Journal of Cleaner Production: Yuting Tang, Fengqi You

Journal of Cleaner Production 181 (2018) 271e292
Contents lists available at ScienceDirect
Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro
Life cycle environmental and economic analysis of pulverized coal

oxy-fuel combustion combining with calcium looping process or
chemical looping air separation
Yuting Tang a, b, Fengqi You b, *
a
Guangdong Province Key Laboratory of Efficient and Clean Energy Utilization, School of Electric Power, South China University of Technology, Guangzhou,
510640, China
b
Robert Frederick Smith School of Chemical and Biomolecular Engineering, 318 Olin Hall, Cornell University, Ithaca, NY, 14853, USA
a r t i c l e i n f o a b s t r a c t
Article history: This paper presents multi-criteria environmental and economic analyses of pulverized coal power plants
Received 15 December 2017 with various advanced CO2 capture and separation (CCS) technologies, including oxy-fuel combustion
Received in revised form (Oxy), calcium looping post-combustion capture (CaL), combination of Oxy with CaL (Oxy-CaL) and Oxy
30 January 2018
with chemical looping air separation (Oxy-CLAS). The life cycle analysis (LCA) and techno-economic
Accepted 31 January 2018
Available online 5 February 2018
analysis (TEA) are integrated with Analytic Hierarchy Process (AHP) and Technique for Order Prefer-
ence by Similarity to Ideal Solution (TOPSIS) approaches. This methodology is applied to forecast the
potential of incorporating CaL or CLAS into oxy-fuel combustion and identify the most promising CCS
Keywords:
Calcium looping process
technology option for pulverized coal power plants from the perspectives of different stakeholders. The
Oxy-fuel combustion results show that application of CCS reduces the ecosystem quality and the human health impacts, but
Chemical looping air separation increases the resources use and yields an economic penalty of $12.76~$33.33 per ton of CO2 avoidance.
Multi-criteria decision making From the perspective of industry only, CCS has an unfavorable effect on the performance of the pul-
Life cycle assessment verized coal power plant, and the promotion in carbon price is critical for CCS to attract the support from
industry. In terms of the four CCS technologies, Oxy-CLAS comprehensively performs the best, followed
by Oxy. Decrement of consumption of Ca-based sorbents is critical for Oxy-CaL to outrank Oxy.
© 2018 Elsevier Ltd. All rights reserved.
1. Introduction power plants could help to address this challenge (Mukherjee

