Economic Analysis
Economic Analysis
Applied Energy
journal homepage: www.elsevier.com/locate/apenergy
Energy and Power, School of Water, Energy and Environment, Cranfield University, Bedford, Bedfordshire MK43 0AL, UK
HIGHLIGHTS
Keywords: Oxy-fuel combustion is currently gathering attention as one of the promising options for capturing CO2 effi-
Carbon capture ciently, when applied to power plants, for subsequent carbon sequestration. However, this option requires a
Oxy-fuel turbine large quantity of high-purity oxygen that is usually produced in an energy-intensive air separation unit (ASU).
Cryogenic ASU Chemical looping combustion (CLC) is a technology with the potential of reducing the costs and energy penalties
Chemical looping combustion
associated with current state-of-the-art cryogenic ASUs. In this work, the techno-economic performance of a
Natural gas combined cycle
Oxygen production
natural gas-fired oxy-combustion cycle with cryogenic ASU is compared with that based on CLC. Two natural
gas-fired cycles are considered: (i) staged oxy-fuel natural gas combined cycle as a reference; and (ii) gas-fired
CLC with supercritical CO2 cycle. The process models were developed in Aspen Plus® in order to evaluate the
thermodynamic performance of the proposed system and to benchmark it against the reference cycle. The results
show that the net efficiency of the proposed cycle, including CO2 compression, is more than 51%, which is
comparable to that of a conventional natural gas combined cycle with CO2 capture and 2.7% points higher than
that of the reference cycle. Moreover, the economic evaluation indicates that a reduction in levelised cost of
electricity from £38.3/MWh to £36.1/MWh can be achieved by replacement of the ASU-based oxy-fuel system
with CLC. Hence, gas-fired CLC with a supercritical CO2 cycle has high potential for commercialisation.
Abbreviations: ASU, Air Separation Unit; CCS, Carbon Capture and Storage; CLC, Chemical Looping Combustion; CPU, Carbon Purification Unit; EGR, Exhaust Gas
Recycle; LCOE, Levelised Cost of Electricity; NGCC, Natural Gas Combined Cycle; OC, Oxygen Carrier; PCC, Post-Combustion Capture; sCO2, Supercritical CO2; SOFC,
Solid Oxide Fuel Cell; SOF-NGCC, Staged Oxy-fuel Natural Gas Combined Cycle; TIP, Turbine Inlet Pressure; TIT, Turbine Inlet Temperature
⁎
Corresponding author.
E-mail address: [email protected] (D.P. Hanak).
https://doi.org/10.1016/j.apenergy.2019.04.096
Received 30 November 2018; Received in revised form 29 March 2019; Accepted 16 April 2019
Available online 04 May 2019
0306-2619/ © 2019 Elsevier Ltd. All rights reserved.
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
In chemical looping combustion (CLC), O2 is separated from the air transport capacity. Furthermore, regarding the performance of different
and then transported to the fuel via an oxygen carrier (OC). In its CLC configurations, Brandvoll and Bolland [13] revealed that a net
preferred embodiment, CLC consists of two interconnected fluidised efficiency of 54% was obtainable with integration of CLC to a humid air
bed reactors: an air reactor and a fuel reactor. O2 separated from the air turbine cycle. Naqvi and Bolland [14] showed that using CLC in the
in the air reactor is transferred by OCs to the fuel reactor [6]. The conventional combined cycle can result in a net efficiency of 53%, in-
chemical reactions in the air and fuel reactors, respectively, are as cluding CO2 compression. Chen et al. [15] analysed the possibilities and
follows: benefits of integration of CLC with supercritical CO2 for combined heat
and power cogeneration. Their proposed cycle reached a net power
Mex Oy 2 (s ) + Air (g ) Mex Oy (s ) + Oxygen depleted air (g ) (R1)
efficiency of 41.3% with a heating efficiency of 40.4%. However, the
Mex Oy (s ) + fuel Mex Oy (R2) economic feasibility of such concept has not yet been proven. Spallina
2 (s ) + Combustion products (g )
et al. [16] performed a thermodynamic analysis on the integration of
The reduced OC is then recycled to the air reactor. The gas stream solid oxide fuel cells (SOFCs) with CLC in natural gas-fired power
leaving the fuel reactor comprises mainly CO2 and water vapour. The plants. The integrated plant showed a net efficiency in the range of
benefit of using CLC is that the combustion products (CO2 and water 63–70%. Hamers et al. [17] analysed a two-stage CLC integrated with a
