of pre-combustion CO2 capture in IGCC plants.
iv. Fuel flexibility targets in the gas turbine, with respect to fuel change due to slip
of CO2 capture unit, could not be accomplished using an identical combustor
designed for H2-rich syngas. This was mainly due to the large difference between
thermal properties of the H2-rich syngas and the syngas produced in the noncapture IGCC plant.
v. The use of existing gas turbine technology, which is designed for NG operation,
would not be appropriate for handling H2-rich syngas. In this regard, a new gas
turbine was designed by other partners, involving some modifications, mainly in
the expander. Accordingly, the boundary conditions generated were used to
update the GT model and the overall IGCC plant.
98
Concluding remarks
6.3. Suggestions for further research
The following topics from different perspectives, i.e. a holistic view on energy conversion
systems to a detailed technological level, are considered by the author as an appropriate
continuation in this field and thus recommended for further research:
In the context of fossil-fuel energy conversion systems
vi. An investigation into the operating strategies of newly built or existing fossil-fuel
power plants under current market conditions is highly essential. Therefore, a
techno-economic study on existing fossil-based technologies for power
generation could be performed, defining different scenarios for increasing the
share of renewable energies and the need for fossil-fuel plants as a backup/balancing option. The major difference from the underlying assumptions
made for the current study would then be operating at part load rather than at
base load. The cyclic operation of the fossil-fuel power plants and its effects on
maintenance costs and lifetime consumption of different parts could be
incorporated to improve such an analysis.
In the context of system integration and analysis of the IGCC power plants
vii. In order to achieve better performance indicators of the IGCC plant, alternative
technologies listed in this thesis, such as ITM for air separation or SEWGS for
shift reaction and CO2 capture, could be integrated into the cycle. However, the
efficiency improvements should be evaluated against the economic implications
and operational challenges. It should be highlighted that different types of
optimizations, such as heat integration, cost-benefit optimization, and desired
level of integration, are required to make the IGCC technology ready for future
application.
viii. In order to have appropriate control over the simulation and modeling of such an
integrated and complex energy conversion system (i.e. IGCC system with CO2
capture), integrating software tools could be beneficial. For this purpose, it could
be an option to generate dynamic link library files, assuming that all software
tools used for this study are available. This option provides all the benefits which
could be gained by using each and every one of the previously mentioned
software tools. In the case of performing a simple techno-economic analysis, it
Concluding remarks
99
would be beneficial to simulate the whole system in e.g. IPSEpro or ASPEN
Plus, assuming the level of uncertainties in the cost assessments.
In the context of hydrogen-rich fueled gas turbine
The dynamic behavior and off-design conditions of the gas turbine when it is fed by a
hydrogen-rich syngas need to be investigated. Off-design modeling of such gas turbines
will be very useful, especially when the gas turbine technology is used as a back-up or
balancing power option for renewable energy sources. Power-to-gas technologies might
be considered for storing a part of intermittent RE and then high hydrogen content NG
might be used as a fuel for GTs, perhaps not at the base load condition.
7. Summary of appended papers
This chapter briefly presents the main findings of the papers appended to this thesis.
These papers are mainly related to the establishment of the baseline IGCC plant for the
purpose of system analysis; an investigation of the effects of coal quality and gasification
process type on the overall performance of the selected IGCC plant; a study of the effects
of fuel flexibility on the performance of the selected gas turbine; and techno-economic
comparatives studies on different fossil-fuel power plants including the selected IGCC
plant.
Paper I Development of H2-rich syngas fuelled GT for future IGCC power plants –
Establishment of a baseline, Presented at ASME Turbo Expo 2011, GT2011-45701,
Vancouver, Canada, June 2011.
This paper presents the establishment of two baseline IGCC power plants, i.e. with and
without pre-combustion CO2 capture. For this purpose, different sub-systems including
gas cleaning, gas turbine, steam turbine, heat recovery steam generator along with the
inputs from the industrial partners Vattenfall/Nuon and E.ON were integrated. The gas
turbine used for this study is based on Ansaldo Energia 94.3A without any dilution of the
syngas. The main goal of this study was to provide a baseline for further investigations
incorporating the necessary changes/modifications related to the gas turbine during the
lifetime of the H2-IGCC project. The secondary objective was to provide the potential of
burning undiluted H2-rich syngas and its effects on enhancement of the efficiency of the
IGCC power plants with CO2 capture.
101
102
Summary of appended papers
The analysis shows that the combustion of H2-rich syngas has the potential for increasing
the overall IGCC efficiency compared to data available in the literature for IGCC plants
with diluted syngas and CO2 capture. The overall efficiencies of the plants are 37.4% and
47.2% (LHV basis) respectively for the IGCC plant with CO 2 capture and for non-capture
IGCC. The difference between two configurations, IGCC with and without CO2 capture,
results in two completely different syngas compositions. Preliminary results of this study
show that combustion of undiluted H2-rich syngas does not impose any significant effects
on the gas turbine, at least from a system perspective. However, the large change in fuel
flow in the case of non-capture IGCC plant generates some challenges for both the
combustion process and the turbo-machinery.
Paper II An EU initiative for future generation of IGCC power plants using hydrogenrich syngas: Simulation results for the baseline configuration, Applied Energy, Vol. 99,
pages 280-290, June 2012.
This paper is in continuation of Paper I to investigate the use of undiluted H 2-rich syngas
in the IGCC plant with CO2 capture. However, simulation of the gas cleaning part of the
IGCC plant including the acid gas removal unit and the CO2 capture system was
performed using ASPEN Plus, unlike Paper I which was in IPSEpro. The main reason for
this was the specific capabilities of ASPEN Plus to simulate gas cleaning processes.
Moreover, additional plant’s components were integrated into the system to provide more
comprehensive and practical plant layout.
The paper presents a detailed thermodynamic model of the baseline IGCC plant with and
without CO2 capture. Realistic performance indicators verified by the operators of
similar/relevant plants were used, as compared to Paper I, which was mainly based on
data available in the literature. In addition to changes in performance indicators of some
plant’s components, some information on the GT provided by other partners was
incorporated into the GT model.
Results revealed that the effects of these changes/modifications on the model presented in
Paper I were negative in terms of the overall plant efficiency. The estimated overall
efficiency of the IGCC power plant without carbon capture is 46.3%, while it is 36.3% for
the plant with carbon capture, somewhat lower than the results presented in Paper I. The
Summary of appended papers
103
results confirm the fact that a significant penalty on efficiency (21.6% relative) is
associated with the capture of CO2.
Through comparison with other published studies, more integration of sub-systems
indicated some potential for better efficiency, although probably at the expense of lower
reliability. Using undiluted syngas in the GT significantly improves GT power. However,
some challenges related to the unstable operating condition of the GT combustor and
compressor, as well as reduced lifetime of the blades of the existing gas turbines when
using undiluted H2-rich syngas, should be addressed by future studies.
Paper III Estimation of performance variation of future generation IGCC with coal
quality and gasification process – Simulation results of EU H2-IGCC project, Applied
Energy, Vol. 113, pages 452-462, August 2013.
This paper presents the effects of gasifier type and coal quality on the overall performance
of the baseline configuration of the IGCC plant. In this regard, four commercially
available gasifiers from Shell, GE, Siemens, and ConocoPhillips have been considered for
this comparative study. The effects of three different types of coals on these gasifiers, as
well as on the overall performance of the IGCC plant, have been investigated. Utilizing
validated models against existing plant data for simulation of gasification block resulted
in more reliable results.
The results confirm that the coal quality considerably influences the cold gas efficiency
for slurry-fed gasifiers, while dry-fed gasifiers are relatively insensitive to the quality of
the input coal. Amongst slurry-fed gasifiers, the coal quality has the greatest impact on the
performance of the GE gasifier. The cold gas efficiency of the GE gasifier gasifying
lignite coal is 29% lower than gasifying bituminous coal. It is also shown that dry-fed
gasifiers are advantageous compared to slurry-fed types with respect to constant quality of
produced syngas even when low-rank coal is gasified. Based on the findings of this paper,
slurry-fed gasifiers investigated in this study, i.e. GE and ConocoPhillips, are suitable for
bituminous and sub-bituminous coals, while dry-fed gasifiers, i.e. Shell and Siemens,
show a relatively constant behavior for a wider range of coal quality.
104
Summary of appended papers
The higher water content of the produced syngas from slurry-fed gasifiers results in an
enhanced ST power output due to reduction of the steam extraction from the steam cycle
for the water-gas shift reaction. However, this power increase cannot compensate for the
increase of ASU power demand and results in lower system efficiency for low-rank coal.
Paper IV Fuel change effects on the gas turbine performance in IGCC application,
Presented at 13th International Conference on Clean Energy (ICEE-2014), Istanbul,
Turkey, June 2014.
The effect of fuel change (i.e. from NG to H2-rich syngas and clean syngas) for the
selected GT is reported in this paper. This study focused on the operation of the gas
turbine as a stand-alone unit. The results of this paper proved the preliminary findings of
Paper I, showing that operation on undiluted H2-rich fuel (syngas produced in the IGCC
plant with CO2 capture) is feasible. However, a reduced surge margin should be accepted
without significant changes made to the gas turbine compared to the NG-fired engine. It
should be noted that the challenges concerning pre-mixed combustion of the H2-rich fuel
and different heat transfer rate to the expander materials when operated with H 2-rich fuel
are not within the scope of this study.
The GT operation on clean syngas (i.e. syngas produced in the IGCC plant without CO 2
capture) results in a significantly low surge margin and high turbine outlet temperature,
which needs different operating conditions and/or engine modification options to be
considered. When operating with a fuel with low calorific value, such as clean syngas,
expected operational hours are very important for the selection of appropriate operating
conditions or modification options. Although several modification options as well as
operating strategies have been suggested in this paper with regard to clean syngas
operation, reduced efficiency and compressor stability can be tolerated for limited
operational hours with clean syngas.
The results revealed that the effect of the altered VIGV angle on maintaining a reasonable
surge margin is not significant for the selected GT. In order to have a minor modification
of the GT compared to the design case engine for clean syngas operation, decreasing the
TIT and keeping the TOT similar to the reference case (NG-fired GT) with fully open
VIGV is a plausible option. However, results show a significant reduction of efficiency
Summary of appended papers
105
and power output. Concluding this paper, using clean syngas requires major modifications
on the GT, including additional compressor stages, air bleed from compressor outlet, and
expander re-staggering, which resulted in putting this option (i.e. operation on the clean
syngas) aside within the H2-IGCC project.
Paper V Techno-economic evaluation of an IGCC power plant with carbon capture,
Presented at ASME Turbo Expo 2013, GT2013-95486, San Antonio, Texas, USA, June
2013.
This paper presents a techno-economic analysis for the selected IGCC plant configuration
with CO2 capture using the cost data and methodology of the U.S. Department of Energy.
The main objective was to generate a database using publicly available literature to
calculate the COE for the IGCC plant. The secondary objective was to compare the COE
for the IGCC plant with other fossil fuel competing technologies, i.e. NGCC and SCPC
plants, all with CO2 capture system. In this paper, the methodology used for the economic
evaluation of the plant, as well as relevant assumptions, calculation methods, and
economic figures are described.
The COE for the IGCC plant with CO2 capture is projected to be 160 US$/MWh. It
should be noted that all economic results are strongly dependent on presented
assumptions. Therefore, a sensitivity analysis was also carried out, showing that the most
influential parameter amongst selected parameters on the COE is the capacity factor. The
fuel price was the second ranked parameter. As the selected IGCC plant is considered
with CO2 capture, most of the CO2 produced is actually captured. Moreover, the costs for
CO2 allowances are very low. Therefore, the effect of CO2 allowances’ costs on the COE
is negligible. Finally, a comparative study was carried out to highlight the cost difference
between various power generation technologies, i.e. IGCC, SCPC, and NGCC plants with
CCS. The total overnight costs for IGCC, SCPC, and NGCC with CO2 capture are
estimated at 4677, 4065, and 1669 US$/kW, respectively.
The results revealed that the investment cost in the IGCC plant is projected to be more
than double that for the NGCC plant. However, other aspects such as the security of the
energy supply may encourage investors to select IGCC plants. Moreover, it is shown that
with the higher capacity factor and CO2 allowances’ cost, which is plausible in the coming
106
Summary of appended papers
years, the IGCC plant could attract more investments compared to the SCPC plant.
Furthermore, income from poly-generation applications might also improve the economic
viability of future IGCC plants.
Paper VI Techno-economic assessment of fossil fuel power plants with CO2 capture ‒
Results of EU H2-IGCC project, International Journal of Hydrogen Energy, Vol. 39,
pages 16771-16784, September 2014.
In this paper the thermodynamic performance indicators of various power plants,
including IGCC, advanced supercritical pulverized coal, and natural gas combined cycle
power plants are presented. The technical indicators of the selected IGCC plant and
NGCC with and without CO2 capture unit are based on the simulations carried out during
the lifetime of the H2-IGCC project. These indicators for the SCPC were adopted from
literature.
Results confirm that the NGCC is the most efficient plant, while the advanced SCPC plant
is the least efficient plant amongst non-capture cases. This trend is similar for the plants
with capture unit. The relative efficiency penalties associated with the capture deployment
(compared to the identical plant with CO2 capture) are 24%, 27%, and 16% for the
selected IGCC, advanced SCPC, and NGCC plants, respectively.
In this article, a comparative study was also conducted, comparing the COE and the cost
of CO2 avoided for the mentioned fossil-based power plants. The economic performance
indicators of each plant were estimated using the model developed in the Microsoft. Excel
environment. It is very important to take into account that such a techno-economic
analysis cannot provide an absolute result, since the cost data and assumptions are
uncertain by nature.
The COE for the IGCC plant with and without capture is 91 and 59 €/MWh, respectively.
The COE for the advanced SCPC is 96 and 59 €/MWh for the capture and non-capture
cases, respectively. The COE for the NGCC with and without capture is 61 and 91
€/MWh, respectively. The results show that the least capital-intensive plant is the NGCC
plant without CO2 capture. However, the high fuel costs for this plant decrease the gap
between the COE for this plant compared to that for the other plants. The COE for the
Summary of appended papers
107
NGCC technology was the most sensitive to changes in the fuel price amongst other
COEs for different technologies. However, the COE for the NGCC technology was also
the least sensitive to variations of the plant’s capacity factor. The estimated costs of CO 2
avoided for the IGCC, SCPC, and NGCC technologies are 51, 57, 99 (€/t CO 2 avoided).
Results highlighted that, based purely on the COE for different plants, it cannot be
concluded which technology is better and more cost-effective than other technologies,
considering the level of uncertainty in the economic results of this study (+/-30%). Other
main drivers such as proven technology and operational flexibility will, therefore, play an
important role in the widespread utilization of these technologies.
Bibliography
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
ExxonMobil, The outlook for energy: A view to 2040, 2013, ExxonMobil
Company: Irving, Texas, USA.
IPCC, Climate Change 2007: Fourth Assessment Report (AR4), 2007,
Intergovernmental Panel on Climate Change.
IEA, IEA statistics 2013 edition: CO2 emissions from fuel combustion highlights,
2013, International Energy Agency: Paris, France.
IPCC, Renewable energy sources and climate change mitigation- Special Report,
2012, Intergovernmental Panel on Climate Change: Cambridge University Press,
Cambridge, United Kingdom.
IEA, World energy outlook. Special report: Are we entering a golden age of
gas?, 2011, International Energy Agency: Paris, France.
Spliethoff, H., Power generation from solid fuels. 1st ed., 2010: Springer. 712.
EU, Energy Roadmap 2050, 2012, European Commission, European Union:
Luxembourg: Publications Office of the European Union.
BP, Statistical review of world energy, 2013, British Petroleum Company.
IPCC, Climate change 2007: Mitigation of climate change. Contribution of
Working Groups III to the Fourth Assessment Report, Metz, B., Davidson, O.R.,
Bosch, P.R., Dave, R.A., Meyer, L.A. (eds.), Editor 2007, Intergovernmental
Panel on Climate Change: Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA. p. 863.
UN, World population prospects: The 2012 revision, key findings and advance
tables, 2013, United Nations Population Division: New York, United States.
Birol, F., The impact of financial and economic crisis on global energy
investment, 2009, IEA: G8 Energy Ministers' Meeting.
BP, Energy outlook 2030, 2012, British Petroleum Company.
IEA, World energy outlook, 2012, International Energy Agency: Paris, France.
IEA, Key world energy statistics, 2012, International Energy Agency: Paris,
France.
IPCC, Climate change 2007: Synthesis report. Contribution of Working Groups
I, II and III to the Fourth Assessment Report, Pachauri, R.K., Reisinger, A.
109
110
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Bibliography
(eds.), Editor 2007, Intergovernmental Panel on Climate Change: Geneva,
Switzerland. p. 104.
IPCC, Climate change 2013: The Physical Science Basis. Working Group I
Contribution to the Fifth Assessment Report, Stocker, T.F., Qin, D. (eds.), Editor
2013, Intergovernmental Panel on Climate Change: New York, USA. p. 1535.
Tans, P., Keeling, R.F., Annual mean growth rate of atmospheric carbon dioxide
at Mauna Loa, 2013, National Oceanic & Atmospheric Administration/Earth
System Research Laboratory and Scripps Institution of Oceanography: Hawaii,
USA.
IPCC, Carbon dioxide capture and storage - Special Report, 2005,
Intergovernmental Panel on Climate Change: Cambridge University Press,
Cambridge, United Kingdom.
UNFCCC. Kyoto Protocol to the United Nations Framework Convention on
Climate Change. [Accessed 23.05.2014]; Available from: http://unfccc.int.
UNFCCC. Doha amendment to the Kyoto Protocol. 2012
[Accessed
23.05.2014];
Available
from:
http://unfccc.int/files/kyoto_protocol/application/pdf/kp_doha_amendment_engli
sh.pdf.
IEA, Key world energy statistics, 2013, International Energy Agency: Paris,
France.
IEA, World energy outlook, 2013, International Energy Agency: Paris, France.
Joskow, P.L., Parsons, J.E, The future of nuclear power after Fukushima, 2012,
MIT Center for Energy and Environmental Policy Research (CEEPR).
WNA. Nuclear plant construction. 2013 [Accessed 23.05.2014]; Available
from:
http://www.world-nuclear.org/info/current-and-future-generation/plansfor-new-reactors-worldwide/.
IPCC, Climate change 2007: Mitigation of climate change. A Report of Working
Group III, Metz, B., Davidson, O, Swart, R, Pan, J (eds.), Editor 2001,
Intergovernmental Panel on Climate Change: Cambridge University Press,
Cambridge, United Kingdom.
IEA, Energy technology perspectives 2012, International Energy Agency: Paris,
France.
IEA, Power generation from coal - Ongoing developments and outlook, 2011,
International Energy Agency: Paris, France.
Wall, T., Stanger, R., Santos, S., Demonstrations of coal-fired oxy-fuel
technology for carbon capture and storage and issues with commercial
deployment. International Journal of Greenhouse Gas Control, 2011. 5S: p. S5S15.
EC, Communication from the Commission to the Council, the European
Parliament, the European economic and social committee and the committee of
the regions, 2007, European Commission: Brussels.
Bibliography
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
111
EC, The EU Emissions Trading System (EU ETS), 2013, European Commission:
Brussels.
EC, Factsheet- Climate change, 2013, European Commission: Brussels.
Eurostat. Renewable energy in the EU28. 2014 [Accessed 17.09.2014];
Available
from:
http://epp.eurostat.ec.europa.eu/cache/ITY_PUBLIC/810032014-AP/EN/8-10032014-AP-EN.PDF.
EC, Press release: 2030 climate and energy goals for a competitive, secure and
low-carbon EU economy, 2014, European Commission: Brussels.
Irons, R., Sekkapan, G., Panesar, R., Gibbins, J., Lucquiaud, M., CO2 capture
ready plants, 2007, International Energy Agency Greenhouse Gas R&D
Programme: Cheltenham, United Kingdom.
Pires, J.C.M., Martins, F.G, Alvim-Ferraz, M.C.M, Simões, M., Recent
developments on carbon capture and storage: An overview. Chemical
Engineering Research and Design, 2011. 89: p. 1446–1460.
GCCSI, Strategic analysis of the global status of carbon capture and storage,
2009, The Global CCS Institute: Canberra, Australia.
IEA, Improvements in power generation with post-combustion capture of CO2,
2004, International Energy Agency Greenhouse Gas R&D Programme: United
Kingdom.
Kunze, C., Spliethoff, H., Assessment of oxy-fuel, pre- and post-combustionbased carbon capture for future IGCC plants. Applied Energy, 2012. 94: p. 109116.
GCCSI, The global status of CCS, 2012, The Global CCS Institute: Canberra,
Australia.
IEA, Roadmapping coal’s future - Zero emissions technologies for fossil fuels,
2005, International Energy Agency: Paris, France.
Higman, C., Van der Burgt, M., Gasification. 1st ed., 2003: Gulf Professional
Publishing, Elsevier Science (USA).
Rubin, E.S., Mantripragada, H., Marks, A., Versteeg, P., Kitchin, J., The outlook
for improved carbon capture technology. Progress in Energy and Combustion
Science, 2012. 38: p. 630-671.
Kanniche, M., Gros-Bonnivard, R., Jaud, P., Valle-Marcos, J., Amann, J.M.,
Bouallou, C., Pre-combustion, post-combustion and oxy-combustion in thermal
power plant for CO2 capture. Applied Thermal Engineering, 2010. 30: p. 53-62.
Lozza, G., Romano, M., Giuffrida, A., Thermodynamic performance of IGCC
with oxy-combustion CO2 capture, in 1st International Conference on
Sustainable Fossil Fuels for Future Energy - S4FE20092009: Rome, Italy.
Robinson, P.J., Luyben, WL, Integrated gasification combined cycle dynamic
model: H2S absorption/stripping, water-gas shift reactors, and CO2
absorption/stripping. Industrial & Engineering Chemistry Research, 2010.
49(10): p. 4766-4781.
112
Bibliography
46.
Chiesa, P., Consonni, S., Kreutz, T., Williams, R., Co-production of hydrogen,
electricity and CO2 from coal with commercially ready technology. Part A:
Performance and emissions. International Journal of Hydrogen Energy, 2005.
30: p. 747-767.
IPCC, Guidelines for national greenhouse gas inventories, Vol.2: Energy,
Eggleston, S., Buendia, L, Miwa, K, Ngara, T, Tanabe, K, Editor 2006,
Intergovernmental Panel on Climate Change: Japan.
IEA, CO2 capture and storage - A key carbon abatement option, 2008,
International Energy Agency: Paris, France.
IEA, Fossil fuel-fired power generation - Case studies of recently constructed
coal- and gas-fired power plants, 2007, International Energy Agency: Paris,
France.
IEA, Power generation from coal- measuring and reporting efficiency
performance and CO2 emissions, 2010, International Energy Agency - Coal
Industry Advisory Board (CIAB): Paris, France.
IEA, Energy technology perspectives 2010, 2010, International Energy Agency:
Paris, France.
IEA, Global gaps in clean energy research, development, and demonstration,
2009, International Energy Agency: Paris, France.
IEA, International coal market & policy developments in 2010, 2011,
International Energy Agency - Coal Industry Advisory Board (CIAB): Paris,
France.
Liang, X., Wang, Z., Zhou, Z., Huang, Z., Zhou, J., Cen, K., Up-to-date life cycle
assessment and comparison study of clean coal power generation technologies in
China. Journal of Cleaner Production, 2013. 39: p. 24-31.
Cormos, C.C., Starr, F., Tzimas, E., Use of lower grade coals in IGCC plants
with carbon capture for the co-production of hydrogen and electricity.
International Journal of Hydrogen Energy, 2010. 35: p. 556-567.
Pettinau, A., Frau, C., Ferrara, F., Performance assessment of a fixed-bed
gasification pilot plant for combined power generation and hydrogen
production. Fuel Processing Technology, 2011. 92: p. 1946-1953.
Henderson, C., Future developments in IGCC, 2008, International Energy
Agency Clean Coal Centre: London, United Kingdom. p. 45.
GE, Syngas turbine technology, 2010, General Electric Company (GE Energy).
Poloczek, V., Hermsmeyer, H., Modern gas turbine with high fuel flexibility, in
Power-GEN Asia2008, Siemens AG, Energy Sector, Germany: Kuala Lumpur,
Malaysia.
IEA, Improved oxygen production technologies, 2007, International Energy
Agency Greenhouse Gas R&D Programme: United Kingdom.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Bibliography
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
113
Huth, M., Heilos, A, Gaio, G, Karg, J, Operation experiences of Siemens IGCC
gas turbines using gasification products from coal and refinery residues, in
ASME paper 2000-GT-26, 2000. ASME Turbo Expo, Munich, Germany.
Lee, J.J., Kim, Y.S., Cha, K.S., Kim, T.S., Sohn, J.L., Joo, Y.J., Influence of
system integration options on the performance of an integrated gasification
combined cycle power plant. Applied Energy, 2009. 86: p. 1788-1796.
Karg, J., IGCC experience and further developments to meet CCS market needs,
in Coal-Gen Europe2009, Siemens AG: Katowice, Poland.
Kanniche, M., Bouallou, C., CO2 capture study in advanced integrated
gasification combined cycle. Applied Thermal Engineering, 2007. 27: p. 26932702.
Chen, C., Rubin, E.S., CO2 control technology effects on IGCC plant
performance and cost. Energy Policy, 2009. 37: p. 915-924.
Geosits, R.F., Schmoe, .LA., IGCC - The challenges of integration, in ASME
paper GT2005-68997, 2005. ASME Turbo Expo, Reno-Tahoe, Nevada, USA:
Bechtel Corporation.
Allam, R.J., Castle-Smith, H., Smith, A.R., Sorensen, J.C., Stein, V.E., Air
separation units, design and future development, in ECOS 20002000: University
of Twente, Enschede, The Netherlands.
Häring, H., Ahner, C (Translator), Belloni, A (Preface), Industrial gases
processing. 1st ed., 2008, Weinheim, Germany: Wiley-VCH Verlag GMBH &
Co. KGaA.
Smith, A.R., Klosek, J., A review of air separation technologies and their
integration with energy conversion processes. Fuel Processing Technologies,
2001. 70: p. 115-134.
Allam, R.J., Russek, S.L., Smith, A.R., Stein, V.E., Cryogenics & ceramic
membranes: Current & future technologies for oxygen supply in gasification
systems, in 4th European Gasification Conference2000, Air Products and
Chemicals, Inc.: Noordwijk, The Netherlands.
Chen, Q., Rao, A., Samuelsen, S., H2 coproduction in IGCC with CCS via coal
and biomass mixture using advanced technologies. Applied Energy, 2014. 118:
p. 258–270.
DOE/NETL, Current and future technologies for gasification-based power
generation- Volume 2: A pathway study focused on carbon capture advanced
power systems r&d using bituminous coal, 2010, United States Department of
Energy, National Energy Technology Laboratory.
EPRI, Coal fleet integrated gasification combined cycle research and
development roadmap, 2011, Electric Power Research Institute (EPRI): Palo
Alto, California, USA.
Collot, A.G., Matching gasification technologies to coal properties. International
Journal of Coal Geology, 2006. 65: p. 191-212.
114
Bibliography
75.
Abadie, L.M., Chamorro, J.M., The economics of gasification: a market-based
approach. Energies, 2009. 2: p. 662-694.
Parulekar, P.S., Comparison between oxygen-blown and air-blown IGCC power
plants: a gas turbine perspective, in ASME paper GT2011-45154, 2011. ASME
Turbo Expo, Vancouver, Canada.
Klara, J.M., Plunkett, J.E., The potential of advanced technologies to reduce
carbon capture costs in future IGCC power plants. International Journal of
Greenhouse Gas Control, 2010. 4: p. 112-118.
Yun, Y., Yoo, Y.D., Chung, S.W., Selection of IGCC candidate coals by pilotscale gasifier production. Fuel Processing Technologies, 2007. 88: p. 107-116.
Harris, D., Roberts, D., ANLEC R&D scoping study: Black coal IGCC, 2010,
CSIRO Report No. EP103810.
Shell, The Shell coal gasification process for sustainable utilisation of coal,
2006, Shell Global Solutions.
Shelton, W., Lyons, J., Texaco gasifier IGCC base cases, 2000, DOE/NETL
PED-IGCC-98-001.
Maurstad, O., Herzog, H., Bolland, O., Beér, J., Impact of coal quality and
gasifier technology on IGCC performance, in 8th International Conference on
Greenhouse Gas Control Technologies (GHGT8)2006: Trondheim, Norway.
Holt, N., Booras, G., Todd, D., A summary of recent IGCC studies of CO2
capture for sequestration, in Gasification Technologies Conference2003: San
Francisco, CA, USA.
DOE/NETL, Cost and performance baseline for fossil energy plants volume 3a:
low rank coal to electricity: IGCC cases, 2011, United States Department of
Energy, National Energy Technology Laboratory.
Moulijn, J.A., , Makkee, M., Van Diepen, A.E., Chemical process technology.
2nd ed., 2013, United Kingdom: John Wiley & Sons Ltd.
E-Gas™ technology for integrated gasification combined cycle: Phillips 66.
Van Der Ploeg, H.J., Chhoa, T., Zuideveld, P.L., The Shell coal gasification
process for the US industry, in Gasification Technology Conference, 2004.
Washington DC.
Kondratiev, A., Jak, E., Predicting coal ash slag flow characteristics (viscosity
model for the Al2O3-CaO-'FeO'-SiO2 system). Fuel, 2001. 80: p. 1989-2000.
Giuffrida, A., Romano, M.C., Lozza, G.G., Thermodynamic assessment of IGCC
power plants with hot fuel gas desulfurization. Applied Energy, 2010. 87: p.
3374-3383.
DOE/NETL, Cost and performance baseline for fossil energy plants, Volume 1:
Bituminous coal and natural gas to electricity, 2010, United States Department
of Energy, National Energy Technology Laboratory.
Pavlish, J.H., Hamre, L.L., Zhuang, Y., Mercury control technologies for coal
combustion and gasification systems. Fuel, 2010. 89: p. 838-847.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
Bibliography
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
115
EEB, Opportunities to reduce mercury emissions through the Thematic Strategy
on Air Pollution (TSAP) review and the revision of the National Emission
Ceilings (NEC) Directive, 2013, European Environmental Bureau.
Mansouri Majoumerd, M., Raas, H., De, S., Assadi, M., Estimation of
performance variation of future generation IGCC with coal quality and
gasification process – Simulation results of EU H2-IGCC project. Applied
Energy, 2014. 113: p. 452-462.
Wright, A., White, V., Hufton, J., Van Selow, E., Hinderink, P., Reduction in the
cost of pre-combustion CO2 capture through advancements in sorption-enhanced
water-gas-shift. Energy Procedia, 2009. 1: p. 707-714.
Carbo, M.C., Jansen, D, Boon, J, Dijkstra, JW, Van Den Brink, RW, Verkooijen,
AHM, Staged water-gas shift configuration: Key to efficiency penalty reduction
during pre-combustion decorbonisation in IGCC. Energy Procedia, 2009. 1: p.
661-668.
DOE/NETL, Evaluation of alternate water gas shift configurations for IGCC
systems, 2009, United States Department of Energy, National Energy
Technology Laboratory.
Sánchez, J.M., Maroño,M., Cillero, D., Montenegro, L., Ruiz, E., Laboratoryand bench-scale studies of a sweet water-gas-shift catalyst for H2 and CO2
production in pre-combustion CO2 capture. Fuel, 2013. 114: p. 191-198.
Strube, R., Manfrida, G., CO2 capture in coal-fired power plants – impact on
plant performance. International Journal of Greenhouse Gas Control, 2011. 5: p.
710-726.
Müller, M., Integration of hot gas cleaning at temperatures above the ash
melting point in IGCC. Fuel, 2013. 108: p. 37-41.
Carbo, M.C., Boon, J., Jansen, D., van Dijk, H.A.J., Dijkstra, J.W., van den
Brink, R.W., Verkooijen, A.H.M., Steam demand reduction of water–gas shift
reaction in IGCC power plants with pre-combustion CO2 capture. International
Journal of Greenhouse Gas Control, 2009. 3: p. 712-719.
Jiang, L., Zhu, H., Razzaq, R., Zhu, M., Li, C., Li, Z., Effect of zirconium
addition on the structure and properties of CuO/CeO 2 catalysts for hightemperature water-gas shift in an IGCC system. International Journal of
Hydrogen Energy, 2012. 37: p. 15914-15924.
Kreutz, T., Martelli, E., Carbo, M., Consonni, S., Jansen, D., Shell gasifier-based
coal IGCC with CO2 capture: Partial water quench vs. novel water-gas shift, in
ASME paper GT2010-22859, 2010. ASME Turbo Expo, Glasgow, UK.
Giuffrida, A., Romano, M.C., Lozza, G., Efficiency enhancement in IGCC power
plants with air-blown gasification and hot gas clean-up. Energy, 2013. 53: p.
221-229.
Korens, N., Simbeck, D.R., Wilhelm, D.J., Process screening analysis of
alternative gas treating and sulfur removal for gasification, Prepared for U.S.
116
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
Bibliography
Department of Energy, National Energy Technology Laboratory: Pittsburg,
Pennsylvania, USA.
Mondal, P., Dang, G.S., Garg, M.O., Syngas production through gasification and
cleanup for downstream applications - Recent developments. Fuel Processing
Technology, 2011. 92: p. 1395-1410.
EPA, Environmental footprints and costs of coal-based integrated gasification
combined cycle and pulverized coal technologies, 2006, United States
Environmental Protection Agency: Washington DC, USA.
Kohl, A., Nielsen, R., Gas purification. 5th ed., 1997, Houston, Texas, USA: Gulf
Publishing Company. 1395.
Breckenridge, W., Holiday, A., Ong, J.O.Y., Sharp, C., Use of SELEXOL®
process in coke gasification to ammonia project, in Laurance Reid Gas
Conditioning Conference2000: The University of Oklahoma, Norman,
Oklahoma, USA.
Kubek, D.J., Polla, E., Wilcher, F.P., Purification and recovery options for
gasification, in Gasification Technologies Conference1996, UOP LLC: San
Francisco, California, USA.
Zheng, L., Furinsky, E., Comparison of Shell, Texaco, BGL and KRW gasifiers
as part of IGCC plant computer simulations. Energy Conversion &
Management, 2005. 46: p. 1767-1779.
Álvarez-Rodríguez, R., Clemente-Jul, C., Hot gas desulphurisation with
dolomite sorbent in coal gasification. Fuel, 2008. 87: p. 3513–3521.
DOE/NETL, Cost and performance baseline for fossil energy plants, Volume 1:
Bituminous coal and natural gas to electricity, 2007, United States Department
of Energy, National Energy Technology Laboratory.
Gazzani, M., Macchi, E., Manzolini, G., CO2 capture in integrated gasification
combined cycle with SEWGS – Part A: Thermodynamic performances. Fuel,
2013. 105: p. 206-219.
Melchior, T., Madlener, R., Economic evaluation of IGCC plants with hot gas
cleaning. Applied Energy, 2012. 97: p. 170-184.
Merkel, T.C., Zhou, M., Baker, R.W., Carbon dioxide capture with membranes
at an IGCC power plant. Journal of Membrane Science, 2012. 389: p. 441-450.
DOE/NETL, DOE/NETL advanced carbon dioxide capture R&D program:
Technology update, 2010, United States Department of Energy, National Energy
Technology Laboratory.
Manzolini, G., Macchi, E., Gazzani, M., CO2 capture in integrated gasification
combined cycle with SEWGS - Part B: Economic assessment. Fuel, 2013. 105: p.
220-227.
Rubin, E.S., Berkenpas, M.B., McCoy, S.T., Development and application of
optimal design capability for coal gasification systems- The economics of CO2
transport by pipeline storage in saline aquifers and oil reserves, 2008,
Bibliography
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
117
Department of Engineering and Public Policy, Center for Energy and
Environmental Studies, Carnegie Mellon University: Pittsburgh, PA, USA.
Bock, B., Rhudy, R., Herzog, H., Klett, M., Davison, J., De La Torre Ugarte,
D.G., Simbeck, D., Economic evaluation of CO2 storage and sink enhancement
options, 2003, United States Department of Energy, National Energy Technology
Laboratory.
Aspelund, A., Jordal, K., Gas conditioning-The interface between CO2 capture
and transport. International Journal of Greenhouse Gas Control, 2007. 1: p. 343354.
Sipöcz, N., Jonshagen, K, Assadi, M, Genrup, M, Novel high-performing singlepressure combined cycle with CO2 capture. Journal of Engineering for Gas
Turbines and Power, 2011. 133/041701: p. 1-8.
Romeo, L.M., Bolea, I., Lara, Y., Escosa, J.M., Optimization of intercooling
compression in CO2 capture systems. Applied Thermal Engineering, 2009. 29: p.
1744-1751.
Li, H., Thermodynamic properties of CO2 mixtures and their applications and
advanced power cycles with CO2 capture processes, in Department of Chemical
Engineering and Technology, 2008, Royal Institute of Technology (KTH):
Stockholm, Sweden.
Coan, C.R., King, A.D., Solubility of water in compressed carbon dioxide,
nitrous oxide, and ethane. Evidence for hydration of carbon dioxide and nitrous
oxide in the gas phase. Journal of the American Chemical Society, 1971. 93(8):
p. 1857-1862.
Chiesa, P., Lozza, G., Mazzocchi, L., Using hydrogen as gas turbine fuel.
Journal of Engineering for Gas Turbines and Power, 2005. 127: p. 73-80.
Romano, M.C., Chiesa, P., Lozza, G., Pre-combustion CO2 capture from natural
gas power plants, with ATR and MDEA processes. International Journal of
Greenhouse Gas Control, 2010. 4: p. 785-797.
Bancalari, E., Chan, P., Diakunchak, I.S., Advanced hydrogen gas turbine
development program, in ASME paper GT2007-27869, 2007. ASME Turbo
Expo, Montreal, Canada: Siemens Power Generation, Inc.
Descamps, C., Bouallou, C., Kanniche, M., Efficiency of an integrated
gasification combined cycle (IGCC) power plant including CO 2 removal.
Energy, 2008. 33: p. 874-881.
Oluyede, E.O., Phillips, J.N., Fundamental impact of firing syngas in gas
turbines, in ASME paper GT2007-27385, 2007. ASME Turbo Expo, Montreal,
Canada.
Kim, Y.S., Lee, J.J., Cha, K.S., Kim, T.S., Sohn, J.L., Joo, Y.J., Analysis of gas
turbine performance in IGCC plants considering compressor operating
condition and turbine metal temperature, in ASME paper GT2009-59860, 2009.
ASME Turbo Expo, Orlando, Florida, USA.
118
Bibliography
131.
Giuffrida, A., Romano, M.C, On the effects of syngas clean-up temperature in
IGCCs, in ASME paper GT2010-22752, 2010. ASME Turbo Expo, Glasgow,
UK.
Krishnan, V., Bharani, S, Kapat , J.S, Sohn, Y.H, Desai, V.H, A simplistic model
to study the influence of film cooling on low temperature hot corrosion rate in
coal gas/syngas fired gas turbines. International Journal of Heat and Mass
Transfer, 2008. 51: p. 1049-1060.
Sreedharan, S.S., Tafti, D.K, Composition dependent model for the prediction of
syngas ash deposition in turbine gas hotpath. International Journal of Heat and
Fluid Flow, 2011. 32: p. 201-211.
Hashimoto, T., Sakamoto, K, Ishii, H, Fujii, T, Koyama, Y, Commercialization
of clean coal technology with CO2 recovery. Mitsubishi Heavy Industries
Technical Review, 2010. 47: p. 9-14.
GE, 9FB gas turbine, new heights in flexible, efficient power generation- Fact
sheet, General Electric Company, GE Energy.
Siemens, Siemens gas turbine SGT5-4000F, 2008, Siemens AG: Erlangen,
Germany.
GTW, Gas Turbine World- Performance specs. 28th ed. Vol. 42. 2012: Pequot
Publishing.
Siemens, Siemens combined cycle reference power plant SCC5-4000F 1S, 2008,
Siemens AG: Erlangen, Germany.
Dennis, R.A., Shelton, W.W., Le, P., Development of baseline performance
values for turbines in existing IGCC applications, in ASME paper GT200728096, 2007. ASME Turbo Expo, Montreal, Canada.
GTC. World gasification database. 2014 [Accessed 17.02.2014]; Available
from: http://www.gasification.org/database1/search.aspx.
MIT. Carbon capture and storage technologies @ MIT. 2014 [Accessed
17.02.2014];
Available
from:
http://sequestration.mit.edu/tools/projects.index.html.
Hoffmann, B.S., Szklo, A, Integrated gasification combined cycle and carbon
capture: A risky option to mitigate CO2 emissions of coal-fired power plants.