et al., 2015). It is estimated that CCS may potentially contribute to
Coal, one of the world's most abundant fossil fuel sources, 15e55% of worldwide CO2 abatement until 2050 (IPCC, 2007).
currently meets about 28.8% of the total world primary energy Amine absorption is a commercially adopted CCS technology in
demand, and 41.3% of global electricity generation (IEA, 2013). Coal existing coal-fired plants. However, regenerating amine absorbents
will continue to dominate power production for at least the next such as monoethanolamine (MEA) requires a large amount of en-
five decades, and its share in the energy portfolio is predicted to be ergy, thus significantly reducing the overall plant efficiency (Kursun
34% in 2040 (U.S. Energy Information Administration, 2016). et al., 2012). Moreover, the degradation of the amine and subse-
However, coal is the most carbon-intensive fossil fuel and con- quent corrosion of the equipment increases the operation and
tributes the largest share of carbon dioxide (CO2) emissions from maintenance costs (Clarens et al., 2016). There are similar concerns
fuel burning (42.9%) (Erlach et al., 2011). Therefore, development of for other CCS technologies, such as adsorption (Chung et al., 2016),
efficient, clean, and economical energy conversion systems has membrane (Ramasubramanian et al., 2012), and algae (Gebreslassie
been an issue of international concern and a challenge for engi- et al., 2013). A calcium looping post-combustion process (CaL post)
neers and researchers (Gong and You, 2015). Incorporating CO2 based on the carbonation/calcination reaction of solid CaO parti-
capture and separation (CCS) technologies into existing coal-fired cles, is found to be a promising alternative to amine absorption
(Hanak et al., 2016). CaL post has several potential advantages over
amine absorption such as relatively lower energy and cost penalties
(Perejon et al., 2016), and the use of cheap and non-toxic sorbent
* Corresponding author.
E-mail address: [email protected] (F. You).
(Rodríguez et al., 2012). Another promising alternative to amine
https://doi.org/10.1016/j.jclepro.2018.01.265
0959-6526/© 2018 Elsevier Ltd. All rights reserved.
272 Y. Tang, F. You / Journal of Cleaner Production 181 (2018) 271e292
post-combustion capture for mitigating CO2 emissions is the oxy- fuel combustion and identify the most promising CCS technology
fuel combustion technology (Jin et al., 2015), which has been suc- option. Moreover, the order preference of CCS technologies is
cessfully demonstrated in large-scale pilot projects (30 MWe) analyzed from the perspectives of local government, industry,
(Escudero et al., 2016a). However, oxy-fuel combustion requires residents, and an 'egalitarian' perspective in which all indicators are
large amounts of oxygen, and the inclusion of conventional cryo- weighted equally, respectively.
genic air separation units (CASU) to meet this demand leads to large The major novelties of this work are summarized as follows:
energy and cost penalties (Kather and Scheffknecht, 2009). Hence,
there is a need for either a simpler and more energy-efficient air Comparative LCA studies of a pulverized coal power plant with
separation technology or a means to decrease the oxygen require- Oxy, CaL post, Oxy-CaL and Oxy-CLAS technologies using
ment for oxy-fuel combustion. The chemical looping air separation comprehensive environmental impact categories.
(CLAS) process and partial oxy-combustion are positioned to fulfill TEA of a pulverized coal power plant with Oxy, CaL post, Oxy-
this need (Moghtaderi, 2010). In the CLAS process, an oxygen car- CaL and Oxy-CLAS technologies, respectively.
rier releases oxygen in steam or flue gas, and the reduced oxygen Comprehensive comparison of novel CCS technology alterna-
carrier absorbs oxygen from air. Partial oxy-combustion is an tives for pulverized coal power plants from different stake-
emerging approach to reduce the pure oxygen requirement, and it holders' perspectives using MCDM.
can be combined with post-combustion capture process to improve
the total CO2 capture efficiency (Vega et al., 2016) These advanced The rest of this paper is organized as follows. First, Section 2
CCS technologies have differences in terms of energy consumption, introduces an existing conventional coal-fired power plant and
environmental impacts, and economic performance. Therefore, for retrofitted power plants with CCS units. Next, Section 3 describes
judicious selection of CCS technologies, rigorous and systematic the LCA, TEA, AHP, and TOPSIS approaches. The proposed meth-
analyses of pulverized coal power plants with various CCS tech- odologies are then applied to coal-fired power plants with four CCS
nologies need to be addressed. technologies in Section 4, and the results are compared with the
Several publications address the life cycle assessments (LCA) conventional pulverized coal power plant without CCS. Conclusions
and/or techno-economic analyses (TEA) of coal power plants with are provided in the end.
different CCS technologies (Korre et al., 2010; Pehnt and Henkel,
2009; Singh et al., 2011a). Most of them have evaluated MEA 2. Process and systems description
post-combustion capture and oxy-fuel combustion with CASU
technologies (Iribarren et al., 2013; Joris Koornneef, 2012; Korre Plants operating with a steam pressure >221 MPa and steam
et al., 2010; Nie et al., 2011; Pehnt and Henkel, 2009; Singh et al., temperature >374 C are super-critical (SC) power plants (Basu and
2011b), while a smaller number of studies have considered CaL Debnath, 2014). Compared with subcritical power plants, they are
post-combustion capture technologies (Clarens et al., 2016; Hurst typically more energy efficient and more environmental friendly
et al., 2012; Petrescu et al., 2017). To the best of our knowledge, (Basu and Debnath, 2014). SC technology is well-developed and
there is no existing literature addressing the environmental and adopted by increasingly more industrial coal-fired power plants
economic analysis of oxy-fuel combustion with CLAS (Oxy-CLAS) (Tan, 2012). Therefore, a SC pulverized coal power plant is chosen as
and the combination of partial oxy-fuel combustion and the CaL the research subject in this work. This conventional power plant
post-combustion capture process (Oxy-CaL) from the whole life without CCS is defined as a baseline (Case 1). Four cases using
cycle perspective. different CCS technologies are considered:
Multi-criteria decision-making (MCDM) is a powerful method-
ology for decision makers to identify the best option from a set of Case 2: SC pulverized coal power plant retrofitted for oxy-fuel
alternatives (Deveci et al., 2015). MCDM methods take various combustion using conventional cryogenic air separation units
criteria into account (Lerche et al., 2017), allowing stakeholders to (Oxy).
participate in decision-making processes and yielding compre- Case 3: SC pulverized coal power plant with calcium looping for
hensive results (Mardani et al., 2016). MCDM approaches have been post combustion capture (CaL post)
used to assess complexity of CCS and reveal the interconnection Case 4: SC pulverized coal power plant with a novel CO2 capture
among complexity factors (Sara et al., 2015), and to identify and system by integrating partial oxy-fuel combustion with the
evaluate the main non-technical factors affecting the CCS chain calcium looping process (Oxy-CaL)
(Jakobsen et al., 2013). MCDM approaches have also been used to Case 5: SC pulverized coal power plant retrofitted for oxy-fuel
compare a power plant using MEA post-combustion capture with combustion using chemical looping air separation with Cu-
the uncontrolled release of CO2 (Fozer et al., 2017). However, to the based oxygen carriers (Oxy-CLAS)
best of our knowledge, there is no systematic MCDM analysis of
other CCS options for pulverized coal-fired power plants that ac- The retrofitted SC power plants for Oxy, CaL post and Oxy-CaL
counts for economic and environmental performance from the life and Oxy-CLAS are shown in Fig. 1. Schematic diagram of the CLAS
cycle perspective. and CASU processes are shown in Fig. 2. The main technical pa-
This paper presents multi-criteria environmental and economic rameters are shown in Table 1.
analyses of pulverized coal-fired power plants with and without Pittsburgh No. 8 coal, which is typical high volatile bituminous
four CCS technologies. The economic and environmental potentials coal, is considered in this paper. The composition of Pittsburgh No.
of incorporating CCS in pulverized coal-fired power plants are 8 coal is listed in Table 2 (U.S. National Energy Technology
systematically investigated by an MCDM methodology. The MCDM Laboratory, 2012).
methodology combines LCA and TEA with Analytic Hierarchy Pro-
cess (AHP) and Technique for Order Preference by Similarity to 2.1. Description of case 1 (without CCS)
Ideal Solution (TOPSIS) approaches. The aforementioned CCS
technologies, namely, conventional oxy-fuel combustion with The base case considers a conventional power plant consisting
CASU (Oxy), CaL post, Oxy-CaL and Oxy-CLAS, are systematically of a pulverized coal boiler with an air preheater and fuel gas
compared by the MCDM methodology to evaluate the economic treatment systems, including a selective catalytic reduction (SCR)
and environmental potentials of incorporating CaL and CLAS in oxy- denitrification device, a cold side electrostatic precipitator (ESP)
Y. Tang, F. You / Journal of Cleaner Production 181 (2018) 271e292 273
Fig. 1. Schematic diagram of a SC pulverized coal power plant without and with CCS: (a) Case 1: without CCS; (b) Case 2: Oxy & Case 5: Oxy-CLAS; (c) Case 3: CaL post; (d) Case 4:
Oxy-CaL.
and a wet flue gas desulfurization (FGD) device. The flue gas leaves consumption of the ESP, SCR, and FGD units are assumed to be
the system through the stack with a CO2 vol concentration of 15%. 0.9 MW, 2.6 MW, and 8.6 MW, respectively (Clarens et al., 2016).
The power plant is equipped with a supercritical steam cycle The boiler efficiency is 90% (Ortiz et al., 2016), and net power output
(Pvv ¼ 290 bar, Tvv ¼ 600/620 C) and a matching turbine electricity is calculated by subtracting all auxiliary power requirements. The
generator. Annual operation time is 7446 h. The energy net power output is 478.57MWe, and other main technical
Fig. 1. (continued).
parameters are shown in Table 2. The data for energy consumption clean, dehydrate, and compress flue gas to 100 bars, so that the final
and emissions are obtained from the references related to a SC coal CO2 enriched flue gas can be easily transported by pipeline. There is
power plant with a similar capacity using the bituminous coal no need for a SCR denitrification device unit in oxy-fuel combustion
(Koornneef et al., 2008; Liang et al., 2013; Ortiz et al., 2017). plants, since the concentration of NOx in this case is much lower
The overall net efficiency (ht ) is calculated as follows: than that in Case 1 because of the avoidance of the formation of
thermal NOx (Hu et al., 2000; Pak et al., 2009), and the reduction of
Wnet Wg Wself NO on char surface through the NO/CO/char reaction (Li et al., 1999;
ht ¼ ¼ (1)
mcoal $QLHV mcoal $QLHV Ohtsuka and Wu, 1999; Tang et al., 2012). For the typical oxy-fuel
combustion, the conversion ratio of nitrogen in coal to NOx is
where Wnet is the net power output (kWe); Wg is total electricity assumed to be 15%, whereas this conversion ratio is 50% in air
generation (kWe); Wself is the total auxiliary system requirements combustion (Koiwanit et al., 2014).
in the whole plant (kWe); mcoal is the consumption of coal (kg/s); Recycled fuel gas and a smaller excess air factor (a) reduce the
and QLHV is lower heating value of coal (kJ/kg). thermal loss of exhaust gas, and consequently result in the incre-
ment of the boiler efficiency in oxy-fuel combustion (Tang and You,
2018). The increase in boiler efficiency is set to 4% in this paper
2.2. Description of case 2 (oxy) following the literature value (Koiwanit et al., 2014). Compared
with Case 1, the main additional units of the oxy-fuel combustion
Oxy-fuel combustion refers to the combustion under pure ox- plant are a CASU, a CPU, and a RFG fan. CASU starts by pressurizing a
ygen provided by a cryogenic air separation unit (CASU). After the large volume of ambient air in its air compressors (Almås, 2012).
combustion process, the flue gas is passed through an ESP unit to The compressed air flows through the two-stage direct contact
remove particulate matter (PM). Between 60% and 80% (78% is used aftercooler, and it is then fed into the distillation process to separate
in this paper) of the flue gas is recycled to the furnace to fill the O2 based on different boiling points of the species (Koiwanit et al.,
volume fraction of lost N2 (Toftegaard et al., 2010). The use of 2014). CPU is used to lower the temperature of flue gas and remove
recycled flue gas (RFG) here is an important and integral part of the acidic gases (93e97% SO2 and 58e78% SO3), moisture (>85%), and
oxy-fuel combustion process because firing pure oxygen in a boiler particulates (>90%) (Koiwanit et al., 2014). NOx, SOx, HCl, moisture
would result in excessively high flame temperatures, which may and heavy metals are further removed as condensate in the
damage the boiler (Moghtaderi, 2012). The rest of the flue gas compression (Yan et al., 2009). The specific energy consumption for
passes through the FGD and comprehensive CO2 purification and CASU and CPU are assumed to be 200 kW h per ton of O2 (Martínez
compression unit (CPU). Some researchers claim that a coal oxy- et al., 2011) and 143 kW h per ton of CO2 (Escudero et al., 2016b),
fuel combustion power plant might not need an individual FGD respectively. A RFG fan is required to overcome the pressure which
device (Croiset E, 2001; MacDowell et al., 2010; Scheffknecht et al., forces the recycled flue gas stream to go back to the boiler (Rubin
2011), but other studies state that a coal oxy-fuel combustion plant et al., 2007), and the power requirement of this RFG fan (ERFG ,
still required a FGD device with comparatively higher-efficiency kWe) is calculated as follows (Rubin et al., 2007).
and lower-cost desulfurization (Xiong et al., 2009). Since Pitts-
burgh No. 8 is medium sulfur coal (see Table 1), the FGD device is 2:67 106 $VRFG $DP
still retained in this oxy-fuel combustion plant. The energy con- ERFG ¼ (2)
hFan
sumption of FGD is assumed to be reduced by 1/3 with a lower
conversion rate from sulfur to SO2 (Nie, 2009). The CPU is used to
Fig. 2. Schematic diagram of CLAS and CASU process.
Table 1
Main technical parameters of the power plant.
Item Unit Case 1: without CCS Case 2: Oxy Case 3: CaL Case 4: Oxy-CaL Case 5: Oxy-CLAS
mass flow of fuel for main plant kg/s 42.2 55.1 42.2 47.5 55.1
mass flow of fuel for second plant kg/s 0.0 0.0 18.8 23.4 0.0
mass flow of air kg/s 475.0 0.0 475.0 100.2 0.0
mass flow of O2 for oxy kg/s 0.0 136.9 0.0 96.4 136.9
mass flow of O2 for CaL kg/s 0.0 0.0 48.0 61.3 0.0
mass flow of total flue gas to CPU/CCU kg/s 0.0 188.0 165.1 214.4 188.0
mass flow of recycle flue gas to boiler kg/s 0.0 666 0.0 616 666
mass flow of flue gas to the carbonator kg/s 0.0 0.0 516 240.0 0.0
CO2 concentration in flue gas % 15.0 89.0 91.1 91.1 89.0
CO2 mass flow to CPU or CCU kg/s 0.0 149.2 150.5 195.4 149.2
CO2 capture efficient % 0.0 90.0 82.7 97.6 90.0
pre-heated water kg/s 311.0 423.0 451.0 553.0 423
net main network MWe 478.6 485.9 478.6 495.8 531.3
net second network MWe 0.0 0.0 75.8 126.4 0.0
efficiency of boiler % 90.0 94.0 90.0 92.31 94.0
thermal efficiency % 38.5 30.0 30.9 29.8 32.8
energy consumption for SCR MWe 2.6 0.0 2.6 0.0 0.0
energy consumption for FGD MWe 8.4 5.5 8.4 5.5 5.5
energy consumption for ESP MWe 0.9 0.9 0.9 0.9 0.9
energy consumption for ASU for oxy MWe 0.0 98.6 0.0 69.0 36.5
energy consumption for RFG fan MWe 0.0 0.6 0.0 0.5 0.6
energy consumption for CPU MWe 0.0 76.8 0.0 0.0 76.8
energy consumption for oxygen carrier transport MWe 0.0 0.0 45.7 51.2 0.0
energy consumption for ASU for CaL MWe 0.0 0.0 34.6 43.9 0.0
energy consumption for flue gas compressor to carbonator MWe 0.0 0.0 12.5 4.9 0.0
energy consumption for CO2 compressor MWe 0.0 0.0 57.6 74.6 0.0
Table 2
Composition of coal.
Proximate analysis (%, as received) Ultimate analysis (%, as received) Heating value (kJ/kg)
Moisture Volatile matter Ash Fixed carbon Carbon Hydrogen Nitrogen Sulfur Chlorine Oxygen HHV LHV
2.63 35.82 9.17 52.38 73.15 4.97 1.46 2.36 0.04 6.22 30,508 29,443
where VRFG is the RFG volume (m3/h) (density of oxygen used is to the one in Case 1, and the CaL post-combustion capture unit is
1.439 kg/m3); DP is the total change of pressure (kPa), and hFan is installed after the SCR, ESP, and FGD units. The CaL process com-
the energy conversion efficiency of fans, which is typically 75% prises of dual interconnected fluidized bed (DFB) reactors operating
(Rubin et al., 2007). at atmospheric pressure with a continuous looping of Ca-based
Application of oxy-fuel combustion technology reduces the net solids. In the carbonator, CO2 chemically reacts with the solid sor-
thermal efficiency of the power plant to 29.98% with 90% CO2 bent (CaO) to form carbonate calcium at around 650 C, so that it is
capture efficiency (Ortiz et al., 2017). For the oxy-fuel combustion separated from the flue gas. In the calciner, the carbonate calcium is
plant, data on the removal efficiencies of PM, SO2, SO3, HCl, HF, Hg further decomposed into CaO and a CO2-rich stream which is sent
and other trace element by ESP, FGD and CPU are taken from Rubin to the CO2 compression unit (CCU) prior to pipeline transportation.
et al. (1991), Manuilova (2011), and Nie (2009), respectively. The carbonation process takes place at a high temperature of
around 950 C and the calcination reaction is endothermic. There-
2.3. Description of case 3 (CaL post) fore, an oxy-fuel combustion in which extra coal is combusted with
pure oxygen from a CASU is required for calcination (Luo et al.,
In Case 3, the combustion section of the power plant is identical 2013). The calcium-based sorbents are cycled with a total flow
rate of 2402 kg/s in the carbonator/calciner system for CO2 capture reaches 30e60% (45% in this paper) instead of the typical 15% vol
until their reactivity is lost. The calcium-based sorbents show sta- concentration obtained from air combustion. The carbonated par-
ble reactivity in limited cycles. However, after some cycles, visible ticles are then circulated to the calciner reactor in which fast
fragmentation, sintering and agglomeration occur, and carbon decomposition of CaCO3 occurs to regenerate the CaO solids and
deposition and sulfur-containing compounds significantly decrease produce a rich CO2 stream to be compressed and transported. The
the reactivity of sorbents. Thus, the whole batch of sorbents are basic principles and process of the downstream section of Oxy-CaL
added into the CaL process and substituted by fresh sorbents on a is the same with the CaL in Case 3, but the capture capacity of CO2 is
regular basis. The allocating consumption of fresh sorbents with significantly enhanced with a higher CO2 concentration in the
time is assumed to be 2% of total flow rate in this paper and carbonation environment. The Oxy-CaL case has a higher CO2
permanently devitalized sorbent needs to be removed (Abanades capture efficiency of 97.6%, compared with 82.7% for the CaL system
et al., 2004). The amount of the solid waste from the CaL process in Case 3 and 90% for total oxy-fuel combustion in Case 2 (Ortiz
(devitalized sorbents) is assumed to be equal to that of fresh sor- et al., 2017). The above CASU also provides pure oxygen with a
bents. The carbonation reaction is strongly exothermic, and there is flow rate of 61.3 kg/s for the calciner. Steam generated from the
generally a good heat recovery potential from the hot gaseous heat sources at the outlets of carbonator and calciner is used to
streams at the outlets of both carbonator and calciner because of generate secondary power, so the gross power output of this whole
the high reactor temperature (Mantripragada and Rubin, 2014). plant increases to 622.22MWe.
Steam generated from these heat sources is used to generate
additional power for the power plant. This secondary power is
added to the gross power output of the base plant to calculate the 2.5. Description of case 5 (Oxy-CLAS)
total gross power output (Mantripragada and Rubin, 2014).
The most important differences in terms of design features be- Most equipment in the power plant considered in Case 5 (oxy-
tween the CaL case and the base case (conventional plant without fuel combustion boiler with RFG unit, FGD, and CPU) is identical to
CCS) include the CaL capture unit, CASU, CCU, and solids handling those in Case 2. The CO2 capture efficiency remains 90%. The dif-
devices (Cormos and Cormos, 2017). It is recognized that the high ference between Case 2 and Case 5 is their different air separation
consumption of fuel and pure oxygen in the calciner imposes an units (CASU & CLAS). The configuration and principle of the CLAS
important energy penalty to the CaL technology, and CaO deacti- unit is similar to the chemical looping oxygen uncoupling (CLOU),
vation and substitutions by fresh Ca-based sorbent increase the and both of them are able to separate oxygen in air through
demand of heat. The specific energy consumption of CASU units in continuous recirculation of metal oxide particles between two
this case is the same as that in Case 2 (200 kW h per ton of pure O2), interconnected reactors (Zhu et al., 2016). The most significant
but the oxygen requirement in CaL case is only a fraction of the difference between CLAS and CLOU is the different applications of
oxygen required in a standalone oxy-fired system in Case 2 (Ozcan oxygen (Zhu et al., 2016). For CLOU, released gas-phase oxygen
et al., 2014). reacts directly with solid fuel in the reduction reactor (Song et al.,
Since a large flow rate of solids are recirculated in the CaL cycle, 2013a), and the oxygen concentration at the outlet should be
a heat exchanger (simplified as a heat transfer between solids with around 4 vol % for ensuring complete combustion of the fuel (Song
a temperature approach of 20 C) is incorporated for transferring et al., 2013b). However, CLAS produces oxygen at a higher con-
sensible heat between the CaO particles leaving the calciner and centration to meet the requirement for oxy-fuel combustion pro-
the solids entering into it to be heated up to the calcination tem- cess (Song et al., 2013b).
€
perature (Ozcan et al., 2017). Energy consumption derived from The reduction reactor and the oxidation reactor are linked
solid transportation has been set at 20 MJ per ton of solids (Edwards together through a loop seal to prevent gas leakage from one
and Materi c, 2012). The final CO2 product is compressed to 100 bar reactor to another (Moghtaderi, 2012). Air is fed into the oxidation
by CCU. Net power output is calculated by subtracting auxiliary reactor, so that the incoming reduced carrier particles can be re-
power requirements. Compared with Case 1, the gross power generated to a higher oxidation state (Moghtaderi, 2012). The re-
output of the plant with CaL case increases to 554.37 MWe. The generated carrier particles, in turn, are transported back to the
overall plant efficiency of the CaL post case is 30.84% with a CO2 reduction reactor, where oxygen decoupling occurs, in which high-
capture efficiency of 82.7%, which implies an energy penalty of temperature flue gas from the combustor flows through it. 78% of
7.67% in the range of previous values reported in literature (Hanak the flue gas is first recycled to the reduction reactor for fluidizing
n et al., 2016).
et al., 2016; Martínez et al., 2016; Perejo the particles and heat-transfer purpose (Shah et al., 2015). Then the
oxygen-enriched flue gas is directly fed into the boiler of the oxy-
fuel plant. The continuous generation of the oxygen enriched flue
2.4. Description of case 4 (Oxy-CaL) gas is achieved by the continuous reduction and oxidation of oxy-
gen carrier. Oxides of Cu, as the most favorable oxygen carriers
A CO2 capture technology based on the integration of partial applied in CLOU, are chosen as the oxygen carrier for CLAS in this
oxy-fuel combustion and the CaL post-combustion capture is pro- paper, because of its higher oxygen transport capacity, higher
posed by Ortiz et al. (2017). The boiler is initiated by partial oxy-fuel reactivity (Wang et al., 2016a), lower inventory, and higher rate of
combustion of coal under a mixture of air, pure oxygen (95% purity), oxygen decoupling (Wang et al., 2016b). The quantity of metal
and recycled flue gas (RFG). Compared with total oxy-fuel com- oxygen carriers (mMOC ) needed per year in CLAS is calculated as
bustion (Case 2), the energy penalty for partial oxy-fuel combustion follows, ignoring physical wear rate (Zhou et al., 2015):
is notably reduced. Since the amount of pure O2 for partial oxy-fuel
combustion is substantially decreased, the total power consump- Hplant
mMOC ¼ mB $NBR ¼ mB $ (3)
tion of the CASU is notably reduced, although the specific electrical LOC
power consumption is unchanged (200 kW h per ton of O2)
(Romano et al., 2012). Moreover, the CPU for CO2 purification is not where NBR is number of bed replacements per year; Hplant is annual
needed, because this step is carried out after the CaL process. operation time of power plant (hours); LOC is life of oxygen carrier
Depending on the air/O2/RFG ratio, the CO2 vol concentration in unit it shows significant loss of reactivity (1000 h is assumed); and
flue gas stream exiting the partial oxy-fuel combustion boiler mB is bed inventory (ton), which is calculated as follows:
and loop seals also have auxiliary power consumption (Shah et al.,
R $60 2013b), which can be provided by electrical power.
mB ¼ S (4)
1000 The specific electrical power consumption required for CLAS is
set to 74 kW h per ton of O2 (Shah et al., 2015). The CASU is a
mO2 distillation process which is carried out at a high pressure (6e8 bar)
RS ¼ (5) and a low temperature (165e170 C), while CLAS is mainly a
ROT *
thermo-chemical process which operates at atmospheric pressure
where RS is solid circulation rate (kg/s); mO2 is actual oxygen mass and high temperatures (900 C for oxidation and 1000 C for
demand (kg/s), ROT * is rate of oxygen transport expressed on the reduction). Therefore, the specific electrical power consumption
basis of the weight loss of CuO corresponding to 80% conversion required for CLAS is much lower than CASU (Moghtaderi, 2009).
(3%/s for the system under investigation) Previous studies have shown that the thermal efficiency penalty
In this paper, the RFG rather than steam is employed in the associated with the Oxy-CLAS layout ranges from 4.1% to 6% (5.74%
reduction phase. The replacement of steam by RFG not only elim- in this paper) (Zhou et al., 2015), much lower than that of oxy-fuel
inates the need for steam generation but also avoids the condenser combustion with CASU (Case 2).
units for separation of O2 from steam (Moghtaderi, 2012). This leads
to a more effective integration of the ASU with the oxy-fuel plant 3. Systems analysis methodology
because of better use of material and energy streams. It also re-
duces the number of unit operations and, thereby, capital cost, as Fig. 3 presents the flowchart of the systems analysis method-
well as operational and running costs. Zhou et al. (2015) also ology of this work. First, we perform LCA to calculate the compre-
confirmed that the power demand and the thermal efficiency hensive resource consumption and environmental impacts of a SC
penalty associated with the CLAS using RFG are lower than that pulverized coal power plant with and without CCS (the five cases
using steam by about 2.5%. described in the previous section). Next, we analyze the economic
From an energy-efficiency point of view, the CLAS technology is performance by evaluating the levelized cost of electricity (LCOE,
quite efficient because of its low energy demands (Zhu et al., 2016). $/MW$h) and CO2 avoidance cost (CAC, $ per ton of CO2 avoidance)
In theory, the thermal energy generated from the exothermic of this SC pulverized coal power plant with CCS. The next step is to
oxidation reaction is equal to that required by endothermic weight each criterion from different stakeholders’ perspectives
reduction. Furthermore, under steady-state operation, much of the using AHP, which models the decision process using a hierarchy,
heat required for preheating air to the desired operating temper- followed by selection of the most suitable CCS technology for the SC
ature is offset by the heat contents of the high-temperature oxygen pulverized coal power plant using TOPSIS. The TOPSIS approach
enriched flue gas leaving the reduction reactor. In practice, external chooses an alternative with the shortest distance from the Positive
energy input is needed for the reduction reactor to compensate the Ideal Solution (PIS) and the longest distance from the Negative Ideal
heat losses to the surroundings (Moghtaderi, 2012). The air blower Solution (NIS).
Fig. 3. Flowchart of the proposed systems analysis methodology. (AHP: Analytic Hierarchy Process; TOPSIS: Technique for Order Preference by Similarity to Ideal Solution; ALO:
agricultural land occupation; ECC: climate change (ecosystems); FET: freshwater ecotoxicity; FE: freshwater eutrophication; MET: marine ecotoxicity; NLT: natural land trans-
formation; TA: terrestrial acidification; TET: terrestrial ecotoxicity; ULO: urban land occupation; HCC: climate change (human health); HT: human toxicity; OR: ionizing radiation;
OD: ozone depletion; PMF: particulate matter formation; POF: photochemical oxidant formation; FD: fossil depletion; MD: metal depletion; LCOE: levelized cost of electricity; CAC:
CO2 avoidance cost).
3.1. Life cycle environmental impact analysis and outputs to and from the system (Kursun et al., 2014). Data
should be adapted and dimensioned to the functional unit. The
Life cycle assessment is a systems analysis tool for compre- present LCA study is based on the energy and material consump-
hensively quantifying the material and energy flows to and from tion of each unit process. Several assumptions are made in the life
the environment over the life cycle of a product or activity (Clarens cycle inventory analysis:
et al., 2016). LCA study is comprised of four phases: goal and scope
definition, inventory analysis, impact assessment, and interpreta- (1) All heat and power consumed in circulating water, SCR, ESP,
tion (ISO, 2006). FGD, oxygen production, and CCS units are provided by this
power generation system itself. The remaining electricity is
3.1.1. Goal definition and scope sold to the electric power grid.
The goal of this LCA study is to quantify and analyze resource (2) The operation lifetime of all capital equipment is designed to
consumption and environmental impacts of electricity generation be 30 years.
and simultaneously capturing CO2 in a SC coal power plant using a (3) The CO2 capture efficiency of Oxy with CASU or CLAS, CaL
bituminous coal with medium sulfur content. The functional unit is post-combustion, and Oxy-CaL technologies are assumed to
selected as 1 MW h of net power produced following the existing be 90%, 82.7%, and 97.6%, respectively (Ortiz et al., 2017).
LCA studies on CCS systems (Yue et al., 2015). The net power pro- (4) The coal supply chain contains the mining construction,
duced is obtained by subtracting the ancillary power consumption mining operation, coal pre-processing operation, coal
from the gross electric power for each case (Leperi et al., 2016). cleaning, and coal transportation. Underground mining is
The system boundary contains: i) the power generation of po- used for coal extraction. The coal pre-processing operation
wer plants coupled with CCS technologies based on individual Oxy, includes the cutting, drilling, blasting, loading, and hauling
CaL post, multiple Oxy-CaL and Oxy-CLAS; ii) upstream processes of processes. The electricity requested for coal pre-processing
the power plant, such as building materials production, building operation varies in the range of 12e124 kWh per ton of
materials transportation, construction of the power plant and CO2 raw coal (Spath et al., 1999), and an average value of 85 kW
capture infrastructure; iv) upstream processes of fuel and chem- per ton of raw coal is used in this paper (Petrescu et al., 2017).
icals, such as mining construction, mining operation, coal pre- A quantity of 0.79 MJ/tone of raw coal is required for coal
operation, coal transportation, chemicals production and chem- washing (Spath et al., 1999). All coals are transported by train
icals transportation; v) downstream processes including decom- (powered by electricity) with a distance of 250 km from the
missioning of the power plant, solid waste treatment, and flue gas coal mine to the plant. The data related to mining con-
treatment with SCR, ESP, FGD and CCS. The following items are struction, mining operation, and coal transportation are ob-
excluded from the system boundaries of this study: i) construction tained from the Ecoinvent 3.3 database (Ecoinvent Centre,
of roads and railways, as well as building of trains and trucks for 2016). It is estimated that 0.02%, 0.02%, and 0.075% of the
transportation; ii) the transmitting of electricity to the transmission coal is lost at the loading, unloading, and transportation
and distribution network, and the delivery of the electricity to the phrases, respectively (Petrescu and Cormos, 2017).
customer; iii) indirect land use; iv) transportation, storage and (5) The oxygen carriers are transported to the power plant by
applications of CO2; v) low-frequency, high-magnitude, non- trucks with Euro III emission standards, and the average
predictable environmental events, such as accidental releases. transportation distance to the plant is 100 km. All building
Major unit processes of this LCA are summarized in Fig. 4. materials are transported by train (electricity) with the dis-
tance of 100 km.
(6) The spent oxygen carrier is reused until it permanently lacks
3.1.2. Life cycle inventory analysis
further CO2 capturing capacity. All solid waste including ash
Inventory analysis includes collecting data about flows of inputs
Fig. 4. Unit processes boundary of the life cycle analysis.