vapour) are inherently separated from other components such as N2 and coal gasification plant. They revealed that a two-stage CLC can achieve
Ar. Hence, unlike in the case of post-combustion capture (PCC), no a net efficiency of about 40% and results in significantly lower reactor
additional energy is required for CO2 separation [7]. cost compared to a single-stage CLC. Olaleye and Wang [18] performed
CLC is a promising technology [8] which has been developed for an economic analysis of a CLC unit integrated with a humid air turbine
combustion of gaseous, liquid and solid fuels and is undergoing sig- that was characterised with an efficiency of 57.1%. Their study has
nificant scale-up at present [9]. The largest (120 kW) gas-based CLC shown that for a 50 MWth plant, the capital cost and net present value
system is located at the Vienna University of Technology [10]. Fur- are £52M and £104M, respectively, assuming a cost of electricity of
thermore, a number of OCs have been thoroughly studied. Källén et al. £77.5/MWh. However, to support commercialisation of the CLC tech-
[11,12] tested calcium manganite in a 10 kW pilot-scale CLC unit and nology, its techno-economic feasibility needs to be further evaluated,
iron/manganese/silicon-based OCs in a 300 W lab-scale CLC unit. All especially considering advanced power cycles, such as the supercritical
performed well with respect to gas conversion, achieving high O2
238
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
CO2 cycle (sCO2) that is considered as a suitable replacement for con- This process continues until O2 is almost entirely (except for 5% excess
ventional steam cycles [19]. O2) consumed at the last combustion stage. The exhaust gas, which is
This work evaluates the techno-economic feasibility of integrating mainly composed of water vapour and CO2, is expanded in the turbine
CLC with the sCO2 cycle to achieve high efficiency with low CO2 to generate power. The flue gas then passes through a heat exchanger
emissions and affordable electricity cost. A novel gas-fired chemical and transfers heat to the O2, fuel and CO2 streams in the sCO2 cycle. The
looping combustion process with supercritical CO2 cycle (CLC-sCO2) is water vapour is easily separated from the cold flue gas and the re-
proposed and its performance is benchmarked against staged oxy-fuel maining CO2 is sent to the carbon purification unit (CPU) to be con-
natural gas combined cycle (SOF-NGCC). The process model of the CLC- ditioned for storage. In the bottoming cycle, the sCO2 stream is first
sCO2 is developed in Aspen Plus®. Parametric studies are performed to pressurised to 300 bar and then preheated in a high-temperature re-
determine the optimum thermodynamic performance, characterised cuperator (HTR) and low-temperature recuperator (LTR), as well as
with the highest net efficiency. Finally, the economic feasibility of the multi-stage combustor, before entering the high-pressure CO2 turbine.
considered cycles is evaluated and compared in terms of capital cost A detailed description of SOF-NGCC is presented in Khallaghi et al.
and levelised cost of electricity. Therefore, this work demonstrates the [23].
advantage of using CLC instead of an ASU-based oxy-fuel combustor for As shown in Fig. 2, the CLC-sCO2 power section is the same as the
high-efficiency low-CO2-emission power generation. It also proves the closed sCO2 cycle used as a bottoming cycle in the SOF-NGCC. Con-
economic feasibility of linking CLC with the sCO2 cycle that is char- versely, the heat is transferred indirectly from two fluidised bed re-
acterised by a higher efficiency and smaller size compared to other actors, oxygen-depleted air and exhaust gas to the sCO2 cycle. Then, the
cycles, such as the conventional steam cycle. Such information on hot sCO2 enters the high-pressure CO2 turbine. Two reactors work at
process design, operation, and techno-economic feasibility will support considerably lower pressure (1.25 bar) than that in SOF-NGCC com-
further development of CLC for power generation. bustors (300 bar). Although this results in no power obtained from
exhaust gas and oxygen-depleted air leaving the reactors, it is expected
that such reactors will have a significantly smaller capital cost. Finally,
2. Process description and simulation to ensure that the separated CO2 leaves the system at the pressure re-
quired for its storage, the CO2 stream is pressurised to 120 bar after
Fig. 1 presents a schematic of the SOF-NGCC. The entire quantity of water separation.
