Professional Documents
Culture Documents
A R T I C L E I N F O A B S T R A C T
Keywords: A growing interest within the Carbon Capture, Utilisation and Storage (CCUS) community is large cluster net
CO2 Capture works that combine captured CO2 from power and industrial sources, in order to reduce the costs associated with
CO2 compression transportation and storage. Several studies have investigated the effects of stream impurities or pipeline oper
CO2 liquefaction
ating conditions, specifically for the compression train, but few sources look at both ends of the conditioning
CCUS
CCUS Cluster
problem.
Within this study, various CO2 sources (post-capture) are used as inputs into process models for two different
conditioning routes. Evaluation of each scenario’s performance are related to the compressor power consumption
and the overall specific energy demand. This highlighted the benefits of higher initial stream pressure and the
increased energy demand due to the small flowrate basis. A sensitivity analysis showed the rate at which the
energy demand fluctuates as a function of changing inlet stream characteristics (pressure and flowrate). Dynamic
simulations showed the effects of transient CO2 production on the conditioning train, based on a realistic power
generation scenario. Overall, the results indicate several conditioning challenges with linking various CO2
sources in the same cluster network, mainly the much higher energy demand of low flowrate (<10 kg/s) systems
and the negative energy demand effects of transient CO2 production.
1. Introduction phase a pump can be used (McCollum and Ogden, 2006). Liquefaction
then pumping is economically beneficial due to the lower capital and
An increasing policy focus for the Carbon Capture, Utilisation and operating costs of liquid pumps. However, due to the low boiling point
Storage (CCUS) industry is cluster networks, interlinking various CO2 of CO2 at elevated pressures, cooling systems must be in place to keep
sources with different stream characteristics: pressure, temperature, and the system under isothermal conditions and below the saturation tem
composition. The sources are mixed and delivered to potentially perature (Martynov et al., 2016).
different storage options, posing technical and operational challenges An early study by Aspelund and Jordal (Aspelund and Jordal, 2007)
for CO2 transportation (Moe et al., 2020). Alongside power industry investigates the interface between CO2 capture and transportation,
focused CO2 capture technologies (post-, pre-, and oxy-combustion), a stating the energy requirement for CO2 conditioning is between 90 and
growing interest is industrial sources such as the production of 120 kWh/tonne CO2. Several studies have looked at optimising and
hydrogen, fertilisers, cement, iron and steel (Bui et al., 2018). comparing conditioning strategies, mainly focused on conventional
CO2 is typically transported and stored in its liquid state in insulated/ multistage compression and liquefaction processes, with early stage
non-insulated storage tanks (EIGA 2008) or through pipelines (Jensen liquefaction consistently showing a lower energy consumption (Wit
et al., 2014). In most cases the CO2 stream pressure requires elevation, kowski and Majkut, 2012, Alabdulkarem et al., 2012, Aspelund, 2010,
typically using centrifugal compressors, and due to the low pressure Muhammad et al., 2020), although in some cases the energy reduction is
ratio (between 1.7-2:1) a multistage system is required. Three factors potentially offset by an increased cooling water demand (IEAGHG
affect compressor power consumption: pressure ratio, efficiency, and 2011).
inlet temperature (Romeo et al., 2009). In its gaseous phase, a There is a wide range of potential sources of CO2, with each tech
compressor is required to increase the pressure, whilst in the liquid nology producing CO2 streams at different pressures, temperatures, and
* Corresponding author.
E-mail address: s.f.brown@sheffield.ac.uk (S. Brown).
https://doi.org/10.1016/j.ijggc.2021.103449
Received 2 March 2021; Received in revised form 15 June 2021; Accepted 1 September 2021
Available online 13 September 2021
1750-5836/© 2021 Elsevier Ltd. All rights reserved.
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Table 1
Captured CO2 source characteristics.
Source Temperature (◦ C) Pressure (bar) Composition (%) Source
CO2 H2O N2
Notes: monoethanolamine (MEA), Ammonia (NH3), Methyl-diethanolamine (MDEA), molten-carbonate fuel cells (MCFC), pressure swing adsorption (PSA), solid
oxide fuel cells (SOFC), and steam methane reforming (SMR).
Fig. 1. Study Overview, highlighting multistage compression and liquefaction routes from IEAGHG (IEAGHG 2011). Included in the Fig. is low, medium, and high
pressure shipping specifications from MEP (MEP 2016).
2
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Fig. 2. Post-combustion CO2 compression system, base case B0 from IEAGHG (IEAGHG 2011).