and deactivated oxygen carriers are dumped back into the collected inventory and represents them in aggregate category in-
coal mine from which the coal is supplied by railway trans- dicators. ReCiPe methodology is used to conduct this assessment.
portation (Kursun et al., 2014). This methodology integrates two LCIA methods, the CML 2000
(7) 158,758 kg concrete, 50,721 kg steel, 419 kg aluminum, and method, developed by the Institute of Environmental Sciences,
619 kg iron per MW plant capacity are used for constructing a Leiden University (NL) and the Ecoindicator 99 method (Goedkoop
pulverized coal power plant without CCS (Spath et al., 1999). e, 2001). Resource consumption and emissions,
et al., 2009; Guine
Due to the lack of data regarding the construction of CaL and including the related up-stream and down-stream activities, are
CLAS systems, 25% of material flow and emissions during the calculated and systematically analyzed. Three damage (endpoint)
construction of the pulverized coal power plant are assumed categories, namely, ecosystem quality, human health, and re-
for the CaL post-combustion system (Petrescu et al., 2017). sources, are taken into account in ReCiPe methodology. Ecosystem
81% of material flow and emissions for the CASU construction quality is comprised of nine midpoint categories: agricultural land
are assumed for the CLAS construction, because the con- occupation (ALO), ecosystems climate change (ECC), freshwater
struction cost of CLAS is about 81% of that of CASU (Zhou ecotoxicity (FET), freshwater eutrophication (FE), marine ecotox-
et al., 2015). The energy and building materials require- icity (MET), natural land transformation (NLT), terrestrial acidifi-
ment for CASU and CO2 comparison construction are ob- cation (TA), terrestrial ecotoxicity (TET), urban land occupation
tained from Althaus et al. (2004) and Faist Emmenegger et al. (ULO). Human health is comprised of six midpoint categories: hu-
(2003), respectively. The remaining data related to building man health climate change (HCC), human toxicity (HT), ionizing
material and chemicals production and transportation are radiation (OR), ozone depletion (OD), particulate matter formation
obtained from the Ecoinvent 3.3 database (Ecoinvent Centre, (PMF), photochemical oxidant formation (POF). Fossil depletion
2016). (FD) and metal depletion (MD) comprise resources.
(8) The values for energy, materials and emissions released
during commissioning are multiplied by 10% to account for
decommissioning (Petrescu and Cormos, 2017). 3.1.4. Results interpretation
The objective of performing life cycle interpretation is to iden-
The energy, materials and emissions for life cycle inventory (LCI) tify, quantify, check, and evaluate information from the results of
of 1 MW h net electricity for the five cases is summarized in Table 3. the life cycle inventory and the life cycle impact assessment, in
accordance with the goal and scope of the study (ISO, 2006). In this
3.1.3. Life cycle impact assessment paper, interpretation provides a complete and understandable
Life cycle impact assessment (LCIA) calculates impacts from the presentation of the life cycle impact results of a SC pulverized coal
Table 3
Energy, materials and emissions inventory for 1 MW h net electricity.
Item Unit Case 1: Case 2: Case 3: Case 4: Case 5:

without CCS Oxy CaL post Oxy-CaL Oxy-CLAS
Raw Materials-Fuel
coal kg/MW h 317.446 407.896 396.385 410.207 373.041
Raw Materials-Chemicals
catalyst for SCR m3/MWe 0.107 0.000 0.134 0.000 0.000
limestone for FGD kg/MW h 27.532 7.783 34.378 18.425 7.118
ammonia for SCR kg/MW h 1.283 0.000 1.608 0.000 0.000
Ca-based sorbents for CaL kg/MW h 0.000 0.000 311.965 314.858 0.000
Cu-based oxygen carriers for CLAS kg/MW h 0.000 0.000 0.000 0.000 0.554
cooling water kg/MW h 92.650 129.710 147.600 139.550 129.710
Raw Materials-Building Materials
concrete kg/MW h 0.711 0.751 0.767 0.722 0.687
steel kg/MW h 0.227 0.241 0.245 0.231 0.221
aluminum kg/MW h 0.002 0.002 0.002 0.002 0.002
iron kg/MW h 0.003 0.003 0.003 0.003 0.003
copper kg/MW h 0.007 0.007 0.007 0.007 0.006
polyethylene kg/MW h 0.004 0.004 0.004 0.004 0.004
Direct Gas Emission
CO2 kg/MW h 860.759 110.546 128.550 17.906 101.100
SO2 kg/MW h 0.270 0.000 0.000 0.000 0.000
PM > 10 kg/MW h 0.177 0.002 0.177 0.177 0.002
PM < 10 kg/MW h 0.067 0.067 0.067 0.067 0.067
NOX kg/MW h 0.714 0.098 0.891 0.099 0.090
CO kg/MW h 0.059 0.059 0.059 0.059 0.059
Pb kg/MW h 2.13E-05 2.13E-06 2.13E-05 2.13E-05 2.13E-06
Hg kg/MW h 6.08E-06 0.00Eþ00 6.08E-06 6.08E-06 0.00Eþ00
As kg/MW h 3.30E-06 3.30E-07 3.30E-06 3.30E-06 3.30E-07
NH3 kg/MW h 4.94E-04 4.72E-04 4.94E-04 4.94E-04 4.72E-04
HF kg/MW h 2.13E-04 0.00Eþ00 2.13E-04 2.13E-04 0.00Eþ00
VOC kg/MW h 2.95E-03 2.95E-03 2.95E-03 2.95E-03 2.95E-03
methane kg/MW h 1.380 1.380 1.380 1.380 1.380
MEA kg/MW h 1.99E-07 1.99E-07 1.99E-07 1.99E-07 1.99E-07
PAH kg/MW h 3.55E-05 3.55E-05 3.55E-05 3.55E-05 3.55E-05
Direct Solid Emission
solid waste kg/MW h 30.090 37.048 25.975 23.143 33.882
gypsum kg/MW h 9.080 0.000 9.080 9.080 0.000
unreactive sorbent/oxygen carrier kg/MW h 0.000 0.000 311.965 314.858 0.554
Table 4 FOM includes operating labor costs, overhead charges, and