high-purity O2, required to ensure complete combustion, enters into the
first combustion stage, whereas natural gas enters into each of three
combustion stages with nearly equal feed rates. The combustion pro- 2.1. Model development
ducts of the first stage, along with unreacted O2, act as a diluent for the
second-stage combustion. Additionally, to maintain the desired com- The process model for CLC-sCO2 has been developed in Aspen Plus®.
bustion temperature of the topping cycle, excess heat is extracted from The package used for the thermodynamic property estimation is the
the first stage to pre-heat the CO2 stream in the sCO2 cycle primarily by Peng Robinson equation of state which is suitable for hydrocarbons and
radiation, as described by Gopan et al. [20,21]. Absence of the exhaust light gases, such as CO2 and H2 [24]. All components are defined as
gas recycle (EGR) results in a high combustion temperature and heat conventional except for OCs which are solid. The reactors are modelled
transfer. This may result in a high surface temperature that may exceed as Gibbs reactors (RGibbs), which assume chemical and phase equili-
allowable operating limits. This challenge is addressed by staging the brium based on Gibbs energy minimisation [25]. Importantly, it is as-
delivery of the fuel and controlling the characteristics of the flame [22]. sumed that heat loss is negligible. All heat exchangers are modelled
Fig. 2. Schematic of CLC-sCO2 (sCO2 gains heat, Depleted Air-1 loses heat, Flue gas-1 loses heat, Air-2 gains heat, sCO2-4 loses heat, sCO2-5 loses heat).
239
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
using the MHeatX block and designed based on the assumption that the sales. Therefore, the LCOE indicates the minimum electricity price re-
minimum temperature approach of the heat exchangers is 5 °C, simi- quired for the system to become economically feasible. This approach
larly to previous studies on the sCO2 cycle [26,27]. Similarly to the correlates the net power output, net thermal efficiency, and capacity
study by Hanak and Manovic [19] and Le Moullec [28], it is assumed factor (CF) with economic performance indicators, such as total capital
that high-pressure CO2 in the sCO2 cycle is heated in the CLC reactors. requirement (TCR), variable (VOM) and fixed (FOM) operating and
All turbines are modelled as individual turbine sections using the Compr maintenance costs, specific fuel cost (SFC) and the fixed charge factor
block. Cu has been selected as OC for this study due to its high reaction (FCF). The main assumptions for the economic analysis are described in
rate, high O2 transport capacity and its complementary combustion Table 3.
thermodynamics that completely convert fuel to CO2 and water vapour TCR × FCF + FOM SFC
[29,30]. It is assumed that the agglomeration rate of Cu is negligible. LCOE = + + VOM
8760 × Wnet × CF (2)
Moreover, due to the high oxidation conversion rate of Cu [31] and its net
high combustion temperature, it is assumed that the complete oxidation Importantly, FCF, which divides the total capital cost into uniform
of Cu occurs [32]. The properties of the natural gas, key assumptions annual amounts over the project lifetime, is calculated using Eq. (3)
used in modelling, the turbomachinery and initial simulation para- considering the project interest rate (r) and project lifetime (T). Finally,
meters for the CLC-sCO2 are summarised in Table 1. the annual net electricity generation, FCF along with the fuel costs, and
Importantly, the SOF-NGCC, which has been thermodynamically operating and maintenance costs, are assumed to be constant for the
analysed by Khallaghi et al. [23], is selected as a reference cycle in this project lifetime.
study. To have a fair comparison between CLC-sCO2 and the reference
r (1 + r ) T
cycle, the SOF-NGCC simulation is adapted to the new assumptions FCF =
(1 + r ) T 1 (3)
mentioned in Table 1.
The capital costs for the considered cases are estimated using the
2.2. Model validation bottom-up approach, considering the individual capital costs of the key
equipment. The capital costs for combustors in the SOF-NGCC and re-
The considered process consists of two parts: (i) the power gen- actors in the CLC-sCO2 are estimated using the component scaling
eration, which is based on the sCO2 cycle, and (ii) CLC that converts the factor, as shown in Eq. (4). In the latter case, both reactors are assumed
chemical energy of natural gas into heat for the sCO2 cycle. to be fluidised bed reactors. Having assumed that the residence times
Importantly, the prediction of the CLC model is comparable to that for the air reactor and fuel reactor are 5 and 60 s [37], respectively, the
presented in the work by Mantripragada and Rubin [10]. The sCO2 volume for each reactor is calculated using the procedure described by
cycle has been developed based on the recompression sCO2 cycle from Lyngfelt et al. [38]. Then, the weight for each reactor is calculated as
Moisseytsev and Sienicki [36] and has been validated by Hanak and presented in Peters and Timmerhaus [39]. Thereafter, by considering
Manovic [19]. As this study considers a pump in place of the main reactors described by NRE [40] as a reference and using the scaling
compressor, the data from Moisseytsev and Sienicki [36] are used as a factor of 0.67, the capital cost of both CLC reactors is estimated. On the
benchmark for the prediction of the sCO2 cycle model used in this study other hand, the reference for capital cost of combustors used in the SOF-
(Table 2).