Applied Energy, 2011. 88: p. 3917-3929.
GCCSI, The global status of CCS, 2013, The Global CCS Institute: Canberra,
Australia.
Marchino, C., Edwardsport IGCC station, in Gasification Technologies
Conference2013, Duke Energy: Colorado Springs, CO, USA.
McCarthy, D., Air separation units for gasification: Learning by doing US and
Asia, in Gasification Technologies Conference2013, Air Products: Colorado
Springs, CO, USA.
Lako, P., Coal-fired power, 2010, International Energy Agency, Energy
Technology Systems Analysis Programme.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
Bibliography
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
119
Low emission gas turbine technology for hydrogen-rich syngas Under the 7th
Framework Programme FP7-239349. [Accessed 17.02.2014]; Available from:
www.h2-igcc.eu.
IPSEpro version 4.0, 2003, Simtech Simulation Technology (Simtech): Graz,
Austria.
Enssim®, 2009, Enssim Software: Doetinchem, The Netherlands.
Aspen Plus version 7.1, 2009, Aspen Technology Inc.: Cambridge, MA, USA.
Rubin, E., Booras, G., Davison, J., Ekstrom, C., Matuszewski, M., McCoy, S.,
Short, C., Toward a common method of cost estimation for CO2 capture and
storage at fossil fuel power plants, 2013, Global CCS Institute: Canberra,
Australia. p. 36.
ZEP, The costs of CO2 capture - Post-demonstration CCS in the EU, 2010,
European Technology Platform for Zero Emission Fossil Fuel Power Plants.
CAESAR, European best practice guidelines for assessment of CO2 capture
technologies, in the European Benchmarking Task Force (EBTF)2011.
DOE/NETL, Quality guidelines for energy system studies - Cost estimation
methodology for NETL assessments of power plant performance, 2011, United
States Department of Energy, National Energy Technology Laboratory.
Bejan, A., Tsatsaronis, G., Moran, M., Thermal design and optimization, 1996,
New York: John Wiley & Sons Inc.
Universal currency converter, September 2013. 2013 [Accessed 27.03.2014];
Available from: http://www.xe.com.
Rubin, E.S., Understanding the pitfalls of CCS cost estimates. International
Journal of Greenhouse Gas Control, 2012. 10: p. 181-190.
EPRI, Power generation technology data for integrated resource plan of South
Africa, 2010, Electric Power Research Institute (EPRI): Palo Alto, California,
USA.
Paper I
Development of H2-rich Syngas fuelled GT for future IGCC
power plants – establishment of a baseline
Nikolett Sipöcz, Mohammad Mansouri, Peter Breuhaus,
Mohsen Assadi
Presented at ASME Turbo Expo 2011, Vancouver, Canada,
June 2011
121
Proceedings of ASME Turbo Expo 2011
GT2011
June 6-10, 2011, Vancouver, British Columbia, Canada
GT2011-45
DEVELOPMENT OF H2-RICH SYNGAS FUELLED GT FOR FUTURE IGCC POWER
PLANTS – ESTABLISHMENT OF A BASELINE
1
1
2
Nikolett Sipöcz , Mohammad Mansouri , Peter Breuhaus , Mohsen Assadi
1
2
Department of Mechanical- and
Structural Engineering and
Materials Science
University of Stavanger
4036 Stavanger, Norway
ABSTRACT
As part of the European Union (EU) funded H2-IGCC
project this work presents the establishment of a baseline
Integrated Gasification Combined Cycle (IGCC) power plant
configuration under a new set of boundary conditions such as
the combustion of undiluted hydrogen-rich syngas and high
fuel flexibility. This means solving the problems with high NOx
emitting diffusion burners, as this technology requires the
costly dilution of the syngas with high flow rates of N2 and/or
H2O. An overall goal of the project is to provide an IGCC
configuration with a state-of-the-art (SOA) gas turbine (GT)
with minor modifications to the existing SOA GT and with the
ability to operate on a variety of fuels (H2-rich, syngas and
natural gas) to meet the requirements of a future clean power
generation. Therefore a detailed thermodynamic analysis of a
SOA IGCC plant based on Shell gasification technology and
Siemens/Ansaldo gas turbine with and without CO2 capture is
presented. A special emphasis has been dedicated to evaluate at
an intermediate stage of the project the GT performance and
identify current technical constraints for the realization of the
targeted fuel flexibility.
The work shows that introduction of the low calorific fuel
(H2 rich fuel more than 89 mol% H2) has rather small impact on
the gas turbine from the system level study point of view. The
study has indicated that the combustion of undiluted syngas has
the potential of increasing the overall IGCC efficiency.
1
International Research Institute of
Stavanger (IRIS)
Postbox 8046
4068 Stavanger, Norway
for the development of reliable, low-emission, cost-competitive
gas turbine technologies for hydrogen-rich syngas combustion.
Integrated gasification combined cycle is currently one of the
most attractive technologies for the use of coal with high
efficiency and it offers the greatest fuel flexibility among the
most advanced technologies for power production. In addition,
gasification also provides an opportunity to control and reduce
gaseous pollutant emissions such as NOx and SOx. It in addition
offers one of the least costly approaches to concentrate carbon
dioxide (CO2) at high pressure to facilitate CO2 capture and
storage (CCS). However, coal-based IGCC plants have still not
achieved any commercial breakthrough, even though research
and development of IGCC plant technology began 40 years
ago. The currently six IGCC power plants in the world,
operating on coal as primary feedstock are demonstration plants
with capacities of 250-400 MW [1].
The design and operational experiences along with the
technical limitations of current state-of-the art IGCC power
plants have been reported in the recent past [2]-[5]. Important
contribution to field highlighting the two design variables
affecting the gas turbine operation i.e. the integration level of
the ASU and the nitrogen supply ratio for dilution of the syngas
has been presented by Kim et al [6]. These two parameters do
also have an influence on the turbine metal temperature. It has
been shown that low integration degree designs cause
overheating of the turbine metal due to higher pressure ratios.
Overheating of the turbine metal also becomes more severe as
the heating value of the syngas decreases. As a consequence of
the increased fuel flow the pressure ratio is increased, which in
turn gives higher temperature of the extracted air for turbine
cooling [7]. Even though higher integration levels results in a
higher IGCC efficiency [6] the operational experience from
1 INTRODUCTION
The continued need to use coal as primary fuel
engenders both increased interest and concern while, in
connection with coal gasification, generating a sincere demand
1
Copyright © 2011 by ASME
with a SOA GT with minor modifications to the existing SOA
GT and with the ability to operate on a variety of fuels (H2-rich,
syngas and natural gas) to meet the requirements of a future
clean power generation. Therefore a detailed thermodynamic
analysis of a SOA IGCC plant based on Shell gasification
technology and Siemens/Ansaldo gas turbine with and without
CO2 capture is presented. A special emphasis has been
dedicated to evaluate the GT performance and identify current
technical constraints for the realization of the targeted fuel
flexibility.
Buggenum has shown that the highly integrated design layouts
are problematic and has a negative effect on the plant
availability.
With the last years growing concern about greenhouse
gas emissions the near-term implementation of pre-combustion
CO2 capture technologies in IGCC applications has drawn
increased R&D interest [8]-[11]. One of the most promising
alternatives to the pre-combustion technology in IGCC power
plants is the oxy-combustion IGCC [12], [13], having the
potential of increasing both efficiency and environmental
characteristics of coal power plants. However, the large oxygen
consumption and required re-design of the gas turbine are still
the main drawbacks [13]. Accordingly, this CO2 abatement
technology along with membranes, adsorption onto solids and
cryogenic separation are different in terms of efficiency and
cost compared to chemical or physical absorption of CO2 and
thus the realization of these are within the mid-long term time
frame. Nevertheless, the capital costs associated with current
SOA IGCC is a major challenge, especially compared to natural
gas combined cycles. Adding the costs for implementing any
near term CCS technology makes the challenge even greater
[14]-[17]. In this context the high operational costs, coming at
the top of the investment, is another drawback deriving from
the currently low reliability and availability of the gasifier,
reduced efficiency due to de-rating of the gas turbine, and the
required syngas pre-treatment in terms of dilution.
Although IGCC offer significant advantages over
pulverized coal (PC) plants in terms of cost effective reduction
of CO2 emissions, the main challenges including cost,
compatibility with alternative technologies and the insecurity of
the implementation of any future CCS remain critical obstacles
for widespread commercialization [18]. Numerous research
projects such as Australia’s COAL21 National Action Plan, the
European funded Clean Coal Technology activities under the 7th
Framework Program, and the Canadian Clean Coal Technology
Roadmap have thus been released in recent years. They are
aiming at reducing these barriers by focusing on new coal
feeding systems, novel H2 production and purification
processes, and CO2 management [19].
In addition to capture, CCS involves two other major
components: transport and storage. One of the biggest
uncertainties in the CCS chain is finding suitable sites for the
storage of CO2 close to the emissions sources. Other storage
issues that need be addressed are: storage capacity estimation,
the potential for storage e.g. in deep saline reservoirs,
understanding the CO2 trapping mechanisms and quantifying
the risks of CO2 geological storage. Even though considerable
progress has been made in understanding many of these issues
trough the many research and demonstration projects around
the world i.e. Sleipner, Weiburn, In Salah and Otway to
mention a few, the regulatory framework and incentives for a
near term implementation of CCS is still to be solved [20]-[22].
As a part of the EU funded H2-IGCC project this work
presents the establishment of a baseline IGCC power plant
configuration under a set of new boundary conditions. An
overall goal of the project is to provide an IGCC configuration
2 H2-IGCC PROJECT
One of the largest barriers towards the usage of syngas
in current IGCC power plants is its inherently variation in
composition and heating value. At the same time the high
content of H2 in syngas derived from gasification of coal
complicates the application of pre-mix burners (Dry Low
Emission of Dry Low NOx burners) , which is current SOA in
natural gas fired GTs. The restriction of using DLE burners is
due to the higher reactivity of H2 compared to natural gas. For
this reason GTs in existing IGCC power plants are utilizing
high NOx emitting diffusion burners that also requires the
hydrogen-rich syngas to be diluted with nitrogen or
water/steam to control the higher adiabatic flame temperature.
Given these limitations the overall objective of the H2IGCC project is to provide and demonstrate technical solutions
which will allow the use of SOA highly efficient, reliable GTs
in the next generation of IGCC plants. The goal is to enable
combustion of undiluted hydrogen-rich syngas with low NOx
emissions and also allowing for high fuel flexibility by enabling
the burning of back-up fuels, such as natural gas, without
adversely affecting the reliability and availability.
The project is divided into the following four technical
subprojects (SP)[23]:
Combustion (SP1) – development and demonstration
of safe and low emission combustion technology for
undiluted, hydrogen-rich syngas.
Materials (SP2) – development and demonstration of
improved materials systems with advanced coatings
able to protect base blade and combustor materials
against the different and potentially more aggressive
temperatures and compositions of exhaust gases.
Turbomachinery (SP3) – investigation of modified
compressor/turbine aerodynamics and hot path cooling
in order to manage the increased mass flow rate of fuel
and the increased heat transfer of exhaust gases.
System Analysis (SP4) – evaluation of optimum
IGCC plant configurations and establishment of
guidelines for optimized full scale integration while
providing detailed system analysis to generate realistic
techno-economical results for future gas turbine based
IGCC plants with CCS.
2
Copyright © 2011 by ASME
3 METHODOLOGY
This work covers the description of the current
thermodynamic model set-up of the whole IGCC cycle
including important aspects of assumptions and limitations as
well as a discussion of the results. A special emphasis in this
regards has been given to the GT since this component is the
major of the overall H2-IGCC project.
The thermodynamic model set up, described by the mass
and energy balances of the IGCC plant with gasification of coal
and pre-combustion CO2 capture has been established based on
commercially available technology:
 oxygen-blown, entrained flow coal gasifier (Shell
technology),
 sour water-gas-shift (WGS) reactors,
 physical absorption using Selexol solvent for acid gas
removal (AGR),
 power island consisting of a 300 MW single shaft gas
turbine based on the Ansaldo Energia 94.3A with a
conventional triple-pressure steam cycle as the
bottoming cycle.
The focus of utilizing SOA technology is an important
element of the overall project. Thus the foundation of the
reference IGCC layout provides a fairly conservative baseline
for future studies. At the end of the project the goal is to find
the optimum combination of commercial gasification units with
modified gas turbines, incorporating solutions to the technical
challenges of burning undiluted hydrogen-rich syngas at an
appropriate level of integration.
Modelling of the IGCC power plant has been made
using three different modelling tools:





Data exchange between these codes was done
manually and iterated for optimal match.
Even though three different tools have been used for
simulating the whole IGCC power plant with as well as without
CO2 capture, the main platform for the simulations is IPSEpro
and the aim is to be able to simulate the whole IGCC except
from the gasification island in the IPSEpro environment by
solving current limitation in terms of pressure of pure gaseous
streams. The main reason for using IPSEpro as basis for the
simulations is the comprehensive model library, which has been
developed as a result of many years work within the research
group of University of Stavanger. This includes detailed and
sophisticated models of various power plant components that
have been developed due to the main advantage of IPSEpro,
allowing for introducing new and modified components in a
very straight-forward and flexible manner. This advantage is
very important in this project as the GT model will need to be
adapted to certain changes based on the results from the
different SPs. IPSEpro also provides additional benefits in
terms of thermo-economical optimization features that will be
of major significance to achieve the overall project target of
finding optimum combination of commercial gasification units
with modified gas turbines and appropriate level of integration.
The schematic outline of the IGCC with CO2 capture is
illustrated in Figure 1.
Enssim – Simulation tool developed by Enssim
Software.
Aspen HYSYS – Commercial process simulator by
AspenTech [24].
IPSEpro- Commercial heat and mass balance
programme by SimTech [25].
The reason for using a combination of several
simulation tools is that each of the selected tools have shown
advantages when simulating different parts of the IGCC plant
in terms of providing reliable results and the possibility of
incorporating detailed component characteristics. Hence, the
simulation tool among these three satisfying these requirements
for each sub-system to the greatest extent has been selected as
described below:



syngas and steam in Aspen HYSYS (Peng-Robinson
EOS) while the subsequent shift and two stage acid
gas removal has been performed in the heat and mass
balance program IPSEpro. In the case when no capture
of CO2 takes place the syngas leaving the wet scrubber
is bypassed to the H2S absorber before entering the
power island (without any dilution).
The clean syngas leaving the CO2 absorber/H2S
absorber is directed to the GT, which together with the
triple-pressure steam cycle is modelled in IPSEpro.
The CO2 captured in the second absorber in the AGR
process is compressed in a seven-stage intercooled
compressor and finally pumped to appropriate
transportation conditions. This part has also been
completed using the Aspen HYSYS modelling tool
(Peng-Robinson EOS).
4 IGCC POWER PLANT DESIGN
4.1 Coal input
Bituminous coal being a mixture of various trade coals
on the world market (mainly Russia, but also USA, Columbia
and South Africa) with the composition according to Table 1. It
is milled and dried to a moisture level of 2%wt, and fed to the
gasifier by means of lockhopper pressurization using
pressurized N2 as conveying gas. Heat for drying is provided by
burning approximately 0.9% of the shifted syngas. The amount
of coal needed is determined by the thermal power required by
the gas turbine model, based on the Ansaldo Energia 94.3A GT.
The resulting coal input is within the range of 1’008 -1’110
MWLHV.
The detailed modelling of the Shell gasification
process including the process components: coal
milling and drying, gasification, raw syngas cooling
and scrubbing have been performed by Nuon using the
Enssim modelling tool.
The required compression work in the air separation
unit (ASU) has been calculated using Aspen HYSYS
(Peng-Robinson equation of state (EOS)).
The syngas cleaning downstream the wet scrubber has
been modelled by first simulate the mixing of raw
3
Copyright © 2011 by ASME
Table 1 – Composition (% by weight) and heating value of as received
of the Bituminous coal used in the calculations.
C
64.10 Moisture
10
H
3.90 Ash
12.50
N
0.70
O
7.21
kJ/kg
S
1.50 HHV
26195
Cl
0.09 LHV
25100
4.2 Air separation unit
The ASU is a stand-alone unit generating oxygen with
a purity of 95mol% (with 2% N2 and 3% Ar) from air supplied
by the non integrated main air compressor (MAC). Selection of
the non- integrated MAC was motivated by negative
experiences concerning plant availability, from partially or fully
integrated ASU systems. The MAC is a seven-stage intercooled
B
Coal
milling &
drying
Raw syngas
A
HRSG
D
B
F
E
F, G
Gasifier
F
C
D
HT
WGS
LT
WGS
H
Wet
Scrubber
Make-up
water
E
Syngas
Cooler
CO2
Slag
I
CO2
removal
flash drums
Demister
C
M
M
D
A
Fly Ash
Claus/
SCOT
GOX compressor
ambient air
H
M
H2S
removal
M
O2
liquified CO2
G
I
H2-rich
syngas
MAC
M
N2
M
ASU
PGAN compressor
HP
IP/LP
Waste N2
ambient air
gas turbine
Condenser
Figure 1 – Plant schematic of the Shell IGCC with CO2 capture and conventional WGS
compressor with a discharge pressure of 5.5 bara. The gaseous
oxygen (GOX) is compressed to 55 bara in a nine-stage
intercooled compressor and fed to the gasifier while the pure
gaseous nitrogen (PGAN) is compressed to 80 bara in a tenstage intercooled compressor used for fuel feeding to the
gasifier. Since the GT is operated on undiluted syngas all
remaining nitrogen from the ASU not needed in the gasification
island is vented to the atmosphere. For further technical
assumptions for the air separation unit please see Table A1 in
Annex 1.
4
4.3 Gasification, syngas cooling and scrubbing
The gasification of the coal is taking place in an O2blown, entrained flow gasifier based on the technology licensed
by Shell [26]. The gasification process (technical assumptions
presented in Table A2 in Annex 1), in which the milled and
dried coal is gasified in the presence of intermediate pressure
(IP) steam and oxygen is modelled assuming full equilibrium at
45 bara and 1600 °C. This condition determines the
composition of the raw syngas and it is achieved by adjusting
the O2 to coal mass ratio while setting the heat loss to the
membrane wall to 2.5% (LHV). The single pass and overall
Copyright © 2011 by ASME
carbon conversion rate is 99.3% (no recycling of fly ash) and
the fine particles that are not captured as fly ash by the ceramic
filter (after syngas cooling) leave the bottom of the gasifier as
vitreous slag.
The raw syngas from the gasifier is first cooled to 900
°C by adding a stream of recycled, cooled, ash-free syngas in
order to lower the gas temperature below the ash melting point.
The raw syngas is then further cooled to 340 °C in syngas
coolers that evaporate high pressure (HP) and IP pressure boiler
feedwater to produce HP steam for the steam cycle and IP
steam to be used in the water-gas-shift process. After passing
the dry particulate filters removing the fly ash, a small part of
the raw syngas is recycled back (0.84%) for cooling the raw
syngas exiting the gasifier. The rest is sent to the wet scrubber
for removal of species soluble in water, and trace particulate
matter such as unconverted carbon, slag and metals. The
quenched and cleaned syngas leaving the scrubber has a
temperature and pressure of 165 °C and 43 bara respectively.
However, the dry-feed characteristics for the Shell gasifier
leaves the raw syngas with a relatively low steam-to-CO ratio
thus requiring injection of steam to insure adequate CO to CO2
conversion during the WGS. The IP steam for this purpose is
partly supplied from the syngas cooler, but since the
requirement is larger than the amount generated in the gasifier
the rest is bled from the HP/IP turbine crossover. In order to
promote the WGS reaction sufficiently and to avoid carbon
formation on the WGS catalyst the steam-to CO ratio has been
adjusted to 2.4 (molar basis).
passed through a demister before being sent to the acid-gas
removal. The total pressure loss of the syngas from the exit of
the wet scrubber to the exit of the demister is 9.1%.
4.5 Acid gas removal
During gasification, sulphur in the raw coal is
converted to H2S and COS. Nevertheless, in the CO2 capture
case most of the COS is converted to H2S during the WGS
reaction. The H2S and CO2 are removed from the shifted syngas
in a two-stage physical absorption system using dimethyl ether
of polyethylene glycol also known as Selexol. The syngas
enters the first absorption column in which the H2S is removed
by a counter current flow of the solvent. The acid gases in the
rich solution exiting the bottom of the absorber column is
flashed and then stripped off in a regenerator for which heat
(approximately 13.6 MWth) is provided from steam bled from
the LP steam turbine. The regenerated solvent is cooled and
recycled back to the top of the absorber while H2S is sent to a
sulphur recovery unit including a Claus plant for oxidizing H2S
to elemental sulphur and a Shell Claus off gas treating (SCOT)
plant for tail gas cleanup.
After leaving the H2S absorber the syngas enters the
second absorber for removal of CO2. Similar to the removal of
H2S the CO2 is absorbed by the solvent flowing downwards the
column and exits the bottom of the column with the CO2 solved
in the solution. This collected rich CO2 solvent exiting the
bottom of the tower is passed through four flash drums
connected in series, where CO2 is released as a result of
lowering the pressure. The lean solvent leaving the last flash
drum is pumped and returned back to the top of the absorber
column. The release of the pressure of the rich solvent between
the column and the different flash drums is achieved by
hydraulic turbines. In this way part of the solvent pumping
power could be recovered. The solubility of CO and H2 in
Selexol is low, but not negligible, hence in order to minimize
the amount of H2 and CO that are co-absorbed with the CO2 in
the absorber and thereby lowering the heating value of the fuel.
The gas leaving the first flash drum is recycled back to the
absorber column, since virtually all H2 and CO absorbed is
released in this drum. The CO2 released in the flash drums two
to four is sent to compression. The CO2 removal rate in the
AGR unit is 96.3% (molar basis), though, the overall CO2
capture rate as defined in Eq. 4 is 88.6% (molar basis).
4.4 Water-gas-shift
The water-gas-shift process is the reaction used to
convert most of the CO in the raw syngas into CO2, by shifting
the CO with water over a bed of catalyst. Besides CO2
hydrogen is generated in this reaction (Eq.1). In IGCC
applications with CO2 capture this is the first step in order to
convert the gasifier product into a hydrogen-rich-syngas. The
CO converter is located upstream of the AGR unit (sour shift)
and is arranged as two reactors in series to meet higher CO2
capture rates. The WGS reaction is exothermic (44 KJ/moleCO)
and it is thermodynamically favoured at lower temperatures,
where reaction rates are comparatively slow. However, catalyst
activity is in general higher at high temperatures.
(1)
The scrubbed and steam mixed syngas is pre-heated to
250 °C before entering the first stage of the WGS unit. The
syngas leaving the first high temperature (HT) reactor is cooled
down from the equilibrium temperature of 463 °C to 250 °C by
generating HP steam and it then enters the low temperature
(LT) WGS reactor. The warm syngas leaving the second reactor
at an equilibrium temperature of 278 °C is cooled to 25°C by
means of preheating the raw syngas entering the first WGS
reactor and by preheating HP boiler feedwater. The resulting
overall adiabatic conversion of CO to CO2 and H2 in the WGS
process is 98.9% (molar basis). The cooled shifted syngas is
(4)
In the case when CO2 is vented, the raw syngas
leaving the wet scrubber is passed through the demister before
entering the H2S absorber. The rich solution leaving the bottom
of the column is regenerated and the sulphur is stripped off
using IP steam produced in the gasification island. Since this
amount is only partly sufficient the rest is extracted from the
HRSG. However, since the solvent flow rate in this case is
5
Copyright © 2011 by ASME
considerably lower (the flow rate of dry raw syngas is lower
than that of dry shifted) the thermal heat input is 3.6 MWth
lower than for the case with CO2 capture. The H2S poor syngas
exiting the absorber top is passed to the GT combustor. For
further technical assumptions for the AGR unit please see Table
A3 in Annex 1.
according to the results provided by the working group for
combustion. Fuel pre-heating has not been included, but will be
considered in the optimization of the whole IGGC plant.
Turbine model - The turbine part has been modelled using a
simplified approach based on the input received. The turbine
model used in this work has been assumed with a constant hot
gas flow, even though the real turbine is cooled and cooling air
is mixed into the hot gas at different stages this was not
considered in the existing model. In order to never the less
cover the overall performance of the turbine the turbine-, inlet
temperature and efficiency are calculated in terms of virtual
measures according to following equations:
4.6 CO2 compression
The CO2 collected in the flash drums in the CO2
removal process is compressed in a seven-stage intercooled
compressor to 60 bara, liquified and then pumped up to final
pressure of 150 bara. The compressor/pumping approach has
been evaluated in a previous work by the authors and found to
be the most efficient approach [27].
(2)
4.7 Gas turbine model
The gas turbine model has been modelled based on
internal project information exchange with the working group
focussing on the GT design (SP3). This information included
initial performance calculation results of a lumped
turbomachinery model of the GT, Ansaldo Energia 94.3A, with
a first version of the compressor map and some turbine data.
All information received was based on natural gas as fuel. The
control algorithm currently adopted when burning undiluted
hydrogen-rich syngas and cleaned syngas is without any major
modifications to the natural gas operation:
and
(3)
This has been done to match the data received from the
turbomachinery working group. By doing so the general
expansion in the turbine (mainly the pressure ratio and
therefore also the power consumption in the compressor) as
well as the overall power output was met. The technical
assumptions for the GT are presented in Table A4 in Annex 1.
The above described simplifications are an often used
approach in the early stage simulation of a GT process. These
models are going to be replaced by more detailed ones as soon
as this information will become available.
 The turbine inlet temperature (TIT) was fixed to 1331
°C.
 The compressor variable inlet guide vanes (VIGV) are
slightly closed to adjust for the increased fuel flow by
reducing the air mass flow.
 Due to the increased fuel flow the model adjusts the
pressure ratio accordingly.
4.8 Heat recovery steam generator design
Downstream the GT is a three pressure level heat
recovery steam generator (HRSG) with reheat. The admission
levels have been set according to internal discussions and
agreements within SP4. The superheating temperature has been
set to 530 °C in order to meet the GT exhaust temperature and
the required amount of HP steam needed to be superheated. The
heat integration represents somewhat a first approach and has
not been optimized. The assumptions of the parameters of the
HRSG are considered to be conservative in terms of pressure
losses, approach temperatures, steam turbine efficiency, etc.
There is potential of increasing the HRSG efficiency in order to
maximize the net electrical output, however the economical
feasibility of such optimization should not be disregarded.
The IP and HP boiler feedwater (BFW) needed in the
gasification island is taken from the HRSG and all HP steam is
returned back to the HRSG and mixed with the HP steam
produced in the WGS and superheated before expanded in the
steam turbine. The IP level has been set to meet the pressure of
the syngas leaving the wet scrubber 43 bara, since a
considerable amount of IP steam is extracted from the HRSG in
the case with CO2 capture and mixed with the raw syngas in
The GT models used at current state have some limitations
for off-design calculations, as only subsections of the
compressor- and turbine maps are implemented, there is no
detailed modelling of the cooling flows, etc., but will be
handled as soon as more information from other teams within
the project will be available. Nevertheless, the current model is
built up accordingly:
Compressor model – In terms of the speed lines, a
characteristic has been used which is, according to the authors,
reflecting state of the art characteristics Besides, cooling air
extraction at different pressure level has not been considered as
this information was not available at this time. However
extractions are already part of the model and can be activated
when needed. It is planned that the characteristic currently in
use is going to be replaced as soon as a more detailed version of
the compressor map is available.
Combustor model – The fuel composition to the combustor
was calculated using the detailed models described above (4.1 4.5). Besides, a pressure loss reflecting current state of the art
technology was used. This will also be updated later on
6
Copyright © 2011 by ASME
Table 2 – Performance results of the IGCC power plant with and
without CO2 capture
GT power out
ST shaft power
HRSG pumping power
AGR turbine power out
AGR pumping/ compr. power req.
ASU compression power req.
Gasification power req.
CO2 compression power req.
Net power out
Net IGCC efficiency (LHV)
IGCC w.
324.07
166.30
3.54
3.42
11.39
40.88
4.96
18.06
414.96
37.40
IGCC w.o.
309.39
211.43
3.13
0.2
37.14
4.50
476.86
47.20
output. Since the HRSG in this study has still not been
optimized the genuine improvement using undiluted syngas is
to be determined. The initial results have though confirmed that
without considering any any modifications of the GT and
keeping the efficiency constant, the power output from the
engine could increase with as much as 30 MW (compared to
the same GT fired with natural gas).
The big differences in fuel composition between
natural gas, hydrogen-rich syngas and cleaned syngas will most
probably result in different designs of the combustion system as
well as compressor and turbine to maintain stable combustion
and to keep the pressure ratio for the different mass flow ratios
in turbine and compressor. The extent of these changes or
requirements will be revealed within the project in a near future
and will be implemented in the GT model, giving the
opportunity to optimise the processes for the various cases.
However, Table 3 summarizing the two different fuel
compositions directly indicates that two different combustor
designs might be essential given the huge differences in the
properties, which are difficult to be covered in a single design.
The resulting difference in turbine inlet flow due to the huge
difference in fuel flow will either require a compressor design
with high efficiency over a wider range of IGV positions, or
also two different designs. This topic, which is closely
connected to transients and operation at off-design, will be
addressed during the next steps within the project.
The turbine outlet temperature (562°C and 588°C
respectively) as well as the turbine flow (700 kg/s and 749 kg/s
respectively) is higher for both IGCC cases compared to natural
gas (576°C /698 kg/s), which favours the steam bottoming
cycle. However, there is an important difference in terms of
extraction of BFW and steam along with returning condensate
and steam from different parts of the IGCC power plant for the
two cases investigated. This will have a major impact on the
investment costs if the targeted fuel flexibility ranging from
natural gas to cleaned syngas is going to be met. The further
optimization of the two cases (with and without CO2 capture)
should thus be performed taking into consideration to reduce
these distinctions to the greatest extent possible. In addition
some aspects of changed conditions for component lifetime
need to be evaluated since the lifetime due to the above
mentioned increases i.e. the gas temperature of the un-cooled
blade rows of the GT under certain operating conditions and
could if not designed for have certain negative impacts on plant
availability and costs.
The current overall CO2 capture rate for the hydrogen
rich case is 88.6 mol%, although the removal rate in the AGR
unit is approximately 96.3 mol%. The reason for this significant
difference is due to CO2 lost in the H2S absorber. The current
outline of the AGR unit has not been optimized; hence a
minimization of absorbed CO2 in the first stage of the AGR will
be further investigated by finding a more convenient
combination of number of flashes as well as the extent of
solvent pre-loading. Currently there is also a deviation in
pressure loss in the H2S absorber for the two cases due to
pressure limitations in IPSEpro for pure gaseous streams which
MW
MW
MW
MW
MW
MW
MW
MW
MW
%
order to perform the WGS reaction. The IP steam produced in
the gasifier island has a pressure of 50 bara, thus the BFW
extracted for this purpose is pumped to appropriate pressure
and heated by utilizing a small part of the heat generated in the
HT part of the WGS. The assumptions made for the HRSG
calculation are presented in Table A5 (Annex 1).
In the case without CO2 capture the HP BFW for the
gasification island is extracted in the same manner as in the
case with CO2 capture, however all the HP steam produced is
returned back to the HRSG and superheated to 500°C. The IP
BFW for is extracted similarly as for the CO2 capture case and
the small amount IP steam not needed in the gasification island
is used for regenerating the solvent in H2S removal unit. Since
the IP steam needed in the H2S removal is higher than the
amount produced in the gasification island the additional
required is bled from the HP/IP crossover. The IP SH/RH
temperature has likewise the HP SH temperature for the CO2
venting case been lowered with 30°C to 500 °C to accomplish
the superheating of all steam produced in the gasifier as well as
the steam no needed for the WGS. All other assumptions for the
HRSG are presented in Table A5 in Annex 1.
5 RESULTS AND DISCUSSION
The performance of the IGCC power plant with and
without CO2 capture based on the calculation using the models
as previously described are presented in Table 2 and the
composition of the syngas for the two cases are given in Table
3. The IGCC without CO2 capture has a somewhat lower
efficiency, even though the syngas in this work is not diluted,
than a similar case presented last year by Kreutz et al [17]. The
main reason for this is that the reference GT used in this work
is less efficient than the General Electric 9 FB even though the
TIT was de-rated to 1327°C in the previous work. Nevertheless,
the syngas considered was highly pre-heated and the HRSG
fully optimized, which are issues within the scope of future
activities within this project. The net efficiency of the case with
CO2 capture and undiluted hydrogen rich syngas is on the
contrary demonstrating a higher efficiency compared to the
same publication. This is due to a slightly difference in the
steam-to- CO ratio between the present study and the one
presented in [17]. In addition the higher heating value of
undiluted syngas results in a significantly higher GT power
7
Copyright © 2011 by ASME
HP
IGCC
IP
LP
mol
NOx
SOx
SOA
SP
ST
TIT
WGS
wt
limits the pressure, to 35 bara, in the case where the syngas is
sent for further removal in second stage. This has an impact on
the removal of CO2, since physical absorption is favoured at
higher partial pressures.
Table 3 – Composition (wt%) and characteristics after AGR of the
hydrogen-rich and cleaned syngas respectively (undiluted)
Hydrogen-rich syngas
Cleaned syngas
CO
0.0448
0.7857
CO2
0.1078
0.0716
H2
0.3595
0.0262
N2
0.4879
0.1165
Fuel flow (kg/s)
17.67
70.78
LHV (kJ/kg)
43641
11100
Temperature °C
25.6
25.24
Pressure (bara)
34.5
42.4
High pressure
Integrated gasification combined cycle
Intermediate pressure
Low pressure
Molar
Nitrogen oxide
Sulphor oxide
State-of-the-art
Subproject
Steam turbine
Turbine inlet temperature
Water-gas- shift
Weight
REFERENCES
[1] Dennis, R.A., Shelton, W.W., Le, P., 2007,
Development of baseline performance values for
turbines in existing IGCC applications. ASME paper
GT2007-28096. ASME Turbo Expo, Montreal,
Canada.
[2] Huth, M., Heilos, A., Gaio, G., Karg, J., 2000,
Operation experiences of Siemens IGCC gas turbine
using gasification products from coal and refinery
residues. ASME paper 2000-GT-26. ASME Turbo
Expo, Munich, Germany.
[3] Hannemann, F., Koestlin, B., Zimmermann, G.,
Morehead, H., García Peña, F., 2003, Pushing forward
IGCC technology at Siemens. Gasification technology
conference, San Francisco, California, USA,
[4] Brdar, R.D., Jones, R.M., 2000, GE IGCC technology
and experience with advanced gas turbines. General
electric report. GER-4207.
[5] Oluyede, E.O., Phillips, J.N., 2007, Fundamental
impact of firing syngas in gas turbines. ASME paper
2007-27385. ASME Turbo Expo, Montreal, Canada
[6] Lee, J.J., Kim, Y.S., Cha, K.S., Kim, T.S., Sohn, J.L.,
Joo, Y.J., 2009, Influence of system integration options
on the performance of an integrated gasification
combined cycle power plant. Applied Energy 86, pp.
1788–1796.
[7] Kim, Y.S., Lee, J.J., Cha, K.S., Kim, T.S., Sohn, J.L.,
Joo, Y.J., 2008, Analysis of gas turbine performance in
IGCC plants considering compressor operating
condition and turbine metal temperature. ASME paper
GT2009-59860. ASME Turbo Expo, Orlando, Florida,
USA.
[8] Pruschek, R., Oeljeklaus, G., Brand, V., Haupt, G.,
Zimmermann, G., Ribberink, J.S., 1995, Combined
cycle power plant with integrated coal gasification,
CO shift and CO2 washing. Energy Conversion
Management 36(6–9), pp.797–800.
[9] Chiesa, P., Consonni, S., 1999 Shift reactors and
physical absorption for low- CO2 emission IGCCs.
Journal of Engineering for Gas Turbines and Power
121(2), pp. 295–305.
6 CONCLUSIONS
As part of the EU-funded H2-IGCC project this work
has described the establishment of two fairly conservative
baseline IGCC cycles aimed for further investigations. The first
IGCC power plant has been modelled with pre-combustion
separation of CO2 while the second is without the application of
CO2 removal resulting in two completely different syngas
compositions. Both IGCC power plants are based on the GT
Ansaldo Energia 94.3A without any dilution of the syngas. By
performing new gasifier calculations including fly-ash recycle,
optimizing the heat integration and implementing the
characteristic GT data there is a potential to increase the net
efficiencies of both plants beyond current values of 37.4% for
the IGCC power plant with CO2 capture and 47.2% for the case
with CO2 venting. The overall CO2 capture rate presented in
this work, 88.6mol% is somewhat low due to lost of CO2 in the
first AGR stage. A more favourable configuration of the H2S
removal unit will be further investigated to demonstrate higher
capture ratios.
ACKNOWLEDGMENTS
The authors wish to acknowledge Nuon and E.ON for their
technical input and truthful discussions in the early phase of
this work. The authors are also grateful to Han Raas at Nuon
for performing the gasification simulations. The authors would
also like to acknowledge the project partners in SP3 for
providing the gas turbine performance data.
NOMENCLATURE
AGR Acid gas removal
ASU
Air separation unit
BFW Boiler feed water
CCS
Carbon capture and storage
CO2
Carbon dioxide
DLE
Dry low NOx emission
EOS
Equation of state
GT
Gas turbine
H2
Hydrogen
H2S
Hydrogen sulfide
8
Copyright © 2011 by ASME
[10] Chiesa, P., Lozza, G., 1999, CO2 emissions abatement
in IGCC power: plants part B—with air blown
combustion and CO2 physical absorption. Journal of
Engineering for Gas Turbines and Power 121(4),
pp.642–648.
[11] Descamps, C., Bouallou, C., Kanniche, M., 2008,
Efficiency of an Integrated Gasification Combined
Cycle (IGCC) power plant including CO2 removal.
Energy 33, pp. 874–881.
[12] Chisea, P., Lozz, G., 1999, CO2 emission abatement in
IGCC power plants by semiclosed cycles: Part A —
with
oxygen-blown
combustion.
Journal
of
Engineering for Gas Turbines and Power 121 (4), pp.
635-641.
[13] Lozza, G., Romano, M., 2009, Thermodynamic
Performance of IGCC with Oxy-combustion CO2
capture. International Conference on Sustainable
Fossil Fuels for Future Energy - S4FE2009. Rome,
Italy.
[14] Ordorica-Garcia, G., Douglas, P., Croiset, E., Zheng,
L., 2006, Technoeconomic evaluation of IGCC power
plants for CO2 avoidance. Energy Conversion
Management 47, pp.2250–2259.
[15] Kanniche, M., Bouallou, C., 2007, CO2 capture study
in advanced integrated gasification combined cycle.
Applied Thermal Engineering 27, pp.2693–2702.
[16] Mondol, J.D., McIlveen-Wright, D., Rezvani, S.,
Huang, Y., Hewitt, N., 2009, Techno-economic
evaluation of advanced IGCC lignite coal fuelled
power plants with CO2 capture. Fuel 88, pp. 2495–
2506.
[17] Kreutz, T., Martelli, E., Carbo, M., Consooni, S.,
Jansen, D., 2010, Shell Gasifier-Based IGCC with
CO2 Capture: Partial Water Quench vs. Novel WaterGas Shift. ASME paper GT2010-22859. ASME Turbo
Expo, Glasgow, UK.
[18] Gasification Technologies Council (GTC), 2008,
Gasification: redefining clean energy.
[19] Liu, H., Ni, W., Li, Z., Ma, L., 2008, Strategic
thinking on IGCC development in China, Energy
Policy 36, pp. 1–11.
[20] IEA Greenhouse Gas R&D Programme ( IEA
GHG), 2008, Geologic Storage of Carbon Dioxide
Staying Safely Underground.
[21] CO2CRC, www.CO2CRC.com.au
[22] Ninth International Conference on Greenhouse Gas
Control Technologies, GHGT 9, Washington D.C.,
November 2008, Conference Summary.
[23] Low emission gas turbine technology for hydrogenrich syngas. Under the 7th Framework Programme
FP7-239349. Project website: www.h2-igcc.eu
[24] Aspen Plus, 2009. Aspen Plus Version 7.1. Aspen
Technology Inc., Cambridge, MA, USA.
[25] IPSEpro v.4.0, 2003, Simtech Simulation Technology
(Simtech), Graz, Austria.
[26] Shell Global Solutions, The Shell Gasification
Process For Sustainable Utilisation of Coal.
[27] Sipöcz, N., Jonshagen, K., Assadi, M., Genrup, M.,
2010, Novel High-Performing Single Pressure
Combined Cycle with CO2 Capture. Paper GT201023259, ASME Turbo Expo, Glasgow, UK.