Capital requirements of main components of the pulverized coal power plant. maintenance costs. Operating labor costs are estimated based on
Item Unit Value Date Source the number of personnel required to operate the plant per shift. The
coal handing system $/kWe 114.81 Zhou et al. (2015)
number of shifts to ensure continuous operation is dependent on
coal prepare and feed system $/kWe 39.59 Zhou et al. (2015) the type of plant. This operating labor costs is estimated as $1.5/
feed water system $/kWe 140.86 Zhou et al. (2015) MW$h (Zhou et al., 2015). All other business services not directly
PC boiler and accessories $/kWe 765.58 Zhou et al. (2015) involved in the operations of the plant fall into this category, such
heat recovery steam generator $/kWe 35.73 Zhou et al. (2015)
as management, administration, personnel services, technical ser-
（HRSG）, ducting and stack
steam turbine generator $/kWe 420.87 Zhou et al. (2015) vices and clerical staff. This cost is estimated as $0.1/MW$h (Zhou
cooling water system $/kWe 101.27 Zhou et al. (2015) et al., 2015). The annual maintenance cost of the plant is esti-
ASH/spent sorbent handing $/kWe 41.00 Zhou et al. (2015) mated as a percentage of the installed capital cost of the facilities,
access electric plant $/kWe 210.62 Zhou et al. (2015)
and the percentage is in the range of 2e2.5%, depending on facility
Instrument and control $/kWe 42.88 Zhou et al. (2015)
system（I&G)
type (Mancuso et al., 2017).
ESP $/kWe 30.56 Hu and Zhai (2017) Variable operating and maintenance cost is dependent on the
SCR $/kWe 47.70 Mancuso et al. (2017) plant operating load and proportional to the amount of production.
FGD for air combustion $/kWe 98.53 Mancuso et al. (2017) The unit purchasing prices of limestone for FGD, ammonia for SCR,
FGD for oxy-fuel combustion $/kWe 63.86 Mancuso et al. (2017)
SCR catalysts, Cu-based oxygen carriers for CLAS, Ca-based sorbents
CASU $/kW-CAUS 4623.58 Zhou et al. (2015)
CLAS $/kW-CLAS 3742.35 Zhou et al. (2015) for CaL, and cooling water are set to be $22.14/ton (Cormos, 2014),
CaL post-combustion $/kWe 254.35 Cormos (2014) $221.37/ton (Cormos and Cormos, 2017), $4.1/kWe/yr (Mancuso
CPU for oxy $/kWe 254.35 Zhou et al. (2015) et al., 2017), $25.76/ton (Mantripragada and Rubin, 2014), $5250/
CO2 compression and $/kWe 55.74 Cormos (2014) ton (Zhou et al., 2015), and $2.84/ton (Khan and Shamim, 2016),
drying for post
respectively. The treatment cost of ash is set to be $11.07/ton
(Cormos and Cormos, 2017). The average annual sale prices of coal
at mines producing bituminous coal is $51.57/ton, and the average
power plant with and without CCS, and explains significant limi- coal transportation cost is $10.75/ton (U.S. Energy Information
tations caused by application of CCS for coal power plants. More- Administration, 2017).
over, interpretation evaluates the opportunities and provides The CO2 avoidance cost (CAC) is an economic indicator that is
insightful recommendations to reduce the life cycle impact of the widely used to measure the increment in generation cost for
SC pulverized coal power plant with various CCS technologies. avoiding the emission of CO2 in the field of techno-economic
evaluation of fossil fuel power plants with CCS (Gong and You,
3.2. Techno-economic analysis 2014). CAC can be calculated by (Wang et al., 2010):

The goal of this techno-economic analysis is to compare the LCOEcapture LCOEref
CAC ¼ (8)
economic performance of a SC coal power plant with and without
mco2 ref mco2 capture
CCS, and to identify the economical optimum CCS technology.
For simplicity, it is assumed that the net output of the power
plant and the operating, maintenance, and fuel costs are constant where mCO2 refers to the CO2 emission amount (ton/MW$h), the
over the life of the plant. Thus, a functional unit based approach is subscript ‘capture’ and ‘ref’ refer to the power plants with carbon
used for the techno-economic analysis (Yue et al., 2013), and the capture units and conventional power plant without CCS capture,
resulting economic metric is the levelized cost of electricity (LCOE, respectively.
$/MW$h), which can be calculated from the following set of The transportation, storage and applications of CO2 are excluded
equations: from the system boundaries of the LCA, so the corresponding costs
for transportation, storage and applications of CO2 is neglected in
ðTCR FCFÞ þ FOM this TEA. The power plant's demolition cost at the end of the life
LCOE ¼ þ VOM þ mF FCF (6) cycle is assumed to be equal to its salvage value, so the decom-
NPO
missioning cost is 0.
where TCR is total capital requirement ($), FOM is fixed operating
and maintenance cost ($/yr), NPO is net power plant output (MWe),
VOM is variable operating and maintenance cost ($/MW$h), mF is 3.3. TOPSIS calculation
the fuel consumption rate (ton/MW$h), and FC is the fuel unit cost
($/ton). MCDM approach can be used to integrate LCA and techno-
Fixed charged factor (FCF) is calculated as follows, economic analysis results and to systematically compare various
alternatives. The TOPSIS approach is one of the most popular
rð1 þ rÞt MCDM approaches. Compared with Linear Programming Technique
FCF ¼ (7)
ð1 þ rÞt 1 for Multidimensional Analysis of Preference (LINMAP) and Viekri-
terijumsko Kompromisno Rangiranje (VIKOR), which require the
where r is the rate of capital discharge (%; assumed to be 7% in this selected solution to be the closest to the positive ideal point, TOPSIS
study), and t is power plant life (yr; assumed to be 30 yr in this is more suitable for the decision makers who want to avoid risk and
study) simultaneously maximize benefit (Wang and Rangaiah, 2017). This
The breakdown of unit capital requirements of the pulverized is because the TOPSIS approach simultaneously considers the dis-
coal power plant is listed in Table 4. All currencies other than USD$ tances to both positive and negative ideal solutions (Ramos et al.,
are converted using the average exchange rate from the year of the 2014). Besides, the number of steps in TOPSIS is entirely unre-
cost data (Dong et al., 2009), and all the capital cost are then lated to the number of attributes (Tansel, 2012), and the TOPSIS
updated to values in 2016 using the Chemical Engineering Plant approach is less complex than the approaches of Elimination and
Cost Index (CEPCI) (Towler and Sinnott, 2012). Choice Translating Priority III (ELECTRE III) and Preference Ranking
Organization Method for Enrichment of Evaluations (PROMETHEE) the highest capture efficiency, the impact values caused by direct
€
(Ozcan et al., 2017). An alternative with a higher relative closeness emissions for the power plant using Oxy-CaL are smaller than those
coefficient (Ci ) value is considered as a better one. The alternative using the other three CCS technologies.
with the highest Ci is considered as an overall priority. The key For the whole life cycle, application of CCS to pulverized coal
steps in TOPSIS are given in Appendix A. power plants alleviates pollution but increases resource con-
In this study, TOPSIS is used to rank the SC coal power plants sumption. Compared with Case 1 (without CCS), the total
without CCS, with Oxy, CaL post, Oxy-CLAS, and Oxy-CaL technol- ecosystem quality impact values of Case 2 (Oxy), Case 3 (CaL), Case
ogies by calculating their corresponding relative closeness 4 (Oxy-CaL), and Case 5 (Oxy-CLAS) at the life cycle level are
coefficient. reduced from 25.00 to 13.06, 14.09, 11.75, and 12.07, respectively, as
shown in Fig. 6. The total human health impact values of Case 2
3.4. AHP calculation (Oxy), Case 3 (CaL), Case 4 (Oxy-CaL) and Case 5 (Oxy-CLAS) are
reduced from 47.21 to 28.32, 31.24, 26.77, and 27.49, respectively.
The main criticism of TOPSIS is that the relative importance of The total resources impact values of Case 2 (Oxy), Case 3 (CaL), Case
the separations is not fully considered, and this criticism raises the 4 (Oxy-CaL) and Case 5 (Oxy-CLAS) are increased from 28.85 to
question on how to combine the subjective preferences of decision 35.66, 36.15, 36.62, and 33.34, respectively. The order of decreasing
makers. To answer this question, the AHP method which allows ecosystem quality performance and human health performance is:
multi-stakeholder inclusion can be combined with TOPSIS (Saaty, Case 4 (Oxy-CaL), Case 5 (Oxy-CLAS), Case 2 (Oxy), Case 3 (CaL) and
1980). AHP is a powerful tool to simplify complicated and ill- Case 1 (CCS), whereas the order of increasing resources perfor-
structured problems by arranging the decision attributes and al- mance is Case 1 (CCS), Case 5 (Oxy-CLAS), Case 2 (Oxy), Case 3 (CaL)
ternatives in a hierarchical structure with the help of a series of and Case 4 (Oxy-CaL). Although direct emissions are the smallest,
pair-wise comparisons (Tavana and Hatami-Marbini, 2011). More- the Oxy-CaL case has the highest energy penalty and worst re-
over, it can easily adjust the size to accommodate decision-making sources performance among the four cases using CCS technologies.
€ Fig. 7 illustrates various life-cycle midpoint impacts. Compared
problems because of its hierarchical structure (Ozcan et al., 2017).
with the case without CCS, the cases with CCS have higher values in
The AHP method has been widely used to aggregate the economic,
the midpoint indicators such as MD, ALO and ULO. The increase is
resource, and environmental objectives (Chen and Shonnard, 2004;
due to the introduction of some additional up-stream processes
Chen et al., 2002). The key steps in AHP are given in Appendix B.
such as CO2 capture facility infrastructure, oxygen carriers’ pro-
In this work, AHP is used to weight resource consumption,
duction and transportation, as well as the thermal efficiency pen-
environment impact and economic criteria for the SC coal power
alty. For the full-life cycle, the most important impact categories for
plant with or without CCS from the perspectives of local govern-
all cases are FD, HCC and ECC, followed by HT and PMF, with MD,
ment, industry, residents, and egalitarian, respectively, in order to
TET, NLT, ALO and ULO ranked behind. Pulverized coal power plants
reflect various stakeholders’ preferences.
have no obvious effect on other impact categories except the above
ten impact categories.
4. Results and discussion
4.1. Resource use and environmental impacts 4.2. Techno-economic performance
Fig. 5 presents the midpoint impacts caused by direct emissions The major components of capital cost of a pulverized coal power
for the five cases considered in this work. Compared with power plant with or without CCS are listed in Table 5. The order of
generation without CCS, pulverized coal power plants with Oxy, CaL increasing capital costs is Case 1 (CCS), Case 5 (Oxy-CLAS), Case 3
post, Oxy-CaL, and Oxy-CLAS reduce the total direct emissions by (CaL), Case 2 (Oxy), and Case 4 (Oxy-CaL). The oxygen required in
79.96%, 75.82%, 89.24%, and 80.96%, respectively. This is mainly due the calciners is only about 1/3 of that required for an oxy-fuel
to sharp decreases in ECC and HCC. Since Oxy-CaL technology has process, reducing the capital cost of CASU (MacKenzie et al.,
Fig. 5. Various ReCiPe midpoint impacts caused by direct emissions.

Fig. 6. Total ecosystem quality, human health and resources impacts in the full life cycle.
2007). The capital cost of the auxiliary CCU in the CaL case is lower fewer equipment units for the CLAS process (essentially only two
than that of CPU in the oxy-fuel combustion case. Besides, the boiler reactors) and lower operation pressure significantly reduces its
in the oxy-fuel combustion system should be upgraded and added capital cost (Shah et al., 2013a, 2013b). The replacement of CASU by
into the flue gas recycle unit and steel frame item, so there is an CLAS decreases the proportion of capital cost of ASU in TCR from
extra investment cost of 7% for the oxy-fuel boiler (Simbeck, 2001; 28.5% to 10.6%, making it no longer the limiting factor for oxy-fuel
Xiong et al., 2009). However, SCR and FGD devices in the post- plant. Therefore, the total capital cost of Case 3 (CaL) is smaller than
combustion plants cost more than those in the oxy-fuel combus- that of Case 2 (Oxy), but larger than that of Case 6 (Oxy-CLAS). The
tion plant, due to a smaller flue gas flow and a different removal capital cost of Case 5 is the highest, because it comprises the most
process of NOx and SOx after gas recycling (Catalanotti et al., 2015). auxiliary units.
CaL plant requires additional fuel consumption for its calciner and Annual O&M costs and total costs of pulverized coal incineration
generate additional electricity by heat recovery, so the coal handing power plant with and without CCS units are also shown in Table 5.
and feed system, heat recovery steam generator and steam turbine The LCOE values of the four cases with CCS units are in the range of
generator in Case 3 and Case 4 cost more than those in oxy-fuel $62.35~$80.74/MW$h, greater than the LCOE of the conventional
combustion plants (Case 2 and Case 5). Compared with the CASU, pulverized coal ($52.65/MW$h). For all cases, about 36e48% of
the capital cost of the CLAS equipment is much smaller because of LCOE is the capital cost, indicating that a promotion in the
Fig. 7. Various ReCiPe midpoint impacts throughout the full life cycle of pulverized coal power plants with and without CCS.
Table 5
Costs of a pulverized coal power plant with and without CCS units ($1 MM ¼ $1,000,000 USD).
Item Unit Case 1： Case 2： Case 3: Case 4: Case 5:

without CCS Oxy CaL post Oxy-CaL Oxy-CLAS
capital cost coal handing system MM$ 54.94 71.67 79.47 92.31 71.67
coal pre and feed system MM$ 18.94 24.71 27.40 31.83 24.71
feed water MM$ 67.41 68.44 78.09 87.64 68.44
PC boiler and accessories MM$ 366.38 398.00 366.38 398.00 398.00
heat recovery steam generator, MM$ 17.10 17.36 19.81 22.23 17.36
ducting and stack
steam turbine generator MM$ 201.42 204.48 233.32 261.88 204.48
cooling water system MM$ 48.46 49.20 56.14 63.01 49.20
ASH/spent sorbent handing MM$ 19.62 19.92 22.73 25.51 21.78
access electric plant MM$ 100.80 102.33 116.76 131.06 102.33
Instrument and control system (I&G) MM$ 20.52 20.83 23.77 26.68 20.83
ESP MM$ 14.62 14.62 14.62 14.62 14.62
SCR MM$ 22.83 0.00 22.83 0.00 0.00
FGD MM$ 47.15 31.03 47.15 39.73 31.03
CPU/CO2 compression MM$ 0.00 123.58 31.97 35.89 123.58
CASU/CLAS MM$ 0.00 455.77 159.79 522.25 136.49
CaL post-combustion MM$ 0.00 0.00 82.61 92.72 0.00
Total PTC MM$ 1000.21 1601.96 1382.86 1845.36 1284.54
fixed O&M cost operator labor $/MW$h 1.50 1.50 1.50 1.50 1.50
admin and support labor $/MW$h 0.10 0.10 0.10 0.10 0.10
maintenance cost of coal MM$/yr 1.85 2.41 2.67 3.10 2.41
handing and feed
maintenance cost of boiler MM$/yr 12.68 13.53 13.24 14.53 13.57
maintenance cost of steam cycle unit MM$/yr 6.69 6.79 7.75 8.70 6.79
maintenance cost of FGD and ESP MM$/yr 1.54 1.14 1.54 1.36 1.14
maintenance cost of SCR MM$/yr 0.57 0.00 0.57 0.00 0.00
maintenance cost of CASU/CLAS MM$/yr 0.00 9.12 3.20 10.44 2.73
maintenance cost of CaL MM$/yr 0.00 0.00 1.65 1.85 0.00
maintenance cost of CPU/CO2 MM$/yr 0.00 3.09 0.80 0.90 3.09
compression
variable O&M cost limestone for FGD $/MW$h 0.61 0.17 0.76 0.41 0.16
ammonia $/MW$h 0.28 0.00 0.36 0.00 0.00
SCR catalysts $/MW$h 0.61 0.00 0.76 0.00 0.00
Cu-based oxygen carriers $/MW$h 0.00 0.00 0.00 0.00 2.91
CaL sorbents $/MW$h 0.00 0.00 8.04 8.11 0.00
ash disposal $/MW$h 0.33 0.41 3.74 3.74 0.38
cooling water $/MW$h 0.26 0.37 0.42 0.40 0.37
Fuel cost $/MW$h 19.78 25.42 24.70 25.56 23.25
LCOE $/MW$h 52.65 73.63 74.99 80.74 62.35
CAC $/ton-CO2 0.00 27.96 30.51 33.33 12.77
investment is the most effective path to reduce the total cost. As work are taken from Dong et al. (2014). The values ‘3’ and ‘5’ mean
shown in Table 5, fuel costs account for about 30% of LCOE, indi- that the residents consider that economic and environmental in-
cating that a promotion in the thermal efficiency is effective to dicators are moderately and much more important than the
improve economic performance. The result shows that Oxy-CLAS resource consumption indicator, respectively (Dong et al., 2014).
has the best economic performance, followed by Oxy and CaL, For the matrix A related to the pair-wise comparison, the diagonal
with Oxy-CaL being the worst. values of the matrix are always ‘1’, and all judgments below the
In the pulverized coal power plant, the CAC values for Case 2 diagonal are the reciprocal of those above. As shown in Table 6, for
(with Oxy), Case 3 (with CaL), Case 4 (with Oxy-CaL) and Case 5 the pair-wise comparison based on the view of residents,
0 1
(with Oxy-CLAS) are $27.96, $30.51, $33.33 and $12.76 per ton of 1 0:2 0:333
CO2 avoidance, respectively. Previous studies showed that the CAC A¼ 5 @ 1 3 A. The weight factors are calculated by the
ranged from $39 to $96 per ton of CO2 avoidance in a coal power 3 0:333 1
plant with chemical amine absorption post-combustion capture eigenvalue method, and the vector w containing the weight factors
n et al., 2016). This indicates that the CCS technologies
(Perejo is the eigenvector of the matrix A. w ¼ ð0:106; 0:633; 0:260T Þ. lmax
considered in this paper have the potential to be economically is its maximal eigenvalue (lmax ¼ 3:04) Aw ¼ lw. According to
competitive with other existing chemical absorption CCS technol- Equations (20) and (21) (in Appendix B), the consistency index (CI)
ogies for power plants. and consistency ratio (CR) are 0.02 and 0.034. The CR is smaller than
0.1, indicating that the consistency of these pair-wise comparisons
is acceptable. Residents give the top priority to the environment,
4.3. AHP-TOPSIS analysis results and might have minimum concern on the resource. Therefore, the
weight factors of environment, economic and resource progres-
Table 6 lists the pair-wise comparison and weighting factors in sively decrease.
AHP based on the view of residents. A criterion on the left is The TOPSIS results from the residents' perspective are summa-
compared with another one indicated at the top. The numerical rize in Table 7. The normalized decision matrix is obtained using
values basically represent how many times or how much a criterion formula (13). The weighted normalized decision matrix is obtained
is more important than the other criteria. The representativeness by multiplying the normalized decision matrix by the weight
for different stakeholders and their pair-wise comparisons in this
Table 6 from the positive ideal and negative ideal solutions (Dþ and D ) are
Pair-wise comparison and weight factors for AHP based on the view of residents.
calculated by Equations (17) and (18). The Ci values calculated by
Resource Environment Economy Weight factor Equation (19) simultaneously reflect the distances to both positive
Resource 1 0.2 0.333 0.106 and negative ideal solutions, and the alternative with the largest Ci
Environment 5 1 3 0.633 value is chosen as the optimum. The comparison of Ci values in
Economy 3 0.333 1 0.260 Table 7 shows that a pulverized coal power plant with CCS outranks
the power plants without CCS from the residents’ standpoint.
Different stakeholder groups put different weights on each in-
Table 7 dividual criterion. Industry gives the top priority to the economic
TOPSIS result for a power plant with or without CCS unit based on the view of indicator, and might have minimum concern on the environment.
residents.
Local governments give preference to the energy consumption with
Item TOPSIS result economy as the least concerned indicator (Dong et al., 2014). Fig. 8
Resource Environment Economy summarizes the final ranking of the AHP-TOPSIS calculation results
consumption impact from the perspective of local government, industry, and residents,
Matrix as well as egalitarian approach (0.333 for each criterion).
Case 1 28.851 72.214 52.650 From the perspective of industry only, CCS has an unfavorable
Case 2 35.659 41.405 73.627 effect on the performance of the pulverized coal power plant, no
Case 3 36.150 45.328 74.989 matter which of the CCS technologies is adopted. Among the four
Case 4 36.617 38.534 80.743
Case 5 33.335 39.555 62.348
cases containing CCS, Case 5 (Oxy-CLAS) always has greater relative
Normalized decision matrix closeness coefficients (Ci ) than other capture technologies, no
Case 1 0.377 0.658 0.338 matter from the perspective of egalitarian, government, industry or
Case 2 0.466 0.377 0.473 residents. From the perspectives of local government and residents,
Case 3 0.472 0.413 0.482
Case 5 (Oxy-CLAS) is the best choice, followed by Case 2 (Oxy) and
Case 4 0.478 0.351 0.519
Case 5 0.435 0.361 0.401 Case 4 (Oxy-CaL), and then Case 3 (CaL), with Case 1 (without CCS)
Weighted normalized decision matrix being the worst. The disparity in comprehensive performance be-
Case 1 0.040 0.417 0.088 tween five cases from the government perspective is the smallest.
Case 2 0.049 0.239 0.123 The main criticism of AHP is that pair-wise comparisons are
Case 3 0.050 0.262 0.126
subjective. If ignoring the subjective pair-wise comparisons in AHP,
Case 4 0.051 0.222 0.135
Case 5 0.046 0.228 0.104 analysis on impacts of altering weight factors on the ranking results
Ideal solution of TOPSIS is conducted in Fig. 9, where the three criteria are rep-
positive value 0.040 0.222 0.088 resented separately on each axis. Different ranking situations of the
negative value 0.051 0.417 0.135
alternatives are distinguished by different colors filled within the
Dþ D Ci
Case 1 0.194 0.048 0.199
dots. As can be seen from the ternary graph, nearly two-thirds of
Case 2 0.040 0.178 0.817 the weight variations show that Case 5 (Oxy-CLAS) has the best
Case 3 0.055 0.156 0.738 comprehensive performance. Case 4 (Oxy-CaL) ranks first only
Case 4 0.048 0.194 0.801 when the weight factor of environmental criterion is increased to 1
Case 5 0.018 0.191 0.912
and 0.9, with the weight factor for resource is 0 and 0.1, respec-
tively. Case 1 (without CCS) ranks first when the sum of weight
factors of resource and economic criteria is equal to and greater
factors from AHP calculation. The economic indicator, resource than 0.9. As shown in Fig (b), Case 1 (without CCS) is the worst case
consumption and environmental impacts are cost criteria, so their in half of the weight variations. Case 3 (CaL) or Case 4 (Oxy-CaL)
positive and negative ideal solutions take minimum and maximum become the worst when increasing the weight factor of resource or
values, respectively. The separation measures of each alternative economic criteria.
Fig. 8. Relative closeness coefficients of the AHP-TOPSIS calculation from the perspectives of local government, industry, residents, and egalitarian.
subsystem in the life cycle with the continuous social development,

and absence of local experimental data on CCS equipment and
related technological parameters are the error sources of
assessment.
In order to better understand the impacts of altering parameters
on the economic performance of the SC pulverized coal power plant
with CCS, sensitivity analysis is conducted. For illustration, Fig. 10
presents the sensitivity analysis result of CO2 avoidance cost (CAC)
based on changes in coal price, energy consumption of ASU, energy
consumption of CCU/CPU, capital investment of ASU, investment of
CCU/CPU and sorbent/oxygen carrier price. Among the six factors,
the energy consumption of ASU is the most critical factor influ-
encing CAC for Oxy (Case 2) and Oxy-CaL (Case 4), but investment of
ASU and sorbent/oxygen carrier price are the second important
factors influencing CAC for Oxy (Case 2) and for Oxy-CaL (Case 4),
respectively. Although sorbent/oxygen carrier price has no influ-
ence on environmental performance of power plants, it has a
slightly greater impact on CAC than energy consumption of ASU, and
becomes the most critical factor influencing CAC for the CaL case
(Case 3). Further research on alternative low-cost and high-
reactivity materials for oxygen carrier benefits the economic per-
formance of CaL (Case 3) and Oxy-CaL (Case 4). The most important
factor for Oxy-CLAS (Case 5) is the energy consumption of the CPU,
and its CAC is less sensitive to energy efficiency of the ASU than that
in other cases, indicating that the energy efficiency of the ASU is no
longer the bottleneck and energy efficiency of the CPU deserves
more attention to be effectively improved for Oxy-CLAS (Case 5).
With the further development of CCS technologies, energy
consumption of CCS devices and the make-up rate of oxygen carrier
gradually decrease. Thus, it is important to evaluate how the
decrease in energy consumption of the ASU and CCU/CPU systems
and consumption of sorbent/oxygen carrier influence the LCOE and
ReCiPe endpoint scores with respect to SC pulverized coal power
plants. Fig. 11 presents the effects of energy consumption of CCS
devices and consumption of sorbent/oxygen carrier on the LCOE,
and ReCiPe endpoint scores. For Case 2 and Case 4, the most critical
factor influencing their LCOE and the three ReCiPe endpoint scores
is energy consumption of the ASU, followed by energy consumption
of the CPU/CCU with consumption of sorbent/oxygen carrier
ranked next. For Case 5, the most critical factor influencing its LCOE
and three ReCiPe endpoint scores is the energy consumption of the
Fig. 9. Effect of weight factors on ranking results of the alternatives: (a) the optimal
case; (b) the worst case. CPU, and the effects of energy consumption of the ASU and con-
sumption of sorbent/oxygen carrier are roughly similar. For the
three ReCiPe endpoint scores in Case 3, the effects of the energy
In LCA and TEA, the results on the resource consumption, consumption of the ASU and CCU are similar, and greater than the
environmental burden, and financial cost are reported separately. effect of consumption of sorbent/oxygen carrier. However, the
Among the four CCS options, Oxy-CLAS has the advantages of the consumption of sorbent/oxygen carrier is the most critical factor
smallest resource consumption and the smallest economic penalty, influencing the LCOE in Case 3. In Case 4 and Case 2, LCOE and the
and Oxy-CaL has the advantages of the best ecosystem quality and three ReCiPe endpoint scores are more sensitive to energy con-
human health performance. Oxy-CaL has the disadvantages of the sumption of the ASU than those in Case 3 and Case 5. The impact of
largest economic penalty and the largest resources consumption, energy consumption of CPU/CCU on LCOE is most obvious in Case 4,
and CaL has the disadvantages of the worst ecosystem quality and but its impacts on the three ReCiPe endpoint scores are most
human health performance. The above separated LCA and TEA re- obvious in Case 2 and Case 5. The variation of LCOE and resources
sults only provide qualitative comparison based on a specific cri- and ecosystem quality performance due to variations of sorbent/
terion. Instead, the weighted data by implementing MCDM oxygen carrier consumption in Case 3 and Case 4 are more pro-
provides a comprehensive quantitative indicator. As shown in nounced than those in Case 2 and Case 5. However, decreasing
Figs. 8 and 9, the Oxy-CLAS technology is more promising than the consumption of sorbent/oxygen carrier has a more significant in-
other three CCS technologies. fluence on human health performance in Case 5 than the other
three cases. The consumption of sorbent/oxygen carrier has no
4.4. Sensitivity analysis impact on Case 2.
Fig. 12 shows the impacts of decreasing energy consumption of
The accurate absolute results of environmental impact and costs the ASU, energy consumption of the CCU/CPU systems, and con-
cannot be expected, since no power plants with Oxy-CaL and Oxy- sumption of sorbent/oxygen carrier on relative closeness co-
CLAS are currently in operation and some techno-economic as- efficients (Ci ) from different standpoints. As shown in Fig. 11, the
sumptions involve in the analysis. Uncertainty of variation of each fluctuations in the energy consumption of the CPU/CCU influence
Fig. 10. Sensitivity analysis result of CO2 avoidance cost (CAC).
Fig. 11. Effects of energy consumption of CCS devices and consumption of sorbent/oxygen carrier on the LCOE, and ReCiPe endpoint scores.
the Ci values, but the differences are not large enough to change the combustion with CLAS (Case 5) is the likeliest CCS candidate to be
ranking result, regardless of perspective. The decrease in the energy applied in power plants, and industry will prefer the case without
consumption of the ASU can change the ranking sequence of Case 4 CCS even if the efficiency of CCS's devices and sorbent/oxygen
and Case 3 from industry's standpoint, but the ranking results from carrier can be significantly improved.
the perspective of residents, government and egalitarian still If the price of industrial CO2 avoidance is considered, the pref-
remain the same even when the energy consumption of the ASU erence of industry will change, as the carbon price has a strong
decreases by 50%. The ranking result of Case 2, Case 3, and Case 4 impact on the economic performance of power plants. Fig. 13 shows
changes from the perspective of residents and industry, if the the minimum carbon price that could make the targeted SC coal
consumption of sorbent/oxygen carrier decreases by 25% or 50%. power plant with CCS outperform the conventional power plant
However, only when the consumption of sorbent/oxygen carrier without CCS under different conditions, from the industry's
decreases by 50%, Case 4 and Case 2 exchange the ordered sequence standpoint. For example, Oxy-CLAS case will rank first no matter
with each other from the perspective of government and egali- from the perspectives of local government, industry, residents, or
tarian. This indicates that reducing the consumption of oxygen egalitarian, if the carbon price reaches $6.1 per ton of CO2 avoid-
carrier makes Oxy-CaL outrank Oxy. The sequencing result of Case 1 ance. A decrease in investment and energy consumption of the CCS
and Case 5 show a strong robustness. Specifically, oxy-fuel devices can decrease the minimum carbon price to make Oxy-CLAS
Fig. 12. Effects of energy consumption of CCS devices and consumption of sorbent/oxygen carrier on the relative closeness coefficients from different standpoints.
as the highest-ranking case. Under different conditions, when the 30% and 60% that of Cu oxides, respectively (Wang et al., 2016). The
carbon price is above $8.0~$27.0, $21.7~$30.4, and $18.3~$30.7 per price of Mn and Co oxides are $2290/ton and $30,300/ton,
ton CO2 avoidance, Oxy (Case 2), CaL (Case 3), and Oxy-CaL (Case 4), respectively (Zou, 2015). Moreover, their production process is
respectively, can outperform the conventional SC power plant from different. Replacement of CuO by Mn2O3 leads to a dramatically
the industry's standpoint. negative impact on the resource performance, making the
Fig. 14 shows how the type, wear rate, and active life of oxygen comprehensive performance of Oxy-CLAS worse than Oxy and Oxy-
carrier influence the LCOE, CAC and ReCiPe endpoint scores of Oxy- CaL from the government's perspective. Replacing CuO by Co3O4
CLAS. Fig. 15 presents the ranking result of Oxy CLAS in MCDM causes the economic performance of Oxy-CLAS to dramatically
based on changes in the characteristics of the oxygen carrier. Cu, deteriorate, mainly because of the high price of Co3O4. Replacement
Mn, and Co oxides can release oxygen through reactions (9), (10) of CuO by Co3O4 makes the ranking result of Oxy-CLAS close to CaL
and (11), so are identified as oxygen carrier candidates for CLAS and and Oxy-CaL from the industry's perspective. The minimum carbon
CLOU (Mattisson et al., 2009; Shah et al., 2012). Among the three price to make the targeted power plant with Oxy-CLAS outrank the
candidates, the oxygen transport capacity of Mn and Co oxides is conventional power plant without CCS increases to $12.9 and $34.6
Fig. 13. Minimum carbon price to make the SC coal power plant with CCS outrank the conventional power plant from the industry's standpoint.
Fig. 14. Impacts of the type, wear rate and active life of oxygen carrier the LCOE, CAC and ReCiPe endpoint scores of Oxy-CLAS.
Fig. 15. Effects of the type, wear rate and active life of oxygen carrier in CLAS on the relative closeness coefficients from different standpoints.
per ton of CO2 avoidance when Mn and Co oxides are chosen as