Table 1
Using the maximum cycle pressure of 200 bar for the stream en-
Main assumptions and turbomachinery specification for simulation.
tering LTR and sCO2 turbine inlet temperature of 472 °C as in
Moisseytsev and Sienicki [36], the temperatures and pressures of the Parameter Value
streams in the sCO2 cycle are compared with the data reported in that
Natural gas composition and conditions [23]
study. The main reasons for such a deviation are different turbo- Methane (%vol) 89
machinery efficiencies and pressure drops considered in this study Ethane (%vol) 7
compared to those in Moisseytsev and Sienicki [36]. Moreover, a pump Propane (%vol) 1
and a compressor are used in this study instead of two compressors for Butane (%vol) 0.1
Pentane (%vol) 0.01
the compression stage. This resulted in a lower temperature of the CO2
CO2 (%vol) 2
stream entering the LTR. Nevertheless, the model prediction was found N2 (%vol) 0.89
to be in good agreement with the literature data. Lower Heating Value (LHV) (MJ/kg) 46.50
Temperature (°C) 15
Pressure (bar) 1.25
3. Techno-economic performance indicators
Turbomachinery specification
Isentropic efficiency of pump (%) [33] 90
To evaluate the thermodynamic performance of the CLC-sCO2, the Isentropic efficiency of turbine (%) [33] 93
thermal efficiency of the system is defined in Eq. (1) as the ratio of the Isentropic efficiency of compressor (%) [33] 89
net power output (Ẇnet), which is calculated as the gross power output Mechanical efficiency of compressors and pump (%) [34] 99.6
less the system’s parasitic load, and the chemical energy input to the Electrical efficiency of generator (%) [35] 98.5
system, that is defined as the product of the fuel consumption rate Initial CLC-sCO2 operating parameters
(ṁfuel) and its lower heating value (LHV). The gross power output of the Oxygen carrier type Cu/CuO
Oxygen carrier mass flow rate (kg/s) 350*
CLC-sCO2 is the electric power output of the generator linked with the
Combustors (oxidation and reduction) pressure (bar) 1.25
sCO2 turbine through the mechanical shaft. Importantly, the parasitic Combustors and reactor pressure drop (mbar) 150
load is the sum of all parasitic loads of the entire cycle associated with Pressure drop in heat exchangers (%) 1
its compressors and pumps. Oxidation reactor temperature (°C) 995
Reduction reactor temperature (°C) 900
Wnet Turbine backpressure (bar) 35
net = sCO2 turbine inlet temperature (°C) 700
mfuel · LHV (1) sCO2 turbine inlet pressure (bar) 300
sCO2 turbine backpressure (bar) 75
The economic performance of the considered cases is evaluated Recompression split fraction (-) 0.3
using the levelised cost of electricity (LCOE), defined in Eq. (2). This
equation is based on the assumption that the capital cost of the pro- * Assumption is made based on fully oxidised oxygen carrier flowing into the
posed process is completely covered by the revenue from electricity fuel reactor.
240
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
Table 2 392. 2
Benchmark of the sCO2 cycle stream data with Moisseytsev and Sienicki [36]. CT = mT · . ·ln ( )(1 + exp (0. 036TT , i 65. 66))
1 T (5)
Stream Temperature Pressure
47. 1
Literature Model Literature Model CC = mC C. ln ( C )
0. 92 C (6)
sCO2-7 31.3 31 74.0 75
sCO2-10 84.4 52 200.0 200 CHE = 2546. 9· AHE 0.67 · PHE 0.28 (7)
sCO2-11 171.8 178 199.6 198
sCO2-1 323.3 317 199.1 196 The capital cost of the CO2 pump is estimated using Eq. (8) [48] that
sCO2-3 471.8 472 198.4 198 considers the brake power (Wp ) and isentropic efficiency ( i, P ).