ANNEX A
TECHNICAL ASSUMPTIONS USED IN THE MODELLING
Table A1 – Technical assumptions for the ASU
Delivery pressure/temperature of O2 and N2 by ASU
1.2/10
Main air compressor polytropic efficiency
87
GOX compressor polytropic efficiency
87
HP PGAN compressor polytropic efficiency
87
Inter-cooling temperature
40
9
bara/°C
%
%
%
°C
Copyright © 2011 by ASME
Table A2 – Technical assumptions for the Shell gasification island including the syngas conditioning downstream to the wet scrubber exit
Dried coal moisture content
Gasification pressure/temperature
Shifted syngas for drying
Steam/coal ratio
O2/coal ratio
HP PGAN/coal ratio
Power requirement
Heat loss to membrane wall
Carbon conversion (single pass/overall)
Syngas cooler pinch-point HP evaporator
Syngas cooler pinch-point IP evaporator
Heat exchanger heat loss
Pressure drop syngas cooler (gas side)
Pressure drop wet scrubber
Water pump mechanical efficiency
Steam-to-CO ratio at WGS inlet
2
45/1600
2.2
0.061
0.7839
0.241
112
2.5
99.3
30
64
0
0.33
1
85
2.4
wt%
bara/°C
wt% (of total flow)
kg / kg coal (ar)
kg / kg coal (ar)
kg / kg coal (ar)
kJel/kg coal (ar)
% coal LHV
%
°C
°C
%
bar
bar
%
Table A3 – Technical assumptions used for the AGR unit
CO2 capture
No CO2 capture
Syngas pressure/temperature at H2S absorber inlet
39.1/25
43.96/25
CO2 co-absorbed in H2S absorber
9.5
8.5
Syngas pressure/temperature at CO2 absorber inlet
35/25.7
Pressure loss in 1st/2nd absorber
4.1/0.5
0.5
H2S stripping duty
13.6
10
H2 co-absorbed (overall)
0.35
0.1
CO co-absorbed (overall)
1.2
0.2
Solvent pumps polytropic efficiency
70
70
Compressor isentropic efficiency (recycle gas)
85
85
Hydraulic expander isentropic efficiency
85
85
Mechanical and electrical efficiency
99
99
Solvent temperature at absorber inlet
25
25
bara/°C
wt% (of inlet)
bara/°C
bar
MWth
wt% (of inlet)
wt%
%
%
%
%
°C
Table A4 – Technical assumptions used for the gas turbine
Ambient air pressure
1.013
bara
Ambient air temperature
15
°C
Moisture in air
60
%
TIT
1331
°C
GT outlet pressure
1.08
bara (total)
Pressure ratio
18.2
(target natural gas)
Electrical/mechanical efficiency
99/99.5
%
Table A5 – Technical assumptions used for the HRSG
HP/IP/LP
140/43/4
bara
SH and RH temperature
530*
°C
SH LP steam
300
°C
HP/IP/LP ST isentropic efficiency
88.5/89/91
%
ST and generator mechanical efficiency
99.5
%
Gas side HRSG pressure drop
0.04
bara
Generator electrical efficiency
98.2
%
Pump polytropic efficiency
70
%
Pump mechanical efficiency
95
%
Evaporator pinch point IP/LP
10/10
°C
Super heater pinch point
32
°C
Economizer pinch point
10
°C
Approach point temperature
5
°C
Condenser pressure
0.04
bara
*
The superheating/reheat temperature for the case without CO2 capture is 500°C, all other
assumptions are the same.
10
Copyright © 2011 by ASME
Paper II
An EU initiative for future generation of IGCC power plants
using hydrogen-rich syngas: Simulation results for the baseline
configuration
Mohammad Mansouri Majoumerd, Sudipta De, Mohsen Assadi,
Peter Breuhaus
Published in Applied Energy, Vol. 99, p. 280-290, June 2012
133
Author's personal copy
Applied Energy 99 (2012) 280–290
Contents lists available at SciVerse ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
An EU initiative for future generation of IGCC power plants using hydrogen-rich
syngas: Simulation results for the baseline configuration
Mohammad Mansouri Majoumerd a,⇑, Sudipta De b, Mohsen Assadi a, Peter Breuhaus c
a
Department of Mechanical and Structural Engineering and Materials Science, University of Stavanger, 4036 Stavanger, Norway
Department of Mechanical Engineering, Jadavpur University, Kolkata 700032, India
c
International Research Institute of Stavanger (IRIS), Postbox 8046, 4068 Stavanger, Norway
b
h i g h l i g h t s
" A baseline IGCC power plant with and without CO2 capture is presented.
" Burning of undiluted hydrogen-rich syngas in the gas turbine is assumed.
" A significant efficiency penalty is associated with the CO2 capture system.
a r t i c l e
i n f o
Article history:
Received 13 February 2012
Received in revised form 23 May 2012
Accepted 25 May 2012
Available online 22 June 2012
Keywords:
IGCC
CO2 capture
Gas turbine
H2-rich fuel
a b s t r a c t
In spite of the rapid development and introduction of renewable and alternative resources, coal still continues to be the most significant fuel to meet the global electricity demand. Emission from existing coal
based power plants is, besides others, identified as one of the major sources of anthropogenic carbon
dioxide, responsible for climate change. Advanced coal based power plants with acceptable efficiency
and low carbon dioxide emission are therefore in sharp focus for current development. The integrated
gasification combined cycle (IGCC) power plant with pre-combustion carbon capture is a prospective
technology option for this purpose. However, such plants currently have limitations regarding fuel flexibility, performance, etc. In an EU initiative (H2-IGCC project), possible improvements of such plants are
being explored. These involve using premix combustion of undiluted hydrogen-rich syngas and improved
fuel flexibility without adversely affecting the availability and reliability of the plant and also making
minor modifications to existing gas turbines for this purpose. In this paper, detailed thermodynamic
models and assumptions of the preliminary configuration of such a plant are reported, with performance
analysis based on available practical data and information. The overall efficiency of the IGCC power plant
with carbon capture is estimated to 36.3% (LHV). The results confirm the fact that a significant penalty on
efficiency is associated with the capture of CO2. This penalty is 21.6% relative to the IGCC without CO2
capture, i.e. 10.0% points. Estimated significant performance indicators as well as comparisons with alternative schemes have been presented. Some possible future developments based on these results and the
overall objective of the project are also discussed.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Use of energy is closely related to the development of an economy. Often per capita consumption of energy is considered as an
index of the living standard of the people of a country. Though
the efficiency of energy usage has a strong impact on energy consumption, the demand for energy is always expected to increase
with the growth in population and living standards. The most
useful form of energy in the modern world is electricity. Thus the
efficient conversion of primary energy to electricity is, besides its
⇑ Corresponding author. Tel.: +47 45 39 19 26; fax: +47 51 83 10 50.
E-mail address: [email protected] (M. Mansouri Majoumerd).
0306-2619/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.apenergy.2012.05.023
efficient use, critical for human civilization [1]. In terms of the economical aspects of available reliable technology, coal is still the
major source of electric power [2]. It is also available in different
parts of the world, safe to store and easy to transport over a long
distance. Thus coal has emerged as the most widely used fossil fuel
for large-scale power generation, though natural gas (NG) use is
also increasing mostly in localities of availability due to the fact
that it is more environmentally friendly [3]. Conventional pulverized fuel (PF) fired thermal power plants have been the most
prevalent technology worldwide over a long period. These plants
are mostly used for large-scale electricity supply through the grid.
Climate change due to anthropogenic greenhouse gas (GHG)
emissions is identified as the greatest threat to mankind [4]. The
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
281
Nomenclature
AGR
Ar
ASU
a.r.
BFW
CCS
CMD
CO
COS
CO2
DEPG
DLN
EOS
EU
GHG
GOX
GT
HHV
H2
H2S
HP
acid gas removal
argon
air separation unit
as received
boiler-feed-water
carbon capture and sequestration
coal milling and drying
carbon monoxide
carbonyl sulfide
carbon dioxide
di-methyl ether of polyethylene glycol
dry low NOx emission
equation-of-state
European Union
greenhouse gas
gaseous oxygen
gas turbine
higher heating value
hydrogen
hydrogen sulfide
high pressure
major source of these GHGs is carbon dioxide emission, and the
power sector is identified as the single largest sector contributing
to this emission. According to the International Energy Agency,
the CO2 emissions from electricity and heat production were about
41% of total global emissions from fossil fuels in the year 2009 [5].
The present challenge for the power sector is to meet the ever
increasing demand for electricity and simultaneously mitigate
the GHG emissions, principally CO2. Besides drastically increasing
the energy efficiency, one possible option is to replace fossil fuel
based power plants by renewable sources (say, solar, hydro, biomass, geothermal, etc.). Unfortunately, available technologies for
producing electricity from renewable sources are still not wholly
mature and/or not installed to the extent to meet the present demand fully in an economic and feasible way. Renewable sources
are undoubtedly the only option for the future, but the estimated
timescale for the complete transformation from fossil fuels to
renewable resources is not definite and is likely to be a significant
time away [3]. Thus the development of suitable technology for
large-scale power generation using coal during this transition is urgently needed.
The major challenge for future generation coal based power
plants is to minimize the emissions of CO2 to the atmosphere while
maintaining acceptable overall plant efficiency. One way of passively reducing this emission per MW power generation is to continue to further increase the efficiency of conversion. However,
incorporating some active measures to reduce CO2 emissions is
also necessary if targets for 2050 are to be met. Several routes have
been identified for this purpose [6–11]. Capturing CO2 from the exhaust flue gas mixture before it is vented to the atmosphere is
known as ‘post combustion carbon capture’. Solutions are usually
used to absorb CO2 from the flue gas mixture [12]. Processes based
on solids to capture CO2 from the flue gas mixture are described in
[13,14]. The small fraction of CO2 in the flue gas, which is mixed
with other combustion products and a large fraction of nitrogen
from atmospheric air, makes capture difficult. An alternative technology is oxy–fuel combustion. This is to use pure oxygen for the
combustion of coal, resulting in mostly CO2 in the exhaust flue
gas, which is easier to capture and transport directly for geological
storage. Large-scale separation of oxygen from the air, which is an
energy-intensive process, is needed and results in high penalties
HRSG
IGCC
IP
LHV
LP
MAC
NG
NOx
PF
PGAN
RH
SCOT
SGC
SGS
SH
ST
SWGS
TIT
TEG
TOT
VIGV
heat recovery steam generator
integrated gasification combined cycle
intermediate pressure
lower heating value
low pressure
main air compressor
natural gas
nitrogen oxide
pulverized fuel
pure gaseous nitrogen
reheating
Shell Claus off-gas treating
syngas cooling
syngas scrubbing
superheating
steam turbine
sour water–gas shift
turbine inlet temperature
tri-ethylene glycol
turbine outlet temperature
variable inlet guide vane
on efficiency. Also the gas turbine has to be redesigned for this process [15]. Significant development of this process at the laboratory
level is reported in the literature [16,17]. Pre-combustion carbon
capture is another good alternative technology [18–21]. Coal is
gasified to produce syngas (primarily a mixture of H2 and CO),
and the carbon monoxide produced in this process is converted
to carbon dioxide by ‘water-shift reaction’ [22]. CO2 is then separated from the syngas before combustion. As CO2 partial pressure
is higher in the gas mixture in this pre-combustion process, it is
comparatively easier to capture. Also other pollutant gases can
be removed in this gas cleaning process before combustion, resulting in minimum emission of pollutants. The resulting hydrogenrich syngas is subsequently used in a combined power plant with
high thermodynamic efficiency. Such integrated gasification combined cycle (IGCC) power plants with pre-combustion carbon capture appear to be promising for using coal and meeting the
environmental standards [23]. However, the overall plant efficiency is reduced, and complex plants for coal gasification and
gas treatment/cleaning may lead to frequent shutdown and reduced reliability. Though high-scale integration may improve the
efficiency of the plant during operation [24], it might also reduce
the reliability of the power supply, according to the practical experience of operating similar equipment in integrated configurations
such as the Buggenum plant operated by Vattenfall. Determining
the optimum degree of integration to obtain acceptable efficiency
and reliable power supply is, therefore, another objective for future
plants.
In the H2-IGCC project (refer to Section 2), the overall objective
is to enable the stable operating conditions of the gas turbine with
combustion of undiluted H2-rich syngas, and to increase the ability
to operate on a variety of fuels (such as cleaned syngas, and natural
gas), a feature known as ‘fuel flexibility’, without adversely affecting the reliability and availability of the plant. Besides, minor modification of existing GTs is one of the project’s goals.
In this paper a detailed thermodynamic model is presented, as
well as the performance analysis of the integrated gasification
combined cycle power plant based on practical flow-sheet and
realistic performance indicators verified by the operators of similar, relevant plants. Further investigations for an optimum configuration, based on the results of this study, are also discussed.
Author's personal copy
282
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
2. H2-IGCC project
Current technologies for coal gasification and the use of the syngas as fuel in combined cycle power plants suffer from two main
limitations. The wide variation of fuel composition (i.e., wide variation of the heating value) causes several difficulties during operation using available gas turbines (GTs). Furthermore, the syngas is
rich in H2 due to pre-combustion carbon capture resulting in challenges for using the premix burners. These burners are the stateof-the-art technology for dry low NOx (DLN) combustion in natural
gas fuelled GTs. Higher reactivity of H2 compared to that of NG
causes problems for using these burners in IGCC plants. Often this
problem is overcome by using conventional diffusion burners
resulting in high emissions of NOx, besides diluting the fuel with
either N2 or water/steam to reduce the effect of the higher adiabatic flame temperature when burning H2-rich fuel.
In November 2009 the H2-IGCC project was started with the
overall goal to develop and demonstrate technological solutions
to overcome the above-mentioned drawbacks while burning H2rich fuels. The developed technology therefore may allow burning
undiluted H2-rich syngas with low NOx emissions, comparable
with that of the state-of-the-art technologies with NG as fuel.
The goal of the project is also to develop an optimized plant layout
that will not only maximize efficiency but also allow fuel flexibility
as well as reliability of power supply. Both options are also to be
explored: with CO2 capture or without since access to CO2 transport, injection and storage infrastructure is not yet guaranteed.
Twenty-four different partners including academia and manufacturers as well as plant operators from ten European countries
are working together to achieve the above-mentioned goals. Four
major research areas are targeted namely combustion, materials,
turbomachinery and system analysis. Results of these activities
should support:
Developing and demonstrating a safe and low emission premixed combustor technology for the undiluted hydrogen-rich
syngas from coal gasification with pre-combustion carbon
capture.
Developing and demonstrating improved materials with
advanced coatings for the turbine blades and combustor.
The target is to achieve lifetimes similar to those of the latest
natural gas fired gas turbines for identical run time in spite of
the potentially more aggressive temperature and composition
of the exhaust gas.
Providing required design for the compressor/expander aerodynamics as well as the cooling of hot gas path components
in order to cope with increased mass flow rate, due to the
higher fuel flow and changed gas properties of the exhaust
gas, which causes changed heat transfer conditions at all surfaces exposed to the hot gas.
Evaluating and optimizing the best IGCC plant configuration
as well as to provide guidelines for optimized full-scale integration in order to satisfy the above-mentioned requirements. Moreover, a detailed systems analysis will be
performed to generate realistic techno-economic results for
IGCC plants with pre-combustion carbon capture. The results
are to be compared with a natural gas fired plant for
benchmarking.
3. System overview
In this section the thermodynamic model of the baseline configuration of the IGCC power plant consisting of several sub-systems is
described. The detailed model includes many sub-systems with
reasonable assumptions based on either commercially available
technology or data provided by other subgroups of the project. Significant sub-systems with relevant assumptions and issues are discussed. The thermodynamic model of the baseline IGCC power
plant was based on commercially available technologies as follows:
Cryogenic air separation unit (ASU);
Oxygen-blown, entrained flow coal gasifier based on Shell
technology;
Sour water–gas shift (SWGS)
Carbonyl sulfide (COS) hydrolysis unit for the case when no
capture of CO2 took place;
Acid gas removal (AGR) unit using physical absorption by
SELEXOLTM system;
CO2 compression and dehydration unit; and
power generation block consisting of a 300 MW single-shaft
gas turbine based on the Siemens/Ansaldo Energia 94.3A
technology and a conventional triple-pressure bottoming
steam cycle.
Theoretical description and integration methods of similar subsystems have been addressed by several studies [21,22,25–31].
Therefore, in this study special emphasis is placed on a practical
plant (Fig. 1) to estimate realistic performance indicators, verified
by project team members with relevant plant operating experiences. The specifications of each sub-system as well as assumptions and boundary conditions are described in the following
sub-sections.
3.1. Coal feed
The assumed coal was a bituminous coal which was a mixture
of various trade coals on the world market. The composition of
the coal (wt.% a.r.) is summarized in Table 1.
The amount of coal required was determined by the thermal
power demand of the gas turbine which was within the range of
1005–1112 MWLHV.
3.2. Air separation unit (ASU)
The cryogenic air separation process, which is currently the
most reliable technology for large-scale production of oxygen
and nitrogen [25], was considered as a stand-alone unit. The purity
of oxygen was set to 95 mol% (2% N2 and 3% Ar) as obtained from
the techno-economic evaluation of such units utilized in real
plants.
Ambient air was initially compressed by the main air compressor (MAC). Although integration of the ASU and the GT would lead
to a higher efficiency, the most appropriate degree of integration
needs to be decided. Better operational safety and greater availability of the plant would be the result of low or no integration
[26]. Experience from the real IGCC plant in Buggenum promoted
a non-integrated approach for improved plant availability [27].
The MAC was a three-stage intercooled compressor with a discharge pressure of 5.5 bar. The delivery pressures of the pure gaseous nitrogen (PGAN) and gaseous oxygen (GOX) from the MAC
were 5 and 1.2 bar, respectively. The discharge pressure of the
PGAN which was used for coal feeding to the gasifier was 80 bar,
while the gaseous oxygen (GOX) was compressed further up to
55 bar and fed to the gasifier. Both compressors for the GOX and
the PGAN were six-stage intercooled compressors. It is worth noting that the lower number of inter-cooled stages used here compared to other studies, as well as the lower component
efficiencies, represent ‘‘common practices’’ in the currently existing power plants. Further technical assumptions for the ASU are
shown in Table 2 below.
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
283
Fig. 1. The schematic configuration of the IGCC power plant with carbon capture.
Table 1
Composition (% by weight) and heating values of the bituminous coal.
C
H
N
O
S
Cl
F
64.10
3.90
0.70
7.21
1.50
0.09
20 ppm
Moisture
Ash
10
12.50
HHV
LHV
kJ/kg
26,195
25,100
Table 2
General assumptions for the ASU.
Delivery pressure/temperature of O2 by ASU (bar/°C)
Delivery pressure/temperature of N2 by ASU (bar/°C)
Main air compressor polytropic efficiency (%)
GOX compressor polytropic efficiency (%)
HP PGAN compressor polytropic efficiency (%)
Inter-cooling temperature (°C)
1.2/10
5.0/10
85
78
78
40
3.3. Gasification, syngas cooling and scrubbing
Gasification of coal took place in the Shell gasifier [28]. The
milled and dried coal was gasified in the presence of intermediate
pressure (IP) steam and oxygen. The gasifier was of single pass
type, and the remaining fine particles that were not captured by
the ceramic filters as fly ash left the bottom of the gasifier as vitreous slag. One of the major sources of inaccuracies in theoretical
studies presented in open literature is due to the non-realistic
model of the gasification unit. The gasification plant model used
in this study was based on the existing plant in Buggenum [27],
and the gas analysis obtained from the model was validated
against real plant data.
The raw syngas was first cooled to 900 °C by recycling the
cooled, ash-free syngas stream and then further cooled to 340 °C
in syngas coolers by generating high pressure (HP) and intermediate pressure (IP) steams, as shown in Fig. 1. After passing through
the dry particulate filters and recycling a part of the syngas stream
for the aforementioned cooling purpose, the rest of the syngas was
sent to the wet scrubber. The quenched and cleaned syngas entering the SWGS unit had a temperature and pressure of 165 °C and
43 bar, respectively. Further technical assumptions and results
are presented in Table 3 below. The syngas composition at the inlet
of SWGS unit is presented in Table 4.
3.4. Sour water–gas shift
The water–gas shift process was a catalytic reaction converting
the CO of the raw syngas to CO2 (Eq. (1)). This was prior to acid gas
removal and was therefore termed as sour water–gas shift. This
reaction was carried out in two sequential reactors.
kJ
ð44moleÞ
COðgÞ þ H2 OðgÞ $ CO2ðgÞ þ H2ð gÞ
ð1Þ
The dry-feed characteristics of the Shell gasifier required the
injection of a considerable amount of steam to ensure adequate
CO to CO2 conversion rate during the SWGS. In order to protect
Author's personal copy
284
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
Table 3
General assumptions for the Shell gasification block including the syngas conditioning
downstream to the wet scrubber.
Dried coal moisture content (wt.%)
Gasification pressure/temperature (bar/°C)
Steam/coal ratio (kg/kg coal a.r.)
O2/coal ratio (kg/kg coal a.r.)
HP PGAN/coal ratio (kg/kg coal a.r.)
Power requirement (kJel/kg coal a.r.)
Heat loss to membrane wall (% coal LHV)
Carbon conversion (single pass/overall) (%)
Syngas cooler pinch-point HP evaporator (°C)
Syngas cooler pinch-point IP evaporator (°C)
Heat exchanger heat loss (%)
Pressure drop syngas cooler (gas side) (bar)
Pressure drop wet scrubber (bar)
Water pump mechanical efficiency (%)
Shifted-desulfurized syngas for CMD in the case with CCS (% of total
flow)
Shifted-desulfurized syngas for CMD in case of CCS unit trip (% of
total flow)
2
45/
1600
0.061
0.7839
0.241
112
2.5
99.3
30
64
0
0.33
1
85
1.7
1.5
Table 4
Syngas composition (mole fraction) at the inlet of
SWGS unit.
Components
Raw syngas
CO
CO2
H2
H2O
N2
H2S
COS (ppmv)
NH3 (ppmv)
0.4895
0.0305
0.2268
0.1751
0.0739
0.0042
341
87
the catalytic bed from carbon deposition as well as to increase the
equilibrium conversion of CO and steam to H2 and CO2, the steamto-CO ratio was set to 2.4. The IP steam for this purpose was partly
supplied from the syngas cooler and the rest was provided from
the HP/IP steam turbine crossover. The inlet temperatures of both
reactors were set to 250 °C. The high temperature syngas leaving
the first reactor was quenched by saturating HP and IP boilerfeed-water (BFW) flows, while the warm syngas leaving the second
reactor was cooled by preheating the raw syngas entering the first
reactor and preheating HP and IP BFWs and generating low-pressure (LP) steam.
The resulting overall adiabatic conversion of CO to CO2 in the
SWGS process was 98% (molar basis). The total pressure loss of
the syngas from the exit of the wet scrubber to the exit of the demister was 7.7%. It is worth noting that one of the advantages of this
unit is the simultaneous conversion of carbonyl sulfide (COS) to
hydrogen sulfide (Eq. (2)) [29] with shift reaction.
COSðgÞ þ H2 OðgÞ
$
3.6. Acid gas removal (AGR) unit
During the gasification process, the sulfur content in the raw
coal was converted to H2S and COS. However, most of the COS
(more than 98%) was converted to H2S during the SWGS reaction.
The pressure of the system, which was dictated by the gasification
process, resulted in high partial pressure of CO2 (around 15.5 bar).
Therefore, two-stage physical absorption of acid gases, using dimethyl ether of polyethylene glycol (DEPG) also known as SELEXOL, was favorable for the AGR unit. High stability, high absorption capacity as well as low corrosive effect, are the favorable
features of SELEXOL.
H2S was removed from the shifted gas in the first stage. The rich
solvent exited at the bottom of the absorber column. Since the concentration of CO2 in the syngas leaving the SWGS unit was high,
and the order of magnitude of solubility of H2S and CO2 in SELEXOL
was similar, a considerable amount of CO2 content was co-absorbed with H2S [31]. Having the appropriate content of H2S (more
than 35% molar basis) in the inlet stream of the Claus unit, which
was used to oxidize H2S to elemental sulfur, a concentrator unit
was considered after the H2S absorber tower. A part of the fuel
gas (2.5%) produced in the AGR unit was recycled back to the concentrator column as a stripping agent. The rich solution exiting at
the bottom of the concentrator column was then stripped off in a
regenerator for which heat was provided by the LP steam generated in the SWGS unit. The regenerated solvent was cooled to
5 °C and recycled back to the H2S absorber. The separated H2S
was sent to a sulfur recovery unit including a Claus plant and a
Shell Claus off-gas treating (SCOT) plant for tail gas clean-up.
In the case of running the gas turbine with cleaned syngas (i.e.
CCS unit trip and without the SWGS unit) the H2S concentrator column was bypassed, since the concentration of CO2 was low in the
syngas leaving the COS hydrolysis unit. Fig. 2 illustrates the configuration of IGCC power plant without carbon capture unit.
Syngas leaving the H2S absorber, after extraction of a small part
of it for coal drying purpose, entered the second stage for CO2 removal. The removal process was similar to that in the first stage.
The rich CO2 solvent was passed through four flash drums connected in series, where CO2 was released as a result of lowering
the pressure. The lean solvent leaving the last flash drum was
cooled to 5 °C to increase the absorption efficiency. However, this
cooling increased the inherent loss of energy [29]. The cooled solvent was then re-circulated back to the absorber tower. The gas
leaving the first flash drum was recycled back to the absorber
tower. The CO2 released in flash drums two to four was sent for
compression. The CO2 removal rate in the AGR unit was 93.65%
(molar basis), while the overall CO2 capture rate as defined in Eq.
(3) was 89.80% (molar basis).
Capture rate ¼
kJ
ð33:6moleÞ
syngas, LP steam generation, and IP BFW saturation. The conversion of COS to H2S in this reaction was 99.8% (molar basis).
H2ð gÞ þ CO2ðgÞ
ð2Þ
3.5. Carbonyl sulfide hydrolysis unit
In order to remove more than 99.99% of the sulfur content in the
produced syngas, it was necessary to add a COS hydrolysis unit to
convert COS to H2S in the case of CCS unit trip (Fig. 2), according to
Eq. (2) [30]. In this case, the scrubbed syngas was preheated to
220 °C before entering the COS hydrolysis reactor. The syngas
leaving this reactor was cooled down by preheating the scrubbed
CO2sent to compression þ C A þ C S
100
CF
ð3Þ
where CA, CS, and CF are carbon content in fly ash, slag, and feed coal,
respectively. The solidification of fly ash and slag took place in the
plant and this was the reason for the relocation of these two terms
from denominator to numerator in the above equation (Eq. (3)).
For cases when CO2 was not captured, the syngas leaving the
COS hydrolysis unit and the demister entered the H2S absorber.
The rich solution leaving the bottom of the tower was regenerated
and the sulfur was stripped off using LP steam produced after the
COS hydrolyser unit. The H2S-free hydrogen-rich syngas exiting
at the top of the absorber was passed to the GT combustor. The
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
285
Fig. 2. The schematic configuration of the IGCC power plant without carbon capture.
Table 5
General assumptions and results of both AGR units with and without CO2 capture.
Syngas pressure/temperature at H2S absorber inlet
(bar/°C)
Solvent pumps polytropic efficiency (%)
Compressor isentropic efficiency (recycle gas) (%)
Mechanical and electrical efficiencies (%)
Coefficient of performance for refrigeration pump
Solvent temperature at absorber inlet (°C)
Pressure loss in 1st/2nd absorber (bar)
H2S stripping duty (MWth)
H2S removal efficiency (%)
CO2 co-absorbed in H2S absorber (mol% of inlet)
Overall H2 co-absorption (mol% of inlet)
Overall CO co-absorption (mol% of inlet)
CO2 removal efficiency(mol% of inlet)
CO2
capture
Without
CO2
capture
39.7/
25
70
85
98
2.2
5
0.5/0.5
44.7
99.99
1.32
2.19
3.26
93.65
39.7/25
70
85
98
2.2
5
0.5/–
22.0
99.99
12.62
0.02
0.06
12.62
general assumptions and results of the AGR units for both cases, i.e.
with and without CCS, are presented in Table 5 below.
3.7. CO2 compression
The CO2 released from the last three flash drums in the CO2 separation process was compressed in a seven-stage intercooled compressor to 60 bar, liquefied and then pumped up to a final pressure
of 150 bar. In order to reduce the corrosion risk in the transport
pipeline, a dehydration unit using tri-ethylene glycol (TEG) was
considered. The upper limit of 20 ppm (mass basis) for water con-
tent in the CO2 stream was specified according to Ref. [32]. The
compressor isentropic efficiency and pump polytropic efficiency
have set to 80% and 70%, respectively.
3.8. Gas turbine model
The gas turbine was modeled using internal project information
exchange focusing on turbomachinery. This information included
initial performance calculation results of a lumped turbomachinery model of the GT, Ansaldo Energia 94.3A, the compressor map
and some turbine data. All information used was based on a natural gas fuelled engine. The control algorithm for burning undiluted
H2-rich syngas and cleaned syngas was without any major modifications, i.e. the same as for natural gas operation. The turbine inlet
temperature (TIT) was fixed and set to 1331 °C, and the model adjusted the pressure ratio due to the increased fuel flow. The current
GT model was built up accordingly:
Compressor model: The generated compressor characteristics
using a lumped GT model were implemented with variable inlet
guide vane (VIGV) opening positions for ISO condition (i.e., 15 °C,
1.01325 bar, 60% relative humidity).
Combustor model: The composition of fuel to the combustor
was calculated using the models described above (Sections 3.2–
3.6). Furthermore, a pressure loss reflecting current state-of-theart technology for dry low NOx combustors was used.
Turbine model: A simplified expander model was used, assuming a constant flow through the turbine. The influence of cooling air
entering the turbine at different rows was considered, using a virtual/mixed turbine inlet temperature as well as virtual mixed polytropic efficiency according to the following equations:
Author's personal copy
286
T t mixed i ¼
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
Pshaft
þ T to
_ total cp
m
ð4Þ
and
gt polytropic mixed ¼
C p lnðT tmixed i =T to Þ
RD lnðpti =pto Þ
ð5Þ
_ is mass flow rate, cp is specific
where T is temperature, P is power, m
heat, g is efficiency, R is gas constant, p is pressure, and subscriptions i, o, and t are expander inlet, expander outlet and total condition, respectively. The general assumptions for the GT including NG
operating conditions are listed in Table 6 below.
3.9. Heat recovery steam generator
Downstream of the GT was a three-pressure heat recovery
steam generator (HRSG) with reheat. The superheating (SH) temperature was set to 530 °C in order to meet both the GT exhaust
temperature and the required amount of HP steam needed to be
superheated. Potential exists for further increasing the HRSG effectiveness in order to maximize the net electrical output. However,
the economic feasibility of such optimization should not be disregarded. The IP level was set to meet the pressure of the syngas
leaving the wet scrubber, i.e. 43 bar, since a considerable amount
of IP steam was extracted from the HRSG and mixed with the
raw syngas in order to perform the SWGS reaction.
For the case without CO2 capture, the required amount of IP
steam is considerably lower compared to the cycle with CCS, due
to the bypassing of SWGS unit. The HP superheating temperature
was lowered to 500 °C to accomplish the superheating of the entire
steam produced in the gasifier as well as the extra IP steam which
is not required in the process. The general assumptions made for
the HRSG calculation are listed in Table 7.
Table 6
General assumptions used for the gas turbine.
Ambient air pressure (bar)
Ambient air temperature (°C)
Moisture in air (%)
TIT (°C)
GT outlet pressure (bar (total))
Pressure ratio
Electrical/mechanical efficiency (%)
NG mass flow (kg/s)
GT power output fuelled with NG (MW)
Turbine outlet temperature (NG fuelled) (°C)
1.013
15
60
1331
1.08
18.2
99/99.5
14.88
292
577
4. Simulation tools
To obtain reliable results and to utilize the possibility of incorporating detailed component characteristics later, a combination of
the following simulation tools was used for modeling the IGCC
power plant:
Enssim: Simulation tool developed by Enssim Software [33];
ASPEN Plus: Commercial process engineering software by
AspenTech [34]; and
IPSEpro: Commercial heat and mass balance programme by
SimTech [35].
Data exchange between software tools was performed manually to find the optimal match. The main reason for utilizing three
software tools was the specific capabilities of each tool to model a
certain sub-system. To simulate each aforementioned sub-system
(Section 3.1), the relevant one of these three software tools was
used as described below.
Detailed modeling of the Shell gasification process, including
components such as coal milling and drying (CMD), gasification, raw syngas cooling (SGC) and scrubbing (SGS), was performed by Vattenfall (Nuon) using the Enssim modeling tool.
It is worth noting that the gasification model was validated
against real plant operational data.
The air separation unit (ASU) was modeled using ASPEN Plus.
The Peng-Robinson properties method was selected as the
equation-of-state (PR EOS).
The sour water–gas shift (SWGS) reaction was modeled in
ASPEN Plus using PR EOS.
The acid gas removal (AGR) unit was modeled in ASPEN Plus.
Two different equations-of-state, i.e., Peng-Robinson and Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT),
were used for simulation. However, based on a benchmarking
study with one of the industrial partners, the simulation
using PC-SAFT equation-of-state was selected.
For cases without CO2 capture, the carbonyl sulfide (COS)
hydrolysis unit and H2S removal (i.e. AGR unit) was modeled
in ASPEN Plus using PR EOS and PC-SAFT EOS, respectively.
The compression of captured CO2 and dehydration of CO2
stream was modeled in ASPEN Plus using PR EOS, and Schwarzentruber and Renon (SR polar) equation-of-state,
respectively.
The power block including the GT, and the triple-pressure
steam cycle were modeled in IPSEpro.
5. Results and discussion
Table 7
General assumptions used for the HRSG.
HP/IP/LP (bar)
SH and RH temperature (°C)
SH LP steam (°C)
HP/IP/LP ST isentropic efficiency (%)
ST and generator mechanical efficiency (%)
Gas side HRSG pressure drop (bar)
Condenser pressure (bar)
Generator electrical efficiency (%)
Pump polytropic efficiency (%)
Pump mechanical efficiency (%)
Evaporator pinch point IP/LP (°C)
Super heater pinch point (°C)
Economizer pinch point (°C)
Approach point temperature (°C)
140/43/4
530a
300
88.5/89/91
99.5
0.04
0.04
98.2
70
95
10/10
32
10
5
a
The superheating/reheating temperature for the case without CO2 capture is
500 °C; all other assumptions are the same.
The overall objective of the H2-IGCC project is to enable the premix combustion of undiluted H2-rich syngas in IGCC power plants.
Developing an optimized plant layout that not only maximizes efficiency but also increases fuel flexibility by enabling the burning of
cleaned syngas with variable composition is another goal. The system analysis research group aims to evaluate and optimize the
IGCC plant configuration. As an initial step toward optimizing the
configuration of the plant, this group has established a realistic
baseline process layout. A detailed thermodynamic model using a
combination of three simulation tools is presented in this paper.
Performance analysis of the IGCC plant based on practical flowsheet and realistic performance indicators is reported in this paper.
Simulation results of the baseline IGCC plant for two different
cases, i.e. with or without CO2 capture (referred to here as Case A
and Case B, respectively), based on calculations using models
as described in Sections 3 and 4 are presented. Estimated
Author's personal copy
287
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
Table 8
Syngas composition (mole fraction) and characteristics prior to the GT burner.
Components
H2-rich syngas
Cleaned syngas
CO
CO2
H2
H2O
N2
Pressure (bar)
Temperature (°C)
LHV (kJ/kg)
HHV (kJ/kg)
Mass flow (kg/s)
Mole flow (kmole/s)
0.0117
0.0403
0.8583
0.0004
0.0893
38.71
30.0
33255.55
39211.07
23.30
3.67
0.5984
0.0334
0.2774
0.0005
0.0904
39.21
30.0
11084.33
11656.48
70.92
3.33
performance indicators as well as specific CO2 emissions in both
cases are also reported. Performance indicators such as gas and
steam turbine power outputs, auxiliary power demands, and efficiencies of processes were validated against the results of feasibility studies conducted by industrial partners of the project and the
results of published literature. The syngas compositions for both
cases are also reported. These compositions were validated against
published data. Some adverse effects of fuel change and the possible measures for those are also briefly discussed.
The composition and other properties of the corresponding fuel
gas in the two cases in this study, i.e. H2-rich syngas (Case A) and
cleaned syngas (Case B) prior to the gas turbine burner, are presented in Table 8.
Estimated performance parameters of this simulation for Cases
A and B together with corresponding operating conditions (points
in Figs. 1 and 2) are shown in Table 9. The gas turbine power outputs in both cases are higher than the power output of the NG
fuelled GT (referred to here as Case C, see Table 6) which is approximately 292 MW [36]. This is due to greater hot gas flow in Cases A
and B than the Case C. It is worth noting that the current GT model
in all cases was using the compressor map based on NG as the fuel,
and the component efficiencies were kept constant. The steam turbine power outputs for Cases A and B show a significant difference.
IP steam consumption decreases in the absence of SWGS for Case B.
This additional IP steam expands in the steam turbine to increase
power output of the steam turbine by 39 MW. Higher integration
of the HRSG and the SWGS in Case A results in 12.4% increase of
the HRSG pumping power demand compared to Case B.
Simulation results confirm that the highest auxiliary power
consumption is in the ASU for both cases (refer to Fig. 3). Solvent
flow in the AGR unit for Case A is higher due to CO2 capture. This
causes approximately 8.5 MW higher power demand for solvent
circulation. Moreover, the higher solvent flow needs 6.8 MW more
power for its refrigeration. The total power demands for pumping,
compression and refrigeration for H2S removal in both cases are
relatively small. These power demands are 7.65 MW and
3.61 MW (about 0.36% and 0.69% of the respective gross thermal
Table 9
Performance results and operating conditions of the IGCC power plant with and without CO2 capture (Fig. 1 and 2).
Point #
IGCC plant with CCS (Case A)
IGCC plant without CCS (Case B)
T (°C)
P (bar)
_ (kg/s)
m
T (°C)
P (bar)
_
m(kg/s)
15.0
15.0
125.8
98.4
164.7
250.0
276.7
25.0
12.4
12.4
33.8
30.0
15.0
419.8
1331.0
563.3
530.0
530.0
241.6
29.1
314.0
338.1
337.0
30.1
31.3
104.8
1.01
1.01
55.00
80.00
43.00
41.72
40.48
39.71
39.21
39.21
150.00
38.71
1.01
19.02
19.02
1.08
140.00
43.00
4.00
0.04
43.00
143.00
140.00
50.00
141.00
1.04
150.09
44.31
36.74
10.68
95.71
177.69
177.69
120.62
119.13
2.00
93.19
23.30
683.01
614.71
638.22
706.52
145.62
95.47
114.67
114.67
78.39
35.52
130.82
21.62
95.30
706.52
15.0
15.0
125.8
98.4
164.7
220.0
220.4
25.0
9.2
9.2
4.4
30.0
15.0
424.7
1331.0
583.3
500.0
500.0
206.7
29.1
263.2
338.1
337.0
257.4
329.8
66.5
1.01
1.01
55.00
80.00
43.00
42.50
42.00
39.71
39.21
39.21
39.71
39.21
1.01
19.48
19.48
1.08
140.00
43.00
4.00
0.04
50.00
143.00
140.00
50.00
43.00
1.04
136.74
40.11
33.26
9.67
86.64
86.63
86.63
73.47
72.03
1.11
121.92
70.92
682.96
607.84
678.76
753.89
118.62
153.98
150.51
150.51
3.25
118.42
118.42
19.61
118.62
753.89
Performance indicators
Case A
Case B
GT power output (MWe)
ST power output (MWe)
HRSG pumping power demand (MWe)
AGR pumping and compression power demand (MWe)
AGR refrigeration power demand (MWe)
ASU compression power demand (MWe)
CO2 compression power demand (MWe)
Gasification power demand (MWe)
Net power out (MWe)
Net IGCC efficiency (% LHV)
Overall CO2 capture (%)
Specific CO2 emissions (g CO2/kWh)
329.22
172.91
3.53
9.12
9.78
50.02
20.83
4.96
403.89
36.32
89.80
78.57
311.27
211.95
3.14
0.66
2.95
45.45
0.00
4.48
466.53
46.34
0.00
716.01
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Author's personal copy
288
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
Fig. 3. The shares of auxiliary power consumptions for both cases (IGCC with and without carbon capture).
Table 10
The main specification and performance results of published studies.