oxygen carrier, respectively. Therefore, Cu oxide is considered to be
the most promising oxygen carrier for CLAS. mMOC ¼ mB $NBR, 1 þ rOC $Hplant
Hplant
¼ mB $ $ 1 þ rOC $Hplant (12)
4CuO⇔2Cu2 O þ O2 (9) LOC
where rOC is the wear rate of oxygen carrier, assumed to be 0.04%/h

6Mn2 O3 ⇔4Mn3 O4 þ O2 (10) in this paper (Zou, 2015).
As shown in Fig. 15, although the economic indicators of Oxy-
CLAS (Case 5) becomes worse than those under the base condi-
tion, they are still better than the other three CCS alternatives. The
2Co3 O4 ⇔6CoO þ O2 (11)
advantage of Oxy-CLAS (Case 5) over Oxy (Case 2) is offset by the
If the wear rate of Cu-based oxygen carrier is considered in CLAS, wear rate of the Cu-based oxygen carrier from the perspective of
the quantity of metal oxygen carriers (mMOC ) needed per year in residents, government and egalitarian. If only the active life of Cu-
CLAS is calculated by Eq. (12). based oxygen carrier (LOC) is shortened form 1,000 he500 h, the
ranking results remain the same, indicating the Oxy-CLAS (Case 5)

still is the likeliest CCS candidate for all the stakeholders. fij
r ij ¼ sffiffiffiffiffiffiffiffiffiffiffiffi; i ¼ 1; 2; :::; m; j ¼ 1; 2; :::n (13)
It is noteworthy that the transportation, storage and applica- Pm
tions of CO2 are excluded from the system boundaries in this paper. f 2ij
j¼1
The aforementioned minimum carbon price and CAC will increase if
the cost, resources consumption, and emissions during trans-
where rij is the normalized value for each criterion, fij is original
portation and storage of CO2 are taken into account. The effects of
value for each criterion, m is the number of alternatives (m ¼ 5), and
applications of CO2 on minimum carbon price and CAC are more
n is the number of criteria (n ¼ 3).
complicated, and this issue is outside this study but is worthy of
further study.
Step 2: Calculate the weighted normalized decision matrix:
This paper presents an analysis for a specific case study and the
results aim to enable quick assessment of the SC pulverized coal vij ¼rij wj ; i ¼ 1; 2; :::; m; j ¼ 1; 2; :::n (14)
power plants firing bituminous coal with and without CCS. How-
ever, the methodology applied in this work is not case specific, and
where vij is the weighted normalized value for each criterion and wj
it also can be adopted for other power plants and other capture
is criterion weight from the matrix in AHP
technologies.
Step 3: Calculate the positive (Aþ ) and negative (A ) ideal

solutions:
5. Conclusion
( ! !)
n o
In this work, we conducted multi-criteria environmental and Aþ ¼ vþ þ þ
1 ; v2 ; :::vn ¼ maxjj2I þ ; minjj2I (15)
economic analyses of SC pulverized coal power plants with CCS. At 1jm 1jm
the full life cycle, the application of CCS to a SC pulverized coal
( ! !)
power plant reduced the ecosystem quality impact value from n o
25.00 to 11.75e14.09, reduced the human health impact value from A ¼ v
1 ; v2 ; :::vn ¼ minjj2I þ ; maxjj2I (16)
47.21 to 26.77e31.24 and increased the resources impact value 1jm 1jm
from 28.85 to 33.34e36.62. The most important impact categories