sCO2-4 362.3 364 77.3 77
3
sCO2-5 190.7 190 76.9 76 1 0.8
sCO2-6 90.2 85 76.3 75 CP = 3531.4·WP 0.71· 1 +
1 i, P (8)
Table 3
Assumptions for the economic analysis [34]. 4. Results and discussion
Parameter Value
4.1. Thermodynamic performance
Variable operating cost as a fraction of total capital cost (%) 2.0
Fixed operating cost as a fraction of total capital cost (%) 1.0 The thermodynamic assessment of the CLC-sCO2, considering initial
Natural gas price (£/GJ) 3.0 design parameters presented in Table 1, revealed that this cycle has a
Plant lifetime (years) 25
net power output of 380.3 MW. Such output corresponded to a net ef-
Project interest rate (%) 8.75
Capacity factor (%) 80
ficiency of 49.3%. The detailed performance summary of the CLC-sCO2
is shown in Table 5. As can be observed, a considerable share of total
system energy input (26.4%) is utilised for the sCO2 compression stage
NGCC is selected from López et al. [41]. Considering the total outlet (compressor and pump).
mass flow rate as a scaling parameter and the scaling factor of 0.6, the The thermodynamic performance of the proposed process is directly
cost for the combustor in the SOF-NGCC is calculated. dependent on the performance of the closed sCO2 cycle. Therefore, a
parametric study of the CLC-sCO2 was performed by varying the sCO2
cost of equipment A = (cost of equipment Aref ) turbine inlet temperature (TIT) and turbine inlet pressure (TIP). In
scaling factor addition, in the sCO2 cycle, there is an imbalance in the specific heat of
capacity of A the hot- and cold-side of the recuperator [49], as the specific heat of the
×
capacity of Aref (4) sCO2 stream is higher at the condition of low temperature and high
pressure [50]. This difference results in a pinch point problem [26]. To
The capital cost of the SOF-NGCC topping cycle is determined based compensate for this imbalance, the sCO2 stream is split after the LTR, as
on the capital cost correlations for specific pieces of equipment, using shown in Fig. 2b. Recompressing one stream without heat rejection
the bottom-up approach. These correlations were taken from the lit- [51] compensates for this imbalance, reducing the amount of waste
erature and are gathered in Table 4. heat in the system, and subsequently leading to a higher net efficiency.
The capital costs associated with the SOF-NGCC bottoming cycle Thus, the effect of the split fraction (SF) on the net efficiency is ana-
and the power cycle of the CLC-sCO2, both of which are sCO2 cycles, are lysed.
determined from the capital cost correlations for each specific piece of Fig. 3 shows the variation in the CLC-sCO2 net efficiency with split
equipment as outlined below. To estimate the costs of the sCO2 turbine fraction. The split fraction varies from 0.25 to 0.4, indicating the
and compressor, the methodology presented by Benjelloun et al. [47] is fraction of total flow entering the HTR. Fig. 3 reveals that the maximum
used. Eqs. (5) and (6), respectively, relate the capital cost of the turbine efficiency is achieved when the split fraction is 0.38. This is because, at
and compressor to their operating parameters, such as mass flow rate this split fraction, the approach temperature is minimised at both ends
(m) , pressure ratio ( ) , isentropic efficiency (η) and turbine inlet tem- of the LTR, maximising the heat transfer rate.
perature (TT , i ). Eq. (7), which is taken from Gabbrielli and Singh [48], is The effect of the sCO2 TIT and TIP on the net efficiency of the cycle
used to calculate the capital cost of any type of heat exchanger. This is presented in Fig. 4. This parametric analysis indicated that the net
capital cost depends on the surface area ( AHE ) and the operating efficiency of the cycle is correlated to the sCO2 TIT and TIP. It can be
pressure (PHE ) of the particular heat exchanger. seen that the correlation with the temperature is nearly linear (Fig. 4a)
while that with the pressure is of the second order (Fig. 4b). The highest
TIT of the sCO2 cycle in this study is set at 700 °C, considering the cost
Table 4
Capital cost estimation for SOF-NGCC topping cycle.