Sub-system/performance
indicators
Kreutz et al. [38]
Kanniche et al. [32,37]
Chiesa and Consonni
[22]
Present study
Gasifier
Shell technology
O2 blown, dry fed based on
PRENFLO technology
27 bar
100% integration with GT
Sweet
SELEXOLTM
150 bar to storage
85% Capture
Siemens V 94.3
O2 blown, slurry fed
based on Texaco
60 bar
Non-integrated
Sour
SELEXOLTM
80 bar to storage
91.7%% Capture
Generic F technology
Shell technology
Diluted with steam for IGCC
without CCS
Diluted with N2 and steam
Pre-heating (125 °C)
Triple pressure level
Triple pressure level
47.7%
43.9%
45.9%
Triple pressure
level
46.3%
36.6%
29.9%
38.8%
36.3%
ASU
WGS type
CO2 capture system
GT technology
38.5 bar
N2 for dilution
Sour
SELEXOLTM
150 bar to storage
93% Capture
GE 9FB
Diluted with N2 and small amount of steam
HRSG
Overall efficiency of IGCC
without carbon capture
Overall efficiency of IGCC with
carbon capture
Pre-heating (350 °C for IGCC w/o CO2 capture &
200 °C for IGCC plus capture)
Triple pressure level
inputs) for Cases A and B, respectively. The total power demand for
CO2 capture and compression for Case A is approximately
32.1 MW, which is about 2.88% of the gross thermal input for that
case. The net IGCC efficiency is 10% (absolute value) or 21.6% (relative value) less for Case A than for Case B. This significant efficiency penalty for capturing CO2 is due to the auxiliary power
demand of the SWGS unit and the CO2 capture and compression
units. However, this loss in efficiency may be justified in future
IGCC plants by more stringent environmental regulations for CO2
emissions.
Simulation results of some configurations of IGCC plants are
reported by other authors [22,32,37,38]. A brief comparison of
the results of this work with a few others available in literature
is described and discussed in the following. The similarities and
deviations in results are explained for the validation of the present
work as well as to gain insight for future improvements. Table 10
shows a brief comparison of the specifications of sub-systems
and some performance parameters of this study with some others
available in literature. Even though the syngas used in this work is
not diluted with N2 as assumed by Kreutz et al. [38], Case B of this
study has about 1% lower efficiency. The cleaned syngas fuel was
strongly (up to 350 °C) pre-heated before the combustor and the
HRSG were fully optimized in Kreutz’s study. However, the net efficiency for Case A is closer to that reported by them. Some factors
43 bar
Non-integrated
Sour
SELEXOLTM
150 bar to storage
89.8% Capture
Siemens/Ansaldo
Energia 94.3A
Undiluted fuel
Saturation with steam
have had increasing effects on the reported efficiency in the present study compared to Ref. [38]. These factors include: (a) Higher
power output from the GT due to using undiluted syngas; (b) lower
steam-to-CO ratio in the SWGS, which results in lower steam
consumption and higher steam bottoming power; (c) lower CO2
capture rate which implies lower loss; and (d) lower power demand in gasification block. On the contrary, the lower optimization/integration level of the HRSG in the current study is likely to
cancel the increasing effects of the aforementioned factors. The differences in efficiencies are not significant when being compared
with the overall plant efficiency. In another study, although full
integration between the GT and ASU was considered (i.e. 100% of
the air feeding the ASU is extracted from the GT), lower net efficiency (LHV basis) in both cases (6.4% for Case A and 2.4% for Case
B) has been reported by Kanniche et al. [32,37]. The steam injection
into the GT for controlling NOx formation results in an efficiency
penalty in both cases. Having a similar low heating value compared
to that produced in Case B, the dilution with N2 in Case A has been
considered, which results in efficiency reduction. Also, the lower
pressure of the gasification block (27 bar) results in lower carbon
capture efficiency and a higher energy penalty in the CCS unit.
However, lower carbon capture efficiency results in lower steam
injection into the SWGS and a lower penalty for the bottoming cycle. Another study conducted by Chiesa and Consonni [22] reported
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
a net efficiency of about 45.9% for Case B which shows similar values to this study. On the contrary, the net efficiency of Case A in
Chiesa’s study is approximately 2.5% higher compared to this
study. One reason is the utilization of the hydraulic expander for
recovery of solvent pumping power from CO2-rich solvent before
first flash drum in the AGR unit. Using a slurry fed gasifier at higher
pressure (60 bar), which results in lower injection of IP steam to
the SWGS is also another reason for this efficiency difference.
The scientific literature confirms that the composition of the
cleaned syngas based on Shell technology shown in Table 8 is within the range of those obtained from commercial gasifiers [30,39].
The similar lower heating value of H2-rich syngas in the current
study and that of Ref. [22] (33.255 MJ/kg and 34.663 MJ/kg, respectively) shows a good agreement for every sub-system upstream of
the GT. Even though in the study conducted by Zheng and Furinsky
[30], the higher O2-to-coal ratio (0.9 instead of 0.78 in this study)
and the higher O2 content in the coal result in lower CO in the
cleaned syngas, the higher gasification temperature employed in
Zheng and Furnisky’s study (2000 °C instead of 1600 °C in this
study) led to similar CO and CO2 contents in both studies. Due to
the lower H2 content in cleaned syngas (Table 8), this fuel has lower HHV and LHV compared to H2-rich syngas. The big difference in
composition, especially in H2 content, will certainly demand different combustor designs to maintain stable combustion. In addition,
the large difference in the cleaned syngas flow compared to the H2rich fuelled GT (70.9 kg/s compared to 23.3 kg/s), leads to the higher turbine inlet flow and consequently higher back pressure. Closing the variable inlet guide vane (VIGV) of the compressor will
solve this problem to some extent. However, keeping a reasonable
surge margin in the compressor and an acceptable efficiency over a
wider range of VIGV positions will require compressor modifications. The turbine outlet temperature (TOT) is higher (583 °C) for
Case B compared to Cases A (563 °C) and C (577 °C), which favor
the steam bottoming cycle. However, the higher TOT for the
cleaned syngas fuelled GT than that of the NG fuelled one raises
one critical design concern, i.e. the lifetime of turbine components.
It may lead to a considerable reduction in the lifetime, especially of
the last and penultimate (un-cooled) stage blades of the turbine.
This problem may be solved in several ways, such as by modification of the compressor and/or expander flow path, by blowing off a
part of the compressed air and by modification of the operating
condition (i.e. different TIT and VIGV position). Some changes in
the design of the combustor and turbomachinery block are inevitable in order to accommodate the switching of fuels (say, from H2rich syngas to cleaned syngas) and the associated problems of the
reduced lifetime of turbine blades. These may evolve with the future progress of the project.
6. Conclusion
The overall objective of the H2-IGCC project is to enable the premix combustion of undiluted H2-rich syngas in IGCC power plants
with carbon capture. The object of the system analysis group of the
project is to evaluate and optimize the IGCC plant configuration. As
an initial step, simulation of a baseline configuration with a detailed thermodynamic model, using realistic assumptions and
internal information from other partners of the project, is reported
in this paper. This simulation will set the framework for technoeconomic analysis in the next step for the commercial feasibility
assessment of the concept.
The estimated overall efficiency of the IGCC power plant without carbon capture is 46.3%. For the plant with carbon capture,
the figure is 36.3%. This confirms the fact that a significant penalty
on efficiency is associated with the capture of CO2. This penalty is
21.6% relative to the IGCC without CO2 capture. However, stricter
289
environmental regulations regarding CO2 emissions as well as
the requirement for a secured energy supply and the lack of mature
alternatives may justify the IGCC technology with CCS in future.
Through comparison with other published studies, more integration of sub-systems indicated some potential for better efficiency but with lesser reliability. Using undiluted syngas in the
GT improves GT power significantly. However, some challenges related to the unstable operating condition of the GT combustor and
compressor as well as reduced lifetime of the blades of the existing
gas turbines when using undiluted H2-rich syngas have to be addressed. Thus optimization of the plant needs more investigation.
This study identifies these areas for future investigation and sets
the framework for techno-economic analysis in the next step of
optimization involving commercial feasibility.
Acknowledgments
The authors wish to acknowledge Vattenfall and E.ON for their
technical inputs. The authors are also grateful to Han Raas at Vattenfall for performing the gasification simulations.
References
[1] Armaroli N, Balzani V. Towards an electricity-powered world. Energy Environ
Sci 2011;4(9):3193–222.
[2] Spliethoff H. Power generation from solid fuels. 1st ed. Springer; 2010.
[3] IEA. World Energy Outlook 2011. International Energy Agency, Paris; 2011.
[4] IPCC. IPCC fourth assessment report: climate change 2007 (AR4); 2007.
<http://www.ipcc.ch>.
[5] IEA. IEA Statistics 2011 Edition: CO2 emissions from fuel combustion
highlights. International Energy Agency, Paris; 2011.
[6] Sen S. An overview of clean coal technologies II: mitigating the environmental
impacts by continuous improvement in coal combustion and CCS technology.
Energy Source Part B: Econ Plan Policy 2011;6(2):118–25.
[7] Olajire AA. CO2 capture and separation technologies for end-of-pipe
applications – a review. Energy 2010;35(6):2610–28.
[8] Blomen E, Hendriks C, Neele F. Capture technologies: improvements and
promising developments. Energy Proc 2009;1:1505–12.
[9] Wall TF. Combustion processes for carbon capture. Proc Combust Inst
2007;31(1):31–47.
[10] Kvamsdal HM, Jordal K, Bolland O. A quantitative comparison of gas turbine
cycles with CO2 capture. Energy 2007;32(1):10–24.
[11] Pennline HW, Luebke DR, Morsi BI, Heintz YJ, Jones KL, Ilconich JB. Carbon
dioxide capture and separation techniques for advanced power generation
point sources. In: 23rd Annual international pittsburgh coal conference,
Pittsburgh, PA, USA. University of Pittsburgh School of Engineering, Pittsburgh,
PA; 2006.
[12] Feron PHM. The potential for improvement of the energy performance of
pulverized coal fired power stations with post-combustion capture of carbon
dioxide. Energy Proc 2009;1:1067–74.
[13] Shimizu T, Hirama T, Hosoda H, Kitano K, Inagaki M, Tejima K. A twin fluid-bed
reactor for removal of CO2 from combustion processes. Chem Eng Res Des
1999;77(1):62–8.
[14] Kunze C, De S, Spliethoff H. A novel IGCC plant with membrane oxygen
separation and carbon capture by carbonation–calcinations loop. Int J Greenh
Gas Control 2011;5(5):1176–83.
[15] Lozza G, Romano M, Giuffrida A. Thermodynamic performance of IGCC with
oxy-combustion CO2 capture. In: 1st International conference on sustainable
fossil fuels for future energy – S4FE2009, Rome, Italy; 2009.
[16] Buhre BJP, Elliott LK, Sheng CD, Gupta RP, Wall TF. Oxy-fuel combustion
technology for coal-fired power generation oxy-fuel combustion technology
for coal-fired power generation. Prog Energy Combust Sci 2005;31:283–307.
[17] Toftegaard MB, Brix J, Jensen PA, Glarborg P, Jensen AD. Oxy-fuel combustion
of solid fuels. Prog Energy Combust Sci 2010;36:581–625.
[18] Carbo MC, Jansen D, Dijkstra JW, Van Buijtenen JP, Verkooijen AHM. Precombustion decarbonisation in IGCC: gas turbine operating window at
variable carbon capture ratios. Energy Proc 2009;1:669–73.
[19] Strube R, Manfrida G. CO2 capture in coal-fired power plants – impact on plant
performance. Int J Greenhouse Gas Control 2011;5:710–26.
[20] Erlach B, Schmidt M, Tsatsaronis G. Comparison of carbon capture IGCC with
pre-combustion decarbonisation and with chemical-looping combustion.
Energy 2011;36. 3804-115.
[21] Robinson PJ, Luyben WL. Integrated gasification combined cycle dynamic
model: H2S absorption/stripping, water-gas shift reactors, and CO2 absorption/
stripping. Ind Eng Chem Res 2010;49:4766–81.
[22] Chiesa P, Consonni S. Shift reactors and physical absorption for low-CO2
emission IGCCs. J Eng Gas Turbin Power 1999;121(2):295–305.
Author's personal copy
290
M. Mansouri Majoumerd et al. / Applied Energy 99 (2012) 280–290
[23] Dennis RA, Shelton WW, Le P. Development of baseline performance values for
turbines in existing IGCC applications. In: ASME paper GT2007-28096. ASME
Turbo Expo, Montreal, Canada; 2007.
[24] Lee JJ, Kim YS, Cha KS, Kim TS, Sohn JL, Joo YJ. Influence of system integration
options on the performance of an integrated gasification combined cycle
power plant. Appl Energy 2009;86:1788–96.
[25] Smith AR, Klosek J. A review of air separation technologies and their
integration with energy conversion processes. Fuel Process Technol
2001;70:115–34.
[26] Harris D, Roberts D, ANLEC R&D scoping study: black coal IGCC, CSIRO, report
no. EP103810; 2010.
[27] Hannemann F, Schiffers U, Karg J, Kanaar M. V94.2 Buggenum experience and
improved concepts for syngas applications. In: GTC conference. San Fransisco,
USA; 2002.
[28] The Shell gasification process for sustainable utilisation of coal, Shell Global
Solutions.
[29] Álvarez-Rodríguez R, Clemente-Jul C. Hot gas desulphurisation with dolomite
sorbent in coal gasification. Fuel 2008;87:3513–21.
[30] Zheng L, Furinsky E. Comparison of shell, Texaco, BGL and KRW gasifiers as
part of IGCC plant computer simulations. Energy Convers Manage 2005;46:
1767–79.
[31] Kohl A, Nielsen R. Gas purification. 5th ed. Houston: Gulf Publishing Company;
1960.
[32] Kanniche M, Bouallou C. CO2 capture study in advanced integrated gasification
combined cycle. Appl Therm Eng 2007;27:2693–702.
[33] EnssimÒ. Enssim Software: Doetinchem, The Netherlands; 2009.
[34] ASPEN Plus version 7.1. Aspen Technology Inc.: Cambridge, MA, USA;
2009.
[35] IPSEpro version 4.0. Simtech Simulation Technology, Simtech, Graz, Austria;
2003.
[36] Ceric H, Slad J, Johnke T. Latest performance upgrade of the Siemens gas
turbine SGT5-4000F. In: Power-Gen Europe, Milan, Italy; 2008.
[37] Kanniche M, Gros-Bonnivard R, Jaud P, Valle-Marcos J, Amann JM, Bouallou C.
Pre-combustion, post-combustion and oxy-combustion in thermal power
plant for CO2 capture. Appl Therm Eng 2010;30:53–62.
[38] Kreutz T, Martelli E, Carbo M, Consonni S, Jansen D. Shell gasifier-based coal
IGCC with CO2 capture: partial water quench vs. novel water-gas shift. In:
ASME paper, GT2010-22859. ASME Turbo Expo, Glasgow, UK; 2010.
[39] Rezaiyan J, Bechtel T, Weisenfeld H, Cheremisinoff N. Assessment of the
commercial potential for small gasification combined cycle and fuel cell
systems phase II-final draft report. HM Associates Inc., Princeton Energy
Resources International (LLC), and TFB Consulting; 2003.
Paper III
Estimation of performance variation of future generation IGCC
with coal quality and gasification process – Simulation results
of EU H2-IGCC project
Mohammad Mansouri Majoumerd, Han Raas, Sudipta De,
Mohsen Assadi
Published in Applied Energy, Vol. 113, p. 452-462, August
2013
147
Author's personal copy
Applied Energy 113 (2014) 452–462
Contents lists available at ScienceDirect
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Estimation of performance variation of future generation IGCC with coal
quality and gasification process – Simulation results of EU H2-IGCC
project
Mohammad Mansouri Majoumerd a,⇑, Han Raas b, Sudipta De c, Mohsen Assadi a
a
b
c
Faculty of Science and Technology, University of Stavanger, 4036 Stavanger, Norway
Asset Development Division, Vattenfall, Westervoortsedijk 73, 6827 AV Arnhem, The Netherlands
Department of Mechanical Engineering, Jadavpur University, Kolkata 700 032, India
h i g h l i g h t s
Coal quality effects on the performance of four commercial gasifiers are reported.
The overall IGCC performance indicators using various coals are presented.
Dry-fed gasifiers are relatively insensitive to the coal quality.
Slurry-fed gasifiers are not suitable for the gasification of low-rank coals.
a r t i c l e
i n f o
Article history:
Received 14 January 2013
Received in revised form 17 June 2013
Accepted 21 July 2013
Available online 23 August 2013
Keywords:
IGCC
Gasification
Dry-fed
Slurry-fed
Coal quality
Performance
a b s t r a c t
The integrated gasification combined cycle (IGCC) power plant delivers environmentally benign power
from coal. The overall objective of the European Union (EU)’s H2-IGCC project is to develop and demonstrate technological solutions for future generation IGCC plants with carbon capture. As a part of the general goal, this study evaluates the effects of coal quality and the selection of gasifiers on the overall
performance of the baseline configuration of the IGCC plant. Four commercially available gasifiers i.e.,
Shell, GE, Siemens, and ConocoPhillips gasifiers are considered for this comparative study. The effects
of three different types of coals on the gasification processes have been investigated, as well as the overall
performance of the plant. Simulation results show that slurry-fed gasifiers are not suitable for lignite coal,
while dry-fed gasifiers are less sensitive to coal quality. Coal quality has the greatest effect on the GE gasifier. The ConocoPhillips gasifier demonstrates the highest cold gas efficiency using bituminous coal. The
coal rank and the gasification process have relatively less effect on gas turbine power output, while steam
turbine power output varies significantly with these. Although steam turbine power output increases
with a reduction in coal quality, especially for slurry-fed gasifiers, the air separation unit power demand
offsets this increase. The highest overall plant efficiency is 37.6% (LHV basis) for the GE gasifier and coal B.
The lowest overall efficiency penalty with coal quality is 5% (LHV basis) for the Shell gasifier with input
changed from bituminous to lignite. Moreover, simulation results show that GE’s gasification technology
has the highest CO2 emissions for lignite coal, i.e. 158 g/kWh.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
The rapid growth of industry and population coupled with improved living standards has led to an ever-increasing demand on
world energy, more specifically for electric power. Fossil fuels,
mostly coal, have been catering to most of this demand for some
decades. However, the ‘climate change’ problem [1] has forced
the development of new technologies for fossil fuel based power
and utility heat supply. According to the International Energy
⇑ Corresponding author. Tel.: +47 45391926; fax: +47 51 83 10 50.
E-mail address: [email protected] (M. Mansouri Majoumerd).
0306-2619/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.apenergy.2013.07.051
Agency (IEA), CO2 emissions from the electricity and heat supply
sector amounted to about 41% of total global CO2 emissions from
fossil fuels in the year 2010 [2]. However, the IEA’s New Policies
Scenario suggests a 25% increase in coal consumption in the year
2035 compared to the 2009 level. This increase will be 65% based
on the current policies scenario [3]. One of the key players within
these policies is ‘Carbon Capture and Sequestration’ (CCS) according to the European Energy Roadmap 2050 [4]. The deployment of
CCS in coal-fired power generation will ensure the growing share
of the coal consumption among other fossil fuels in the coming
years with more restricted emissions regulations.
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
453
Nomenclature
AGR
Ar
ASU
C
CCS
CCU
CH4
CMD
CO
COS
CO2
EU
GE
GT
HHV
HP
HRSG
acid gas removal
argon
air separation unit
carbon
carbon capture and sequestration
combined cycle unit
methane
coal milling and drying
carbon monoxide
carbonyl sulfide
carbon dioxide
European Union
General Electric
gas turbine
higher heating value
high pressure
heat recovery steam generator
Integrated gasification combined cycle (IGCC) has been identified as an attractive coal technology which offers less expensive
pre-combustion carbon capture than conventional plants with
post-combustion capture. Moreover, the IGCC technology provides
opportunities for the production of steam, H2, and synthetic chemicals such as Fischer–Tropsch fuels [5]. Effort is devoted to develop
IGCC technology as a possible good alternative power using coal
but lower CO2 emissions. Thermodynamic estimation of possible
improvements of the performance of an IGCC plant with an entrained-flow, dry-fed, oxygen-blown gasifier with hot gas desulfurization in comparison with conventional process has been
discussed by Giuffrida et.al. [6]. They also reported thermodynamic
performance estimation vis-à-vis comparison through simulation
between air-blown and oxygen-blown gasifiers for IGCC applications [7]. Cost effectiveness of power from IGCC plants with respect
to conventional coal fired plants and nuclear power plants against
the backdrop of penalty of CO2 emission has been exhaustively explored [8]. Results show relative advantages and disadvantages
depending on several conditions of operation as well as regulations. With rapid growth of renewable energy, future coal based
power plants may have to accommodate wide range of ratio of
renewable and non-renewable power mix with efficient use of excess power for some other utility using IGCC power plants. Simulation results of an IGCC plant integrated with CaO based CO2
absorption using Shell and Texaco gasifiers are analyzed to estimate the expected performance [9]. It was noted that though plant
efficiency may be low (30–33%), CO2 capture may improve (97%).
Two innovative options for reducing the penalty in efficiency of
shell coal IGCC plants during CO2 capture are simulated and optimized by Martelli et al. [10].
With a distinct view towards the development of IGCC technology, the European Union has sponsored a well-integrated and coordinated research endeavor under its FP7 Framework Research
Program. Fifteen academic and nine industrial partners of the EU
are involved in this H2-IGCC project to develop and demonstrate
a future generation IGCC plant with pre-combustion CO2 capture
[11]. Four different sub-groups are working in well-defined and
coherent work packages towards this common goal. The simulation sub-group is working not only to achieve optimization
through detailed system analysis at different stages of development but also to ensure a realistic techno-economic evaluation.
Gasification plays the key role in an IGCC plant [12,13]. It not
only makes it possible to use coal or other solid fuels in an efficient
H2
H2O
H2S
IEA
IGCC
IP
LHV
NOx
N2
O2
PFD
SCGP
SFG
SOx
ST
SWGS
hydrogen
water
hydrogen sulfide
international energy agency
integrated gasification combined cycle
intermediate pressure
lower heating value
nitrogen oxides
nitrogen
oxygen
process flow diagram
Shell Coal Gasification Process
Siemens Fuel Gasification
sulfur oxides
steam turbine
sour water–gas shift
combined power cycle but also provides opportunities to capture
most of the pollutants, including CO2, efficiently. Employing proper
gasification technology is essential for optimizing the operation of
a future generation IGCC plant. Moreover, gasification performance
is significantly affected by coal quality. This quality varies widely
depending on the geographical location of coal source [14]. High
ash, sulfur, chlorine, alkali metals etc. in addition to low heat value
and ash melting point are some typical characteristics of low quality coals [15]. Unfortunately, about 53% of global coal reserves are
of low rank, i.e. sub-bituminous and lignite [16]. Thus to explore a
useful real-life future generation IGCC plant, the effects of the gasification process, as well as that of coal quality, on the performance
of the plant must be investigated. Several previous studies [17–21]
have reported on IGCC plants using bituminous coals without any
reference to low rank coals.
In an effort to evolve the optimum IGCC plant configuration
based on data provided by other sub-groups of the H2-IGCC project, the simulation sub-group previously reported detailed simulation results for a baseline configuration of the IGCC plant as
developed in this project [22]. In this paper, subsequent simulation
studies on the effects of the gasification process as well as of coal
quality on the performance of that baseline configuration are reported. Comparative performance evaluations of the gasification
process of three different types of coal, viz. bituminous, sub-bituminous and lignite in four different commercially available gasifiers are reported in this paper. Two coals typically represent low
rank coal and the results for these are compared with that for bituminous coal as reference. Subsequently, the performance evaluation is extended to the whole plant using the same coals and
gasification technologies. The results of this work show some distinct implications for the optimum configuration of the future IGCC
plant that may evolve through subsequent studies of different subgroups of the project.
2. Gasification and gasifiers for power generation
Conventional gasification is the process of conversion of a solid
or liquid through sub-stoichiometric reaction with oxidants, either
air or O2 at a temperature exceeding 700 °C to produce a synthetic
gaseous product [23]. Compared to conventional pulverized coal
firing, gasification offers great opportunities for both higher efficiency and improved capture of pollutants. According to the flow
geometry, the commercial gasification technologies can be classi-
Author's personal copy
454
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
fied into three categories, viz. entrained-flow, fluidized bed, and
moving bed gasification technologies [24]. However, the need of
modern power plants for large capacity gasifiers demands the
shortest residence time and hence entrained-flow gasifiers are
the most favored for this purpose. Entrained-flow gasifiers allow
high operating pressures (20–80 bar) and temperatures (1200–
1600 °C). High operating temperatures enable a favorable slagging
process to remove ash and render gasification almost tar-free.
These conditions are very desirable for large-scale power generation. Hence, almost all the commercially useful coal gasifiers deployed for large-scale power generation are of the entrained-flow
type.
Although entrained-flow gasifier can process many varieties of
feedstock e.g., high/intermediate/low quality coals, and heavy liquid fuels [12,25], the feedstock characteristics significantly influence the gasification performance and consequently the
performance of the whole IGCC plant [12,24,26]. The existing gasifiers show a substantial increase in cost combined with a drastic
reduction in performance when operating on low-rank feedstock
e.g., lignite coals [19]. Nevertheless, the utilization of such types
of coals can improve flexibility of supply and consequently improve the security of the energy supply [15]. The main parameters
for the selection of coal type for IGCC plant are ash content, slag
viscosity, and coal reactivity.
The feedstock can be fed either wet (using slurry water) or dry
(using N2 as a conveying gas) into the entrained-flow gasifier. The
high pressure and temperature environment of the gasifier facilitates the gasification of the fed coal [27]. The released heat results
in the melting of the ash content and the production of molten, inert slag. Meanwhile, the carbon content in the coal is converted
mainly to CO and H2 rather than to the normal products of combustion, CO2 and H2O, due to the reducing environment of the gasifier. The content of these products in the syngas depends heavily
on the gasifier lambda value (the fraction of stoichiometric O2 demand of the reactor when conceived as a partial oxidation burner).
Due to the high operating pressure and other advantages of the
entrained-flow regime, this type of gasifier provides a high H2/CO
ratio syngas and an inert slag. In the case of high rank coal, steam
is added to the gasifier to moderate the temperature of the process
while maintaining good carbon conversion. Under the extremely
hot condition, coal devolatilizes (pyrolysis) almost instantaneously
into gaseous elements like CH4, aromatics, CO2, CO, and solid char
residue (C).The volatiles formed are immediately combusted with
the supplied oxygen leading to an enormous rise in temperature.
1
Volatiles þ x þ y O2 ! xCO2 þ yH2 O
2
ð1Þ
Depending on the oxidant factor of the gasification process,
either the volatiles consume all O2 in the equation or some excess
O2 is left. In the case of there being no oxygen left after combustion
of the volatiles, the following endothermic gasification reactions
(Eqs. (2)–(4)) of residual char with the oxidants CO2 and H2O (so
called moderating agents) take place.
C þ H2 O $ CO þ H2
ð2Þ
C þ 2H2 O $ CO2 þ 2H2
ð3Þ
C þ CO2 $ 2CO
ð4Þ
These moderation reactions are relatively slow compared to the
devolatilization reaction (Eq. (1)) and cause a temperature drop. In
the case where there is excess O2 after combustion of the volatiles,
some of the residual char will be completely (Eq. (5)) or partially
combusted (Eq. (6)) and release heat prior to moderation. These
reactions (i.e. Eqs. (5) and (6)) are exothermic irreversible
reactions.
C þ O2 ! CO2
ð5Þ
1
C þ O2 ! CO
2
ð6Þ
Another important reaction is the gasification’s CO-shift reaction (Eq. (7)), which is an exothermic reaction to convert CO to CO2.
CO þ H2 O $ CO2 þ H2
ð7Þ
The methane content of the produced syngas is increased
through reactions (8) and (9) [23]. These are exothermic methanation reactions (below); however, they are more prevalent in gasifiers operating at lower temperatures. A higher gasifier pressure also
increases the methane content. In addition to the mentioned reactions, most of the sulfur content in the coal is converted to H2S,
with a small fraction being converted to COS. Moreover, nitrogen
is converted to ammonia in the reducing environment of the gasifier. This ammonia also breaks down into N2 and H2 in the high
temperature environment. A small amount of hydrogen cyanide
is also produced.
CO þ 3H2 $ CH4 þ H2 O
ð8Þ
C þ 2H2 $ CH4
ð9Þ
2.1. Gasifiers assumed for this simulation
The gasification of coal has gained special significance in the
context of future generation IGCC plants. Several industrial research groups are developing coal gasifiers, specifically those of
the entrained-flow type, on a commercial scale. Some of the leading companies in the power sector have patented their technologies in this field. To assess the significance of the process of
gasification on the performance of future generation IGCC plants,
four common commercially-matured gasifiers with known specifications have been used for this simulation. These are Shell Coal
Gasification Process (SCGP), General Electric (GE) gasifier (formerly
Texaco), Siemens Fuel Gasification (SFG), and ConocoPhillips (EGas™) gasifier. The main characteristics of these gasification technologies are shown in Table 1. All of these gasifiers are oxygenblown and entrained-flow type, each having some specific features
as discussed below.
2.1.1. Shell Coal Gasification Process (SGCP technology)
The SCGP gasifier typically operates at around 45 bar, with a
temperature range of 1400–1600 °C, well above the ash melting
point to ensure that molten ash has a low viscosity to flow easily
out of the gasifier [28,29]. Since it has a dry-fed system, no water
must be evaporated in the gasifier leading to high cold gas efficiencies compared to (single stage) slurry-fed entrained flow gasifiers
[29]. Coal is pulverized and dried to 2% residual moisture in a roller
mill system featuring a hot drying gas recycle loop. Drying heat is
supplied by an in-line burner in a hot drying gas recirculation loop,
burning a small fraction of the cleaned syngas flow downstream
the syngas desulphurization unit (1–2%). The dry, pulverized coal
is subsequently pressurized using a lock-hopper system. The pressurized coal is pneumatically fed to the gasifier in dense phase
mode using pure nitrogen as a conveying gas. Gasifier lambda value is low (0.30–0.33) which, at least for hard coal gasification,
would lead to both unacceptably high gasifier working temperatures and low carbon conversion. To solve this problem, intermediate pressure (IP) moderation steam is admixed to the gasifier
oxygen feed in a ratio of typically 0.1 (kg steam/kg coal dust feed).
When gasifying lignite, lambda values are high such that external
moderation steam supply is not required. The raw syngas leaving
the gasifier is first rapidly cooled to around 900 °C by recycling
Author's personal copy
455
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
Table 1
The main characteristics of investigated gasifiers.
Specification
Shell Coal Gasification
Process (SCGP)
GE (formerly Texaco)
Siemens fuel gasification (SFG)
Conoco-Philips (E-Gas™)
Flow regime
Type of ash
Oxidant
Dry/slurry
Feed type
Pressurization
Entrained-flow
Slag
O2-blown
Dry-fed
Pulverized coal
Lock hopper (pneumatic
feeding)
Single
Lock-hopper (bottom)
Entrained-flow
Slag
O2-blown
Slurry-fed
Pulverized coal
Slurry pump
Entrained-flow
Slag
O2-blown
Dry-fed
Pulverized coal
Pneumatic feeding
Entrained-flow
Slag
O2-blown
Slurry-fed
Pulverized coal
Slurry pump
Single
Lock-hopper (bottom)
Single
Lock-hopper (bottom)
Double
Continuous pressure let-down
system (bottom) [23]
Upward flow
Side-fired
Quenching with recycle gas
and radiant cooler
Membrane-wall type
reactor [29]
78–83% [29]
Downward flow
Top-fired
Full water quench, radiant cooler, and
radiant/convective coolers
Refractory-lined reactor
Downward flow
Top-fired
Built-in full water quench
Upward flow
Side-fired
Two-stage gasification
Refractory-lined reactor
70–75%
Both membrane-wall type reactor and
refractory-lined reactor [31]
75–80% [15]
78–83%
Above 99% [29]
Above 96%
Above 98% [31]
Above 99% [35]
Tar free
92%/96% [36]
Tar free
88–90% (availability) [19]
Tar free
90%/94%[32]
Tar free
92% (availability)
Number of stages
Slag removal
system
(position)
Flow direction
Boiler position
Quenching type
Reactor type
Cold gas
efficiencya
Carbon
conversion
Tar formation
Availability/
Reliability
targets
a
The definition of cold gas efficiency is presented in Section 5.
cooled, ash-free syngas. This cooling process is called dry quenching. The purpose of this type of quenching is to solidify the slag
particles and it is essential to reduce the fouling risk in the downstream syngas cooler. The syngas is then further cooled to 340 °C
while generating high pressure (HP) and IP steam. Downstream
the syngas cooler, the syngas is thoroughly dedusted in a filter system consisting of a cyclone and a ceramic candle filter in series.
Part of the dedusted syngas (<10 mg/N m3 residual dust) is recycled to the dry quench section. The rest of the syngas is sent to a
wet scrubber upstream of the acid gas removal (AGR) unit for removal of halogens (Cl and F compounds), trace elements, and fine
particulate matter. The produced, fully dedusted and dehalogenized syngas is then sent to downstream sub-systems in the IGCC
plant. There is currently one IGCC plant in operation using SCGP
technology: the Vattenfall Buggenum IGCC plant in the
Netherlands.
2.1.2. General Electric gasifier
Similar to the SCGP, the GE gasifier uses pulverized coal. However, in this case pulverized coal is mixed with water to produce a
slurry feed. The typical range of slurry (ratio of solid to whole mixture) varies from 35 to 70 wt% depending on the coal’s characteristics [25,28,30]. The slurry type of gasifiers utilize a slurry pump to
feed the slurry into the gasifier enabling the process to have a higher operating pressure compared to dry-fed systems (up to 70 bar).
Lambda value is relatively high (0.40) caused by the fact that some
CO and H2 burning is required to vaporize the slurry water. The
syngas therefore shows relatively high contents of the products
of combustion (i.e. CO2 and H2O). The main disadvantage of this
technology is the limited lifetime of the refractory and the associated cost due to refractory replacement. Therefore, such plants are
designed with a spare gasifier to enable them to achieve 90% target
availability [19]. Contrary to shell process, slag and syngas leave
the gasifier co-currently at 1260–1480 °C.
Application of a dry quench system is not possible due to the
large amount of slag that should be solidified. Instead, a radiant
syngas cooler is applied followed by a convective syngas cooler.
Both syngas coolers raise saturated HP steam. In between both
types of syngas coolers, the solidified slag is separated using a
water quench bath for further quenching. Finally, the solidified slag
is sluiced out of the gasification system. Syngas leaving the convective syngas cooler then enters the scrubber before being sent to the
AGR unit. In the current study, this type of cooling is used in
simulations.
The type of syngas cooling described above is just one way of
cooling for GE gasifier produced syngas. Often GE type gasification
systems are designed based on a so-called wet quench system
where syngas and slag leaving the gasifier are directly quenched
in a wet quenching system. The wet quench design consists of a
large water pool that cools the syngas and removes slag and ash
particles. The quenched raw syngas then enters a wet scrubber.
The dry syngas cooling type of heat recovery results in a higher
overall plant efficiency and steam turbine (ST) power output compared to the wet quench design. However, a gasification system
based on a radiant/convective cooler is considerably more expensive than a wet quench design (by a factor of two).
The higher operating pressure compared to dry-fed gasification
system results in a smaller sized CO2 removal system and reduces
the corresponding cost. The water content of the syngas leaving the
wet scrubber is relatively high compared to dry-fed gasifiers. The
produced syngas needs a certain level of water content to carry
out the CO-shift conversion at the sour water–gas shift (SWGS)
unit and this level is controlled by the steam extraction from the
heat recovery steam generator (HRSG) of the combined cycle unit
(CCU). This extraction is, therefore, lower compared to dry-fed gasifiers [19].
There is presently one IGCC plant using the GE gasifier: the
Tampa Electric Polk IGCC power station in the USA.
2.1.3. Siemens gasifier (SFG technology)
Similar to the SCGP, the SFG technology features a dry-fed system which results in high cold gas efficiency [15]. Coal milling and
drying and pulverized coal dry-fed systems are similar to the one
described for the SGCP process. This technology is commercially
Author's personal copy
456
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
available with both reactor types, refractory-lined for lower ash
content (<2%) and membrane-wall for higher ash content (>2%)
[31]. The gasification temperature range is 1450–1750 °C [32]
and the operating pressure is approximately 40 bar [33]. The SFG
utilizes full wet quenching which results in a simple process compared to radiant or convective coolers and ensures optimal conditions for CO-shift conversion in the subsequent process unit (i.e.
SWGS). In the current study, the quenching process is based on
the direct wet quench type.
The cooled and saturated raw syngas leaves the gasification island for the SWGS unit at a temperature of 220–230 °C [15].
Though there is currently no full-scale SFG for IGCC application,
the power market is increasingly showing interest in this type of
gasifier [31].
(1) Air separation unit (ASU): the cryogenic ASU is a stand-alone
unit generating O2 (95% purity) from air supplied by an
intercooled main air compressor for the gasification of coal.
This compressor is not integrated with the gas turbine of the
CCU.
(2) Gasification island and syngas cooling and scrubbing: the
gasification of the coal takes place in various O2-blown,
entrained-flow gasifiers using technologies described in Section 2.
(3) Sour water–gas shift (SWGS) reaction unit: the SWGS process is the reaction used to convert the CO in the raw syngas
to CO2 by shifting the CO with water over a catalytic bed
(usually alumina supported cobalt molybdate) according to
the following equation:
MJ
2.1.4. ConocoPhillips (E-Gas™) gasifier
Similar to the GE gasifier, this E-Gas™ gasifier is a slurry-fed
type [19]. However, this type of gasification has two stages of gasification and incorporates a proprietary slag removal system, char
recycle and syngas cooling schemes. Both gasification stages are
fed with coal slurry and are provided with refractory inner walls.
The operating pressure is around 40 bar [34]. Operating conditions
of the first stage gasifier resemble the one for the GE gasifier. However, instead of co-current flow of slag and raw syngas in GE gasifier, the major part of the slag is kept separated from the raw
syngas flow by letting the slag flow to be quenched in a water bath
located beneath the gasifier. In the second stage of gasifier, only a
very little amount of oxygen is applied. Most of the gasification of
the entering slurry is accomplished by the moderating agents, CO2
and H2O, present in relatively high concentration in the entering
syngas from the first stage of the gasifier. Since gasification using
CO2 and H2O as the main gasifying agents is endothermic, the gasification of the slurry feed results in a large syngas temperature
drop (from >1300 to 900 °C). This temperature drop also causes
fly ash particles still present in syngas flow from the first stage gasifier to solidify, alike in the dry quench section of the SGCP. Therefore, this second stage gasifier is sometimes denoted as chemical
quench. The chemical quench results in a lower overall lambda value and an improvement of cold gas efficiency. On the other hand,
it reduces the available sensible heat transfer in the downstream
syngas cooler [23]. Therefore, a radiant cooler is no longer needed.
The second stage gasifier is less effective in carbon conversion than
the first one. Therefore, a lower carbon conversion is assumed for
the second stage. A filter downstream the convective syngas cooler
removes the ash particles and unconverted carbon. Because of its
high carbon content, the separated ash flow is recycled to the first
gasification stage in order to be gasified.
There is currently one IGCC plant using the E-Gas™ gasifier: the
Wabash River IGCC plant in the USA.
3. IGCC plant for this simulation
The impact of coal quality on the gasification process has been
investigated for the four assumed gasifiers. However, the principal
objective of the simulation sub-group of the H2-IGCC project is to
explore optimized configuration of the plant through simulation
and using data available in open literature or from other group
partners. Hence, the effects of coal quality and the gasification process on the performance of the baseline configuration of the plant,
as reported in [22], are also investigated subsequently. The schematic of the IGCC configuration using the GE gasifier is shown in
Fig. 1. Details of this scheme (except the gasification unit) may
be obtained from [22]. However, a brief outline of the scheme is
also presented here. The plant consists of seven major sub-systems
as discussed below:
44kmol
COðgÞ þ H2 OðgÞ $ CO2ðgÞ þ H2ðgÞ
ð10Þ
(4) Acid gas removal (AGR) unit: a two-stage SELEXOL system
for H2S removal and CO2 capture is used. Due to the high
partial pressure of acid gases, physical absorption of H2S
and CO2 is preferred to chemical, amine-based absorption
processes.
(5) CO2 compression and dehydration unit: the CO2 captured
from the process (90% capture rate) is compressed by an
intercooled compressor, aftercoooled, liquefied and finally
pumped up to a final pressure (150 bar). In order to reduce
the corrosion risk in the transport pipeline, a dehydration
unit using tri-ethylene glycol is considered (H2O water content in the captured CO2 line is less than 20 mg/kg).
(6) Gas turbine (GT): the GT block including compression, combustion, and expansion generates electric power using a
generator. Simulation has been performed using characteristics and boundary conditions of a gas turbine which is
designed for combustion of the H2-rich fuel produced from
sub-systems 1–4.