for all cases are FD, HCC and ECC, followed by HT and PMF, and then where I þ and I are benefit criteria and cost criteria, respectively.
MD, TET, NLT, ALO and ULO. Among the four CCS options for power
plants, Oxy-CLAS (Case 5) has the best economic performance, Step 4: Calculate the separation measures of each alternative:
followed by Oxy (Case 2) and CaL (Case 3), with Oxy-CaL (Case 4) vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX
being the worst. Their LCOE and economic penalty of CO2 avoidance u n 2
Dþ
i
¼t vij vþ j
(17)
values were in the range of $62.35~$80.74/MW$h, and
j¼1
$12.76~$33.33/ton, respectively. AHP-TOPSIS results showed that
regardless of stakeholder perspective, Oxy-CLAS performed better vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uX
than other capture technologies. Though Oxy-CaL (Case 4) had the u n 2
best ecosystem quality and human health performance, it D
i ¼t vij v j
(18)
j¼1
comprehensively performed worse than Oxy (Case 2) in MCDM. It
was found that decreasing the consumption of Ca-based sorbent
was critical for Oxy-CaL (Case 4) to outrank the Oxy (Case 2). From where Dþ i
and D i are the separation from positive ideal and
the perspective of industry only, CCS had an unfavorable effect on negative ideal solutions, respectively.
the performance of the pulverized coal power plant, so promoting
the carbon price is still necessary to attract more support from Step 5: Calculate the relative closeness coefficient (Ci ) for each
industry. Future extensions of this work can consider the social alternative:
dimension of sustainability through economic input-output anal-
ysis (You et al., 2012), the multiobjective optimization approach of D
i
Ci ¼ (19)
integrating LCA and techno-economic analysis (Yue et al., 2016), Dþ
i
þ D
i
multi-stakeholder economic and environmental optimization (Gao
and You, 2017), and a consequential perspective of the analysis
(Gong and You, 2017). Step 6: Rank alternatives based on their Ci values
Acknowledgments Appendix B
The authors acknowledge financial support from National Sci- The key steps in AHP are as following (Saaty, 1980):
ence Foundation CAREER Award (CBET-1643244), Natural Science
Foundation of China (51606071) and Natural Science Foundation of Step 1: Stratify the decision-making process in a hierarchy of
Guangdong Province (2016A030310424). levels with goal at the top level, followed by criteria, sub-
criteria, and alternatives at the bottom level.
Step 2: Saaty's 1e9 scale is used to judge the relative importance
Appendix A of each criterion in Pair-wise comparisons (Saaty, 1980). A larger
value in pair-wise comparison means larger differences be-
The key steps involved in TOPSIS methodology are: tween criteria levels.
Step 3: For each comparison matrix, maximum eigenvalue,
Step 1: Calculate the normalized decision matrix: consistency index (CI), consistency ratio (CR), and normalized
eigenvector are calculated to obtain priority weights for each ROT * rate of oxygen transport expressed on the basis of the
criteria/alternative. weight loss of CuO corresponding to 80% conversion
(3%/s)
ðlmax nÞ t power plant life (yr; assumed to be 30 yr in this study)
CI ¼ (20)
ðn 1Þ TCR total capital requirement ($)
vij weighted normalized value for each criterion
where lmax is obtained by calculating the scalar product of the VRFG RFG volume (m3/h)
principal eigenvector and the vector of column sums of the matrix VOM variable operating and maintenance cost ($/MW$h)
(Saaty, 2007). wj criterion weight from the matrix in AHP
CI Wg total electricity generation (kWe)
CR ¼ (21) Wnet net power output (kWe)
RI
Wself total auxiliary system requirements in the whole plant
where RI is the Random Consistency Index. (kWe)
DP total change of pressure (kPa)
Step 4: A consistency test is used to evaluate judgments to avoid
the unreasonable subjectivity in pair-wise comparison. Pair- Greek Symbols
wise comparisons with the CR value greater than 10% must be a excess air factor
reset. The inconsistency with the CR value lower than 10% is ht overall net efficiency (%)
acceptable. hFan energy conversion efficiency of fans (75%)
lmax maximum egenvalue of comparison matrix
Nomenclature References
Abanades, J.C., Rubin, E.S., Anthony, E.J., 2004. Sorbent cost and performance in CO2
A matrix related to pair-wise comparison capture systems. Ind. Eng. Chem. Res. 43 (13), 3462e3466.
Aþ positive ideal solutions Almås, K., 2012. Coal-fired Power Plants Based on Oxy-combustion with Carbon
Capture: Combustion Conditions and Water Consumption. Institutt for energi-
A negative ideal solutions
og prosessteknikk. Norwegian University of Science and Technology, Trond-
CAC CO2 avoidance cost ($ per ton of CO2 avoidance) heim, Norway.
Ci relative closeness coefficient Althaus, H.J., Chudacoff, M., Hellweg, S., Hischiert, R., Jungbluth, N., Osses, M.,
Primas, A., 2004. Life Cycle Inventories of Chemicals. Ecoinvent Report No. 8.
CI consistency index
Swiss Centre for Life Cycle Inventories, Dubendorf, Switzerland.
CR consistency ratio Basu, S., Debnath, A., 2014. Power Plant Instrumentation and Control Handbook: a
Dþ i
separation from positive ideal solution Guide to Thermal Power Plants. Elsevier, Amsterdam, Netherland.
Catalanotti, E., Hughes, K.J., Porter, R.T.J., Price, J., Pourkashanian, M., 2015. Evalu-
D i separation from negative ideal solution ation of performance and cost of combustion-based power plants with CO2
ERFG power requirement of this RFG fan is (kWe) capture in the United Kingdom. Environ. Prog. Sustain. Energy 33 (4),
fij original value for each criterion 1425e1431.
Chen, H., Shonnard, D.R., 2004. Systematic framework for environmentally
FC fuel unit cost ($/ton) conscious chemical process design: early and detailed design stages. Ind. Eng.
FCF fixed charged factor Chem. Res. 43 (2), 535e552.
Chen, H., Wen, Y., Waters, M.D., Shonnard, D.R., 2002. Design guidance for chemical
FOM fixed operating and maintenance cost ($/yr)
processes using environmental and economic assessments. Ind. Eng. Chem. Res.
Hplant annual operation time of power plant (hours; assumed 41 (18), 4503e4513.
to be 7446 h in this study) Chung, Y.G., Gomez-Gualdron, D.A., Li, P., Leperi, K.T., Deria, P., Zhang, H.,
Vermeulen, N.A., Stoddart, J.F., You, F., Hupp, J.T., Farha, O.K., Snurr, R.Q., 2016. In
Iþ benefit criteria silico discovery of metal-organic frameworks for precombustion CO2 capture
I cost criteria using a genetic algorithm. Sci. Adv. 2 (10), e1600909.
LOC life of oxygen carrier unit it shows significant loss of Clarens, F., Espí, J.J., Giraldi, M.R., Rovira, M., Vega, L.F., 2016. Life cycle assessment of
CaO looping versus amine-based absorption for capturing CO2 in a subcritical
reactivity (hours; assumed to be 1000 h in this study)
coal power plant. Int. J. Greenh. Gas. Con. 46, 18e27.
LCOE levelized cost of electricity ($/MW$h) Cormos, A.M., Cormos, C.C., 2017. Reducing the carbon footprint of cement industry
m number of alternatives (set to be 5) by post-combustion CO2 capture: techno-economic and environmental
assessment of a CCS project in Romania. Chem. Eng. Res. Des. 123, 230e239.
mB bed inventory (ton)
Cormos, C.C., 2014. Economic evaluations of coal-based combustion and gasification
mcoal consumption of coal (kg/s) power plants with post-combustion CO2 capture using calcium looping cycle.
mCO2 mass of CO2 (ton/MW$h) Energy 78, 665e673.
Croiset, E.T.K., 2001. NOx and SO2 emissions from O2/CO2 recycle coal combustion.
mF fuel consumption rate (ton/MW$h)
Fuel 80, 2117e2121.
mMOC quantity of metal oxygen carriers needed per year in Deveci, M., Demirel, N.Ç., John, R., Ozcan, € E., 2015. Fuzzy multi-criteria decision
CLAS making for carbon dioxide geological storage in Turkey. J. Nat. Gas Sci. Eng. 27,
mO2 actual oxygen mass demand (kg/s) 692e705.
Dong, C.Q., Yang, Y.P., Zhang, J.J., Lu, X.F., 2009. Gaseous emissions from the com-
n number of criteria (set to be 3) bustion of a waste mixture containing a high concentration of N(2)O. Waste
NBR number of bed replacements per year (set to be 8) Manag. 29 (1), 272e276.
NPO net power plant output (MWe) Dong, J., Chi, Y., Zou, D., Fu, C., Huang, Q.X., Ni, M.J., 2014. Energy-environment-
economy assessment of Waste Manage systems from a life cycle perspective:
QLHV lower heating value of coal (kJ/kg) model development and case study. Appl. Energy 114, 400e408.
r rate of capital discharge (%; assumed to be 7% in this Ecoinvent Centre, 2016. Ecoinvent Database Version 3.3. Bern, Switzerland.
study) Edwards, S.E.B., Materi c, V., 2012. Calcium looping in solar power generation plants.
Sol. Energy 86 (9), 2494e2503.
rij normalized value for each criterion Erlach, B., Schmidt, M., Tsatsaronis, G., 2011. Comparison of carbon capture IGCC
rOC wear rate of oxygen carrier (%/h; assumed to be 0.04%/h with pre-combustion decarbonisation and with chemical-looping combustion.
in this study) Energy 36 (6), 3804e3815.
Escudero, A.I., Espatolero, S., Romeo, L.M., Lara, Y., Paufique, C., Lesort, A.L.,
RS solid circulation rate (kg/s) Liszka, M., 2016a. Minimization of CO2 capture energy penalty in second gen-
RI random consistency index eration oxy-fuel power plants. Appl. Therm. Eng. 103, 274e281.
Escudero, A.I., Espatolero, S., Romeo, L.M., 2016b. Oxy-combustion power plant capture. Ind. Eng. Chem. Res. 55 (12), 3338e3350.
integration in an oil refinery to reduce CO2 emissions. Int. J. Greenh. Gas. Con. Lerche, N., Wilkens, I., Schmehl, M., Eigner-Thiel, S., Geldermann, J., 2017. Using
45, 118e129. methods of Multi-Criteria Decision Making to provide decision support con-
Faist Emmenegger, M., Heck, T., Jungbluth, N., 2003. Erdgas -Sachbilanzen von cerning local bioenergy projects. Soc. Plann. Sci. https://doi.org/10.1016/
Energiesystemen. Swiss Centre for Life Cycle Inventories, Paul Scherrer Insti- j.seps.2017.08.002.
tute, Dubendorf and Villigen, Switzerland. Li, Y.H., Radovic, L.R., Lu, G.Q., Rudolph, V., 1999. A new kinetic model for the
Fozer, D., Sziraky, F.Z., Racz, L., Nagy, T., Tarjani, A.J., Toth, A.J., Haaz, E., Benko, T., NOecarbon reaction. Chem. Eng. Sci. 54 (19), 4125e4136.
Mizsey, P., 2017. Life cycle, PESTLE and multi-criteria decision analysis of CCS Liang, X., Wang, Z., Zhou, Z., Huang, Z., Zhou, J., Cen, K., 2013. Up-to-date life cycle
process alternatives. J. Clean. Prod. 147, 75e85. assessment and comparison study of clean coal power generation technologies
Gao, J., You, F., 2017. Economic and environmental life cycle optimization of in China. J. Clean. Prod. 39, 24e31.
noncooperative supply chains and product systems: modeling framework, Luo, C., Zheng, Y., Yin, J., Qin, C., Ding, N., Zheng, C., Feng, B., 2013. Effect of sulfation
mixed-integer bilevel fractional programming algorithm, and shale gas appli- during oxy-fuel calcination stage in calcium looping on CO2 capture perfor-
cation. ACS Sustain. Chem. Eng. 5 (4), 3362e3381. mance of CaO-based sorbents. Energy Fuel 27 (2), 1008e1014.
Gebreslassie, B.H., Waymire, R., You, F., 2013. Sustainable design and synthesis of MacDowell, N., Florin, N., Buchard, A., Hallett, J., Galindo, A., Jackson, G.,
algae-based biorefinery for simultaneous hydrocarbon biofuel production and Adjiman, C.S., Williams, C.K., Shah, N., Fennell, P., 2010. An overview of CO2
carbon sequestration. AIChE J. 59 (5), 1599e1621. capture technologies. Energy Environ. Sci. 3 (11), 1645e1669.
Goedkoop, M., Heijungs, R., Huijbregts, M., De Schryver, A., Struijs, J., Van Zelm, R., MacKenzie, A., Granatstein, D., Anthony, E., Abanades, J., 2007. Economics of CO2
2009. ReCiPe: a Life Cycle Impact Assessment Method Which Comprises capture using the calcium cycle with a pressurized fluidized bed combustor.
Harmonised Category Indicators at the Midpoint and the Endpoint Level. Den Energy Fuel 21 (2), 920e926.
Haag, Netherland. Mancuso, L., Cloete, S., Chiesa, P., Amini, S., 2017. Economic assessment of packed
Gong, J., You, F., 2014. Global optimization for sustainable design and synthesis of bed chemical looping combustion and suitable benchmarks. Int. J. Greenh. Gas.
algae processing network for CO2 mitigation and biofuel production using life Con. 64, 223e233.
cycle optimization. AIChE J. 60 (9), 3195e3210. Mantripragada, H.C., Rubin, E.S., 2014. Calcium looping cycle for CO2 capture: per-
Gong, J., You, F., 2015. Sustainable design and synthesis of energy systems. Curr. formance, cost and feasibility analysis. Energy Procedia 63, 2199e2206.
Opinion Chem. Eng. 10, 77e86. Manuilova, A., 2011. Evaluation of Environmental Performance of Carbon Capture
Gong, J., You, F., 2017. Consequential life cycle optimization: general conceptual and Storage Project in Canada Using Life Cycle Assessment Methodology. Uni-
framework and application to algal renewable diesel production. ACS Sustain. versity of Regina, Regina, SK, Canada.
Chem. Eng. 5 (7), 5887e5911. Mardani, A., Zavadskas, E.K., Khalifah, Z., Zakuan, N., Jusoh, A., Nor, K.M.,
Guine e, J.B., Gorree, M., Heijungs, R., Huppes, G., Kleijn, R., de Koning, A., van Khoshnoudi, M., 2016. A review of multi-criteria decision-making applications
Oers, L., Sleeswijk, A.W., Suh, S., Udo de Haes, H.A., de Bruijn, H., van Duin, R., to solve energy management problems: two decades from 1995 to 2015. Renew.
Huijbregts, M.A.J., 2001. Life Cycle Assessment. An Operational Guide to the ISO Sustain. Energy Rev. 71, 216e256.
Standards. Final Report. Ministry of Housing, Planning and the Environment Martínez, I., Grasa, G., Parkkinen, J., Tynja €l€
a, T., Hyppa €nen, T., Murillo, R.,
(VROM) and the Centre of Environmental Science e Leiden University (CML), Romano, M.C., 2016. Review and research needs of Ca-Looping systems
Leiden, Netherland. modelling for post-combustion CO2 capture applications. Int. J. Greenh. Gas.
Hanak, D.P., Kolios, A.J., Manovic, V., 2016. Comparison of probabilistic performance Con. 50, 271e304.
of calcium looping and chemical solvent scrubbing retrofits for CO2 capture Martínez, I., Murillo, R., Grasa, G., Abanades, J., 2011. Integration of a Ca-looping
from coal-fired power plant. Appl. Energy 172, 323e336. system for CO2 capture in an existing power plant. Energy Procedia 4,
Hu, B., Zhai, H., 2017. The cost of carbon capture and storage for coal-fired power 1699e1706.
plants in China. Int. J. Greenh. Gas. Con. 65, 23e31. Mattisson, T., Lyngfelt, A., Leion, H., 2009. Chemical-looping with oxygen uncou-
Hu, Y., Naito, S., Kobayashi, N., Hasatani, M., 2000. CO2, NOx and SO2 emissions from pling for combustion of solid fuels. Int. J. Greenh. Gas. Con. 3 (1), 11e19.
the combustion of coal with high oxygen concentration gases. Fuel 79 (15), Moghtaderi, B., 2009. Application of chemical looping concept for air separation at
1925e1932. high temperatures. Energy Fuel 24 (1), 190e198.
Hurst, T.F., Cockerill, T.T., Florin, N.H., 2012. Life cycle greenhouse gas assessment of Moghtaderi, B., 2010. Application of chemical looping concept for air separation at
a coal-fired power station with calcium looping CO2 capture and offshore high temperatures. Energy Fuel 24 (1), 190e198.
geological storage. Energy Environ. Sci. 5 (5), 7132e7150. Moghtaderi, B., 2012. Review of the recent chemical looping process developments
IEA (International Energy Agency), 2013. Key World Energy Statistics 2013. http:// for novel energy and fuel applications. Energy Fuel 26 (1), 15e40.
www.iea.org/publications/freepublications/. Mukherjee, S., Kumar, P., Yang, A., Fennell, P., 2015. Energy and exergy analysis of
IPCC, 2007. IPCC Fourth Assessment Report: Climate Change 2007, vols. 1340e1356 chemical looping combustion technology and comparison with pre-combustion
(Geneva, Switzerland). and oxy-fuel combustion technologies for CO2 capture. J. Environ. Chem. Eng. 3
Iribarren, D., Petrakopoulou, F., Dufour, J., 2013. Environmental and thermodynamic (3), 2104e2114.
evaluation of CO2 capture, transport and storage with and without enhanced Nie, Z., 2009. Life Cycle Modelling of Carbon Dioxide Capture and Geological Storage
resource recovery. Energy 50 (1), 477e485. in Energy Production. Imperial College, London, UK.
ISO, 2006. Environmental Management e Life Cycle Assessment ePrinciples and Nie, Z., Korre, A., Durucan, S., 2011. Life cycle modelling and comparative assess-
Framework. International Standards Organisation, Geneva in: ISO 14040. ment of the environmental impacts of oxy-fuel and post-combustion CO2
Jakobsen, J.P., Roussanaly, S., Mølnvik, M.J., Tangen, G., 2013. A standardized capture, transport and injection processes. Energy Procedia 4 (1), 2510e2517.
approach to multi-criteria assessment of CCS chains. Energy Procedia 37, Ohtsuka, Y., Wu, Z., 1999. Nitrogen release during fixed-bed gasification of several
2765e2774. coals with CO2: factors controlling formation of N2. Fuel 78 (5), 521e527.
Jin, B., Zhao, H., Zheng, C., 2015. Thermoeconomic cost analysis of CO2 compression Ortiz, C., Valverde, J.M., Chacartegui, R., 2016. Energy consumption for CO2 capture
and purification unit in oxy-combustion power plants. Energy Convers. Manag. by means of the calcium looping process: a comparative analysis using lime-
106 (2015), 53e60. stone, dolomite, and steel slag. Energy Technol. 4 (10), 1317e1327.
Kather, A., Scheffknecht, G., 2009. The oxycoal process with cryogenic oxygen Ortiz, C., Valverde, J.M., Chacartegui, R., Benítez-Guerrero, M., Perejo n, A.,
supply. Naturwissenschaften 96 (9), 993e1010. Romeo, L.M., 2017. The Oxy-CaL process: a novel CO2 capture system by inte-
Khan, M.N., Shamim, T., 2016. Techno-economic assessment of a plant based on a grating partial oxy-combustion with the Calcium-Looping process. Appl. Energy
three reactor chemical looping reforming system. Int. J. Hydrogen Energy 41 196, 1e17.
(48), 22677e22688. Ozcan, D.C., Alonso, M., Ahn, H., Abanades, J.C., Brandani, S., 2014. Process and cost
Koiwanit, J., Manuilova, A., Chan, C., Wilson, M., Tontiwachwuthikul, P., 2014. A life analysis of a biomass power plant with in situ calcium looping CO2 capture
cycle assessment study of a hypothetical Canadian oxy-fuel combustion carbon process. Ind. Eng. Chem. Res. 53 (26), 10721e10733.
dioxide capture process. Int. J. Greenh. Gas. Con. 28, 257e274. €
Ozcan, E.C., Ünlüsoy, B., Eren, T., 2017. A combined goal programmingeAHP
Koornneef, J., Ramírez, A., Turkenburg, W., Faaij, A., 2012. The environmental impact approach supported with TOPSIS for maintenance strategy selection in hydro-
and risk assessment of CO2 capture, transport and storage an evaluation of the electric power plants. Renew. Sustain. Energy Rev. 78, 1410e1423.
knowledge base. Prog. Energy Combust. 38, 62e86. Pak, P.S., Lee, Y.D., Ahn, K.Y., 2009. Characteristics and economic evaluation of a CO2
Koornneef, J., van Keulen, T., Faaij, A., Turkenburg, W., 2008. Life cycle assessment of -capturing repowering system with oxy-fuel combustion for utilizing exhaust
a pulverized coal power plant with post-combustion capture, transport and gas of molten carbonate fuel cell (MCFC). Energy 34 (11), 1903e1909.
storage of CO2. Int. J. Greenh. Gas. Con. 2 (4), 448e467. Pehnt, M., Henkel, J., 2009. Life cycle assessment of carbon dioxide capture and
Korre, A., Nie, Z., Durucan, S., 2010. Life cycle modelling of fossil fuel power gen- storage from lignite power plants. Int. J. Greenh. Gas. Con. 3 (1), 49e66.
eration with post-combustion CO2 capture. Int. J. Greenh. Gas. Con. 4 (2), Perejo n, A., Romeo, L.M., Lara, Y., Lisbona, P., Martínez, A., Valverde, J.M., 2016. The
289e300. calcium-looping technology for CO2 capture: on the important roles of energy
Kursun, B., Ramkumar, S., Bakshi, B.R., Fan, L.-S., 2014. Life cycle comparison of coal integration and sorbent behavior. Appl. Energy 162, 787e807.
gasification by conventional versus calcium looping processes. Ind. Eng. Chem. Perejon, A., Romeo, L.M., Lara, Y., Lisbona, P., Martinez, A., Valverde, J.M., 2016.
Res. 53 (49), 18910e18919. ChemInform abstract: the calcium-looping technology for CO2 capture: on the
Kursun, B., Ramkumar, S., Bakshi, B.R., Fan, L.S., 2012. Life cycle comparison of coal important roles of energy integration and sorbent behavior. Appl. Energy 162
gasification by conventional versus calcium looping processes. Ind. Eng. Chem. (15), 787e807.
Res. 53 (49), 18910e18919. Petrescu, L., Bonalumi, D., Valenti, G., Cormos, A.M., Cormos, C.C., 2017. Life cycle
Leperi, K.T., Snurr, R.Q., You, F., 2016. Optimization of two-stage pressure/vacuum assessment for supercritical pulverized coal power plants with post-
swing adsorption with variable dehydration level for postcombustion carbon combustion carbon capture and storage. J. Clean. Prod. 157.
Petrescu, L., Cormos, C.C., 2017. Environmental assessment of IGCC power plants Tang, Y., You, F., 2018. Multi-criteria environmental and economic analysis of
with pre-combustion CO2 capture by chemical & calcium looping methods. municipal solid waste incineration power plant with carbon capture and sep-
J. Clean. Prod. 158, 233e244. aration from the life cycle perspective. ACS Sustain. Chem. Eng. 6, 937e956.
Ramasubramanian, K., Verweij, H., Ho, W.W., 2012. Membrane processes for carbon Tansel, Y., 2012. An experimental design approach using TOPSIS method for the
capture from coal-fired power plant flue gas: a modeling and cost study. selection of computer-integrated manufacturing technologies. Robot. Com. Int.
J. Membr. Sci. 421, 299e310. Manuf. 28 (2), 245e256.
Ramos, M.A., Boix, M., Montastruc, L., Domenech, S., 2014. Multiobjective optimi- Tavana, M., Hatami-Marbini, A., 2011. A group AHP-TOPSIS framework for human
zation using goal programming for industrial water network design. Ind. Eng. spaceflight mission planning at NASA. Expert Syst. Appl. 38 (11), 13588e13603.
Chem. Res. 53 (45), 17722e17735. Toftegaard, M.B., Brix, J., Jensen, P.A., Glarborg, P., Jensen, A.D., 2010. Oxy-fuel
Rodríguez, N., Murillo, R., Abanades, J.C., 2012. CO2 capture from cement plants combustion of solid fuels. Prog. Energy Combust. 36 (5), 581e625.
using oxyfired precalcination and/or calcium looping. Environ. Sci. Technol. 46 Towler, G., Sinnott, R.K., 2012. Chemical Engineering Design: Principles, Practice and
(4), 2460. Economics of Plant and Process Design. Elsevier, Butterworth-Heinemann,
Romano, M.C., Martínez, I., Murillo, R., Arstad, B., Blom, R., Ozcan, D.C., Ahn, H., Netherland.
Brandani, S., 2012. Guidelines for Modeling and Simulation of Ca-looping Pro- U.S. Energy Information Administration, 2016. Annual Energy Outlook 2016: with
cesses. European Energy Research Alliance, Brussels, Belgium. Projections to 2040. Washington, DC. https://www.eia.gov/outlooks/aeo/.
Rubin, E.S., Rao, A.B., Berkenpas, M.B., 2007. Development and Application of U.S. Energy Information Administration, 2017. Coal Prices and Outlook. https://
Optimal Design Capability for Coal Gasification Systems. Technical Documen- www.eia.gov/energyexplained/index.cfm?page¼coal_prices.
tation: Oxygen-based Combustion Systems (Oxyfuels) with Carbon Capture and U.S. National Energy Technology Laboratory, 2012. Detailed Coal Specifications
Storage (CCS) Carbon, Pittsburgh, PA, USA. Quality Guidelines for Energy System Studies (Pittsburgh, PA, USA).
Rubin, E.S., Salmento, J.S., Frey, H.C., Abu-Baker, A., Berkenpas, M., 1991. Modeling of Vega, F., Sanna, A., Maroto-Valer, M., Navarrete, B., Abad-Correa, D., 2016. Study of
Integrated Environmental Control Systems for Coal-fired Electrical Generating the MEA degradation in a CO2 capture process based on partial oxy-combustion
Stations. Department of Energy, Pittsburgh, PA, USA. approach. Int. J. Greenh. Gas. Con. 54, 160e167.
Saaty, T.L., 1980. The Analytic Hierarchy Process: Planning, Priority Setting. Resource Wang, K., Yu, Q., Qin, Q., Zuo, Z., Wu, T., 2016a. Evaluation of Cu-based oxygen
Allocation. McGraw-Hill, New York, NY, USA. carrier for chemical looping air separation in a fixed-bed reactor. Chem. Eng. J.
Saaty, T.L., 2007. Time dependent decision-making; dynamic priorities in the AHP/ 287, 292e301.
ANP: generalizing from points to functions and from real to complex variables. Wang, K., Yu, Q., Zuo, Z., Hou, L., Qin, Q., 2016b. Redox performance of Cu-based
Math. Comput. Model. 46 (7e8), 860e891. oxygen carrier used in chemical looping air separation combined oxy-fuel
Sara, J., Stikkelman, R.M., Herder, P.M., 2015. Assessing complexity of carbon capture combustion technology. Appl. Therm. Eng. 98, 440e448.
and storage using multi-criteria decision-making methods. In: 2015 9th Annual Wang, Y., Zhao, Y.C., Zhang, J.Y., Zheng, C.G., 2010. Technical-economic evaluation of
IEEE International Systems Conference (SysCon), pp. 145e152. O2/CO2 recycle combustion power plant based on life-cycle. Sci. China Technol.
Scheffknecht, G., Al-Makhadmeh, L., Schnell, U., Maier, J., 2011. Oxy-fuel coal Sci. 53, 3284e3293.
combustion-A review of the current state-of-the-art. Int. J. Greenh. Gas. Con. 5, Wang, Z., Rangaiah, G.P., 2017. Application and analysis of methods for selecting an
S16eS35. optimal solution from the pareto-optimal front obtained by multiobjective
Shah, K., Moghtaderi, B., Wall, T., 2012. Selection of suitable oxygen carriers for optimization. Ind. Eng. Chem. Res. 56 (2), 560e574.
chemical looping air separation: a thermodynamic approach. Energy Fuel 26 Xiong, J., Zhao, H.B., Zheng, C.G., Liu, Z.H., Zeng, L.D., Liu, H., Qiu, J.R., 2009. An
(4), 2038e2045. economic feasibility study of O2/CO2 recycle combustion technology based on
Shah, K., Moghtaderi, B., Wall, T., 2013a. Effect of flue gas impurities on the per- existing coal-fired power plants in China. Fuel 88, 1135e1142.
formance of a chemical looping based air separation process for oxy-fuel Yan, J., Faber, R., Schmidt, T., Ross, G., Jacoby, J., Anheden, M., 2009. Flue-gas cleaning
combustion. Fuel 103, 932e942. processes for CO2 capture from oxyfuel combustion - experience of FGD and
Shah, K., Moghtaderi, B., Zanganeh, J., Wall, T., 2013b. Integration options for novel FGC at 30 MWth oxyfuel combustion plant. In: The 1st IEA GHGT International
chemical looping air separation (ICLAS) process for oxygen production in oxy- Oxyfuel Combustion Conference (Cottbus, Germany).
fuel coal fired power plants. Fuel 107, 356e370. You, F., Tao, L., Graziano, D.J., Snyder, S.W., 2012. Optimal design of sustainable
Shah, K., Zhou, C., Song, H., Doroodchi, E., Moghtaderi, B., 2015. A novel hybrid cellulosic biofuel supply chains: multiobjective optimization coupled with life
chemical-looping oxy combustor process for the combustion of solid and cycle assessment and input-output analysis. AIChE J. 58 (4), 1157e1180.
gaseous fuels: thermodynamic analysis. Energy Fuel 29 (2), 602e617. Yue, D., Gong, J., You, F., 2015. Synergies between geological sequestration and
Simbeck, D.R., 2001. CO2 mitigation economics for existing coal-fired power plants. microalgae biofixation for greenhouse gas abatement: life cycle design of car-
In: First National Conference on Carbon Sequestration. Washington DC, USA). bon capture, utilization, and storage supply chains. ACS Sustain. Chem. Eng. 3
Singh, B., Stromman, A.H., Hertwich, E., 2011a. Life cycle assessment of natural gas (5), 841e861.
combined cycle power plant with post-combustion carbon capture, transport Yue, D., Kim, M.A., You, F., 2013. Design of sustainable product systems and supply
and storage. Int. J. Greenh. Gas. Con. 5 (3), 457e466. chains with life cycle optimization based on functional unit: general modeling
Singh, B., Strømman, A.H., Hertwich, E.G., 2011b. Comparative life cycle environ- framework, mixed-integer nonlinear programming algorithms and case study
mental assessment of CCS technologies. Int. J. Greenh. Gas. Con. 5 (4), 911e921. on hydrocarbon biofuels. ACS Sustain. Chem. Eng. 1 (8), 1003e1014.
Song, H., Shah, K., Doroodchi, E., Moghtaderi, B., 2013a. Development of a CueMg- Yue, D., Pandya, S., You, F., 2016. Integrating hybrid life cycle assessment with
based oxygen carrier with SiO2 as a support for chemical looping air separation. multiobjective optimization: a modeling framework. Environ. Sci. Technol. 50
Energy Fuel 28 (1), 163e172. (3), 1501e1509.
Song, H., Shah, K., Doroodchi, E., Wall, T., Moghtaderi, B., 2013b. Analysis on Zhou, C., Shah, K., Moghtaderi, B., 2015. Techno-economic assessment of integrated
chemical reaction kinetics of CuO/SiO2 oxygen carriers for chemical looping air chemical looping air separation for oxy-fuel combustion: an Australian case
separation. Energy Fuel 28 (1), 173e182. study. Energy Fuel 29 (4), 2074e2088.
Spath, P.L., Mann, M.K., Kerr, D.R., 1999. Life Cycle Assessment of Coal-fired Power Zhu, L., Wang, F., Zhang, Z., 2016. Thermodynamic evaluation of a conceptual pro-
Production. National Renewable Energy Lab, Golden, CO, USA. cess for coal gasification coupled with chemical looping air separation. Chem.
Tan, X., 2012. Supercritical and ultrasupercritical coal-fired power generation. Bus. Eng. Process: Process Intensification 106, 33e41.
Pub. Admin. Stud. 7 (1), 53. Zou, X., 2015. Process Simulation and Analysis of Several Chemical Looping Com-
Tang, Y., Ma, X., Lai, Z., Zhou, D., Lin, H., Chen, Y., 2012. NOx and SO2 emissions from bustion Systems. Huazhong University of Science & Technology, Wuhan, China
municipal solid waste (MSW) combustion in CO2/O2 atmosphere. Energy 40 (1), (In Chinese).
300e306.