Equipment [scaling parameter] Correlation
Fuel compressor [Brake power requirement, WFC , BRK (kW)] [42,43] WFC , BRK
0.67
CC=91, 562
445
Air/O2 compressor [Brake power requirement, WAC ,BRK (kW)] [42,43] WAC , BRK
0.67
CC=91, 562
445
Air separation unit [O2 production rate, mO2 (kg/s)] [44]
CASU = 2. 926e 7 ( )
mO2 0.7
28 . 9
Oxygen pressure booster pump [Break power output WOP, BRK (kW), Isentropic efficiency OP (-)] [45] COP = 623.22(WOP ,BRK )0.95 (1 +
0.2
)
1 OP
Turbine [Break power output, WT , BRK (kW)] [46] CT = 3744. 3(WT , BRK )0.7 61. 3(WT , BRK )0.95
Generator [Break power output, WT , BRK (kW) [46] CGen = 26. 18(WT , BRK )0.95
241
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
Table 5 Table 6
Performance summary of CLC-sCO2. Thermodynamic performance comparison between CLC-sCO2 and SOF-NGCC.
Component Value Component CLC-sCO2 SOF-NGCC
Thermal energy input (MW) 768.3 Thermal energy input (MW) 768.3 768.3
s-CO2 turbine power output (MW) 609.6 High-pressure turbine power output, W1 (MW) – 64.4
Air compression power consumption (MW) 5.4 sCO2 turbine power output, W2 (MW) 658.7 576.7
s-CO2 cycle compression stage power consumption (MW) 203 Natural gas compressor power consumption, W3 (MW) – 3.8
CO2 compression for storage power consumption (MW) 23.6 Air separation unit power consumption, W4 (MW) – 51.9
Net power output (MW) 377.6 Air/O2 compression power consumption, W5 (MW) 5.4 41.4
Net efficiency (%) 49.1 sCO2 compression stage power consumption, W6 (MW) 234.7 165.9
CO2 purification and compression power consumption, – 4.0
W7 (MW)
54 CO2 compression (for storage) power consumption, W8 23.6 –
(MW)
Gross power output, Wgross (MW) 424.0 475.2
53
Net power output, Wnet (MW) 395.0 374.1
Net efficiency (%) 51.4 48.7
Net efficiency (%)
52
51
50
49
48
0.27 0.29 0.31 0.33 0.35 0.37 0.39 0.41 0.43
Split fraction (-)
56
54
Net efficiency (%)
52
50
48
46
44
500 550 600 650 700 750 800 850
Turbine inlet temperature (°C)
a)
56
54
Net efficiency (bar)
52
50 Fig. 5. Breakdown of the capital cost for (a) SOF-NGCC and (b) CLC-sCO2.
48
and lifetime of the materials under high-pressure and high-temperature 4.2. Techno-economic performance comparison
conditions. Operation under such conditions was found to result in a net
efficiency of 51.4%. However, further development of materials for The thermodynamic performance of the SOF-NGCC and the CLC-
high-temperature application would enable even higher efficiencies as sCO2 is summarised in Table 6. The same natural gas input (59,470 kg/
242
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
h) is used for both cycles. The analysis shows that the CLC-sCO2 has a highest contribution to the total capital cost.
net power of 395 MW (Eq. (9)) with a net efficiency of 51.4%. This Importantly, the specific capital cost of the CLC-sCO2 and the SOF-
performance is worse than that of the state-of-the-art NGCC without NGCC are estimated to be £761/kWgross and £684/kWgross. Such spe-
CO2 capture with a net efficiency of above 62% [53]. However, it is cific capital costs are comparable with the range reported for NGCCs
comparable with NGCCs with CO2 capture as PCC implementation was without CO2 capture (£400/kWgross–£700/kWgross) and figures reported
reported to result in an efficiency penalty of more than 8% [54]. Im- for NGCCs with CO2 capture (£730/kWgross–£1010/kWgross) [34]. This
portantly, the net efficiencies of other configurations of CLC have been indicates that both cycles are economically competitive to more mature
reported as below: technologies.