(7) Heat recovery steam generator (HRSG) and steam cycle:
downstream of the GT is a triple pressure HRSG (140 bar/
530/530 °C) and steam turbine to generate steam and power.
4. Simulation tools and methods
In this study the performance of various gasification technologies (refer to Section 2) using three different types of coal have
been investigated. To obtain reliable results, the following software
tools based on their specific capabilities were utilized for simulation of the entire IGCC plant:
Enssim: simulation tool developed by Enssim Software [37];
ASPEN Plus: commercial process engineering software developed by AspenTech [38]; and
IPSEpro: commercial heat and mass balance program developed
by SimTech [39].
Data exchange between software tools was performed manually to find the optimal match. The simulation of all sub-systems
(refer to Section 3) except gasification block was performed using
commercial tools, i.e. ASPEN Plus and IPSEpro. The relevant one
of these two software tools was used as described below:
The ASU and SWGS reaction were modeled using ASPEN Plus.
The Peng-Robinson properties method was selected as the
equation-of-state (PR EOS).
The AGR unit was modeled in ASPEN Plus with Perturbed-Chain
Statistical Associating Fluid Theory (PC-SAFT) as the equationof-state.
Author's personal copy
457
IP Steam
Feed water
HP steam
Water
Coal
Make-up
water
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
SWGS
Particulate
removal
Slurry tank
CO+H2O→ CO2 +H2
Convective cooler
GE
Particulates
Syngas
Feed water
(CO, H 2)
Slurry pump
H2 -rich syngas
SELEXOL
HP steam
ASU
Air
AGR, CO2
capture,
compression
&
dehydration
CO 2 to storage
O2
Slag
Generator
Sulfur by-product
Air
HRSG
Stack
Generator
Fig. 1. The schematic configuration of the IGCC power plant with carbon capture (using GE gasifier).
The compression of captured CO2 and dehydration of CO2
stream were modeled in ASPEN Plus using PR EOS, and
Schwarzentruber and Renon (SR polar) equation-of-state,
respectively.
The power block including the GT, and the triple-pressure
steam cycle were modeled in IPSEpro.
The simulation of the gasification block was carried out using
the Enssim tool. All simulations were static design calculations
for all equipment. Gasification modeling started from a purely
thermodynamic analysis of the process. Gasification kinetics was
not included in the calculations. Nevertheless, chemical equilibrium is attained for the most important equilibrium reactions
due to the high operating temperatures at entrained flow gasifiers.
Non-equilibrium conditions are taken into account by temperature
differences between actual and equilibrium temperatures for the
various simultaneous gasification reactions. Realistic design data
and data for chemical equilibrium deviations for the Shell gasifier
have been deducted from the design and operational data of the
well-known Nuon (Buggenum) IGCC plant. It is worth noting that
the dry-fed gasification model (based on the Shell technology)
using this in-house tool was validated against real plant operational data and design data for the Magnum plant gasifier (refer
to Table 3). In Tables 2 and 3 below, common input variable data
for validation of the model and a comparison of results provided
by the gasification technology licensor and the Enssim tool are
given.
As seen in Table 3, calculated syngas compositions are quite
close. There is a relatively small difference in calculated GOX to
dry coal dust ratio. The relative difference in calculated gasifier
cold gas efficiency is even smaller. Therefore, Enssim tool is
Table 2
Common input variable data for validation of the Shell gasification model.
Input data
Value
Remarks
Coal designation
Moisture content dried
coal (kg/kg)
Dry coal dust feed
temperature (°C)
Coal dust conveying N2
ratio (kg/kg)
Drayton
0.02
Bituminous Australian coal
O2 purity (mole%)
O2 temperature (°C)
Moderator steam/
burner feed ratio
(kg/kg)
Gasification
temperature (°C)
Relative heat transfer to
membrane wall (%)
Carbon conversion (%)
80
0.07
0.07 kg N2 needed to transport 1 kg of
dry coal dust (in pneumatic dense
phase mode)
99.5
200
0.0769
1650
2.0
As a percentage of gasifier coal feed
thermal flow (LHV based)
99.3
regarded as a reliable tool for fitting gasification data provided
by the licensor.
The simulation of SCGP using this tool is given here as an example. For the SGCP process the Enssim tool was used to analyze the
subset of conceptual process flow diagram (PFD) consisting of coal
milling and drying section (CMD), the dry pulverized coal feeding
section and the gasification section. The latter section consisted
of gasifier, dry quench section, syngas cooler section, dedusting
section, and wet scrubber section.
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
Table 3
Comparison between gasification technology licensor and Enssim results for gasifier
calculation.
Quantity
Licensor result
Gasifier syngas composition (mole fraction)
H2O
0.0205
H2
0.2851
CO
0.6327
0.0107
CO2
CH4
118a
H2S
0.00313
N2
0.0462
Ar
885a
HCl
99.5a
NH3
120a
COS
343a
CS2
–
S2
–
HCN
120a
C2H4
–
GOX to dry coal dust ratio (kg/kg)
0.798
Gasifier cold gas efficiency (%)
79.8
a
Enssim result
0.0201
0.2802
0.6350
0.0112
101a
0.00317
0.0488
834a
99a
102a
244a
2a
5a
110a
9a
0.754
80.4
The reported values are based on ppmv.
Calculations of the PFD for CMD included secant iteration blocks
for:
Drying gas purge temperature, O2 content and relative
humidity.
Blending coal with fluxing agent in the CMD such that the
approach to the flow point of the resulting slag in the gasifier
(100 K) was maintained.
Calculations of the gasifier were carried out using the following
input design parameters:
gasifier working temperature and pressure,
carbon conversion,
relative (to pulverized coal thermal input) heat transfer of the
membrane wall or refractory wall.
The chemical gasifier calculations first start with a calculation
of slag, ash composition, and a first estimate of syngas composition. It also takes into account the amount of CO2 resulting from
calcination of limestone or dolomite if these minerals were added
as fluxing agents in the coal milling and drying unit. Subsequently,
Table 4
Composition and thermal properties of investigated coals.
Coal sample
code
Coal A
(Bituminous)
Coal B (Subbituminous)
Coal C
(Lignite)
Proximate analysis (wt%, dry basis)
Moisture
10
Ash
12.50
Volatile matter
27.00
Fixed carbon
50.50
LHV (kJ/kg)
25,100
HHV (kJ/kg)
26,195
27.40
4.50
31.40
36.70
19,691
20,469
31.24
17.92
28.08
22.76
14,127
14,682
Ultimate analysis (wt%, a.r.)
C
64.10
H
5.02
N
0.70
O
16.09
S
1.50
Cl
0.09
50.25
3.41
0.65
13.55
0.22
0.02
36.27
2.42
0.71
10.76
0.64
0.04
Main ash composition (wt%)
SiO2
55.00
Al2O3
24.00
Fe2O3
5.50
CaO
4.50
33.40
16.30
5.20
21.50
56.96
19.01
3.49
8.39
Table 5
Technical variables for various gasification technologies.
Coal sample code
Coal A
Coal B
Coal C
Shell Coal Gasification Process (SCGP)
Operating temperature (°C)
Operating pressure (bar)
Specific O2 demand (kg/kg coal a.r.)
Specific moderator steam demand (kg/kg coal a.r.)
Specific N2 demand (kg/kg coal a.r.)
Carbon conversion (%)
1550
45.0
0.773
0.060
0.232
99.3
1550
45.0
0.549
0.000
0.186
99.3
1550
45.0
0.419
0.000
0.170
99.3
GE gasifier
Operating temperature (°C)
Operating pressure (bar)
Slurry solid contents (wt%)
Specific O2 demand (kg/kg coal a.r.)
Carbon conversion (%)
1450
60.0
64.5
0.881
99.0
1450
60.0
56.0
0.708
99.0
1450
60.0
45.0
0.721
99.0
Siemens fuel gasifier (SFG)
Operating temperature (°C)
Operating pressure (bar)
Specific O2 demand (kg/kg coal a.r.)
Specific moderator steam demand (kg/kg coal a.r.)
Specific N2 demand (kg/kg coal a.r.)
Carbon conversion (%)
1550
45.0
0.770
0.060
0.160
99.0
1550
45.0
0.554
0.039
0.132
99.0
1550
45.0
0.418
0.000
0.126
99.0
ConocoPhillips (E-Gas™) gasifier
Operating temperature (°C) 1st stage
Operating temperature (°C) 2nd stage
Operating pressure (bar)
Slurry solid contents (wt%)
Specific O2 demand (kg/kg coal a.r.)
Carbon conversion 1st stage (%)
Carbon conversion 2nd stage (%)
1450
991
43.0
64.5
0.712
99.0
95.0
1450
991
43.0
56.0
0.548
99.0
95.0
1450
991
43.0
45.0
0.586
99.0
95.0
sixteen simultaneous homogeneous equilibrium reactions also
including sulfurous and nitrogenous compounds are used to equilibrate the initial syngas composition. These equilibrium calculations are carried out in a secant iteration block which determines
the gasifier lambda value to arrive at the design gasifier heat transfer for the given gasifier design operating temperature. For a number of reactions, non-equilibrium situations were simulated using
approaches to the equivalent equilibrium temperature. Following
the gasifier computations all equipment downstream the gasifier
is analyzed. In the end, the code outputs the syngas composition
at the outlet of the wet scrubber and outputs a list of ratios of
the quantities demand mass flow rates, required and produced
heat rates, power demands, production mass flow rates, and produced power and the quantity fuel flow rate (as received) to the
CMD.
To investigate the effect of coal characteristics on the performance of gasifiers as well as the power plant, three coals with various characteristics including bituminous, sub-bituminous and
lignite coals are considered in this work. It is evident from Table 4
that these coals differ significantly in moisture, ash content, and
heating value.
As previously mentioned, four different gasification technologies such as SCGP, GE, SFG, and E-Gas™ were selected to assess
Cold gas efficiency (%)
458
90.0
80.0
SCGP
70.0
GE
60.0
SFG
50.0
E-GAS
40.0
30.0
Coal A
Coal B
Coal C
Coal type
Fig. 2. Effects of coal rank on cold gas efficiencies.
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
the gasification performance for different coals for the IGCC plant
with CO2 capture. The operating conditions and technical assumptions used for simulation of investigated gasification technologies
are given in Table 5.
It is worth noting that the GT has been recently optimized for a
hydrogen-rich fuel produced by the Shell Coal Gasification Process
under the H2-IGCC project’s framework. Hence, GT characteristics
(e.g. turbine inlet temperature, compressor and expander’s isentropic efficiencies, combustion efficiency, compressor map, etc.) for all
aforementioned gasification technologies are the same as the optimized GT for the Shell system. Thus, the fuel mass flow rates (for
the GT) and consequently pressure ratios vary depending on the
fuel composition produced by each gasification technology.
As seen in Table 5, the operating pressure of the GE gasifier is
significantly higher than the one of the other gasifiers. To maintain
the GT fuel at a pressure level similar to that of the other IGCC
plants using different gasifiers, an expander downstream of the
AGR unit has been assumed for the IGCC plant using the GE
gasifier.
5. Results and discussion
Based on the objective of this work, results obtained from simulation as described above are presented in this section in different
subsections to describe the effects of coal quality and type of gasifiers on the process of gasification as well as possible overall performance of the baseline configuration of the plant [22].
5.1. Effects of coal quality on gasification
Coal properties that have significant effects on the gasification
process are mostly ash content, slag viscosity, and coal reactivity.
A coal with low ash content is favorable for the IGCC power plant
since it produces smaller amounts of fly ash and bottom slag. It reduces the risk of the plugging of exit pipes and the fouling of downstream heat transfer surfaces [12]. Moreover, it also reduces coal
feed for the same amount of produced gas. The slag viscosity directly determines the operating conditions of a gasifier. Although
the calculation of slag viscosity for the three coals assumed in this
simulation is a part of the gasification model, the detailed investigation of this topic is outside the scope of the present work. In this
regard, the slag viscosity is set to 25 Pa.s for the calculation of the
amount of fluxing agent which, depending on the ash composition
of the coal to be gasified, is either basic limestone or refractory,
acidic fly ash. The amount of oxidant agent is directly influenced
by the gasification temperature which is determined by coal reactivity. To evaluate the effects of coal quality on the gasification process, two principal characteristics viz., cold gas efficiency and the
properties of produced syngas including its composition have been
investigated.
5.1.1. Coal quality and cold gas efficiency of gasifiers
One of the main parameters used to describe gasifier performance is cold gas efficiency. This parameter indicates how much
of the energy input has been recovered as chemical energy in syngas [12]. The gasification efficiency (cold gas efficiency) is defined
as:
ggasifier ¼
LHV sg Q sg
LHV f Rf
ð11Þ
where ggasifier is the cold gas efficiency of gasification (%), LHVsg is
the lower heating value of the syngas (kJ/m3), Qsg is the volumetric
flowrate of the syngas (m3/s), LHVf is the lower heating value of the
coal input (kJ/kg), and Rf is the gasifier coal consumption rate (kg/s).
459
The comparison between cold gas efficiencies for the investigated gasification technologies using three different coals is shown
in Fig. 2.
According to Fig. 2, the coal quality significantly influences the
gasification efficiencies of slurry-fed gasifiers i.e. GE and E-Gas™
gasifiers. Amongst slurry-fed gasifiers, the coal quality has the
greatest impact on the GE gasifier. The cold gas efficiency of the
GE gasifier with lignite coal is 29% lower than that of the same gasifier with bituminous coal. However, it is noted from Fig. 2 that utilization of the second-stage gasification in the E-Gas™ gasifier
resulted in higher cold gas efficiency compared to the other slurry-fed gasifier, i.e. GE gasifier. Although the ash content of coal affects the cold gas efficiency of slurry-fed gasifiers, the lower ash
content of coal B could not offset the lower dry solid content of
the slurry compared to coal A. Therefore, the cold gas efficiency
is lower for coal B than for coal A. As observed from Fig. 2, cold
gas efficiencies of dry-fed gasifiers are relatively insensitive to
the coal quality which is a significant advantage compared to slurry-fed gasifiers. The higher cold gas efficiency for coal B than that
for coal A in SCGP and SFG gasifiers may be due to the lower ash
content of coal B.
5.1.2. Coal quality and properties of raw syngas from gasifiers
The composition and characteristics of raw syngas produced by
different gasification technologies are presented in Tables 6 and 7.
The presented data correspond to upstream of the sour water–gas
shift unit.
Results of the simulation show that raw syngas from dry-fed
gasifiers has higher CO and lower CO2 content compared to slurry-fed gasifiers. The higher water content in the slurry-fed gasifier
results in conversion of CO to CO2 and H2 through CO-shift reaction
(refer to Eq. (7)). In addition, the higher rate of oxygen consumption in the GE gasifier (refer to Table 5, Section 4) compared to
SFG and SCGP, caused by the need to evaporate the slurry water
in the gasifier, yields higher carbon dioxide in the raw syngas. Even
though the O2 consumption for bituminous and sub-bituminous
coals is lower in the E-Gas™ than that in the SFG and SCGP, the
CO2 content is higher in the raw syngas produced by the former
gasifier. This is caused by more intense CO-shifting (refer to Eq.
(7)) during gasification at a lower employed temperature in the
case of the E-Gas™ technology.
Results in Tables 6 and 7 show another important difference between slurry-fed and dry-fed gasifiers for low-rank coals. The produced syngas from gasification of lignite coal in slurry-fed gasifiers
shows a very low heat value. Therefore, it demands higher feedstock consumption to produce the same energy input for the
downstream GT block compared to other coals. Thus, slurry-fed
gasifiers are unsuitable for lignite coal. On the other hand, the
capability of the dry-fed gasifiers to produce syngas from lignite
coal with relatively closer energy density to that from bituminous
coal obviously establishes them as better options for low-rank
coals.
The CH4 content from Shell, GE, and Siemens gasifiers is very
small (ppm level). On the contrary, methane formation within
the E-Gas™ gasifier is prominent. This is a result of the lower temperature employed in the second stage of this gasifier that favors
exothermic methanation reactions (refer to Eqs. (8) and (9)). Unlike
the other gasification technologies, the E-Gas™ gasifier also produces some ethylene caused by the lower temperature of the second stage of gasification.
5.2. Effects of coal quality and gasifier on overall performance of the
plant
Estimated performance parameters of the simulation for various gasification technologies using different coal quality are shown
Author's personal copy
460
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
Table 6
Composition and properties of produced syngas by SCGP and GE gasifiers (prior to the
SWGS unit).
Component (mol%)
Ar
H2
CO
H2O
CO2
H2S
COS
NH3
N2
CH4
Pressure (bar)
Temperature (°C)
HHV (MJ/kg)
LHV (MJ/kg)
a
SCGP
Parameter
GE
Coal A
Coal B
Coal C
Coal A
Coal B
Coal C
0.71
23.29
50.04
16.38
2.45
0.42
436a
203a
6.56
237a
43.0
161.4
10.14
9.64
0.66
23.80
50.92
15.46
2.05
794a
84a
213a
6.78
781a
43.0
159.1
10.34
9.83
0.67
20.35
48.61
18.29
3.62
0.31
353a
193a
8.07
781a
43.0
165.7
9.19
8.77
0.70
22.97
34.16
30.23
10.78
0.04
250a
152a
0.66
651a
58.1
199.7
7.93
7.43
0.62
20.67
25.43
39.25
13.31
608a
28a
34a
0.62
239a
58.1
212.2
6.29
5.85
0.62
10.61
9.82
59.94
18.18
0.17
73a
47
0.64
7a
58.1
232.7
2.66
2.44
GT power (MW)
ST power (MW)
Expander powera (MW)
Generator power output
(MW)
Gasification power
demand (MW)
ASU compression power
demand (MW)
Syngas compression and
pumping power
demand (MW)
AGR refrigeration power
demand (MW)
CO2 compression power
demand (MW)
HRSG pumping power
demand (MW)
Auxiliary power demand
(MW)
Net power output (MW)
The reported values are based on ppmv.
Table 7
Composition and properties of produced syngas by SFG and E-Gas™ gasifiers (prior to
the SWGS unit).
Component (mol%)
Ar
H2
CO
H2O
CO2
H2S
COS
NH3
N2
CH4
C2H4
Pressure
Temperature
HHV (MJ/kg)
LHV (MJ/kg)
a
Table 8
Performance results of IGCC plants using SCGP and GE gasifiers.
SFG
E-Gas™
Coal A
Coal B
Coal C
Coal A
Coal B
Coal C
0.45
14.23
32.27
49.07
0.53
0.21
779a
134a
2.56
339a
0.00
43.3
210.7
6.90
6.57
0.41
14.97
30.87
49.29
1.69
0.04
135a
147a
2.64
205a
0.00
43.3
211
6.70
6.36
0.42
12.85
30.14
50.59
2.40
0.15
587a
134a
3.33
106a
0.00
43.3
212.1
6.15
5.86
0.53
20.83
26.99
35.28
11.81
0.37
93a
984a
0.75
2.30
1.01
41.7
192.7
8.34
7.75
0.51
20.73
25.86
36.91
12.33
669a
16a
936a
0.69
2.01
0.79
41.7
194.8
7.83
7.27
0.51
14.85
12.80
55.00
15.51
0.18
31a
631a
0.87
0.20
150a
41.7
213.7
3.87
3.55
The reported values are based on ppmv.
in Tables 8 and 9. It is observed from these tables that the GT
power output does not vary much either with coal type or gasification technology. This is due to using the same GT technology with
fixed characteristics in simulations. The major difference in the
generator power output between different plants (i.e. using different gasification technology and coal quality) comes from the power
produced by the steam turbine. The higher ST power outputs in GE
and E-Gas™ technologies are due to the type of gasification process, i.e. slurry-fed. The higher water content of the produced syngas from these gasifiers not only reduces the steam extraction from
the steam cycle required to achieve the desired CO conversion
within the SWGS unit but also produces more steam in the syngas
cooler. Although the water content in the produced syngas from
the Siemens gasifier is high due to the direct water quench, the
ST power output is not as high as slurry-fed gasifiers. The reason
is the production of HP steam within slurry-fed gasifiers which is
absent in the SFG technology.
It can also be seen from Tables 8 and 9 that the ST power outputs for both dry- and slurry-fed gasifiers are increased by the
reduction in coal quality. The combination of higher gasifier raw
syngas flow rates and the higher moisture content of slurry (by
reduction in coal quality) results in a higher steam production in
syngas coolers downstream of the gasifiers. However, an exception
SCGP
GE
Coal
A
Coal B
Coal
C
Coal
A
Coal B
Coal
C
324.0
176.6
0.0
500.6
323.61
163.79
0.0
487.4
326.3
177.8
0.0
504.1
317.6
221.7
1.0
540.3
318.60
273.49
1.0
593.1
334.2
330.2
1.2
665.5
4.9
6.1
8.5
4.0
5.6
12.9
48.9
46.4
54.7
54.5
61.0
142.3
10.8
10.7
11.3
9.0
9.2
16.0
8.5
11.3
11.3
11.1
11.2
25.8
20.4
20.2
21.6
20.6
21.8
36.3
3.7
3.9
3.4
3.43
4.33
4.2
97.0
98.1
111.1
103.2
113.1
237.4
403.6
389.4
393.0
437.1
480.1
428.1
a
This value is related to the output power of the gas expander upstream of the
GT for the GE IGCC.
Table 9
Performance results of IGCC plants using SFG and E-Gas™ gasifiers.
Parameter
GT power (MW)
ST power (MW)
Generator power output
(MW)
Gasification power
demand (MW)
ASU compression power
demand (MW)
Syngas compression and
pumping power
demand (MW)
AGR refrigeration power
demand (MW)
CO2 compression power
demand (MW)
HRSG pumping power
demand (MW)
Auxiliary power demand
(MW)
Net power output (MW)
SFG
E-Gas™
Coal
A
Coal
B
Coal
C
Coal
A
Coal B
Coal
C
320.3
170.0
490.3
320.5
174.5
495.0
321.1
183.5
504.6
318.6
186.3
504.9
318.46
201.22
519.7
330.5
287.9
618.3
7.6
9.8
14.5
3.8
5.0
9.1
47.0
45.5
52.0
43.6
43.6
84.0
10.6
10.8
11.3
11.5
11.7
15.1
19.4
20.4
49.8
10.5
10.7
22.8
20.0
20.0
21.5
20.1
20.5
27.3
2.4
2.4
2.4
3.0
107.0
108.9
151.6
92.4
94.5
162.0
383.3
386.2
353.0
412.6
425.2
456.3
2.94
3.6
is the SCGP for coal A and coal B. According to Table 6 (previous
section), due to the lower water content of the produced syngas
from coal B, the steam extraction from the steam cycle should be
more compared to coal A to fulfill the SWGS reaction. This extraction results in a lower ST power output in the case of coal B
utilization.
The increase of ST power output for lower quality coal, however, could not compensate for the increase of auxiliary power demand, more specifically for slurry-fed gasifiers. The higher
moisture content combined with the increased ash content of coal
C compared to coal A in slurry-fed gasifiers results in higher oxygen demand to maintain the gasifier temperature. Therefore, ASU
power demand is drastically increased for coal C compared to coal
A. Results in Tables 8 and 9 confirm that the auxiliary power is in-
Author's personal copy
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
Fig. 3. Oxygen consumption of different gasification technologies with various coal
qualities.
461
Fig. 5. Effects of coal rank and gasification technology on the CO2 emissions of the
IGCC plant.
Fig. 5 shows CO2 emissions from the IGCC plant using different
gasification technologies and for various coal qualities. Although
the carbon content of the coal principally determines CO2 emissions from the IGCC plant, the use of dry-fed gasifiers and the EGas™ shows a relatively constant trend for different coal qualities.
In terms of CO2 emissions of the plant, the GE technology, amongst
various gasification technologies, does not appear to be the correct
option for gasification of lignite coal (i.e. coal C).
6. Conclusions
Fig. 4. Effects of coal rank and gasification technology on the IGCC plant efficiency
and heat rate.
creases with low coal quality. This is primarily due to the increasing oxygen demand but also because of the increasing CO2 content
which should be captured and compressed. The oxygen consumption, which is an important factor to give an indication of the capital cost of the ASU, is shown in Fig. 3.
The effect of coal rank and gasification technology on the thermal efficiency and the net heat rate (the reciprocal of net
efficiency) of the IGCC plant with CO2 capture is shown in Fig. 4.
It is obvious that the coal quality significantly influences the overall plant efficiency. It is noted from Fig. 4 that the slurry-fed GE
gasifier is more affected by the low-rank coal C compared to
dry-fed systems (i.e. SCGP and SFG). The other slurry-fed gasifier,
the E-Gas™, is less sensitive to the coal quality. This is due to better
utilization of the input feedstock with the second gasification
stage. As mentioned previously, the lower overall plant efficiency
for various gasification technologies using coal C is primarily due
to decreased gasification efficiencies.
It can be seen from Fig. 4 that the overall plant efficiency for coal B
is higher than that for coal A in both GE and E-Gas™ gasifiers. Multiple reasons contribute to the mitigation of the impact of the relatively lower cold gas efficiency for coal B compared to coal A (refer
to Fig. 2) on power plant net efficiency. These are as follows:
(a) The ash content of coal B is lower compared to coal A, which
results in a low difference between the dry and ash-free solid content of both coals. It should be noted that this fraction of coal, i.e.
the dry and ash-free solid content, provides the required energy
to vaporize the slurry water; (b) coal B raises more HP steam in
the syngas cooler than coal A; and (c) the produced syngas from
the gasification of coal B is more pre-CO-shifted than syngas from
coal A due to the higher water content. This leads to less exergy
loss in the SWGS section. The higher CO conversion (refer to Eq.
(7)) also results in higher plant efficiencies for coal B in slurryfed gasifiers compared to dry-fed gasifiers.
The EU’s H2-IGCC project is aiming to develop and demonstrate
technological solutions for future generation IGCC plants with carbon capture. In the current study, under the framework of the project which aims for the evaluation and optimization of the best
plant configuration, the effects of coal quality and the selection
of gasifiers on the overall performance of the baseline configuration of the IGCC plant have been reported. Four commercially available gasifiers from Shell, GE, Siemens, and ConocoPhillips have
been considered for this comparative study. The effects of three
different types of coals on these gasifiers, as well as on the overall
performance of the IGCC plant, have been investigated. Given the
fact currently there is not any operating IGCC power plant with
carbon capture, the validation of the overall system performance
is not feasible. However, utilization of validated tools and models
against existing plant data for simulation of different IGCC sub-systems in this study has resulted in more reliable results. Several
conclusions can be inferred from the results as follows:
(1) The coal quality considerably influences the cold gas efficiency for slurry-fed gasifiers i.e. GE and ConocoPhillips gasifiers. Amongst slurry-fed gasifiers, the coal quality has the
greatest impact on the GE gasifier. The cold gas efficiency
of the GE gasifier gasifying lignite coal is 29% lower than gasifying bituminous coal. On the contrary, dry-fed gasifiers are
relatively insensitive to the quality of the input coal.
(2) Results confirm that one of the main advantages of dry-fed
gasifiers compared to slurry-fed types is a relatively constant quality of produced syngas even when low-rank coal
is gasified.
(3) The higher water content of the produced syngas from
slurry-fed gasifiers increases the ST power output due to
reduction of the steam extraction from the steam cycle for
the SWGS reaction. However, this power increase cannot
compensate for the increase of ASU power demand and
results in lower system efficiency for low-rank coal.
(4) The overall performance of the whole IGCC is slightly
affected by the GT performance using different syngas compositions from various gasification technologies. However,
Author's personal copy
462
M. Mansouri Majoumerd et al. / Applied Energy 113 (2014) 452–462
paper findings including characteristics of the inlet syngas to
the gas turbine are very important in terms of the GT design.
(5) Summarizing, slurry-fed gasifiers in this study, i.e. GE and
ConocoPhillips, are suitable for bituminous and sub-bituminous coals, while dry-fed gasifiers, i.e. Shell and Siemens,
show a relatively constant behavior for a wider range of coal
quality.
Acknowledgments
The authors are grateful to the European Commission’s Directorate-General for Energy for financial support of the Low Emission
Gas Turbine Technology for Hydrogen-rich Syngas (H2-IGCC)
project.
References
[1] IPCC. IPCC fourth assessment report: climate change 2007 (AR4); 2007.
<http://www.ipcc.ch>.
[2] IEA. IEA Statistics 2012 Edition: CO2 emissions from fuel combustion
highlights. Paris: International Energy Agency; 2012.
[3] IEA. World Energy Outlook 2011. Paris: International Energy Agency; 2011.
[4] EU Energy roadmap 2050, European Union, European Commission; 2012.
[5] Tremel A, Haselsteiner T, Kunze C, Spliethoff H. Experimental investigation of
high temperature and high pressure coal gasification. Appl Energy
2012;92:279–85.
[6] Giuffrida A, Romano MC, Lozza GG. Thermodynamic assessment of IGCC power
plants with hot fuel gas desulfurization. Appl Energy 2010;87(11):3374–83.
[7] Giuffrida A, Romano MC, Lozza GG. Thermodynamic analysis of air-blown
gasification for IGCC applications. Appl Energy 2011;88(11):3949–58.
[8] Melchior T, Madlener R. Economic evaluation of IGCC plants with hot gas
cleaning. Appl Energy 2012;97:170–84.
[9] Chen S, Xiang W, Wang D, Xue Z. Incorporating IGCC and CaO sorptionenhanced process for power generation with CO2 capture. Appl Energy
2012;95:285–94.
[10] Martelli E, Kreutz T, Carbo M, Consonni S, Jansen D. Shell coal IGCCS with
carbon capture: conventional gas quench vs. innovative configurations. Appl
Energy 2011;88(11):3978–89.
[11] Low emission gas turbine technology for hydrogen-rich syngas Under the 7th
Framework Programme FP7-239349. Available from: Project website:
www.h2-igcc.eu.
[12] Yun Y, Yoo YD, Chung SW. Selection of IGCC candidate coals by pilot-scale
gasifier production. Fuel Process Technol 2007;88:107–16.
[13] Abadie LM, Chamorro JM. The economics of gasification: a market-based
approach. Energies 2009;2:662–94.
[14] Spliethoff H. Power generation from solid fuels. 1st ed. Springer; 2010. 712.
[15] Cormos CC, Starr F, Tzimas E. Use of lower grade coals in IGCC plants with
carbon capture for the co-production of hydrogen and electricity. Int J Hydrog
Energy 2010;35:556–67.
[16] BP Statistical Review of World Energy. British Petroleum company; 2012.
[17] Chiesa P, Consonni S. Shift reactors and physical absorption for low-CO2
emission IGCCs. J Eng Gas Turbine Power 1999;121(2):295–305.
[18] Chiesa P, Consonni S, Kreutz T, Williams R. Co-production of hydrogen,
electricity and CO2 from coal with commercially ready technology. Part A:
performance and emissions. Int J Hydrogen Energy 2005;30:747–67.
[19] Holt N, Booras G, Todd D. A summary of recent IGCC studies of CO2 capture for
sequestration. In: Gasification technologies conference2003: San Francisco,
CA, USA.
[20] Kreutz T, Martelli E, Carbo M, Consonni S, Jansen D. Shell gasifier-based coal
IGCC with CO2 capture: partial water quench vs. novel water-gas shift. In:
ASME paper, GT2010-22859. 2010. ASME Turbo Expo, Glasgow, UK.
[21] Sipöcz N, Mansouri M, Breuhaus P, Assadi M. Development of H2-rich syngas
fuelled GT for future IGCC power plants – establishment of a baseline. In:
ASME paper GT2011-45701. 2011. ASME Turbo Expo, Vancouver, Canada.
[22] Mansouri Majoumerd M, De S, Assadi M, Breuhaus P. An EU initiative for future
generation of IGCC power plants using hydrogen-rich syngas: Simulation
results for the baseline configuration. Appl Energy 2012;99:280–90.
[23] Higman C, Van der Burgt M. Gasification. 1st ed. Gulf Professional Publishing,
Elsevier Science (USA); 2003.
[24] Collot AG. Matching gasification technologies to coal properties. Int J Coal Geol
2006;65:191–212.
[25] Chen C, Rubin ES. CO2 control technology effects on IGCC plant performance
and cost. Energy Policy 2009;37:915–24.
[26] Pettinau A, Frau C, Ferrara F. Performance assessment of a fixed-bed
gasification pilot plant for combined power generation and hydrogen
production. Fuel Process Technol 2011;92:1946–53.
[27] Parulekar PS. Comparison between oxygen-blown and air-blown igcc power
plants: a gas turbine perspective. In: ASME paper GT2011-45154. ASME Turbo
Expo, Vancouver, Canada; 2011.
[28] Maurstad O, Herzog H, Bolland O, Beér J. Impact of coal quality and gasifier
technology on IGCC performance. In: 8th international conference on
greenhouse gas control technologies (GHGT8), Trondheim, Norway; 2006.
[29] The Shell coal gasification process for sustainable utilisation of coal. Shell
Global Solutions; 2006.
[30] Shelton W, Lyons J. Texaco gasifier IGCC base cases. DOE/NETL PED-IGCC-98001; 2000.
[31] Morehead H. Siemens gasification and IGCC update. In: Gasification
technologies conference. Washington, DC, USA: Siemens Energy Inc.; 2008.
[32] Schmid C. Siemens fuel gasification: update for power and industrial
applications. In: Gasification technologies conference. San Francisco, USA:
Siemens AG; 2007.
[33] Klemmer K-D. The Siemens gasification process and its application in the
Chinese market. In: Gasification Technologies conference. Washington, DC:
Siemens Power Generation; 2006.
[34] NETL. Cost and performance baseline for fossil energy plants, Volume 1:
Bituminous coal and natural gas to electricity, 2010, 2nd Revision; DOE/NETL2010/1397.
[35] E-Gas™ technology for integrated gasification combined cycle. <http://
www.phillips66.com/EN/tech/e-gas/Pages/advantage.aspx>.
[36] Van Der Ploeg HJ, Chhoa T, Zuideveld PL. The Shell coal gasification process for
the US industry. In: Gasification technologies conference, Washington DC;
2004.
[37] EnssimÒ. Enssim Software: Doetinchem, The Netherlands; 2009.
[38] ASPEN Plus version 7.1. Aspen Technology Inc.: Cambridge, MA, USA; 2009.
[39] IPSEpro version 4.0. Simtech Simulation Technology (Simtech): Graz, Austria;
2003.
Paper IV
Fuel change effects on the gas turbine performance in IGCC
application
Mohammad Mansouri Majoumerd, Mohsen Assadi
Presented at 13th International Conference on Clean Energy
(ICCE 2014), Istanbul, Turkey, June 2014
161
FUEL CHANGE EFFECTS ON THE GAS TURBINE PERFORMANCE IN IGCC APPLICATION
Mohammad Mansouri Majoumerd1, Mohsen Assadi1,2
1Faculty
2International
of Science and Technology, University of Stavanger, 4036 Stavanger, Norway
Research Institute of Stavanger (IRIS), P.O. Box 8046, 4068 Stavanger, Norway
E-mail: [email protected]; [email protected]
ABSTRACT
Improved security of energy supply by utilizing clean coal technology with and without CO 2 capture results in
changed fuel composition at integrated gasification combined cycle (IGCC) plants. Gas turbine modifications might
be therefore necessary to cope with such changes. As part of the H2-IGCC project, a European Union co-funded
project, this study presents a detailed analysis of the effect of using various fuel(s) compositions on the
performance of the selected gas turbine in IGCC plants considering different operating conditions.
For realistic analysis of the gas turbine behavior, component characteristic maps were generated and
implemented into a detailed thermodynamic model in a commercial heat and mass balance program, IPSEpro.
The fuels studied in this paper are undiluted hydrogen-rich syngas (i.e. 87 mol% H2 content), clean syngas
(without CO2 capture), and natural gas. The impact of the fuel change on the gas turbine performance has been
investigated and the results are presented and discussed in this paper. Moreover, technical solutions for
realization of the targeted fuel flexibility under certain limitations and boundary conditions are presented.
Calculation results show that for the given boundary conditions, the surge margin of the compressor was
slightly reduced when natural gas was replaced by H2-rich syngas. The use of clean syngas instead of H2-rich
syngas resulted in a considerable reduction of the surge margin and elevation of the turbine outlet temperature at
design point conditions, when keeping the turbine inlet temperature and compressor inlet mass flow unchanged. In
order to maintain the exhaust temperature and improve the surge margin, when operating the engine with clean
syngas, a combination of adjustment of variable inlet guide vanes and reduced turbine inlet temperature was
finally considered. Results of this study confirm that using clean syngas requires major gas turbine modifications
such as air bleed from compressor outlet and multiple fuel feed systems.
Key words: IGCC, gas turbine, performance analysis, fuel flexibility, hydrogen-rich syngas, clean syngas
1. INTRODUCTION
The world’s demand for electricity is ever increasing mainly due to the population growth and improved living
standards. Currently, the share of electricity generation is 37% of the global primary energy consumption. In 2012,
the global electricity production was 22,126 TWh [1] with an annual average growth rate of 2.95 % from 1990 [2].
Fossil-based electricity production accounted for 68 % of the total generation and coal, the most carbon-intensive
fossil fuel, was the largest contributor (41%) to the electricity supply in 2012 [1]. For future, electricity demand is
projected to grow more rapidly than the total energy consumption [3, 4]. This demand will be almost 70 % higher in
2035 than the current demand [5].
On the other hand, needs to reduce greenhouse gas (GHG) emissions require considerable efforts to be
directed towards the utilization of clean power generation technologies. The emissions of CO2 from the electricity
and heat supply sector using fossil fuels were about 42% of the total global CO2 emissions in the year 2011 [6].
Several options should be considered in a comprehensive package to reduce the global GHG emissions per unit
of energy consumption. Amongst those options are energy conservation and efficiency improvement,
transformation/replacement of carbon-intensive fossil fuels by cleaner technologies (such as switch from coal to
natural gas (NG), enhanced use of renewable energy sources and utilization of nuclear energy) and reduction of
CO2 emissions using carbon capture and storage (CCS) for fossil-based energy.
The integrated gasification combined cycle (IGCC) has been one of the most promising coal-derived
technologies in terms of higher efficiency and lower environmental impact, compared to conventional pulverized
coal plants [7]. In addition, the well-established high temperature coal gasification technology may facilitate the
control and reduction of gaseous pollutants ( e.g. NOx and SOx) to the level of NG-fuelled plants [8]. Furthermore,
using IGCC provides one of the least costly approaches for CO2 abatement through pre-combustion carbon
capture [8, 9]. However, one of the largest barriers towards widespread utilization of the IGCC technology is its
higher capital costs compared to a conventional pulverized coal plant [10, 11]. In addition, the high H2 content in
the syngas derived from coal gasification (more specifically when a pre-combustion CO2 capture unit is considered
in the cycle) complicates the application of pre-mixed burners, which is the current state-of-the-art (SOA)
technology in NG-fired gas turbines (GTs) [12]. The restriction of using such burners is the flammability limits of
H2-rich fuels, which are much larger than that for natural gas [13]. Moreover, high hydrogen content syngas has
higher adiabatic flame temperature, higher flame speed, and higher flashback potential compared to NG, which
complicate its pre-mixed combustion [14, 15]. For this reason, high NOx emitting diffusion burners have been
employed for the existing IGCC power plants, which require the hydrogen-rich syngas to be diluted with nitrogen or
water/steam to control the higher adiabatic flame temperature.
The other persistent challenge in the IGCC plants is the variation in composition and heating value of the
produced syngas, which needs to be combusted in the downstream GT. This variation is mainly due to the varying
feedstock quality [16], and the process and operational causes (e.g. switch from a plant with CO2 capture to a noncapture mode).
1
Several studies highlighted the effect of varying composition of a specific type of fuel. Nag et al.
investigated the effects of using different compositions of liquefied natural gas (LNG) on gas turbine operation.
They found that change of lower heating value (LHV) and composition of the concerned fuel (i.e. LNG) may lead to
increased emissions and different component lifetime [17]. Chishty mainly studied the combustion and
corresponding design challenges in the combustor when using different fuels [18]. Experiences concerning the
continuous use of different fuel composition, rather than switching fuels during operation, have been also
investigated in various papers from original equipment manufacturers (OEMs) [19-21].