Journal of Cleaner Production: Yuting Tang, Fengqi You

Uploaded by

Copyright:

Available Formats

Journal of Cleaner Production: Yuting Tang, Fengqi You

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Cleaner Production: Yuting Tang, Fengqi You

Uploaded by

Copyright:

Available Formats

Journal of Cleaner Production 181 (2018) 271e292

Contents lists available at ScienceDirect

Journal of Cleaner Production

Life cycle environmental and economic analysis of pulverized coal

1. Introduction power plants could help to address this challenge (Mukherjee

Fig. 2. Schematic diagram of CLAS and CASU process.

Fig. 4. Unit processes boundary of the life cycle analysis.

Item Unit Case 1: Case 2: Case 3: Case 4: Case 5:

Table 4 FOM includes operating labor costs, overhead charges, and

4.1. Resource use and environmental impacts 4.2. Techno-economic performance

Fig. 5. Various ReCiPe midpoint impacts caused by direct emissions.

Item Unit Case 1： Case 2： Case 3: Case 4: Case 5:

subsystem in the life cycle with the continuous social development,

Fig. 10. Sensitivity analysis result of CO2 avoidance cost (CAC).

per ton of CO2 avoidance when Mn and Co oxides are chosen as

where rOC is the wear rate of oxygen carrier, assumed to be 0.04%/h

ranking results remain the same, indicating the Oxy-CLAS (Case 5)

Step 3: Calculate the positive (Aþ ) and negative (A ) ideal

from 28.85 to 33.34e36.62. The most important impact categories

You might also like