Fig. 3. Post-combustion CO2 compression and liquefaction system, case D2c from IEAGHG (IEAGHG 2011).
into one system (Porter et al., 2017), therefore it is not included in this conditioning are multistage compression based on the IEAGHG base case
study. B0, and multistage compression and liquefaction based on the IEAGHG
CO2 can be transported via road/rail/shipping in insulated/non- case D2c. As moisture content is a limitation for transport, utilisation,
insulated storage vessels, as well as through pipelines (MEP 2016). and storage, three moisture levels are analysed within this study to
This study focuses on pipeline transportation for its potential role in compare different end-point composition guidelines:
CCUS clusters (Moe et al., 2020). Table 2 highlights the specifications
used for various pipeline projects globally. Kinder Morgan is one of the • Low - 20 ppm specification for the food and beverages industry
largest CO2 pipeline operators in the world, and their standards for (EIGA 2016).
operating pressure and temperature are >10.3 MPa (103 bar) and • Medium – 300 ppm cost-optimal level for pipeline transportation
<120◦ F (49◦ C) respectively (Jensen et al., 2014), ensuring the CO2 from Brunsvold et al. (Brunsvold et al., 2016).
stream is kept above the critical pressure point and below the critical • High - 600 ppm specification, as used by Kinder Morgan for
temperature point, preventing supercritical fluid flow. Variable and enhanced oil recovery projects (Jensen et al., 2014).
transient mass flow will affect the temperature and pressure within the
pipeline, causing operational challenges and maintenance problems Large CCUS clusters will have a range of flowrates attached, with
(Kahlke et al., 2020). Therefore, it is imperative the conditioning system some having transient production; this is a key challenge highlighted in
minimises fluctuations in the CO2 stream characteristics. Moe et al (Moe et al., 2020). From our earlier study (Wilkes et al., 2021),
the high energy demand at low flowrates raised an important question
1.1. Aims and objectives on the size of units attached in CCUS clusters, specifically the higher
energy demand of conditioning small CO2 flowrates. Hence, a sensitivity
One of the issues with designing the compression train is the end analysis focussed on inlet flowrate and pressure for both conditioning
point specifications. Alongside the various transportation methods, routes is included to identify the relationship between system size and
there are also different composition guidelines depending on the end use power demand. A growing focus for power generation CO2 capture is
of the CO2 (geological storage or utilisation) (Bui et al., 2018). Various transient/flexible operation, due to a growing renewable capacity
sources mentioned in the literature review have examined the effects of driving fossil fuel based sources to act as balancing services (Bui et al.,
impurities in the CO2 feed, or the effects of reaching different end points 2018). Therefore, a dynamic study based on realistic load-following
in terms of composition, pressure, and temperature; however, few capture plant operation is also included, comparing both conditioning
sources look at both ends of the problem. trains and highlighting the issues of transient operation.
This simulation-based study considers the impact of different CO2
sources (Table 1) and end point specifications on the power demand for 1.1.1. Compression case studies
the conditioning (compression and moisture removal) train. For each The IEAGHG compression base case (B0) for post-combustion cap
technology, the compression system will elevate the CO2 stream pres ture (see Figure 2) assumes the CO2 stream comes from an ultra-
sure to 111 bar and 70◦ C as specified in IEAGHG (IEAGHG 2011). An supercritical pulverised coal fired power plant, with NOX, SOX and
overview of the study is illustrated in Figure 1. The two processes for CO2 treatment. The vapour stream leaving the CO2 solvent regeneration
3
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Table 3 given impeller diameter, with constant inlet temperature and specific
IEAGHG cases B0 and D2c unit consumption (IEAGHG 2011). heat ratio (Lüdtke, 2004):
Parameter B0 D2c V in
st
φ=π (1)
1 Compression Stage Outlet Pressure (bar) 7 7.0 d 2
u
4 i i
2nd Compression Stage Outlet Pressure (bar) 34 34
3rd Compression Stage Outlet Pressure (bar) 70 66.0
4th Compression Stage/Liquid Pump Outlet Pressure (bar) 111 111
Where Vin is the inlet volumetric flowrate (m3 /s), di is the impeller
Cooling water demand (t/h) 5392 8932 diameter (m), and ui is the impeller tip speed (m2 /s). Centrifugal com
Condensate Pre-heating (MWth) 34.0 31.4 pressors can be considered polytropic processes (Stewart, 2018). The
1st Compression Stage Power (MWe) 21.7 21.7 polytropic head coefficient (ψ p ) relates the polytropic work and the
2nd Compression Stage Power (MWe) 24.1 23.6
3rd Compression Stage Power (MWe) 8.0 7.3 impeller tip speed (Aungier, 2000):
4th Compression Stage/Liquid Pump Power (MWe) 3.7 1.2
Δhp
Total power consumption (MWe) 57.5 53.8 ψp = (2)
u2i
unit comprises of 95.88 vol.% CO2, 4.11 vol.% H2O, and 0.01 vol.% N2, Where Δhp is the polytropic work/enthalpy change (J/kg). It can also be
at 556451 kg/h, 38◦ C and 1.6 bar. A four stage compression system, related to the polytropic efficiency (ηp ) and work coefficient (w)
with intercooling and dehydration produces a product stream at 546855 (Lüdtke, 2004):
kg/h, 73◦ C, 111 bar and 50 ppm moisture. The intercooling is performed
ψ p = w ηp (3)
in two stages: one stream produces water for condensate pre-heating, the
other stream is cooling water. The dehydration unit consists of three The work co-efficient is similar to the polytropic head coefficient,
beds, two in drying mode and one in regeneration mode, using molec but, instead of using the polytropic enthalpy change, it uses the actual
ular sieves as the solid desiccants. Regeneration is carried out using 10% enthalpy change (Aungier, 2000):
of the dried gas stream, at 250◦ C (IEAGHG 2011).