In addition to the areas covered in the publications mentioned above, the operation of an IGCC plant with
pre-combustion CO2 capture, burning undiluted hydrogen-rich syngas, raises other matters of increased
importance, such as the requirement for secure electricity production, which could be threatened by various
disturbances. Compared to standard combined cycles, a H2-rich fuelled IGCC power plant has a complex plant
layout with an increased number of components. Consequently, a higher probability of disturbances, such as
failure or planned maintenance of the components/sub-systems of the cycle could be expected in such a plant. A
changed operational window might also be caused by economic reasons, e.g. when CO2 capture is not beneficial
due to low prices on the CO2 trading markets resulting in the bypassing of the CO2 capture unit. Furthermore,
bypassing the CO2 capture unit will result in an overall higher power output, which requires efficient operation of
the plant under these operating conditions. As a consequence, the GT should cover a wide range of fuel types
from undiluted H2-rich fuel (as the design fuel) to low-LHV gaseous fuels (in the case of burning clean syngas) and
also natural gas (as the back-up fuel) [21]. These fuels have different composition and LHV, which result in a
different volume flow and consequently the mass flow into the combustor to reach the same order of turbine inlet
temperature (TIT) and thereby similar efficiency level for a given compressed air flow [22]. Change of fuel flow rate
affects the compressor/expander matching [13], induces higher back pressure to the compressor [12], and
reduces available surge margin [22] if no adjustments are implemented to compensate for the mass flow change.
Therefore, it is necessary to adjust the operational parameters of the gas turbine in such a way that a safe
operation with reasonable performance can be offered. Given all these technological challenges, a secure
provision of electricity using a fuel-flexible gas turbine with high operational flexibility is an essential need in IGCC
power plants [23].
In 2009, the H2-IGCC project was co-funded by European Union to develop knowledge that would allow
the use of SOA gas turbines in the next generation of IGCC power plants with deployment of CO2 capture. The
overall objective of the project was to enable the stable operating conditions of the GT with pre-mixed combustion
of undiluted H2-rich syngas. The secondary objective was to increase the fuel flexibility without adversely affecting
the reliability and availability of the entire system by minor modifications to existing GTs [24]. As part of the H2IGCC project, this work presents the consequences of fuel change on the performance of the gas turbine at
various operating conditions. The main objective is to see whether the targeted fuel flexibility or ability to operate
on a variety of fuels (i.e. H2-rich syngas, non-captured clean syngas and natural gas) is achievable under
presumed boundary conditions and limitations or not.
In this paper, the baseline configuration of the selected IGCC plant with and without CO2 capture unit is
briefly presented to provide an overview of the entire cycle’s layout. This will assist the readers for better
understanding of how the fuel properties are affected by different operational/process changes (or disturbances).
The effects of fuel change on the performance of the selected gas turbine, as an isolated sub-system, are then
assessed. Accordingly, different operating conditions and adjustments to mitigate the negative effects of fuel
change are thoroughly investigated and discussed followed by necessary modifications/strategies to minimize the
negative effects of fuel change during the lifetime of a gas turbine in IGCC application.
2. THE SELECTED IGCC CONFIGURATION
In order to investigate the impact of fuel change on the gas turbine performance, it is essential to have the
expected fuel properties and composition prior to the GT. For this purpose, a baseline IGCC plant with and without
CO2 capture unit has been established and thermodynamically analyzed. For better understanding of the process
change from the IGCC plant with CO2 capture to a plant without capture, the plant description is briefly presented
in this section. However, detailed technical assumptions and specification for modeling of the entire IGCC plant
may be obtained from authors previous publication [25]. It should be noted that the fuel compositions in both
cases, i.e. the plant with capture (H2-rich syngas) and the plant without capture have been adopted from a
previous study [25].
2.1. The selected IGCC plant with CO2 capture
The block flow diagram of the IGCC plant with capture unit is shown in Fig. 1. The feedstock considered for the
selected IGCC plant is bituminous coal. The plant consists of seven major sub-systems as explained below:
(1) Air separation unit (ASU): the cryogenic ASU is a stand-alone unit generating O2 with 95% purity from air
supplied by an intercooled main air compressor (MAC) for the gasification of coal. The main advantage of nonintegrated ASU is higher plant availability, operability, and flexibility. However, notably is that the overall plant
efficiency increases with the degree of integration between ASU and the gas turbine compressor due to the higher
isentropic efficiency of the GT compressor [26]. Nevertheless, lower efficiency of the non-integrated GT-ASU case
could be balanced with selection of an intercooled MAC.
Often for IGCC plants either syngas dilution with N2 or steam or syngas saturation with water is
considered to control the NOx emissions from diffusion flame burners. However, here this strategy has been
2
eliminated due to the use of undiluted pre-mixed combustion of the H2-rich syngas. As there is no need for
injection of diluent gaseous nitrogen into the GT for dry-low NOx combustion, heat integration between the GT
compressor bleed air and diluent nitrogen from the ASU is not an option in order to enhance the overall plant
efficiency.
To Sulfur
recovery
Coal
Gasification
Shift
reaction
Acid gas
removal
CO2
O2
Air
Slag
To atmosphere
CO2
compression &
dehydration
CO2
capture
Air
separation
unit
Heat recovery steam
generator
H2-rich
syngas
Stack
HP
IP/LP
Gas turbine
Air
Fig.1. The block flow diagram of an IGCC power plant with CO2 capture
(2) Gasification island and syngas cooling and scrubbing: the gasification of coal takes place in an entrained-flow,
oxygen-blown, dry-fed gasifier based on the Shell Coal Gasification Process (SCGP). Such a technology was
selected due to its high cold gas efficiency and its operating pressure level. A key parameter governing the overall
plant pressure is the operating pressure of the GT combustor. The pressure prior to the combustion chamber was
fixed at about 30 bar to overcome the pressure loss over the fuel valves and required turbulent conditions for premixed combustion. The pressure of the gasification block was then calculated to be at 45 bar considering all
pressure losses from the gasifier to the combustion chamber for eliminating any supplementary syngas
compression. The selection of the SCGP technology was also justified by availability of a validated gasification
model provided by the operator of the Buggennum IGCC plant within the H2-IGCC project consortium. The
validation results are available in [27].
(3) Sour water-gas shift (SWGS) reaction unit: the SWGS process converts the CO in the raw syngas to CO2 by
shifting the CO with water over a catalytic bed according to the following reaction:
(44
𝐶𝑂(𝑔) + 𝐻2 𝑂(𝑔) ↔
𝑀𝐽
)
𝑘𝑚𝑜𝑙𝑒
𝐶𝑂2 (𝑔) + 𝐻2 (𝑔)
(1)
(4) Acid gas removal (AGR) unit and CO2 capture unit: a two-stage SELEXOL system for H2S removal and CO2
capture was used. Due to the high partial pressure of acid gases, physical absorption of H 2S and CO2 is preferred
to chemical, amine-based absorption processes. The H2S is removed by a counter-current flow of solvent in the
first stage. The syngas leaving the H2S absorber enters the second stage where the CO2 is captured. The overall
CO2 capture rate is approximately 90% (molar basis).
(5) CO2 compression and dehydration unit: the CO2 captured from the process is compressed by an intercooled
compressor, aftercooled, liquefied and finally pumped up to a final pressure of 110 bar. In order to reduce the
corrosion risk in the transport pipeline, a dehydration unit using tri-ethylene glycol is considered resulting in water
content in the captured CO2 line less than 20 mg/kg.
(6) Gas turbine: the GT block including compression, combustion, and expansion generates electric power using a
generator. The GT model is further discussed in Section 3.
(7) Heat recovery steam generator (HRSG) and steam cycle: downstream of the GT is a triple pressure level
HRSG with reheat (140bar/530/530°C) and a steam turbine to generate steam and power.
2.2. The selected non-capture IGCC plant
In the plant without CO2 capture, the water-gas shift reaction is bypassed and the raw syngas leaving the wet
scrubber is passed through a COS hydrolysis unit before entering the H2S absorber. Fig.2 illustrates the block flow
3
diagram of the non-capture IGCC plant. In order to remove more than 99.9% of the sulfur content in produced
syngas, it is necessary to add the COS hydrolysis unit to convert the COS to H2S [8]. In addition to this change,
few other sub-systems of the plant with capture should be out of operation such as CO2 capture unit (i.e. the 2nd
absorption stage), CO2 compression and dehydration unit. The H2S-free syngas exiting the top of the H2S
absorber is then sent to the GT combustor.
To sulfur
recovery
To atmosphere
Coal
Gasification
O2
COS
hydrolysis
Acid gas
removal
Heat recovery steam
generator
Non-captured,
clean syngas
Slag
Stack
Air
Air
separation
unit
HP
IP/LP
Gas turbine
Air
Fig.2. The block flow diagram of an IGCC power plant without CO2 capture
3. METHODOLOGY
This work aims to investigate the effects of fuel change on the performance of a selected gas turbine under
various operating conditions in IGCC application. It should be noted that the consequences of fuel change and
necessary adjustments to the IGCC plant are not covered in this study. However, it was necessary to simulate the
entire IGCC system with and without capture unit to reach the composition of different fuels upstream of the gas
turbine. For this purpose, simulation results of the selected IGCC plant available in the recent publications by the
authors have been used [25, 27].
Detailed modeling of the gasification process, based on the Shell Coal Gasification Process (SCGP)
including coal milling and drying, gasification, raw syngas cooling and scrubbing, has been performed by Vattenfall
(Nuon) using an in-house model in the Enssim software, which validated against operational data from Buggenum
power plant [28]. The modeling of upstream and downstream units of the gasification block, such as the air
separation unit, the sour water-gas shift reaction unit, the acid gas removal unit, the CO2 compression and
dehydration unit, and the COS hydrolysis unit has been performed using ASPEN Plus [29]. Details of the
thermodynamic models of the aforementioned units are also available in [25, 27]; hence, are not repeated here. A
special emphasis is dedicated to the gas turbine in this study. The thermodynamic model of the GT has been
established using IPSEpro, a heat and mass balance software [30]. This model is briefly discussed in the following
sub-section.
3.1. Gas turbine model
In order to investigate the effects of fuel change from NG, which is the design fuel for the GT, to undiluted H2-rich
syngas for the IGCC plant with CO2 capture and clean syngas for the non-capture IGCC plant, a reference GT
design should be selected. Accordingly, a Siemens SGT5-4000F/Ansaldo Energia V94.3A type of gas turbine was
selected as the manufacturers being partners of the H2-IGCC consortium.
Using a one-dimensional lumped model, detailed engine specifications including the main GT components
characteristics maps were generated for NG as fuel and made available for the system modeling and simulation.
An advanced thermodynamic GT model in IPSEpro software was then modified followed by model validation
against performance data published by the manufacturer [31]. The main components of the gas turbine model are
described below:
(1) Compressor model: The compressor map generated by aforementioned lumped model, provided by Roma Tre
University (refer to Fig.3), has been implemented into the thermodynamic compressor model in IPSEpro software
tool as look-up tables. The axes’ labels in Fig.3 are not shown for reasons of confidentiality. The look-up table
contains the changes of pressure ratio, corrected inlet mass flow, and isentropic efficiency based on corrected
rotational speed. The effect of the variable inlet guide vane (VIGV) on aforementioned parameters (i.e. look-up
table’s parameters) was also considered. However, the limited dimensions (i.e. number of rows and columns) of
such a table, implemented in IPSEpro, are among limiting factors affecting the accuracy of the information retrieval
from the look-up table.
4
Isentropic efficiency (%)
Pressure ratio [-]
Pressure ratio
Corrected mass flow [-]
IGV 100%
IGV 70%
IGV 47%
IGV 90%
IGV 60%
IGV 100%
IGV 70%
IGV 47%
IGV 80%
IGV 50%
(a)
IGV 90%
IGV 60%
IGV 80%
IGV 50%
(b)
Fig.3. The compressor characteristics maps, (a) pressure ratio versus corrected mass flow and (b) isentropic efficiency
versus pressure ratio
In terms of compressor stability, surge margin calculation has been incorporated into the compressor model.
Surge is defined as a transient condition involving reverse flow through the compressor path and can occur when
the pressure ratio increases beyond a safety margin. The surge margin is defined as:
𝑃𝑅𝑠𝑢𝑟𝑔𝑒 −𝑃𝑅𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛
𝑆𝑢𝑟𝑔𝑒 𝑚𝑎𝑟𝑔𝑖𝑛 (%) =
× 100
(2)
𝑃𝑅𝑜𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛
(2) Combustor model: The fuel composition entering the combustor was obtained from system simulation reported
in previous publication [25] based on the IGCC plant described in Section 2. To achieve realistic results, pressure
losses reflecting the current SOA combustor technology were used.
(3) Expander model: Once H2-rich syngas or non-captured syngas is used as the GT fuel in the existing GT (i.e.
SGT5-4000F/Ansaldo Energia V94.3A) designed for NG operation, the operating conditions and performance of
the GT deviates from the original design. Therefore, an off-design analysis was required in order to provide
information about necessary changes. A simplified off-design model has been considered for modeling the
expander. The turbine off-design operation was modeled assuming a constant swallowing capacity at choking
condition, which is a reasonable assumption for heavy duty gas turbines:
Swallowing capacity = Constant =
𝛾
𝜅 =√ 𝑖(
𝑅𝑖
2
)
𝑚̇𝑖 √𝑇𝑖
(3)
𝜅 𝐴 𝑖 𝑝𝑖
𝛾𝑖 +1
𝛾𝑖 −1
(4)
𝛾𝑖 +1
As shown in Eq.3, the syngas flow rate at the expander inlet is proportional to the square root of the
temperature. In a GT designed for NG, once the fuel flow rate is increased due to the change of fuel composition
and LHV different components and operating conditions might be affected such as expander lifetime and the
compressor stability. The influence of the cooling air entering the turbine at different rows was considered using a
virtual/mixed turbine inlet temperature and a virtual/mixed polytropic efficiency according to the following
equations:
𝑇𝑡𝑚𝑖𝑥𝑒𝑑 𝑖 =
𝑃𝑠ℎ𝑎𝑓𝑡
𝑚̇𝑡𝑜𝑡𝑎𝑙 𝑐𝑝
𝜂𝑡 𝑝𝑜𝑙𝑦𝑡𝑟𝑜𝑝𝑖𝑐 𝑚𝑖𝑥𝑒𝑑 =
+ 𝑇𝑡𝑜
(5)
𝑇𝑡𝑚𝑖𝑥𝑒𝑑 𝑖
⁄𝑇 )
𝑐𝑝 𝑙𝑛(
𝑡𝑜
𝑝
𝑅𝑜
𝑙𝑛( 𝑡𝑖⁄𝑝𝑡0 )
(6)
In order to simplify the GT model, the effects of fuel change on the amount of cooling flows required to
maintain the blade wall temperature at certain level was not considered at this stage. Therefore, it was assumed
that the cooling flows were unchanged when the fuel composition was altered.
5
Table 1. Technical assumptions for GT modeling at design condition
Parameter
Unit Value
Air flow at the compressor inlet
kg/s 685
Pressure ratio
18.2
Cooling flow 1st expander stage
kg/s 83.7
Cooling flow 2nd expander stage
kg/s 52.7
Cooling flow 3rd expander stage
kg/s 26.9
Shaft cooling
kg/s 13.7
Compressor isentropic efficiency
%
88.2
Combustor outlet temperature
ᵒC
1500
Turbine inlet temperature
ᵒC
1266
Expander isentropic efficiency
%
92.1
Expander total inlet pressure
bar
17.9
Expander static outlet pressure
bar
1.1
Mechanical efficiency
%
88.7
Generator electrical/mechanical efficiency %
99/99.5
Sizing of the entire IGCC plant is governed by the gas turbine as it requires a specific amount of fuel depending on
the fuel properties and composition. The operating condition of the GT is determined by matching the operating
characteristics of the compressor and the expander. Thus, if the gas flow rate varies at the expander inlet, e.g. due
to the change of syngas composition, the operating condition of the GT adapts to this change. This could result in
change of pressure ratio even at similar firing temperature. The technical assumptions for GT modeling at its
design condition (NG-fired) are presented in Table 1.
3.2. Boundary conditions
For modeling of the gas turbine, ISO standard conditions have been considered. The ambient air conditions and
composition are shown in Table 2.
Table 2. Ambient air composition and conditions
Components
Unit Value
H2O
wt% 0.63
N2
wt% 75.10
O2
wt% 23.01
Ar
wt% 1.21
CO2
wt% 0.05
Ambient air pressure
bar
1.013
Ambient air temperature °C
15
Relative humidity
%
60
The investigated fuels in this study included (A) natural gas, (B) H 2-rich syngas, and (C) clean (non-captured)
syngas. The corresponding composition and characteristics of each fuel based on the previously described plants’
layouts and thermodynamic models in sections 2 and 3 are given in Table 3.
Table 3. Composition a and characteristics of investigated fuels
Components
Fuel A
Fuel B
Fuel C
(NG)
(H2-rich syngas) (clean syngas)
wt% mol% wt%
mol%
wt% mol%
CO
0.0
0.0
5.4
1.2
79.7 60.4
CO2
0.4
0.1
24.1
3.3
5.6
2.7
H2
2.6
17.9
28.9
86.8
2.7
28.1
H2O
0.0
0.0
0.1
0.0
0.0
0.0
N2
0.0
0.0
41.5
8.7
12.0 8.8
CH4
93.0 80.6
0.0
0.0
0.0
0.0
C3H8
4.0
1.3
0.0
0.0
0.0
0.0
Pressure (bar)
30
30
30
Temperature (°C) 30
30
30
LHV (MJ/kg)
49.7
35.3
11.3
To evaluate the impact of the fuel change on the selected gas turbine technology in IGCC application, various
simulation setups with different boundary conditions such as TIT, turbine outlet temperature (TOT), etc. have been
considered, as shown in Table 4. As mentioned earlier, the reference GT was chosen as being NG fuelled.
However, it should be mentioned that the target of the H2-IGCC project is a plant operated with H2-rich fuel
(without major changes to the original gas turbine design) and NG operation is only considered as backup. The
clean syngas (non-captured) operation is considered as an off-design alternative when the carbon capture unit is
bypassed.
6
Table 4. Different setups for investigation of the fuel change effects
on the gas turbine performance
Parameter Case I
Case II
Case III
Case IV
Case V
TIT (°C)
1266
Calculated
1266
Calculated Calculated
VIGV
Fully open
Fully open
Calculated Fully open Calculated
PR
Calculated a
18.2
18.2
Calculated 18.2
TOT (°C) Calculated
Calculated
Calculated 577
577
In addition to those parameters marked as ‘Calculated’ in Table 4, other variable parameters including GT gross
power, GT efficiency, and the surge margin were also studied. The simulation results are presented and discussed
in the next section.
4. RESULTS AND DISCUSSION
The main focus of this study was evaluation of the impact of fuel change on the GT performance. In this section, A,
B, and C represent different fuels, namely NG, H2-rich syngas, and clean (non-captured) syngas, respectively. The
effects of different parameter setups listed in Table 4 are the same for the NG-fired gas turbine, as these
parameters are the design parameters of the reference engine. Therefore, the effect of varying operating
conditions on NG-fired GT is only reported as Case A.
1.2
1.1
5
Inlet air flow
Gross power
SM
TIT
TOT
Fuel flow
PR
GT efficiency
4.5
4
1
0.9
3
0.8
2.5
2
0.7
Relative fuel flow
Relative value
3.5
1.5
0.6
1
0.5
0.5
0.4
0
Case A
Case BI
Case CI
Fig.4. Effect of fuel change on relevant performance parameters
at fully open VIGV and constant TIT
Fig.4 shows the effect of fuel change on relevant performance parameters, such as pressure ratio, gross GT
power, TOT, fuel flow, and surge margin when the TIT is kept constant and the VIGV is fully open.
Change of fuel from NG to H2-rich syngas, when keeping the VIGV fully open and TIT constant (A to BI), increases
the pressure ratio by 0.6 over the compressor. The pressure ratio of Case CI increases by 1.3 compared to the
reference Case A. The reasons for pressure increase in both cases are the higher fuel mass flow in combination
with unchanged compressor air flow, which leads to an increased total mass flow through the turbine, to maintain
the TIT unchanged.
When using H2-rich syngas, the power output increases compared to the NG operation due to the higher
hot gas flow rate through the expander at a constant TIT (i.e. 1266 °C). This is because of the higher enthalpy
drop through the expander due to the higher H2O content in the H2-rich syngas and also in the flue gas according
to [13]. Comparison of the gross power output for all cases, presented in Fig.4, shows that operating with H2-rich
fuel (Case BI) results in the highest power output, 311.6 MW, which is 7% higher than the NG-fired case, i.e. 291.2
MW. Operating with the clean syngas at a similar setup (Case CI) shows 5% higher gross power output than for
the reference case. Nevertheless, Case CI delivers 6 MW less power compared to Case BI due to changing hot
gas composition and properties, such as lower H2O content.
One of the major design concerns for the engine lifetime is the TOT. This indicator is not affected very
much by the fuel change from NG to H2-rich syngas. However, the marginally lower TOT for Case BI compared to
that for Case A will result in an insignificant drop of power output from the steam cycle downstream of the GT.
Despite the increased pressure ratio for Case CI compared to the NG-fired GT (A), 10 °C higher TOT (587 °C) is
observed for Case CI. Higher TOT would lead to considerable reduction of the lifetime of the last expander stage.
The fuel flow increases with fuel change from NG to H2-rich syngas from 14.9 kg/s to 21.9 kg/s due to the
lower calorific value of the H2-rich fuel compared to NG. The fuel mass flow difference becomes significantly higher
for the case with clean syngas fuel reaching 69.3 kg/s. Referring to the IGCC cycle, where the CO2 capture unit
7
has been bypassed, the syngas production rate increases even though the amount of coal remains unchanged.
This is mainly related to changed details inside the overall fuel gas treatment/preparation process.
Concerning the surge margin, which is mandatory to ensure stable operation of the compressor, a relative
reduction of 22% is observed when using H2-rich fuel instead of NG. However, the remaining surge margin would
still be sufficient. Operation of the engine with clean syngas at the same operating setup (i.e. CI) results in a
relative reduction of the surge margin by 50%, which brings the compressor close to the surge limit.
αIGV
MControl
PR
TOT
Fuel
ṁf
Ballast
VIGV
Mc1
Me
Mc2
Compressor
Expander
Mc3
Air
Exhaust gas
To ASU
Fig.5. Different gas turbine modifications to reduce fuel change effects
In remedy of the main problems caused by switching from NG to clean syngas (and to some extent to H2-rich
syngas), such as unstable operation and reduced lifetime of the turbine blades, various options might be
considered. These options are illustrated in Fig. 5 and described below.
Modification of the control rules
Running the GT safely under different combinations of VIGV, pressure ratio, TIT and TOT is usually an appropriate
option. In order to solve the problem of increased pressure ratio and reduced surge margin when the GT is
operated on H2-rich and clean syngas, the second and third parameter setups, i.e. Case II and III (refer to Table 4)
are considered. The corresponding results to the fuel change effect on various performance indicators using
aforementioned simulation setups are shown in Fig.6.
Keeping the pressure ratio at its design value (i.e. 18.2) and VIGV fully open results in significant
reduction of TIT and thereby GT efficiency drop, which is more pronounced for Case CII compared to that for Case
BII. As shown in Fig.6, the gross power output for both H2-rich syngas and clean syngas operations decreases for
parameter setup II. The gross power outputs of Case BII and Case CII are reduced by 8.6 and 57.1 MW,
respectively compared to that of NG-fired GT (Case A). The negative effect of this parameter setup is not limited to
the GT block alone. The TOT also significantly drops for Case BII (533 °C) and Case CII (522 °C) compared to
Case A (577 °C) leading to lower overall plant efficiency due to the reduced steam cycle efficiency and power
output.
8
1.1
Inlet air flow
TOT
TIT
GT efficiency
PR
SM
5
Gross power
Fuel flow
4.5
1
4
3.5
3
0.8
2.5
0.7
2
Relative fuel value
Relative value
0.9
1.5
0.6
1
0.5
0.5
0.4
0
Case A
Case BII
Case BIII
Case CII
Case CIII
Fig.6. Effect of fuel change on relevant performance parameters (at fixed PR)
Parameter setup III (fixed TIT and PR), which is considered to see the effect of VIGV position on the performance
of the GT, shows no improvement. It is also evident that such a setup is not useful to keep the necessary margin
to the surge condition (refer to Case BIII and Case CIII in Fig.6). Compared to Case A, the surge margin is reduced
by 30% and more than 50% even though the VIGV is closed by 9% and 20% for cases BIII and CIII, respectively. In
addition, TOT increases considerably at setup III for both H2-rich syngas and clean syngas by 4 and 20 °C,
respectively.
To avoid unstable operational conditions, to reduce power output, and to eliminate the risk of reduced
lifetime of the expander last stage blades in Case C (clean syngas), the parameter setups IV and V (refer to Table
4) are also considered. The TOT of the GT is fixed at its design value for NG-fired case (i.e., 577 °C). It should be
noted that simulation results for Case BIV and Case BV are not shown in Fig.7 since none of them offer better
performance compared to BI. However, the results relevant to clean syngas operation are shown in Fig.7.
Obviously, parameter setup IV would be a better solution due to the increased surge margin and power output
compared to setup V. Although, the gross power output of the GT increases for Case CIV compared to Case A, the
surge margin is still very low and shows 42% reduction compared to the NG-fired GT.
1.1
Inlet air flow
TIT
PR
Gross power
TOT
GT efficiency
SM
Fuel flow
5
4.5
1
4
3.5
3
0.8
2.5
0.7
2
Relative fuel value
Relative value
0.9
1.5
0.6
1
0.5
0.5
0.4
0
Case A
Case CIV
Case CV
Fig.7. Effect of fuel change on relevant performance parameters (at fixed PR and TOT)
Modification of the compressor flow path
In order to reduce the air mass flow, while keeping the pressure ratio close to the design value, the following
alternatives (shown as Mc in Fig.5) could be considered:
9
Mc1: Modification of the first compressor stator vanes to reduce the air mass flow. This modification would also
change the last stage height rather extensively. Moreover, a ballast flow (N2 or steam) has to be injected when H2rich syngas is used as fuel (which has a higher LHV compared to non-captured syngas) into the combustor. The
flow of the ballast needs to be reduced, when the LHV of the fuel gas decreases and the syngas flow increases
(e.g. at clean syngas operation).
Mc2: The addition of one or more rear stages to the compressor with adapted clearances to allow the reduction of
the compressed air mass flow rate when the pressure rises. An intrinsic internal flow control would be required to
maintain the high pressure while the mass flow reduces. However, it would result in reduced compression
efficiency and consequently in reduced GT efficiency. In general, this option would result in a lower penalty to the
GT efficiency and stable operating conditions. However, this requires major modifications to the compressor and
expander as well as adjustment of the vanes’ and blades’ cooling paths.
Mc3: In order to compensate for the increased fuel mass flow when a syngas with low calorific value is used, a
fraction of the compressed air could be discharged at the compressor outlet. A loss of efficiency is foreseeable if
an internal use of compressed air is not considered. The most efficient use of the bleed stream is in the ASU. The
integration of the GT compressor and ASU would reduce the power demand of the main air compressor and
slightly increase the overall plant efficiency. However, as mentioned previously, this integration has not been
considered within the H2-IGCC project as it results in reduced plant availability.
Modification of the expander flow path
The other option to reduce the effect of fuel change is to modify the expander. The expander nozzle guide vanes
(NGVs) could be re-staggered (shown as Me in Fig.5) to increase the swallowing capacity of the expander. The
vane and blade cooling paths should be modified accordingly. However, this might also result in a reduction of the
peak efficiency. Adoption of such an option reflects the fact that industry prefers modifications to the expander
side. Nevertheless, extensive modifications to the expander should be also avoided as it will be costly.
Other secondary alternatives
In order to compensate for the increased clean syngas flow, supplementary firing of a portion of the fuel could be
considered to increase the total power output of the entire plant. It means in the case of trip of CCS unit, excess
cleaned syngas not needed in the GT could be used for supplementary firing leading to increased overall power
output of the combined cycle. Another option to combust the extra fuel together with some air blown off at the
compressor exit is the utilization of a second expander. Nevertheless, techno-economic evaluations are required
to predict the penalties imposed by the use of a second combustor or expander. The results will mainly depend on
the expected operating hours of this additional unit, as well as fuel costs.
5. CONCLUSIONS
The effect of fuel change (i.e. from NG to H2-rich syngas and clean syngas) on the selected GT was reported in
this paper. Based on the results, focusing only on the gas turbine as a stand-alone unit, operation with H2-rich fuel
is feasible if a reduced surge margin would be acceptable. The clean syngas operation (in non-capture IGCC
plant) results in significantly lower surge margin and higher turbine outlet temperature compared to the reference
case especially at off-design conditions, which requires engine modification. Results confirm that running the
engine with H2-rich fuel without significant changes compared to the NG-fired engine can be carried out from a
turbomachinery point of view. However, it should be noted that the challenges concerning pre-mixed combustion
of the H2-rich fuel and different heat transfer rate to the expander materials when operated with H2-rich fuel are not
covered in this paper.
When operating with a fuel with low calorific value, such as clean syngas, expected operational hours are
very important for the selection of appropriate operating conditions or modification options. Although several
modification options as well as operating strategies have been suggested in case of clean syngas operation,
reduced efficiency and compressor stability range could be tolerated for limited operational hours with clean
syngas. Nevertheless, the best combination of the GT power output and efficiency has to be searched for, taking
into account the whole IGCC performance as well as investment and maintenance costs, when clean syngas
operation is one of the requirements of the plant.
Moreover, it was demonstrated that altered VIGV angle doesn’t provide acceptable surge margin for the
selected GT. In order to have only minor modifications to the GT for clean syngas operation compared to the
design case, decreasing the TIT and maintaining similar TOT as for the reference case (NG-fired GT) with fully
open VIGV could be a plausible option. However, results showed significant reduction of efficiency and power
output for this specific option. It should be highlighted that using clean syngas requires major modifications to the
GT including additional compressor stages, air bleed from compressor outlet, and re-staggering of the expander
nozzle wanes. Therefore this option was omitted from the list of possible alternatives for the H2-IGCC project.
ACKNOWLEDGMENT
The authors are grateful to the European Commission’s Directorate-General for Energy for financial support of the
Low Emission Gas Turbine Technology for Hydrogen-rich Syngas (H2-IGCC) project. The authors also wish to
acknowledge Han Raas at Vattenfall for performing the gasification simulations. Constructive discussions and data
exchange with Professor G. Cerri and his group at Roma Tre University are also acknowledged.
10
NOMENCLATURE
A
cross-sectional area
AGR
acid gas removal
ASU
air separation unit
CCS
carbon capture and sequestration
CO
carbon monoxide
COS
carbonyl sulfide
CO2
carbon dioxide
cp
specific heat
GHG greenhouse gas
GT
gas turbine
HRSG heat recovery steam generator
H2
hydrogen
H2S
hydrogen sulfide
IGCC integrated gasification combined cycle
LHV
lower heating value
LNG
liquefied natural gas
LP
low pressure
MAC
main air compressor
ṁ
mass flow rate
NG
natural gas
NOx
nitrogen oxide
N2
nitrogen
OEM original equipment manufacturer
O2
oxygen
P
power
p
pressure
PGAN pure gaseous nitrogen
PR
pressure ratio
R
gas constant
SCGP Shell Coal Gasification Process
SOA
state-of-the-art
SOx
sulfur oxide
SWGS sour water-gas shift
T
temperature
TIT
turbine inlet temperature
TOT
turbine outlet temperature
VIGV variable inlet guide vane
Greek Letters
γ
specific heat ratio
κ
choked flow coefficient
η
efficiency
Subscripts
i
expander inlet
o
expander outlet
t
total condition
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
IEA, Key world energy statistics. International Energy Agency, 2013: Paris, France.
BP, Statistical review of world energy, 2013, British Petroleum Company.
BP, Energy outlook 2030, 2012, British Petroleum Company.
ExxonMobil, The outlook for energy: A view to 2040, 2013, ExxonMobil Company.
IEA, World energy outlook. International Energy Agency, 2013: Paris, France.
IEA, IEA statistics 2013 edition: CO2 emissions from fuel combustion highlights. International
Energy Agency, 2013: Paris, France.
Kanniche, M., Bouallou, C, CO2 capture study in advanced integrated gasification combined cycle.
Applied Thermal Engineering, 2007. 27: p. 2693-2702.
Zheng, L., Furinsky, E, Comparison of Shell, Texaco, BGL and KRW gasifiers as part of IGCC plant
computer simulations. Energy Conversion & Management, 2005. 46: p. 1767-1779.
Sipöcz, N., Mansouri, M, Breuhaus, P, Assadi, M. Development of H2-rich syngas fuelled GT for
future IGCC power plants – establishment of a baseline. in ASME paper GT2011-45701. 2011.
ASME Turbo Expo, Vancouver, Canada.
11
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
Carbo, M.C., Jansen, D, Boon, J, Dijkstra, JW, Van Den Brink, RW, Verkooijen, AHM, Staged
water-gas shift configuration: Key to efficiency penalty reduction during pre-combustion
decorbonisation in IGCC. Energy Procedia, 2009. 1: p. 661-668.
Kanniche, M., Gros-Bonnivard, R, Jaud, P, Valle-Marcos, J, Amann, JM, Bouallou, C, Precombustion, post-combustion and oxy-combustion in thermal power plant for CO2 capture. Applied
Thermal Engineering, 2010. 30: p. 53-62.
Chiesa, P., Lozza, G, Mazzocchi, L, Using hydrogen as gas turbine fuel. Journal of Engineering for
Gas Turbines and Power, 2005. 127: p. 73-80.
Romano, M.C., Chiesa, P, Lozza, G, Pre-combustion CO2 capture from natural gas power plants,
with ATR and MDEA processes. International Journal of Greenhouse Gas Control, 2010. 4: p. 785797.
IEA, Improved oxygen production technologies. IEA Greenhouse Gas R&D Programme.
International Energy Agency, 2007: United Kingdom.
Bancalari, E., Chan, P, Diakunchak, I.S. Advanced hydrogen gas turbine development program. in
ASME paper GT2007-27869. 2007. ASME Turbo Expo, Montreal, Canada: Siemens Power
Generation, Inc.
Amick, P., Geosits, R., Herbanek, R., Kramer, S., Tam, S., A large coal IGCC power plant, in 19th
Annual International Pittsburgh Coal Conference2002.
Nag, P., LaGrow, M, Wu, J, Abou-Jaoude, K,Engel, J, LNG fuel flexibility in Siemens’ land-based
gas turbine operations, in Electric Power Conference2007, Siemens Power Generation, Inc.:
Chicago, Illinois, USA.
Chishty, W.A., Fuel flexibility effects on gas turbine operation, in Industrial Application of Gas
Turbines (IAGT)2010: Hamilton, Ontario, Canada.
McMillan, R., Marriott, D, Su, R.H, Fuel flexible gas turbine cogeneration, in Power-Gen Asia2008,
Siemens AG: Kuala Lumpur, Malaysia.
Rahm, S., Goldmeer, J, Moliere, M, Eranki, A, Addressing gas turbine fuel flexibility- GE white
paper GER4601 (06/09), in POWER-GEN Middle East conference2009: Manama, Bahrain.
Jones, R., Goldmeer, J., Monetti, B., Addressing gas turbine fuel flexibility- GE white paper
GER4601 (05/11) revB. 2011.
Kim, Y.S., Lee, J.J, Cha, K.S, Kim, T.S, Sohn, J.L, Joo, Y.J. Analysis of gas turbine performance in
IGCC plants considering compressor operating condition and turbine metal temperature. in ASME
paper GT2009-59860. 2009. ASME Turbo Expo, Orlando, Florida, USA.
Gas turbine fuel flexibility for zero emission power plants, 2007, European Turbine Network
(Position Paper).
Low emission gas turbine technology for hydrogen-rich syngas Under the 7th Framework
Programme FP7-239349. Available from: www.h2-igcc.eu.
Mansouri Majoumerd, M., De, S, Assadi, M, Breuhaus, P, An EU initiative for future generation of
IGCC power plants using hydrogen-rich syngas: Simulation results for the baseline configuration.
Applied Energy, 2012. 99: p. 280-290.
Gazzani, M., Macchi, E, Manzolini, G, CO2 capture in integrated gasification combined cycle with
SEWGS – Part A: Thermodynamic performances. Fuel, 2013. 105: p. 206-219.
Mansouri Majoumerd, M., Raas, H, De, S, Assadi, M, Estimation of performance variation of future
generation IGCC with coal quality and gasification process – Simulation results of EU H2-IGCC
project. Applied Energy, 2014. 113: p. 452-462.
Enssim®, 2009, Enssim Software: Doetinchem, The Netherlands.
Aspen Plus version 7.1, 2009, Aspen Technology Inc.: Cambridge, MA, USA.
IPSEpro version 4.0, 2003, Simtech Simulation Technology (Simtech): Graz, Austria.
Siemens, Siemens gas turbine SGT5-4000F, 2008, Siemens AG: Erlangen, Germany.
12
Paper V
Techno-economic evaluation of an IGCC power plant with
carbon capture
Mohammad Mansouri Majoumerd, Mohsen Assadi, Peter
Breuhaus
Presented at ASME Turbo Expo 2013, San Antonio, Texas,
USA, June 2013
175
Proceedings of ASME Turbo Expo 2013: Turbine Technical Conference and Exposition
GT2013
June 3-7, 2013, San Antonio, Texas, USA
GT2013-95486
TECHNO-ECONOMIC EVALUATION OF AN IGCC
POWER PLANT WITH CARBON CAPTURE
Mohammad Mansouri Majoumerd1, Mohsen Assadi1,2, Peter Breuhaus2
1
2
Faculty of Science and Technology
International Research Institute of Stavanger (IRIS)
University of Stavanger
Postbox 8046
4036 Stavanger, Norway
4068 Stavanger, Norway
Abstract
Nomenclature
Most of the scenarios presented by different actors and
organizations in the energy sector predict an increasing
power demand in the coming years mainly due to the world’s
population growth. Meanwhile, global warming is still one of
the planet’s main concerns and carbon capture and
sequestration is considered one of the key alternatives to
mitigate greenhouse gas emissions. The integrated
gasification combined cycle (IGCC) power plant is a coalderived power production technology which facilitates the
pre-combustion capture of CO2 emissions.
AGR
ASU
BEC
BOP
CCF
CCS
CEPCI
CF
COE
CO
CO2
EPCC
After the establishment of the baseline configuration of
the IGCC plant with CO2 capture (reported in GT201145701), a techno-economic evaluation of the whole IGCC
system is presented in this paper. Based on publicly available
literature, a database was established to evaluate the cost of
electricity (COE) for the plant using relevant cost scaling
factors for the existing sub-systems, cost index, and financial
parameters (such as discount rate and inflation rate).
Moreover, an economic comparison has been carried out
between the baseline IGCC plant, a natural gas combined
cycle (NGCC), and a supercritical pulverized coal (SCPC)
plant.
GHG
GT
HHV
HRSG
H2
H2O
IGCC
LHV
NG
NGCC
NOAK
OECD
The calculation results confirm that an IGCC plant is
180% more expensive than the NGCC. The overall efficiency
of the IGCC plant with CO2 capture is 35.7% (LHV basis),
the total plant cost (TPC) is 3,786 US$/kW, and the COE is
160 US$/MWh.
O2
O&M
SCGP
SCPC
ST
SWGS
1
Acid gas removal
Air separation unit
Bare erected cost
Balance of plant
Capital charge factor
Carbon capture and sequestration
Chemical Engineering Plant Cost Index
Capacity factor
Cost of electricity
Carbon monoxide
Carbon dioxide
Engineering, procurement, and construction
cost
Greenhouse gas
Gas turbine
Higher heating value
Heat recovery steam generator
Hydrogen
Water
Integrated gasification combined cycle
Lower heating value
Natural gas
Natural gas combined cycle
nth-of-a-kind
Organization for Economic Co-operation
and Development
Oxygen
Operation and maintenance costs
Shell coal gasification process
Super critical pulverized coal
Steam turbine
Sour water-gas shift
Copyright © 2013 by ASME
TASC
TOC
TPC
TS&M
Total as-spent cost
Total overnight cost
Total plant cost
Transport, storage, and monitoring
water-gas shift and the capture system, there is currently no
full-scale IGCC power plant built with CCS.
Therefore, further investigation of the various methods
to increase the availability of the system, to evaluate the
associated operational risks, and to reduce the efficiency
penalty and the cost of CO2 avoidance in IGCC plants is
necessary.
1. Introduction
The global financial situation during the past few years has
resulted in falling investments in the power sector, especially
in OECD (Organization for Economic Co-operation and
Development) countries [1]. However, global energy demand
is steadily increasing, mainly due to the world’s population
growth. A greater part of this increasing demand comes from
non-OECD countries. It is assumed that electricity generation
will be the largest source of primary energy consumption in
the coming decades [2]. Among various fossil fuels, coal had
the fastest consumption growth rates in 2011 compared to
2010 [3]. According to the International Energy Agency’s
“new policies” scenario, coal consumption in the year 2035
will have a 25% increase compared to 2009. However, based
on the “current policies” scenario, this increase will be 65%
compared to the level of 2009 [4]. The main reasons for this
are the abundant resources of coal (more than 100 years with
current proved reserves) and its widespread availability
compared to the other fossil fuels [3, 5-7].