The LCOE for the CLC-sCO2 is £36.1/MWh, compared to £38.3/
• CLC integrated with the humid air turbine considered by Brandvoll MWh for the reference SOF-NGCC cycle. This is because of the lower
and Bolland (54%) [13] and by Olaleye and Wang (57.1%) [18]; capital cost and higher net efficiency of the CLC-sCO2 compared to
• CLC-based NGCCs comprising the conventional air turbine cycle and those of the SOF-NGCC, which results in lower fuel costs per unit of
steam cycle considered by Mantripragada and Rubin (48.9–53.2%) generated power. The LCOE of the CLC-sCO2 (£36.1/MWh) is within
[10] and Naqvi and Bolland (53%) [14]; and the range reported for conventional fossil fuel power plants (£28/
• CLC-sCO2 for combined heat and power (net power efficiency of MWh–£55/MWh) [34]. It also needs to be highlighted that it is almost
41.3% and total efficiency of 81.7%) proposed by Chen et al. [15]. half of the cost of electricity considered for CLC with the humid air
turbine (£77.5/MWh). In addition, it is superior to the LCOE of the
Although the sCO2 cycle compared to the conventional power cy- Allam cycle (£50/MWh) [55] which is known to have the best perfor-
cles, such as the air turbine cycle and steam cycle, and advanced power mance among all NG-fired oxy-combustion cycles with the maximum
cycles, such as the humid air turbine cycle, has a higher thermal effi- net efficiency of 55.1% [56] and 54.6% [57] and other fossil fuel power
ciency, the CLC-sCO2 in this study has lower efficiency compared to the plants with CO2 capture (£39/MWh–£78/MWh) [34]. Therefore, fur-
studies above. This is mainly because of different turbomachinery as- ther development of the CLC-sCO2 would contribute to decarbonisation
sumptions considered in this study. It also needs to be highlighted that, of the power sector at an affordable cost.
based on the turbomachinery efficiencies considered in this study, the
SOF-NGCC has a net power output of 374.1 MW (Eq. (10)) with an 5. Conclusions
efficiency of 48.7%, which is lower by 20.9 MW compared to that of the
CLC-sCO2. This is mainly because of the power consumption of the This study presents a novel concept of gas firing using chemical
cryogenic ASU and subsequent O2 compression to 300 bar, which is the looping combustion (CLC) for oxygen supply rather than an air se-
combustion pressure in the SOF-NGCC. In addition, compared to the paration unit (ASU) in oxy-combustion systems, and the supercritical
required power for CO2 compression for storage in the CLC-sCO2, the CO2 cycle for power generation (CLC-sCO2). A process model of the
power consumption for CO2 purification and compression in the SOF- proposed system was developed in Aspen Plus®, and a parametric study
NGCC is much lower, 23.6 MW and 4.0 MW, respectively. This is mainly was conducted by varying the inlet sCO2 turbine conditions and split
due to the higher pressure ratio for CO2 compression (for storage) in the fraction to achieve the optimal performance of the CLC-sCO2. This was
CLC-sCO2 than in the SOF-NGCC, 120 and 3.4, respectively. It is worth followed by an economic assessment to evaluate the economic perfor-
mentioning that there is a slight difference in the CO2 purity for storage mance of the CLC-sCO2 and staged oxy-fuel natural gas combined cycle
in both cycles, 96.7% for CLC-sCO2 and 97.6% for SOF-NGCC. This is (SOF-NGCC). It was found that the CLC-sCO2 is characterised with a net
mainly because of CPU implementation in the SOF-NGCC case. Im- efficiency of 51.4%, which is higher than the SOF-NGCC, which was
portantly, the thermodynamic assessment of the CLC-sCO2 considered considered as a reference, with a net efficiency of 48.7%. Lower capital
in this study indicates that such concept could be a feasible option that cost of the CLC-sCO2 compared to that of the SOF-NGCC (£323M and
would support achieving the emission reduction targets by 2050. Im- £325M, respectively), along with its lower power consumption, mainly
portantly, when natural gas is substituted with biogas, such concept due to no ASU, results in lower levelised cost of electricity (LCOE) by
will become carbon negative. Yet, the economic feasibility of such £2.2/MWh (£36.1/MWh and £38.3/MWh, respectively). Importantly,
concept needs to be proven. the techno-economic performance of the CLC-sCO2 has been shown to
be superior to other high-efficiency low-emission power generation
Wnet , CLC sCO2 = W2 W5 W6 W8 (9)
cycles, such as the Allam cycle and CLC integrated with a humid air
Wnet , SOF NGCC = W1 + W2 W3 W4 W5 W6 (10) turbine, which attracts profound interest in its commercialisation.
243
N. Khallaghi, et al. Applied Energy 249 (2019) 237–244
244