Besides technical issues, economic figures play a major
role in the commercialization of a technology. The
commercial investment in an IGCC plant requires
competitive cost of electricity (COE) compared to other
competing power generation technologies [13]. To find out
the COE for the IGCC plant, a techno-economic investigation
is vital. However, the economic analysis of this plant is more
complicated than for other power generation technologies
e.g. natural gas combined cycle (NGCC), because of the
large number of components used in an IGCC plant.
Moreover, there are several alternative sub-systems to select,
e.g. for the gasification block (e.g. Shell, GE, Siemens, etc.)
with their own specific costs and characteristics. Therefore,
an IGCC plant is not considered as a standardized
commercial technology with well-established costs [14].
Moreover, other factors such as market situation, fuel price,
CO2 allowances cost and currency fluctuations increase the
level of uncertainty in cost estimation.
The necessity to abate greenhouse gas (GHG) emissions
will result in the deployment of carbon capture and
sequestration (CCS) in the power production sector. CCS
will play a key role in curbing CO2 emissions, according to
the European Energy Roadmap 2050 [8]. Moreover, the
deployment of CCS in coal-fired power production will
ensure that coal will have its share in fossil fuel consumption
in future years with more restricted emissions’ regulations,
even though the CO2 emissions from coal combustion are
significantly higher than those for natural gas (NG).
Though the prediction of the investment costs for the
IGCC plant is a difficult task, this study attempts to provide a
rough estimate in order to illustrate the plant’s economic
status. The objectives of this study are to analyze the
economic indicators of the IGCC plant with CO2 capture and
to compare them with other fossil fuel power generation
technologies, i.e. NGCC and supercritical pulverized coal
plant (SCPC) with CCS. Since the emphasis in this work has
been dedicated to the economic evaluation of the selected
IGCC plant, the detailed description of thermodynamic
modeling is not repeated here. The description of simulations
using Enssim software [15], ASPEN Plus [16], and IPSEpro
[17] could be found in references [12, 18].
During the last two decades, the integrated gasification
combined cycle (IGCC) has been assumed to be one of the
most attractive coal-based technologies for generating
electricity from coal in terms of low environmental impact
[9-11]. Although each of the major sub-systems of the IGCC
system has been widely used in industrial applications, their
integration in the IGCC plant is not well-matured yet, and
such a plant is considered to be complex from the plant
owner’s perspective. This complexity may prevent
investments in IGCC plants due to higher risk for low
availability and consequently higher cost of electricity
compared to other fossil fuel-based power technologies.
2. IGCC configuration
The integrated gasification combined cycle plant with CO2
capture (refer to Fig. 1) comprises the following sub-systems:
 Cryogenic air separation unit (ASU) to provide O2
for the gasification process and N2 as the conveying
gas.
 Shell coal gasification process (SCGP) to produce
syngas from coal using O2 and intermediate pressure
steam [19].
 Sour water-gas shift (SWGS) reaction unit to
convert CO and steam to CO2 and H2 using
exothermic catalytic reaction [20].
Incorporating a water-gas shift reaction unit into the
IGCC plant facilitates the pre-combustion capture of CO2
[11, 12]. However, due to the associated efficiency penalty
imposed by CO2 capture and the high investment cost of
2
Copyright © 2013 by ASME
 Acid gas removal (AGR) unit to remove the H2S
content of the syngas using physical solvent
(SELEXOL) [18].
 CO2 capture unit to remove the CO2 content of the
syngas.
 CO2 compression and dehydration unit to ensure the
final CO2 conditions for transport and storage.
 Gas turbine (GT) to produce electricity by
combustion of H2-rich syngas [18].
 Heat recovery steam generator (HRSG) to utilize the
energy in the hot exhaust gas from the GT for steam
and electricity production.
 Steam turbine (ST) to produce electricity from steam
produced in the HRSG.
The main technical specifications for simulation of the whole
IGCC system are shown in Table A.1 (Appendix I).
Reference [18] contains further technical assumptions.
Fig. 1: The schematic figure of the IGCC power plant with CO2 capture
3) Operation and maintenance (O&M) costs
3.1) Fixed costs e.g. labor cost
3.2) Variable costs e.g. chemicals, solid waste
handling, CO2 emissions cost, etc.
4) CO2 transport, storage and monitoring costs
3. Plant economics
The methodology used for cost estimation for the IGCC
plant is based on the “Quality Guidelines for Energy System
Studies” by the United States’ National Energy Technology
Laboratory (NETL) [21]. An estimation of the COE for an
IGCC plant with CCS comprises the following items:
The following sub-sections describe the aforementioned cost
components.
1) Capital costs
2) Fuel cost
3
Copyright © 2013 by ASME
3.1. Capital costs
EPCC
The capital costs have been defined based on the following
five different cost levels:
The EPCC for all major components of the IGCC plant are
about 8% of the BECs of the corresponding components
[22].
1) Bare erected cost (BEC): This cost comprises
process equipment items, supporting facilities
(e.g. labs, roads, etc.), and the direct and indirect
labor required for the construction and installation
of equipment items.
2) Engineering, procurement and construction cost
(EPCC): This cost comprises the BEC plus the
cost of the engineering, procurement and
construction services.
3) Total plant cost (TPC): This cost comprises the
EPCC plus project and process contingencies.
4) Total overnight costs (TOC): This cost comprises
the TPC plus owner’s costs.
5) Total as-spent cost (TASC): This cost is the sum
of all capital expenditures as they are incurred
during the capital expenditure period including
their escalation. TASC also includes interest
during construction.
TPC
The project and process contingencies to calculate TPC are
assumed to be 18% and 5% of the BECs, respectively [22].
TOC
The sum of the TPC and the owner’s costs is the total
overnight cost of a plant. The assumptions for owner’s costs
are shown in Table 1 below.
Table 1: Assumptions for owner’s costs
Parameter
Pre-production costs
BEC
The estimation of the BEC for the major components of the
IGCC plant (except the gas turbine) is derived from the
detailed study of NETL [22].
Inventory capital
The cost of the gas turbine is derived from the 2009
Gas Turbine World Handbook [23]. Since the GT in this
study has some add-on options such as a new burner design
to combust H2-rich fuel (instead of natural gas), new cooling
air flow design, and sophisticated materials for the expander
section, a 15% cost increase is assumed here. Although a GT
with these characteristics is not available on the market, the
cost for a mature nth-of-a-kind (NOAK) GT was considered.
Initial cost for catalyst
and chemicals
Plant site area
Other owner’s costs
Financing costs
The overnight cost of a component ( ) with specific
size ( ) based on a reference component ( ) with reference
size ( ) is shown by the following Eq. 1:
a
Comment
6 months’ operating labor costs
1 month’s maintenance materials
cost at 100% CFa
1 month non-fuel consumables at
100% CF
1 month waste disposal
25% of 1 month fuel cost at 100%
CF
2% of TPC
60-day supply of fuel and non-fuel
consumables at 100% CF
0.5% of TPC (spare parts)
US$7413/hectare for a land
(greenfield without seismic
consideration) with area of 121
hectares
15% of TPC
2.7% of TPC
Capacity factor
TASC
(Eq. 1)
This cost vary based on the capital expenditure period and
the financing scenario (for further information refer to [22]).
The interests during the construction period are included in
the TASC. The ratio of TASC/TOC for the IGCC plant with
CO2 capture for five years of construction is set to 1.140.
where is the number of equally sized equipment trains
operating at 100%/n,
is the cost scaling exponent for
multiple trains of the component, and is the cost scaling
factor. Adjustments of costs (except for the GT) have been
carried out using the Chemical Engineering Plant Cost Index
(CEPCI) of April 2012 [24]. The graph of the simple cycle
price change available in reference [25] has been used for
the fluctuation of the gas turbine price.
3.2. Other assumptions
It should be noted that all costs are limited to “within the
fence line”, except the costs of CO2 transport, storage and
monitoring (TS&M). Other economic assumptions
employed for the cost estimation are shown in Table 2.
4
Copyright © 2013 by ASME
sensitivity analysis, illuminating the impact of cost/price
variations. Finally, results of economic evaluation for
various power plants i.e. IGCC, NGCC, and SCPC plants
are presented and discussed.
Calculation of the cost of electricity is based on the
following Eq. 2:
(Eq. 2)
where
is the capital charge factor,
is the sum of
all fixed annual operating costs,
is the sum of all
variable annual operating costs, and
is annual net
megawatt-hours of power generated at 100% capacity factor
(CF).
4.1. Thermodynamic results
The performance indicators of the IGCC plant with CO2
capture are given in Table 3.
Table 3: Performance indicators of the IGCC power plant with
carbon capture
Table 2: Economic assumptions
Parameter
Coal price1,2
NG price3
IGCC capacity factor
NGCC capacity factor
SCPC capacity factor
NGCC net power output
SCPC net power output
NGCC overall efficiency
SCPC overall efficiency
NGCC CO2 capture rate
SCPC CO2 capture rate
NGCC plant site area
SCPC plant site area
TASC/TOC for NGCC
TASC/TOC for SCPC
CCF4 for IGCC and SCPC
CCF for NGCC
CO2 allowances cost5
CO2 TS&M for IGCC6
CO2 TS&M for NGCC
CO2 TS&M for SCPC
Labor work
Operating labor cost7
Discount rate
Inflation rate
Real escalation rate
Value
110
32.79
80
85
85
473.6
550.0
47.5
29.5
90.7
90.2
1/3
1
1.078
1.140
0.124
0.111
7.36
4.3
3.2
5.6
50
31.18
12
3
0
Unit
US$/tonne
US$/MWh
%
%
%
MW
MW
% LHV
% LHV
%
%
of IGCC land
similar to IGCC
€/tonne of CO2
US$/MWh
US$/MWh
US$/MWh
h/work.week
US$/h.capita
%
%
%
Performance indicators
GT shaft power output
ST shaft power output
Generator power output
ASU compression power demand
Gasification power demand
Syngas cleaning compression and pumping demand
Syngas cleaning refrigeration power demand
CO2 compression power demand
HRSG pumping power demand
Total auxiliary power demand
Net power output
Overall IGCC efficiency (%LHV)
MW
324.0
177.4
501.4
49.8
5.0
10.9
9.2
20.8
3.4
99.1
402.2
35.7
Fig. 2 shows the share of auxiliary power demands for the
IGCC plant. Simulation results confirm that the ASU has the
largest auxiliary power demand with 50.3% of the total
auxiliary power demand.
Gasification power demand
4 % 5 %
ASU compression power
demand
21 %
AGR pumping and
compression power demand
9 %
50 %
1
Coal price is based on API N2, 6000 kCal NAR (CIF) [26].
The USD to Euro exchange rate is 0.7527 [27].
3
NG price is based on Zeebrugge price [26].
4
Capital charge factor
5
The cost of CO2 emissions is based on European Union emission
trading scheme [26].
6
This cost is derived from reference [22], adjusted using CO2
captured and assumed to be constant and free of fluctuations
(from 2007 to 2012) due to the technology improvement. The
captured CO2 is transported 80 km.
7
Labor cost have been updated based on EU Labor Cost for EU27 countries from European Commission reports [28, 29].
2
11 %
AGR refrigeration power
demand
CO2 compression power
demand
HRSG pumping power demand
Fig. 2: The share of auxiliary power consumptions
4.2. Cost estimations
The results of the estimation of overnight costs for major
components of the IGCC plant with CO2 capture using
(Eq.1) and reference [22] are shown in Table 4. It should be
mentioned that the uncertainty of this type of cost estimation
is about ±30%. It is worth noting that the information given
in Table 4 is based on June 2007 US$, without incorporating
4. Results and discussion
In this section, results of thermodynamic simulation are
presented followed by results of the plant’s economic and
5
Copyright © 2013 by ASME
has been adjusted to April 2012. Moreover, the GT price
which is given in Table 4 is the price for 2009. As it was
mentioned in Section 3.1, all of these costs have been
adjusted using relevant cost indexes.
the installation labor cost. Given the same labor cost for the
installation of a component with two different sizes, this cost
has been considered to be the same as in reference [22].
However, using the US Bureau of Labor Statistics (BLS),
data for chemical manufacturing, the installation labor cost
Table 4: Capital costs for major components of the IGCC power plant
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Plant component
Coal handling, preparation and feed
Gasifier and accessories
ASU
Gas clean-up
CO2 compression and drying
GT1
HRSG, ducting and stack
ST generator, condenser, aux.
Cooling water, aux.
Feed water and miscellaneous BOP
Ash and slag handling
Instrumentation and control2
Building and structures
Other (improvements to site,
accessory electric plant)
Scaling parameter
Coal input (tonne/h)
Coal input (tonne/h)
O2 production (kmol/h)
Syngas flow rate (tonne/h)
Compression power (MWe)
GT net power (MWe)
ST gross power (MWe)
ST gross power (MWe)
ST gross power (MWe)
Ash content of the coal (tonne/h)
Plant net power (MWe)
Plant net power (MWe)
3
211040
211040
4642
575
30.2
464
209.4
209.4
209.4
21.1
496.9
496.9
0.67
0.67
0.67
0.67
0.67
0.67
0.67
0.67
0.67
0.67
1
0.67
0.67
3
16461
180256
173504
95090
17811
161436
161436
4004
439
20.8
39499
33914
19404
16678
20018
13026
6461
47953
324
177.4
177.4
177.4
20.2
402.2
402.2
13756
150635
154999
79346
13852
73373.6
31052
30344
17362
14922
19410
13026
5608
41622
1
The price for the GT was derived from Gas Turbine World handbook [23] and the reference GT for this study is Siemens/Ansaldo Energia V94.3A.
2
A similar cost has been used for instrumentation and control cost.
3
All costs (i.e. C and C) in US$×1000.
Table 5 shows results of the economic analysis for the IGCC
plant with CCS, based on the information given in Table 4
and assumptions described in Section 3.
plant owners, confirm the range of TPC reported in this
study.
Table 5: Various cost indicators for the IGCC plant
In order to evaluate the sensitivity of the economic results to
variations in the inputs, a sensitivity analysis has been
performed to identify the parameters which have the
strongest impact on the results. Results highlight possible
drivers which may influence market attention on this
technology.
Parameter
Total plant cost (TPC)
Total overnight cost
Total as-spent cost (TASC)
COE (base year-2012)
COE (1st year of operation-2017)
4.2.1. Economic sensitivity
Unit
US$*1000 US$/kW
1,523,051
3,786
1,881,257
4,677
2,144,633
5,332
US$/MWh
160
186
Table 6: Economic parameters and their variation range for
sensitivity analysis
Parameter
Coal price
Capacity factor
CO2 allowances cost
The TPC for the IGCC plant with CO2 capture has been
reported to be in the range of 1,414-2,513 in references [3032]. The range of COE has been reported to be 54-95.8
US$/MWh [30-32]. The above-mentioned reported figures
are far below the range of COE and TPC calculated in the
current study. This difference may be due to the different
coal price, power plant size, capacity factor and financial
assumptions used. However, the Electric Power Research
Institute (EPRI) report [33] on IGCC shows a TPC of 3,683
US$/kW, which is very close to the number reported in
Table 5. Moreover, personal communications with power
Absolute range
60 ‒ 160 US$/tonne
40 ‒ 90%
3.7 ‒ 11.0 €/tonne
Relative range
-45.5 ‒ 45.5%
-50 ‒ 12.5%
-50 ‒ 50%
Three parameters which are usually considered as the most
uncertain parameters for calculation of the COE were
selected. These parameters and their variation for sensitivity
analysis are shown in Table 6. The results of the sensitivity
analysis are presented in Fig. 3.
6
Copyright © 2013 by ASME
is the coal price, while the impact of CO2 allowances cost is
negligible. The minimal effect of the CO2 price is due to the
low absolute value of CO2 allowances cost. However, its
influence will change depending on the CO2 market
development and change in global mitigation policies.
80.0
70.0
COE Change (%)
60.0
50.0
40.0
30.0
20.0
4.2.2. Results of the comparative plant economics
10.0
0.0
‐10.0‐60.0
‐40.0
‐20.0
0.0
20.0
40.0
Using assumptions presented in Table 2, Fig. 4 shows
results of the comparative study of various plants’ capital
cost items.
60.0
‐20.0
Capacity Factor
Parameter Variation (%)
Fuel Cost
CO2 Allowances cost
The TOCs of IGCC, SCPC, and NGCC plants are 4677,
4065, and 1669 US$/kW, respectively. According to Fig. 4,
the highest capital cost is required for the IGCC plant. The
TOC and TASC for the IGCC plant are 15% higher than the
corresponding values for the SCPC plant (the basis for
comparison is the SCPC).
Fig. 3: Sensitivity response on COE for the IGCC plant
under variation of fuel cost, capacity factor, and CO2
allowances cost
It is evident from Fig .3 that the capacity factor has the
largest influence on the COE. The second ranked parameter
5331.7
TOC and TASC, US$/kW (2012$)
6000.0
4633.6
5000.0
4000.0
3000.0
1799.7
2000.0
1000.0
0.0
Total overnight
cost (TOC)
TASC
TOC
IGCC
TASC
TOC
SCPC
TASC
NGCC
Total as‐spent cost (TASC)
Owner's cost
Process contingency
Project contingency
Engineering, procurement, and construction cost (EPCC)
Bare erected cost (BEC)
Fig. 4: Plant capital costs for IGCC, SCPC, NGCC plants with CO2 capture
The TOC and TASC for the IGCC plant are 180% and
196% higher than corresponding values for the NGCC plant
(the basis for comparison is the NGCC). Even though there
is a large gap between the IGCC and the NGCC, the future
investment in the IGCC plant is very plausible due to the
security of energy supply.
Despite the higher cost for the IGCC plant compared to the
SCPC plant, investment in such a plant may be beneficial in
the coming years with more stringent environmental
regulations due to: a) advantages of gasification such as
easier control of gaseous pollutants, and b) advantages of the
pre-combustion capture, such as high CO2 concentration and
smaller equipment size because of high fuel gas pressure.
7
Copyright © 2013 by ASME
plant, as well as relevant assumptions, calculation methods,
and economic figures.
Results of the COE for the IGCC, SCPC, and NGCC plants
are shown in Fig. 5. The total COEs for IGCC, SCPC, and
NGCC plants are 160, 148, and 114 US$/MWh,
respectively. The COE for the IGCC plant is 8% and 41%
higher than COEs for SCPC and NGCC plants (bases for
comparisons are COEs of the SCPC and NGCC,
respectively).
The COE for the IGCC plant with CCS is 160
US$/MWh. It should be noted that all economic results are
strongly dependent on presented assumptions. A sensitivity
analysis was also carried out showing that the most
influential parameter on the COE was the capacity factor.
The fuel price was the second ranked parameter, while the
effect of CO2 allowances cost was negligible due to the low
cost of CO2 emissions. Finally, a comparative study was
carried out to highlight the cost difference between various
power generation technologies i.e. IGCC, SCPC, and NGCC
plants with CCS. The total overnight costs for IGCC, SCPC,
and NGCC are 4677, 4065, and 1669 US$/kW, respectively.
The results of the COE for the IGCC, SCPC, and NGCC are
160, 148, and 114 US$/MWh, respectively. Even though the
investment cost in the IGCC plant is more than double that
for the NGCC plant, the security of the energy supply may
encourage investors to select IGCC plants.
180.0
COE, US$/MWh (2012$)
160.0
140.0
120.0
82.8
67.7
24.9
100.0
80.0
60.0
40.0
5.8
3.2
13.4
10.4
18.0
11.0
76.6
51.3
44.1
20.0
3.2
5.6
4.3
0.0
IGCC
SCPC
Capital costs
Fixed costs
Fuel costs
CO2 TS&M
Moreover, it was shown that with the higher capacity
factor and CO2 allowances cost, which is plausible in the
coming years, the IGCC plant could attract more
investments compared to the SCPC plant. Furthermore,
income from poly-generation applications might also
improve the economic status of future IGCC plants.
NGCC
Variable costs
Fig. 5: COE for IGCC, NGCC, and SCPC plants with
CO2 capture
Acknowledgment
As mentioned before (refer to Fig. 3), one of the most
effective solutions for reducing the difference in COEs for
IGCC and SCPC plants is to increase the capacity factor of
both IGCC and SCPC plants so that they have a similar
value.
The authors are grateful to the European Commission’s
Directorate-General for Energy for financial support of the
Low Emission Gas Turbine Technology for Hydrogen-rich
Syngas (H2-IGCC) project.
The combination of this increase and the increased
CO2 allowances cost may improve investment in an IGCC
plant where coal is the most available fuel for power
production. The COE for both plants is equal when the
capacity factors of both plants and the CO2 allowances cost
are set to 90% and 74 €/tonne of CO2, respectively.
References
5. Conclusion
1.
Birol, F., The impact of financial and economic crisis
on global energy investment, 2009, IEA: G8 Energy
Ministers' Meeting.
2.
BP energy outlook 2030, 2012, British Petroleum
Company.
3.
BP Statistical Review of World Energy, 2012, British
Petroleum Company.
The aim of this paper was to carry out a techno-economic
analysis for a specific IGCC plant configuration with CO2
capture, and to compare the COE for this plant with
competing technologies which are NGCC and SCPC plants.
The main objective was to generate a database using
publicly available literature to calculate the COE for this
plant.
4.
IEA, World Energy Outlook 2011. International
Energy Agency, 2011: Paris.
5.
The initial part of the paper describes the configuration
of the IGCC plant. The second part of the paper describes
the methodology used for the economic evaluation of the
Chiesa, P., Consonni, S, Kreutz, T, Williams, R, Coproduction of hydrogen, electricity and CO2 from coal
with commercially ready technology. Part A:
Performance and emissions. International Journal of
Hydrogen Energy, 2005. 30: p. 747-767.
6.
Robinson, P.J., Luyben, WL, Integrated gasification
combined
cycle
dynamic
model:
H 2S
absorption/stripping, water-gas shift reactors, and CO2
8
Copyright © 2013 by ASME
absorption/stripping. Industrial & Engineering
Chemistry research, 2010. 49: p. 4766-4781.
Engineering for Gas Turbines and Power, 1999.
121(2): p. 295-305.
7.
Wall, T.F., Combustion processes for carbon capture.
Proceedings of the Combustion Institute, 2007. 31: p.
31-47.
8.
EU Energy Roadmap 2050, 2012, European Union,
European Commission.
21. NETL, Quality guidelines for energy system studiesCost estimation methodology for NETL assessments of
power plant performance, 2011, DOE/NETL2011/1455.
9.
22. NETL, Cost and performance baseline for fossil
energy plants, Volume 1: Bituminous coal and natural
gas to electricity, 2010, 2nd Revision, DOE/NETL2010/1397.
Kanniche, M., Bouallou, C, CO2 capture study in
advanced integrated gasification combined cycle.
Applied Thermal Engineering, 2007. 27: p. 2693-2702.
23. Gas Turbine World Handbook. Vol. 27. 2009: Pequot
Publishing.
10. Holt, N., Booras, G, Todd, D, A Summary of Recent
IGCC Studies of CO2 Capture for Sequestration, in
Gasification Technologies Conference2003: San
Francisco, CA, USA.
24. Chemical Engineering Plant Cost Index (CEPCI).
Chemical Engineering, August 2012. 126.
11. Zheng, L., Furinsky, E, Comparison of Shell, Texaco,
BGL and KRW gasifiers as part of IGCC plant
computer simulations. Energy Conversion &
Management, 2005. 46: p. 1767-1779.
25. Gas Turbine World Handbook. Vol. 29. 2011: Pequot
Publishing.
26. Energy statistics-week 17. 2012; Available from:
http://www.energymarketprice.com.
12. Sipöcz, N., Mansouri, M, Breuhaus, P, Assadi, M.
Development of H2-rich syngas fuelled GT for future
IGCC power plants – establishment of a baseline. in
ASME paper GT2011-45701. 2011. ASME Turbo
Expo, Vancouver, Canada.
27. Universal currency converter. 2012; Available from:
http://www.xe.com.
28. Labour cost index. 2012;
http://epp.eurostat.ec.europa.eu.
13. Rosenberg, W.G., Alpern, D.C., Walker, M.R.,
Deploying IGCC in this decade with 3 party covenant
financing, 2004: John F. Kennedy School of
Government, Harvard University.
Available
from:
29. Eurostat newsrelease- Euro indicators, 2012, Eurostat
press office, 134/2012.
30. Rubin, E.S., Chen, C., Rao, A.B, Cost and
performance of fossil fuel power plants with CO2
capture and storage. Energy Policy, 2007. 35: p. 4444–
4454.
14. Rosenberg, W.G., Alpern, D.C., Walker, M.R.,
Financing IGCC – 3party covenant, 2004: John F.
Kennedy School of Government, Harvard University.
31. Chen, C., Rubin, ES, CO2 control technology effects on
IGCC plant performance and cost. Energy Policy,
2009. 37: p. 915-924.
15. Enssim®, 2009, Enssim Software: Doetinchem, The
Netherlands.
16. ASPEN Plus version 7.1, 2009, Aspen Technology
Inc.: Cambridge, MA, USA.
32. Huang, Y., Rezvani, S, McIlveen-Wright, D,
Minchener, A, Hewitt, N, Techno-economic study of
CO2 capture and storage in coal fired oxygen fed
entrained flow IGCC power plants. Fuel Processing
Technology, 2008. 89: p. 916–925.
17. IPSEpro version 4.0, 2003, Simtech Simulation
Technology (Simtech): Graz, Austria.
18. Mansouri Majoumerd, M., De, S, Assadi, M, Breuhaus,
P, An EU initiative for future generation of IGCC
power plants using hydrogen-rich syngas: Simulation
results for the baseline configuration. Applied Energy,
2012. 99: p. 280-290.
33. Schoff, R., Marasigan, J, Coal fleet integrated
gasification combined cycle research and development
roadmap, 2011, Electric Power Research Institute
(EPRI).
19. The Shell coal gasification process for sustainable
utilisation of coal, 2006, Shell Global Solutions.
20. Chiesa, P., Consonni, S, Shift reactors and physical
absorption for low-CO2 emission IGCCs. Journal of
9
Copyright © 2013 by ASME
Appendix I:
Table A.1: The main technical specifications of the IGCC plant
Air separation unit
O2 purity: 95%
Main air compressor: Three-stage inter-cooled to 5.5 bar
Gaseous O2 compressor: Six-stage intercooled to 55 bar
Pure gaseous N2 compressor: Six-stage intercooled to 80 bar
Acid gas removal unit and CO2 capture
Solvent type: physical
Solvent: SELEXOL
Number of absorption stages: 2
Solvent inlet temperature to the absorber: 5 °C
CO2 capture rate: 90% (molar basis)
Feedstock properties
Type: Bituminous coal
HHV: 26195 kJ/kg
LHV: 25100 kJ/kg
CO2 compression and dehydration
Final pressure: 150 bar
Drying agent: Tri-ethylene glycol
Final water content in the CO2 stream: 20 ppm (mass)
Gasifier
Type: SCGP (O2-blown, entrained flow, dry-fed)
Pressure: 45 bar
Temperature: 1600 °C
Gas turbine
Sour water-gas shift reaction unit
Heat recovery steam generator
Pressure level: 140/43/4 bar
Superheating/reheat temperature: 530/530 °C
Reaction: CO
H O
CO
H
Compressor pressure ratio: 18.2
Firing temperature: 1440 °C
Compressor isentropic efficiency: 92.3%
Expander isentropic efficiency: 89.2%
GT outlet pressure (bar (total)): 1.08 bar
Electrical/mechanical efficiency: 99/99.5%
Inlet syngas temperature to the reactor: 250 °C
steam-to-CO ratio: 2.4
10
Copyright © 2013 by ASME
Paper VI
Techno-economic assessment of fossil fuel power plants with
CO2 capture ‒ Results of EU H2-IGCC project
Mohammad Mansouri Majoumerd and Mohsen Assadi
Published in International Journal of Hydrogen Energy, Vol. 39,
p. 16771-16784, September 2014
187
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Available online at www.sciencedirect.com
ScienceDirect
journal homepage: www.elsevier.com/locate/he
Techno-economic assessment of fossil fuel power
plants with CO2 capture e Results of EU H2-IGCC
project
Mohammad Mansouri Majoumerd a,*, Mohsen Assadi a,b
a
b
Faculty of Science and Technology, University of Stavanger, 4036 Stavanger, Norway
International Research Institute of Stavanger, Postbox 8046, 4068 Stavanger, Norway
article info
abstract
Article history:
In order to address the ever-increasing demand for electricity, need for security of energy
Received 25 June 2014
supply, and to stabilize global warming, the European Union co-funded the H2-IGCC
Received in revised form
project, which aimed to develop and demonstrate technological solutions for future gen-
6 August 2014
eration integrated gasification combined cycle (IGCC1) plants with carbon capture. As a part
Accepted 10 August 2014
of the main goal, this study evaluates the performance of the selected IGCC plant with CO2
Available online 8 September 2014
capture from a techno-economic perspective. In addition, a comparison of technoeconomic performance between the IGCC plant and other dominant fossil-based power
Keywords:
generation technologies, i.e. an advanced supercritical pulverized coal (SCPC2) and a nat-
Techno-economy
ural gas combined cycle (NGCC3), have been performed and the results are presented and
IGCC
discussed here. Different plants are economically compared with each other using the cost
Pulverized coal
of electricity and the cost of CO2 avoided. Moreover, an economic sensitivity analysis of
NGCC
every plant considering the realistic variation of the most uncertain parameters is given.
CO2 capture
Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.
Cost of electricity
Introduction
World total primary energy consumption was 12,470 Mtoe in
2012 [1]. Global population, global economy, energy-intensity
of the global economy, and living standard are the main
drivers of the world's energy demand [2]. Except for decreasing
energy-intensity (energy consumption per capita) [3], other
fundamental drivers of energy demand will grow continuously in the coming decades [4e6]. Nevertheless, improved
energy efficiency cannot outpace the effects of other drivers,
resulting in a growing energy demand over the coming decades [5e7].
Currently, about 37% of global primary energy is
consumed by electricity generation. In 2012 global electricity
generation stood at 22,126 TWh [8], with an annual average
growth rate of 3.0% from 1990 to 2012 [1]. Fossil fuel-based
electricity generation accounted for 68% of the total generation and coal, the most carbon-intensive fossil fuel, was the
largest contributor to the supply of electricity in 2012. Ever-
* Corresponding author. Tel.: þ47 453 91 926; fax: þ47 51 83 10 50.
E-mail addresses: [email protected], [email protected] (M. Mansouri Majoumerd).
1
Integrated gasification combined cycle.
2
Advanced supercritical pulverized coal.
3
Natural gas combined cycle.
http://dx.doi.org/10.1016/j.ijhydene.2014.08.020
0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
16772
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Nomenclature
AGR
ASU
BFD
BUA
CCS
CEPCI
COE
DCF
EBTF
EPCC
EU
FP7
GHG
GT
HHV
HRSG
IGCC
LHV
MEA
NGCC
NPV
O&M
SCGP
SCPC
SCR
SOTA
ST
SWGS
TDA
TDPC
TPC
acid gas removal
air separation unit
block flow diagram
bottom-up approach
carbon capture and storage
Chemical Engineering Plant Cost Index
cost of electricity
discounted cash flow
European Benchmarking Task Force
engineering, procurement, and construction
costs
European Union
Seventh Framework Programme
greenhouse gas
gas turbine
higher heating value
heat recovery steam generator
integrated gasification combined cycle
lower heating value
monoethanolamine
natural gas combined cycle
net present value
operation and maintenance
Shell Coal Gasification Process
supercritical pulverized coal
selective catalytic reduction
state-of-the-art
steam turbine
sour wateregas shift
top-down approach
total direct plant costs
total plant costs
increasing world demand for electricity represents the largest
driver of demand for primary energy consumption. Electricity
demand is projected to grow more rapidly than total energy
consumption over the next few decades [5,6]. In 2035, the
demand for electricity will be almost 70% higher than the
current demand [9].
The growing use of fossil-fuel power plants has resulted in
many environmental concerns over the past decade. The
power sector is identified as the single largest sector contributing to the emission of CO2, the most important greenhouse
gas (GHG4). Carbon dioxide emissions from the electricity and
heat supply sector were about 42% of total global CO2 emissions from fossil fuels in the year 2011 [10]. Minimizing the
negative effects of growing GHG emissions resulted in the
development of environmentally-friendly technologies for
electricity production. The share of renewable energies,
therefore, has been growing significantly, thanks to governmental supports and subsidies around the globe. However, the
estimated timescale for the complete transformation to
renewable resources is likely to be a substantial time away [11]
and fossil fuels are forecasted to steadily cover a major part of
the energy mix. Thus the development of suitable
technologies such as clean fossil fuel-based power technologies is urgently needed during this transition phase.
Natural gas (NG5) power generation offers less CO2 emissions
compared to coal-based systems. Increasing shale gas exploration and production in the United States has meant a shift in
the U.S. energy market towards higher natural gas and lower
coal consumption for the electricity generation sector [12]. This
shift resulted in more coal export from U.S., cheaper price of
coal in other continents, e.g. Europe, and consequently higher
coal consumption. In addition, wide geographical distribution,
abundant reserves, convenient transportation and storage of
coal are still maintaining the level of coal consumption in this
sector [13]. Nevertheless, in both coal- and NG-based power
plants, CO2 emissions need to be mitigated by means of carbon
capture and storage (CCS6) in order to achieve targeted global
GHG emissions [14]. The deployment of CCS in fossil-fired
power plants can prevent the sharp reduction of the fossilfuel consumption in the coming years with more restrictive
emissions regulations and higher renewable energy share.
The integrated gasification combined cycle is currently one
of the most promising technologies for the efficient use of
coal. IGCC technology benefits from its widely known environmental credentials such as low emissions of SO2 and NOx
[15]. Although this technology suffers from high capital costs
and is perceived to be more complex than other technologies
e.g. pulverized coal plants, its significantly better emissions
performance is interesting for future large-scale deployment
[16,17]. In addition, the IGCC technology offers the opportunity
for co-gasification of biomass, good performance with lower
grade coals and other feedstock [18], and the co-production of
H2 and electricity [19]. Moreover, IGCC technology is technically well suited to CO2 capture. If CCS becomes necessary for
the next generation of fossil-based power plants, precombustion carbon capture methods can be easily incorporated into the IGCC system. The additional cost due to the
capture unit will be significant, but probably lower than for
pulverized coal combustion systems [20].
With a distinct view towards the development of IGCC
technology, the European Union has sponsored the H2-IGCC
project under its Seventh Framework Programme (FP77). This
project aims to develop and demonstrate a complete design of a
burner for the combustion of hydrogen-rich syngas for future
generation IGCC plant with pre-combustion CO2 capture in 2014
[21]. In an effort to evolve the new generation of IGCC plant
configuration, the simulation sub-group of the H2-IGCC project
previously reported detailed simulation results for a baseline
configuration of the IGCC plant as developed in this project [22].
In another paper, subsequent simulation studies on the effects
of various gasification processes, as well as of coal quality on the
performance of the selected configuration, were reported [23].
In addition to favorable technical performance, the viability
of the IGCC technology strongly depends on the overall economic figures compared to other competing technologies. The
commercial investment in an IGCC plant requires a competitive cost of electricity (COE8) [24]. An IGCC plant is not
5
6
7
4
Greenhouse gas.
8
Natural gas.
Carbon capture and storage.
Seventh Framework Programme.
Cost of electricity.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
considered as a standardized commercial technology with
well-established costs compared to e.g. natural gas combined
cycle [25]. However, several researchers and energy organizations assessed the COE for power generation by IGCC technology [26e32]. In spite of massive economic assessments, the
breakdown of the costs and assumptions is clear and well
documented in only a few of them [27,28]. In addition, the
majority of open literature did not touch upon a consistent
comparison with other fossil fuel-based technologies. Moreover, the sensitivity of the calculated COE to the variations of
most important input parameters is missing in most of the
recently published studies. Such plausible changes may result
in improved competitiveness of the IGCC technology compared
to other fossil-based power generation technologies and
should be clearly addressed in economic evaluations.
The main purpose of this study is to present the results of a
techno-economic assessment of the selected IGCC technology
with carbon capture performed using a tool developed by the
simulation sub-group of the H2-IGCC project. Such a tool has
provided the opportunity to modify or change the input parameters during economic assessment. The assessment is
based on a practical flow-sheet and realistic technical/economic
performance indicators verified by the plant's operators. The
other objective is to make a consistent and reliable comparison
between the selected IGCC plant and other fossil fuel-based
competing technologies, i.e. an NGCC and an SCPC plant based
on the same sort of economic assumptions. The natural gas
plant was selected for its wide utilization during past years and
for its bright future in a low gas price regime due to the widespread unconventional gas production in coming years [11]. The
pulverized coal plant was selected since it has been the most
prevalent coal technology worldwide over a long period [17].
Moreover, an analysis based on a literature review has been
performed to collect realistic economic data of the aforementioned state-of-the-art (SOTA9) technologies as new-built plants
on a commercial scale. Calculations of the COE were performed
using a set of parameters to ensure that the comparison is made
in a consistent and fair way. All assumptions and sources of
data, more specifically those for economic assessment, have
been carefully gathered and listed in this work.
Note that the results for the IGCC plant are generally based
on the realistic performance of a series of SOTA components/
sub-systems. However, to achieve more realistic results, these
performance indicators have been justified and verified by the
operators of similar plants. The degree of confidence in the
presented results for the NGCC and SCPC is significantly higher
due to the larger number of plants in operation. Nevertheless,
the lack of full-scale capture plants in operation and the lack of
data validation should be assumed as sources of uncertainties
when considering the presented results for all plants.
i.e. an advanced SCPC power plant and an NGCC plant both
with and without CO2 capture. This section concisely presents
the selected IGCC plant configuration. Further details concerning the selected IGCC plant can be obtained from a previous study [22]. The configurations of the SCPC and NGCC
plants are also briefly described here.
IGCC configuration
The block flow diagram (BFD10) of the selected IGCC configuration with capture unit is shown in Fig. 1. The thermodynamic model of the selected IGCC power plant was based on
commercially available technologies representing various
sub-systems. A cryogenic air separation unit (ASU11) is
considered as a stand-alone unit generating O2 (95% purity)
from air supplied by an intercooled main air compressor for
the coal gasification. The gasification of the coal takes place in
an O2-blown, dry-fed, entrained-flow gasifier based on the
Shell Coal Gasification Process (SCGP12). A sour water-gas shift
(SWGS13) reaction unit is used to convert the CO content in the
raw syngas to CO2 by shifting the CO with steam over a catalytic bed. The acid gas removal (AGR14) unit is based on a
double-stage SELEXOL system for H2S removal in the first
absorption-regeneration stage and for CO2 capture with a rate
of 90% in the second absorption stage. The physical absorption was selected over the chemical, amine-based process due
to the high partial pressure of CO2 in the syngas downstream
of the SWGS unit. In the case of a non-capture IGCC plant, a
COS hydrolyzer is considered to convert the COS content of
the raw syngas to H2S which will be removed in the downstream AGR unit without CO2 capture stage.
The CO2 captured from the process (90% capture rate) is
pressurized by an intercooled compressor, aftercoooled, liquefied and finally pumped up to a final pressure (110 bar). A
dehydration unit using tri-ethylene glycol is considered (H2O
content in the captured CO2 line is less than 20 mg/kg) to
prevent the corrosive effects on the transport pipeline. The
gas turbine (GT15) block including compression, combustion,
and expansion generates electric power using a generator. Gas
turbine modeling has been performed using the characteristics and boundary conditions of a gas turbine which is
designed for the combustion of the H2-rich fuel produced from
upstream sub-systems. The heat recovery steam generator
(HRSG16) is based on a triple pressure level with reheat
(140 bar/530 C/530 C) and a steam turbine (ST17) is considered
to generate power from the steam produced at the HRSG.
Advanced SCPC configuration
As mentioned previously, the thermodynamic characteristics
of the advanced SCPC plant has been adopted from Ref. [28].
The BFD of the supercritical pulverized coal power plant is
Power plant configurations
10
The principal objective of this study is to present the results of
a techno-economic assessment of the selected IGCC plant
with CO2 capture. The final goal is to provide a fair and
consistent comparison with other competing technologies,
11
12
13
14
15
16
9
State-of-the-art.
16773
17
Block flow diagram.
Air separation unit.
Shell Coal Gasification Process.
Sour wateregas shift.
Acid gas removal.
Gas turbine.
Heat recovery steam generator.
Steam turbine.
16774
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Compressed CO2
Fuel
Raw
Gas
syngas cleaning
Gasification
O2
Air
H2S
removal
SWGS
CO2
capture
To atmosphere
Slag
ASU
Heat recovery steam generator
Stack
HP
Air
IP/LP
Gas turbine
Condenser
Fig. 1 e The block flow diagram of the selected IGCC plant with CO2 capture.
Steam/water to and from capture plant
Feedwater
heater
system
Air
Ammonia
Coal
Limestone
Oxidation air
Flue
gas
Electrostatic
precipitator
Pulverized coal boiler
Water
Coal and ash
handling
DeNOx plant
Flue
gas
Pre-heated air
Compressed CO2
CO2 product
FGD
CO2 Capture
Bottom ash
Fly ash
Effluent
HP
turbine
IP
turbine
LP
turbine
Gypsum
Condenser
Flue gas
Stack
Fig. 2 e The block flow diagram of the advanced SCPC plant with CO2 capture.
shown in Fig. 2. The plant consists of a steam turbine, steam
generator with coal bunker bay and central switch gear. The
steam cycle consists of a triple pressure level with reheat
(300 bar/600 C/620 C) with extraction points for regenerative
heating of feed water and condensate. The steam boiler is
based on the SOTA Doosan Babcock two-pass single reheat
BENSON boiler. The boiler is equipped with an SOTA combustion system comprising 30 Doosan Babcock low NOx axial
swirl burners and an in-furnace air-staging system for primary control of NOx emissions. A selective catalytic reduction
(SCR18) unit to control NOx emissions located between the
boiler's exit and the air heater inlet. In addition, electrostatic
precipitators and the desulphurization plant (wet limestone
base) are placed before the flue gas stack. The CO2 removal
18
Selective catalytic reduction.
unit is based on SOTA post-combustion capture technology
using chemical absorption with a capture rate of 90%. The
chemical solvent is based on a 30 wt% aqueous solution of
monoethanolamine (MEA19). Further details can be found in
Ref. [28].
NGCC configuration
The selected NGCC with CO2 capture unit is based on a heavy
duty F-class gas turbine, the Siemens SGT5-4000F, a 300 MW
single-shaft gas turbine as topping cycle [33]. This GT is
directly connected to a 50 Hz air-cooled generator running at a
fixed speed of 3000 rpm. Downstream of the GT is a triple
pressure level HRSG with reheat (120 bar/560 C/560 C). The
19
Monoethanolamine.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
16775
GT block has been modeled similarly to that presented for the
GT in the selected IGCC plant. However, compressor and
expander characteristics and the boundary conditions of the
GT are based on the original design, i.e. natural gas operation.
The BFD of the NGCC is illustrated in Fig. 3.
The CO2 capture unit is similar to the SCPC plant, an SOTA
post-combustion unit using standard MEA (30 wt%) with 90%
capture rate. In addition to the capture unit, an exhaust gas
condenser is considered in which the flue gas is cooled before
entering the capture unit. Moreover, a flue gas blower is
considered to compensate for pressure losses in the subsequent capture unit. In the current study, it is also assumed
that seawater at a temperature of 15 C is available to satisfy
the needs for cooling in both the capture process and the
compression stage.
Methodology
In the current section, various simulation tools and assumptions made for thermodynamic modeling of the selected
power plants are briefly described. This section presents the
methodology, assumptions, and the scope of the economic
evaluation used.
Thermodynamic modeling
In order to obtain reliable results and to utilize the possibility
of incorporating detailed component characteristics, a combination of different simulation tools was used for modeling
the selected IGCC power plant. Enssim tool [34] has been used
for the modeling of coal milling and drying, the gasification
process, raw syngas cooling and scrubbing. The modeling of
ASU, AGR, SWGS, CO2 compression and dehydration has been
performed using ASPEN Plus [35]. The power block including
the GT, and the triple-pressure steam cycle were modeled in
IPSEpro [36]. Further details concerning simulation works can
be found in Refs. [22,23]. The list of assumptions for each subsystem of the selected IGCC plant is not repeated here and can
be found in Ref. [22].
Avoiding repetition of the simulation works and utilizing
previous EU20 projects' findings, the technical performance
specifications of the advanced SCPC have been adopted from
the European Benchmarking Task Force (EBTF21) report [28];
hence, the thermodynamic modeling has not been performed
in this work. Preliminary calculations performed by the
simulation sub-group of the H2-IGCC project have shown that
the coal characteristics used for the H2-IGCC project have a
negligible effect on the performance of the SCPC plant. Hence,
technical performance indicators of the advanced SCPC
remained the same as those reported by Ref. [28].
All performance data refer to plants operating at nominal
full load, in new and clean conditions. The detailed models of
the selected IGCC and NGCC plants include many sub-systems
with reasonable assumptions based on either commercially
20
21
22
23
European Union.
European Benchmarking Task Force.
Lower heating value.
Higher heating value.
Fig. 3 e The block flow diagram of the NGCC power plant
with CO2 capture.
available technology or data provided by other subgroups of
the project. The modeling of the gas turbine in IGCC and NGCC
power plants has been performed using ISO standard conditions (1.013 bar, 15 C, 60% relative humidity). The ambient air
composition together with the characteristics of bituminous
coal and natural gas used for simulation works are shown in
Table 1.
The modeling and simulation of the NGCC was performed
using IPSEpro. The model used for CO2 capture simulation is
based on the calculation model proposed by Kohl and Nielsen
€ ller [38]. A detailed description of the
[37] and developed by Mo
model is not further presented here and can be found in Refs.
[39,40]. The specifications of the power block in the NGCC
plant are presented in Table 2.
The model acquires some input data such as information
about the exhaust gas characteristics (e.g. composition and
flow rate) and the pressure needed for the prediction of the
reboiler requirements. Due to the upper pressure limit in
IPSEpro for gaseous streams, simulation of the CO2 compression and dehydration unit has been performed using ASPEN
Plus. Table 3, which follows, shows the assumptions made for
the simulation of the post-combustion CO2 capture unit in the
NGCC plant using IPSEpro software tool.
Economic assessment
The cost estimation methodology for all investigated plants
with and without CO2 capture is described in this section. The
economic evaluation comprises different stages including
estimations of capital investment, fixed and variable operation and maintenance (O&M24) costs and fuel costs to calculate the cost of electricity.
A publicly available report has been initially selected as a
database for economic calculations [28]. The benefit of
choosing this study is that the figures used reflect the cost of
electricity in the European power market. Furthermore, the
24
Operation and maintenance.
16776
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Table 1 e Ambient air composition, composition and
thermal properties of bituminous coal and natural gas.
Table 3 e Assumptions made for simulation of the postcombustion CO2 capture in the NGCC plant.
Feed
Value
Parameter
0.63
75.10
23.01
1.21
0.05
CO2 capture rate
Absorber pressure drop
Regeneration temperature
Reflux ratio
Reboiler approach temperature
Lean/rich amine heat exchanger
approach temperature
Absorber solvent inlet temperature
Solvent
Heat of reaction
Inlet CO2 pipeline pressure
Water content in the CO2 pipeline
Air
Coal
Natural gas
Parameter/component
Unit
H2O
wt%
N2
wt%
wt%
O2
Ar
wt%
wt%
CO2
Proximate analysis (dry basis)
Moisture
wt%
Ash
wt%
Volatile matter
wt%
Fixed carbon
wt%
kJ/kg
LHV22
kJ/kg
HHV23
C3H8
wt%
wt%
CH4
wt%
CO2
wt%
N2
LHV
kJ/kg
10
12.50
27.00
50.50
25,100
26,195
4.02
95.53
0.40
0.05
49,702
level of the technical performance of different plants, more
specifically the advanced SCPC, is closer to existing power
plants in Europe. A set of assumptions have been considered
in order to analyze the economic indicators of different cycles
based on a consistent basis. A Microsoft Excel-based model
has been developed containing cost data for different power
generation technologies as well as assumptions made for
economic evaluation. Such a tool provided the opportunity to
modify or change input parameters during the economic
assessment.
The economic viability of the selected cycles has been
assessed through the cost of electricity and the cost of CO2
avoided. Due to the volatility of some input parameters, a
number of sensitivity analyses have been carried out to
disclose the effect of those parameters on the economic attributes of the cycle.
Table 2 e Technical specifications of the power island in
the NGCC.
Parameter
Unit
Compressor inlet air flow rate
Pressure ratio
Cooling percentage to the
compressor inlet air flow
Temperature increase for
1st cooling flow
Temperature increase for
2nd and 3rd cooling flow
Fuel flow
Combustor outlet temperature
Expander inlet pressure
Exhaust gas temperature
Exhaust gas flow rate to HRSG
GT gross efficiency
Triple pressure level HRSG
Steam superheating/reheat temperature
HRSG pressure drop (hot side)
HP steam turbine isentropic efficiency
IP steam turbine isentropic efficiency
LP steam turbine isentropic efficiency
kg/s
e
%
685.4
18.2
23.8
C
0
C
20
kg/s
C
bar
C
kg/s
%
bar
C
bar
%
%
%
Value
14.9
1500
17.9
577
700.3
39.5
120/27/4.6
560
0.04
91
90
89
Unit
Value
%
bar
C
moleH2 O =moleCO2
C
C
90
0.15
120
1.0
10
10
40
MEA
85
110
20
C
30%
kJ=moleCO2
bar
mg/kg
The COE is a standard indicator (metric) employed in the
assessment of project economics which represents the revenue per unit of product that must be met to reach break-even
over the life time of a plant. It is, hence, the selling price of
electricity that generates a zero profit. For this purpose, the
net present value (NPV25) computation has been performed to
determine the COE. The cost of CO2 avoided is also a standard
cost metric indicating the cost of CO2 avoidance, which is
defined as:
Cost of CO2 avoided½V=tCO2 ¼
COECapture COEref ½V=MWh
CO2specific CO2specificcapture ½tCO2 =MWh
(1)
ref
where COE is cost of electricity generation, CO2specific is tonne of
CO2 emissions to the atmosphere per MWh (based on the net
capacity of each power plant), and the subscripts “capture”
and “ref” refer to plants with and without capture unit,
respectively. Even though the cost of CO2 avoided should
contain costs associated with transport and storage, these
costs are omitted in this study. However, the omitted costs
have no impact on the comparison outcome as applied to all
power plants in a consistent basis. Furthermore, it should be
noted that the reference plant is a similar plant to the one with
capture unit; e.g., the reference plant of the NGCC with capture unit is the non-capture NGCC plant. The selection of a
similar reference plant has been made assuming that all
investigated technologies have a similar chance to be built in
future under a no carbon constraint scenario.
The economic assessment is based on the commercial
installation of each power plant (or nth-of-a-kind technology)
and does not cover the costs for the demonstration plants. The
following considerations have also been taken into account:
All economic assessments are based on the reference year
2013.
Cost estimation represents a complete power plant on a
generic greenfield site, and site-specific considerations are
not taken into account.
The plant boundary is defined as the total power plant facility within the “fence line”.
25
Net present value.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Costs associated with CO2 transport, storage, and monitoring are not included in the reported capital cost and
O&M costs, while the CO2 compression cost is included.
All taxes with the exception of property taxes (property
taxes are included with the fixed O&M costs) are excluded.
Any labor incentives are excluded.
Each power plant is designed to operate at base load
operation.
The costs associated with the plant's decommissioning are
excluded.
It should be clearly highlighted that the techno-economic
analysis presented in this article cannot provide an absolute
result, since i) there is no full-scale carbon capture unit for
power generation application, ii) the used cost data for
equipment cost calculations have some uncertainties, and iii)
assumptions made for cost calculations (e.g. capacity factor,
fuel prices, etc.) are market-dependent and uncertain by nature; they can change a great deal as a function of time and
geographic location. However, the results of this work will
provide a good comparative insight highlighting the competitiveness of different fossil-based power generation
technologies.
Capital costs
The capital cost assessment for the selected IGCC and NGCC
plants is based on a bottom-up approach (BUA26), while, for
the advanced SCPC plant, it is based on a top-down approach
(TDA27). The BUA is the step-count exponential costing
method using dominant parameters or a combination of parameters derived from the mass and energy balance simulation. The BUA for capital costs assessment has three levels, i.e.
total direct plant costs (TDPC28), engineering, procurement
and construction costs (EPCC29), and total plant costs (TPC30)
as shown in Fig. 4(a).
The TDA is based on equipment supplier estimates of
entire EPC costs. The capital costs calculations for the
advanced SCPC are also shown in Fig. 4(b). The cost estimate
for the advanced SCPC plant without CO2 capture is based on
the TDA, while the cost estimate for the capture unit is based
on the BUA. The overall EPCC of the plant is formed by the sum
of the EPCC for the plant without CO2 capture and the EPCC for
capture unit.
The calculation of the equipment costs for a certain process unit, based on utilization of the cost data for different
components' sizes, was performed using the following
equation:
f
Ci ¼ Ci;ref Si Si;ref $IR
(2)
where Ci is the cost of a component (sub-system), Ci,ref is the
known cost of a reference component (sub-system) of the
same type and of the same order of magnitude, Si is the scaling
parameter, f is the reference cost scaling exponent, and IR is
the cost index ratio. The term (f) incorporates economies of
26
27
28
29
30
Bottom-up approach.
Top-down approach.
Total direct plant costs.
Engineering, procurement and construction costs.
Total plant costs.
16777
scale in the equation and indicates that the percentage change
in cost is smaller than the percentage change in size for each
major component. Typical values of the scaling exponent are
reported in Ref. [41]. The typical values for power utilities vary
between 0.6 and 0.7, and the value used in this study, based on
internal discussions, is 0.67. All economic assessments are
based on the reference year 2013 and the IR was adopted from
the Chemical Engineering Plant Cost Index31 (CEPCI) [42] to
incorporate the economic ups and downs (market fluctuations) in the cost assessment. In the economic calculations
carried out in this study, all figures extracted from the literature given in U.S. dollars (US$) were recalculated to European
Euros (V) using the universal currency conversion XE rates
[43].
Assumptions made for economic assessment
The assumptions made for the estimation of the capital costs,
O&M costs, and fuel costs for different power generation
technologies, i.e. the selected IGCC, the advanced SCPC, and
the NGCC power plants are presented here. Table 4, which
follows, shows the economic assumptions made for the
evaluation of capital costs for different plants within the H2IGCC project.
The assumptions made to estimate the O&M costs for
different power generation technologies are listed in the
following Table 5.
The coal price is 2.5 V/GJ, while the NG price is 7.3 V/GJ. The
fuel prices are for September 2013 and based on data available
in Ref. [46]. The coal price is based on API N2, 6000 kCal NAR
(CIF) while the NG price is based on the Zeebrugge price.
Results and discussion
The main objective of this study is to assess the competitiveness of various fossil-based power generation technologies from a techno-economic point of view. The final aim is to
differentiate between these competing technologies by means
of the plant's performance and cost of electricity defined by
capital investment and production cost. Accordingly, thermodynamic performance indicators of the selected IGCC,
advanced SCPC, and NGCC power plants using the assumptions mentioned above are presented in the current section. It
should be observed that technical performance indicators of
the advanced SCPC plant have been adopted from the EBTF
report [28]. The second part of this section is dedicated to the
economic indicators of concerned power plants based on the
economic assumptions presented in Section methodology.
Thermodynamic performance
The overall efficiency as well as the net and gross power
outputs of the IGCC, advanced SCPC, and NGCC power plants
with and without CO2 capture is illustrated in Fig. 5. It should
be noted that the technical performances of the IGCC and
NGCC are based on the assumptions presented in Section
methodology and the simulation carried out by the system
31
32
Chemical Engineering Plant Cost Index.
Discounted cash flow.
16778
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Fig. 4 e Capital costs levels and their elements for (a) the selected IGCC and NGCC plants and (b) the advanced SCPC plant.
analysis sub-group of the H2-IGCC project, while data for the
advanced SCPC plant are adopted from Ref. [28].
As shown in Fig. 5, regarding the overall efficiency of the
non-capture plants (or reference plants), the NGCC is the most
efficient plant, while the advanced SCPC plant is the least
efficient plant with 12.5 percentage points difference. This
trend is the same for the plants with capture unit, although
the difference between the NGCC and the advanced SCPC is
larger with 15.3 percentage points. The relative efficiency
penalty associated with the capture deployment (compared to
the reference plant of each technology) is 24%, 27%, and 16%
for the IGCC, advanced SCPC, and NGCC plants, respectively.
The following Table 6 shows the other performance indicators
for the IGCC and NGCC power plants. Further information
concerning the advanced SCPC is available in Ref. [28]; hence,
it is not repeated here.
Economic performance
In the current section the results of capital costs estimation,
O&M costs, fuel costs, COE, cost of CO2 avoided, and economic
sensitivity for the selected IGCC, advanced SCPC, and NGCC
Table 4 e Assumptions made for capital costs calculations of different power plants with CO2 capture.
Parameter
Base year
Equipment costsa
Scaling exponent (f in Eq. (1))
Escalation of equipment cost
Installation costs
Escalation of installation costs
The average currency exchange rate for September 2012
Construction period
Capital investment distribution
Indirect costsb
Project contingency
Process contingency
Owners' costsc
Discounted cash flow32 (DCF) rate
Inflation rate
a
Assumption
2013
Adopted from Refs. [27,28,44]
0.67
CEPCI
Proportional to the equipment costs
Same as inflation rate
V0.7485/US$ [43]
4 years for IGCC and SCPC with capture
3 ½ years for non-capture IGCC and SCPC
3 years for NGCC þ CCS
2 ½ for non-capture NGCC
4 years: 20%, 30%, 30%, 20%
3 ½ years: 20%, 35%, 35%, 10% (for last half a year)
3 years: 40%, 30%, 30%
2 ½ years: 40%, 40%, 20% (for last half a year)
14% of TDPC
15% of the EPCC for the IGCC with capture
10% of the EPCC for the SCPC and NGCC with capture, and non-capture
IGCC
5% of the EPCC for the non-capture SCPC and NGCC
5% of the EPCC for all plants
5% of EPCC for all plants
8% (real discount rate)
3%
It should be mentioned that additional costs for the GT modifications due to the combustion of H2 rich syngas in the IGCC plant with CO2
capture is assumed to be 15%. Nevertheless, the increase in capital cost for the modified GT would have little impact on the total plant costs of
the IGCC plant with CO2 capture.
b
The indirect costs are associated with the costs for yard improvement, buildings, sundries, and engineering services.
c
This cost includes pre-production costs, inventory capital, and other owners' costs (excluding any financing costs).
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
16779
Table 5 e Assumptions made for O&M costs calculations of different power plants with CO2 capture.
Parameter
Unit
Capacity factora
%
Plant life timeb
Labor costc
Number of labors per shift
Year
V/h
Person
Escalation of the variable O&M costs
Maintenance cost
%
% EPC
Property taxes and insurance cost
CO2 allowances price
% TPC
V/t CO2
Assumptions
1st year of operation for IGCC and SCPC: 40
2nd year of operations for IGCC and SCPC: 65
Rest of plant's operational life time: 80 for IGCC and 85 for SCPC
1st year of operation for NGCC: 65
Rest of plant's operational life time for NGCC: 85
30 for coal-based and 25 for NG-based plants
43.8 [28,45]
30 for IGCC
25 for SCPC
22 for NGCC þ CCS
3
1.5 for IGCC and SCPC with capture
1.3 for non-capture IGCC and SCPC
1.25 for NGCC with capture
1.0 for non-capture NGCC
1.5 for all plants
0
a
This percentage shows the operating hours of the power plant in a year at full load. It should be noted that 80% is a conservative assumption
for the capacity factor of the IGCC plant. Historical data from current IGCC plants (without CCS) showed successful operational hours up to this
level.
b
This period starts from commissioning and extends up to decommissioning.
c
Although the labor rate seems rather high, this cost includes overheads, training, etc.
power plants are presented. It should be highlighted again
that the cost estimation performed includes a level of uncertainty (±30%) given the fact that there is no power plant with a
CO2 capture unit in operation and based on the methodology
selected for cost assessments, i.e. using available equipment
cost data.
Capital costs
Fig. 6 illustrates different components of the total plant costs
for the selected IGCC, advanced SCPC, and NGCC power plants
with and without CO2 capture unit. As shown in Fig. 6, the
lowest capital investment is required for the NGCC plant
without CO2 capture unit. Even with the CO2 capture feature,
the NGCC requires less capital investment compared to the
cheapest coal technology, i.e. the advanced non-capture
SCPC. The results also show one important advantage of the
IGCC cycle compared to the other coal technology, the
advanced SCPC plant, when the CO2 capture is incorporated
into the system and that is less capital requirement for CO2
capture deployment.
The following Table 7 shows different cost indicators such
as total plant costs and specific investment for the concerned
power plants. The updated cost figures for the advanced SCPC
plant are also given in Table 7 based on the assumptions
presented in Section methodology. As mentioned, the calculations of capital costs for the selected IGCC and NGCC plants
are based on a bottom-up approach. In contrast, the capital
costs calculation for the SCPC is based on a top-down
approach. Therefore, the total equipment costs and installation costs are not explicitly given for the advanced SCPC plant.
As shown in Table 7, the highest absolute capital investment is required for the SCPC plant with CO2 capture. However, it should be noted that the production capacity (or power
output) of the advanced SCPC is higher than those for the IGCC
and NGCC plants. As shown in Fig. 6, a better indicator is the
specific investment of a certain plant, in which the capacity of
the plant is embedded in the value. Please note that indirect
costs, owner's costs and contingencies are based on the assumptions given in Section methodology, and these cost indicators for the SCPC plants (with and without capture) have
been calculated backward using the updated EPC costs.
O&M costs and fuel costs
Fig. 5 e The overall plant efficiency and power output of
various fossil-based power plants.
Table 8 presents the calculated operation and maintenance
costs for the selected IGCC plant, advanced SCPC, and NGCC
power plants. The calculated labor costs seem on the high
side. However, it should be mentioned that any training,
holidays, pension, overheads, etc. are included in the
assumed labor costs. Since the details of variable O&M costs
were not available for the advanced SCPC, the variable cost
items have been escalated using the same rate as the inflation
rate from the year of cost data to the base year of this study
(i.e. 2013). Amongst plants with capture, the selected IGCC
16780
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Table 6 e Thermodynamic performance indicators of the IGCC and NGCC plants with(out) CO2 capture.
Parameter
Gross plant's power output
Net plant's power output
Overall efficiency
Fuel flow
Gas turbine net power output
Steam turbine net power output
Total auxiliaries
- HRSG pumping power
- Gasification power demand
- ASU compression power demand
- AGR pumping and compression power demand
- AGR refrigeration power demand
- CO2 compression power demand
- CO2 capture plant including blower, etc.
Specific CO2 emissions
Unit
MW
MW
%LHV
kg/s
MW
MW
MW
MW
MW
MW
MW
MW
MW
MW
t CO2/MWh
plant has the highest specific fixed O&M costs, while the NGCC
has the lowest costs. This trend is similar for the non-capture
plants. The advanced SCPC in both capture and non-capture
cases requires 5% and 13% less fixed O&M costs compared to
the corresponding IGCC cases, respectively.
With respect to the variable O&M costs, the advanced SCPC
with and without capture unit has the highest costs. Similarly
to the fixed O&M costs, the NGCC in both cases (capture and
non-capture) has the lowest costs compared to the corresponding capture or non-capture cases in other plants.
The fuel costs for the selected IGCC, advanced SCPC, and
NGCC power plants with and without CO2 capture are illustrated in Fig. 7. Please note that the amount of fuel (or annual
fuel costs) from capture case to non-capture case only
changes for the IGCC plant, as the capture takes place upstream of the GT. The estimation results of the fuel costs
shown in Fig. 7 confirm that the specific fuel costs (V/MWh
net) are more than two times higher for the NGCC plant than
for other technologies in both capture and non-capture cases.
Fig. 7 also shows that clean fossil fuel-based power generation
(i.e. plants with capture unit) will cause increased primary
energy consumption (i.e. an increased CO2 production) but a
Fig. 6 e The breakdown of the specific total plant costs for
various power plants.
IGCC
NGCC
Non-capture
Capture
Non-capture
Capture
521.5
461.7
47.0
39.1
304.4
217.1
59.8
2.7
4.4
43.5
0.7
8.5
e
e
0.70
490.7
394.4
35.7
44.0
314.0
176.7
96.3
3.4
4.9
48.9
10.8
8.5
19.7
e
0.08
429.4
428.0
58.0
14.9
288.6
140.8
1.3
1.3
e
e
e
e
e
e
0.35
384.7
359.3
48.7
14.9
288.6
96.1
25.4
1.3
e
e
e
e
11.3
12.8
0.04
lower CO2 emission compared to corresponding non-capture
plants.
Costs of electricity generation and CO2 avoidance
As mentioned before, the cost of electricity in this study is
calculated as the break-even point for the electricity selling
price. Fig. 8, which follows, shows the breakdown of the cost
of electricity into different cost elements for various investigated power plants, viz. the selected IGCC, the advanced SCPC,
and the NGCC power plants. As shown in Fig. 8, the cost of
electricity for the advanced SCPC plant with capture unit is
higher than the corresponding values for the selected IGCC
and NGCC plants with capture unit. The COE for the noncapture NGCC plant is the highest among other non-capture
plants. The difference in the COE between the most expensive non-capture technology (i.e. the non-capture NGCC) and
the cheapest power generation technology (i.e. the advanced
SCPC without capture unit) is only 4%. The difference between
the most expensive and cheapest technologies with capture
plants is only 6%. Nevertheless, supporting any decision
against or in favor of one of the concerned technologies based
purely on the estimated COE is not wise as the difference
between the COEs is marginal (when plants from one category, i.e. either with capture or without, are compared
together).
The other important aspect which is not shown in Fig. 8 is
the share (percentage) of each cost element in the COE for a
certain technology. The highest capital costs share is related
to the IGCC plant without capture unit (~55% of the COE). The
share of fuel costs is about 60e70% of the cost of electricity for
the NGCC plant (with and without capture unit), while it is
approximately 30% of the COE for the other power plants. The
highest fixed O&M costs share is related to the selected IGCC
plant with capture unit (~15% of the COE), while the COE for
the advanced SCPC with capture has the highest share of
variable O&M costs (~8% of the COE).
The cost of CO2 avoided is also calculated based on the
equation presented in Section methodology. The cost for CO2
avoided is in fact the break-even price for CO2, where it starts
to make economic sense to build plants with carbon capture.
16781
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Table 7 e The capital costs for the selected IGCC, SCPC, and NGCC plants with(out) CO2 capture.
Cost components
a
TPC þ DCF (8%)
TPC
Owner's costs and contingencies
EPCC
Indirect costs
TDPC
Total equipment costsb
Installation costsb
Specific investment (gross)
Specific investment (net)
a
b
Unit
MV
MV
MV
MV
MV
MV
MV
MV
V/kW gross
V/kW net
IGCC
SCPC
NGCC
Non-capture
Capture
Non-capture
Capture
Non-capture
Capture
1094
899
150
750
M
657
356
302
1725
1948
1392
1144
229
916
112
803
427
376
2332
2902
1697
1434
187
1247
153
1093
e
e
1750
1901
2066
1699
283
1416
174
1242
e
e
2482
3091
347
304
40
264
32
232
136
96
708
710
729
619
103
515
63
452
237
215
1608
1723
This value is to consider the escalation during the construction period.
This value is not available for the advanced SCPC plant since it is based on a top-down approach.
This cost for different technologies with capture unit when
compared to similar reference plants without capture unit is
illustrated in Fig. 9. Please note that the costs associated with
transport and storage are excluded from the cost of CO2
avoided. The specific CO2 emissions for each technology with
or without CO2 capture unit are also shown in Fig. 9.
The reason for such a high avoidance cost for the NGCC is
the low specific emissions from the plant without capture unit
which, according to Eq. (1) (presented in Section economic
assessment), results in a high CO2 avoided cost. The CO2
avoided cost shown in Fig. 9 might be a better indicator
compared to the COE to support which technology is better for
CO2 mitigation. However, it should be noted that costs associated with transport and storage will increase the presented
cost of CO2 avoided and will change the trend shown in Fig. 9.
Economic sensitivity
In order to evaluate the sensitivity of the COE to variations in
the inputs for each plant, an economic sensitivity analysis has
been performed to identify the most influential parameter
with the strongest impact on the results. Two parameters are
considered as the most uncertain parameters for the calculation of the COE. These parameters are the capacity factor (or
load factor) of the plant and the fuel price. The variation range
of capacity factor is 40e90%. The upper limit is selected based
on the technical limitations such as minimum time required
for any overhaul and maintenance activities. The lower limit
is selected based on the experience during recent years which
confirms decreasing electricity production form fossil-based
plants. It is assumed that the load of power plants remains
constant at the design condition. The fuel prices also vary 50%
from the prices presented in Section methodology. The results
of the sensitivity analysis under the variation of the plant's
capacity factor and the fuel price are shown in Fig. 10(a) and
(b), respectively. Please note that the capacity factor for the
IGCC plant (with and without capture) is 80%, while it is 85%
for the other technologies. Please also note that each plant is
compared to its reference COE (at mentioned capacity factor
or fuel price in Section methodology).
Results in Fig. 10 highlight possible drivers which may influence market attention on different technologies. It is
evident from this figure that the COE for both NGCC plants
(with and without capture unit) is less sensitive to changes of
capacity factor compared to other plants. The COE for the
advanced SCPC is the most sensitive amongst concerned
technologies. As shown in Fig. 10(b), the change of fuel price
has a significant effect on the COE for the NGCC with and
without capture unit. Any reduction in the NG price will
change the market tendency towards higher electricity production from NGCC plants compared to coal-based technologies, even at a similar reduction in the coal price.
It should be highlighted that economic sensitivity analysis
could be performed based on variation in the cost of CO2 allowances. However, the impact of CO2 allowances cost on the
COE would be negligible (more specifically for the plant with
capture unit as most of the carbon emissions were already
Table 8 e The operation and maintenance costs for the IGCC, USCPC, and NGCC plants with(out) CO2 capture.
Cost item
Fixed O&M costs
Labor costs
Maintenance costs
Property taxes and insurance costs
Total fixed O&M costs
Specific fixed O&M costs
Variable O&M costs
Total variable O&M costs
Specific variable O&M costs
Unit
MV/a
MV/a
MV/a
MV/a
V/kW gross/a
MV/a
V/MWh net
IGCC
SCPC
NGCC
Non-capture
Capture
Non-capture
Capture
Non-capture
Capture
9.6
9.9
13.7
33.1
63.5
11.5
13.8
17.2
42.5
86.6
7.7
16.2
21.5
45.4
55.4
9.6
21.2
25.5
56.3
82.2
6.5
2.7
4.6
13.8
32.2
8.4
6.5
9.3
24.3
63.1
4.0
1.2
5.2
1.9
10.7
1.9
28.4
6.9
2.8
0.9
4.7
1.7
16782
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
Fig. 7 e Annual fuel costs and specific fuel costs for
different power plants.
Fig. 10 e Sensitivity response of the COE for different
power plants under variation of (a) the capacity factor and
(b) fuel price.
Fig. 8 e The breakdown of the cost of electricity for the
selected IGCC, the advanced SCPC, and the NGCC power
plants with(out) CO2 capture.
captured from the plants), as the absolute value of CO2 allowances cost is very low (about 5.5 V/t CO2 [46]) and currently
any large variation could not be expected. Nevertheless, its
influence will change depending on the CO2 market development and changes in global mitigation policies.
Conclusion
Fig. 9 e The cost of CO2 avoided, COE, and specific CO2
emissions for the selected IGCC, advanced SCPC, and
NGCC.
The thermodynamic performance indicators of various power
plants including IGCC, advanced supercritical pulverized coal
and natural gas combined cycle power plants were presented
in this article. The NGCC is the most efficient plant, while the
advanced SCPC plant is the least efficient plant amongst noncapture cases. This trend is similar for the plants with capture
unit. The relative efficiency penalty associated with the capture deployment (compared to the reference plant of each
technology) is 24%, 27%, and 16% for the IGCC, advanced SCPC,
and NGCC plants, respectively.
A comparative study was also conducted, comparing the
COE and the cost of CO2 avoided for the mentioned fossil-based
power plants. The economic performance indicators of each
plant were estimated using the developed model and the
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
results are presented and thoroughly discussed. It should be
highlighted that the estimation is highly dependent on the
selected assumptions. It is also important to note that technoeconomic analysis cannot provide an absolute result, since the
cost data and assumptions are uncertain by nature. The COE
for the IGCC plant with and without capture is 91 and 59 V/
MWh, respectively. The COE for the advanced SCPC is 96 and
59 V/MWh for the capture and non-capture cases, respectively.
The COE for the NGCC with and without capture is 61 and 91 V/
MWh, respectively. The results show that the less capitalintensive plant is the NGCC plant without CO2 capture. However, the high fuel costs for this plant decrease the gap between
the COE for this plant compared to that for the other plants.
The COE for the NGCC technology was the most sensitive to
changes in the fuel price amongst other COEs for different
technologies. However, the COE for the NGCC technology was
also the least sensitive to variations of the plant's capacity
factor. The estimated costs of CO2 avoided for the IGCC, SCPC,
and NGCC technologies are 51, 57, 99 (V/t CO2 avoided).
Results highlighted that, based purely on the COE for
different plants, it cannot be concluded which technology is
better and more cost-effective than other technologies,
considering the level of uncertainty in the economic results of
this study (±30%). Other main drivers such as proven technology and operational flexibility will, therefore, play an
important role in the widespread utilization of these
technologies.
Acknowledgment
The authors are grateful to the European Commission's
Directorate-General for Energy for financial support of the Low
Emission Gas Turbine Technology for Hydrogen-rich Syngas
(H2-IGCC) project with the grant agreement number of 239349.
The authors wish to acknowledge Chris Lappee at Vattenfall
for sharing constructive opinions about the selection of cost
estimating methodology as well as results and discussion of
this article. The authors are also thankful to Han Raas at
Vattenfall for performing the gasification simulations, Karel
Dvorak and Adam Al-Azki at E.ON UK for being involved in
many discussions at different stages of the H2-IGCC project.
references
[1] BP. Statistical review of world energy. British Petroleum
Company; 2013.
[2] IPCC. Climate change 2007: mitigation of climate change.
Contribution of working groups III to the fourth assessment
report of the intergovernmental panel on climate change. In:
Metz B, Davidson OR, Bosch PR, Dave RA, Meyer LA, editors.
Intergovernmental panel on climate change. Cambridge,
United Kingdom and New York, NY, USA: Cambridge
University Press; 2007. p. 863.
[3] IEA. World energy outlook. Paris, France: International
Energy Agency; 2012.
[4] UN. World population prospects: the 2012 revision, key
findings and advance tables. New York, United States: United
Nations Population Division; 2013.
16783
[5] BP. Energy outlook 2030. British Petroleum Company; 2012.
[6] ExxonMobil. The outlook for energy: a view to 2040. Irving,
Texas, USA: ExxonMobil Company; 2013.
[7] IEA. Key world energy statistics. Paris, France: International
Energy Agency; 2012.
[8] IEA. Key world energy statistics. Paris, France: International
Energy Agency; 2013.
[9] IEA. World energy outlook. Paris, France: International
Energy Agency; 2013.
[10] IEA. IEA statistics 2013 edition: CO2 emissions from fuel
combustion highlights. Paris, France: International Energy
Agency; 2013.
[11] IEA. World energy outlook 2011. Special report: are we
entering a golden age of gas?. Paris, France: International
Energy Agency; 2011.
[12] EIA. Annual energy outlook 2013 with projections to 2040.
Washington, DC: US Energy Information Administration; 2013.
[13] BP. Statistical review of world energy. British Petroleum
Company; 2012.
[14] EU. Energy roadmap 2050. Luxembourg: European
Commission, European Union; 2012 [Publications Office of
the European Union].
[15] Liang X, Wang Z, Zhou Z, Huang Z, Zhou J, Cen K. Up-to-date
life cycle assessment and comparison study of clean coal
power generation technologies in China. J Clean Prod
2013;39:24e31.
[16] IEA. Fossil fuel-fired power generationecase studies of
recently constructed coal- and gas-fired power plants. Paris,
France: International Energy Agency; 2007.
[17] IEA. Power generation from coaleongoing developments and
outlook. Paris, France: International Energy Agency; 2011.
[18] Cormos CC, Starr F, Tzimas E. Use of lower grade coals in
IGCC plants with carbon capture for the co-production of
hydrogen and electricity. Int J Hydrogen Energy
2010;35:556e67.
[19] Pettinau A, Frau C, Ferrara F. Performance assessment of a
fixed-bed gasification pilot plant for combined power
generation and hydrogen production. Fuel Process Technol
2011;92:1946e53.
[20] Henderson C. Future developments in IGCC. London, United
Kingdom: IEA Clean Coal Centre; 2008. p. 45.
[21] Low emission gas turbine technology for hydrogen-rich
syngas under the 7th framework programme FP7-239349.
Available from: www.h2-igcc.eu.
[22] Mansouri Majoumerd M, De S, Assadi M, Breuhaus P. An EU
initiative for future generation of IGCC power plants using
hydrogen-rich syngas: simulation results for the baseline
configuration. Appl Energy 2012;99:280e90.
[23] Mansouri Majoumerd M, Raas H, De S, Assadi M. Estimation
of performance variation of future generation IGCC with coal
quality and gasification process e simulation results of EU
H2-IGCC project. Appl Energy 2014;113:452e62.
[24] Rosenberg WG, Alpern DC, Walker MR. Deploying IGCC in
this decade with 3 party covenant financing. John F. Kennedy
School of Government, Harvard University; 2004.
[25] Rosenberg WG, Alpern DC, Walker MR. Financing IGCC e
3party covenant. John F. Kennedy School of Government,
Harvard University; 2004.
[26] EPRI. Coal fleet integrated gasification combined cycle
research and development roadmap. Palo Alto, California,
USA: Electric Power Research Institute (EPRI); 2011.
[27] DOE/NETL. Cost and performance baseline for fossil energy
plants. Bituminous coal and natural gas to electricity, vol. 1.
United States Department of Energy, National Energy
Technology Laboratory; 2010.
[28] CAESAR. European best practice guidelines for assessment of
CO2 capture technologies, in the European Benchmarking
Task Force (EBTF); 2011.
16784
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 6 7 7 1 e1 6 7 8 4
[29] ZEP. The costs of CO2 captureepost-demonstration CCS in
the EU. European Technology Platform for Zero Emission
Fossil Fuel Power Plants; 2010.
[30] Baufume S, Hake JF, Linssen J, Markewitz P. Carbon capture
and storage: a possible bridge to a future hydrogen
infrastructure for Germany? Int J Hydrogen Energy
2011;36:8809e21.
[31] Chen C, Rubin ES. CO2 control technology effects on IGCC
plant performance and cost. Energy Policy 2009;37:915e24.
[32] Cormos CC. Techno-economic and environmental
evaluations of large scale gasification-based CCS project in
Romania. Int J Hydrogen Energy 2014;39:13e27.
[33] Siemens. Siemens gas turbine SGT5-4000F. Erlangen,
Germany: Siemens AG; 2008.
[34] Enssim. Enssim software; 2009 [Doetinchem, The Netherlands].
[35] Aspen plus version 7.1. Cambridge, MA, USA: Aspen
Technology Inc.; 2009.
[36] IPSEpro version 4.0. Graz, Austria: Simtech Simulation
Technology (Simtech); 2003.
[37] Kohl A, Nielsen R. Gas purification. 5th ed. Houston, Texas,
USA: Gulf Publishing Company; 1997. p. 1395.
€ ller BF. A thermoeconomic evaluation of CO2 capture with
[38] Mo
focus on gas turbine-based power plants. Lund, Sweden:
Lund Institute of Technology; 2005. p. 182.
€ ller BF, Assadi M, Linder U. CO2-free power generation e a
[39] Mo
study of three conceptually different plant layouts. In: ASME
paper GT2003-38413. Atlanta, Georgia, USA: ASME Turbo
Expo; 2003.
€ ller BF, Obana M, Assadi M, Mitakakis A. Optimisation
[40] Mo
of hat-cycles e with and without CO2 capture. In: ASME
paper GT2004-53734. Vienna, Austria: Asme Turbo Expo;
2004.
[41] Bejan A, Tsatsaronis G, Moran M. Thermal design and
optimization. New York: John Wiley & Sons Inc; 1996.
[42] CEPCI. Chemical engineering plant cost index. Chem Eng
2013;121(1):60.
[43] Universal currency converter; September 2013. Available
from: http://www.xe.com.
[44] GTW. Gas turbine world handbook, vol. 30. Pequot
Publishing; 2013.
[45] Eurostat. Labour cost index e recent trends; 2014
[2014.03.03]; Available from: http://epp.eurostat.ec.europa.
eu/statistics_explained/index.php/Labour_cost_index_-_
recent_trends#Further_Eurostat_information.
[46] EnergyMarketPrice. Weekly report e energy statistics. Week
39; 2013 [2014.03.03]; Available from:, http://www.
energymarketprice.com.