The IEAGHG compression and sub-critical liquefaction case D2c Δht
w= (4)
(Figure 3) for post-combustion capture assumes the same CO2 source as u2i
the base case B0. Conditioning up to the 3rd compression stage is iden
Where (Δht ) is the actual enthalpy change (J/kg). It is the difference
tical to base case B0. The 3rd compression stage pressurises the stream to
between the inlet flow enthalpy (ht, in ) and the outlet flow enthalpy
66 bar, after cooling to 20 ◦ C it can be pumped to 111 bar. The lower
(ht, out ):
pressure ratio for the 3rd compression stage and use of liquid pumps
reduces the overall energy penalty by 3.7 MWe, compared to the base Δht = ht, in − ht,out (5)
case. This is offset by the reduction in available waste heat for
condensate preheating and the higher cooling water demand (IEAGHG Whereas the polytropic enthalpy change includes leakage and fric
2011). Table 3 shows the main consumptions from the B0 and D2c cases. tional head losses (Aungier, 2000):
It is worth noting that case D2c is subject to sufficiently low cooling (Δs)(ΔT)
water availability (12◦ C), with high ambient conditions full sub-critical Δhp = Δht − (6)
ln(T out − T in )
liquefaction is unachievable.
Where Δs is the entropy change between the outlet and inlet compressor
2. Modelling CO2 compression flows (J/K/kg), and ΔT is the temperature difference between the outlet
and inlet compressor flows (K). Therefore, combining equations 2, 3 and
There are several important characteristics that influence the design 4 the polytropic efficiency (ηp ) compares the polytropic work to the
and operation of centrifugal compressors, explained in Aungier (Aun actual work, and is independent of the pressure ratio:
gier, 2000) and Lüdtke (Lüdtke, 2004). The flow coefficient (φ) de
termines the size of the volumetric flow through the compressor for a
Fig. 4. a) model topology for conventional multistage CO2 compression, b) model topology for multistage CO2 compression and liquefaction.
4
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Table 4
Design characteristics for the individual compression and pumping stages.
Compressor Stage Flow Coefficient (φ) Polytropic Coefficient (ψ p ) Polytropic Efficiency (ηp ) Pressure Ratio (Pout /Pin )
Δhp steps are lumped into one and modelled via a heat exchanger unit. The
ηp = (7)
Δht heat exchanger units operate in counter-current flow, with a constant
heat transfer area used to deliver a specific outlet CO2 stream temper
The impeller tip speed Mach number (Mi ) explains the compress
ature. The heat exchanger models assume adiabatic operation with
ibility of the fluid; it is a ratio of the impeller tip speed and sonic inlet
constant pressure drop. The knock-out drums remove condensed water,
velocity (Aungier, 2000):
and the models are based on vapour-liquid phase equilibrium with no
Mi =
ui
(8) liquid entrainment. The dehydration unit removes H2O from the CO2
a0 stream to a specified moisture content, and consists of three molecular
sieve beds, two used for drying and one in regeneration mode, operating
Where a0 is the speed of sound in the inlet flow (m/s) (Modekurti et al.,
simultaneously. The dehydrator unit assumes no degradation or loss in
2017). High Mach numbers result in the choke and surge points being performance of the molecular sieve over time. The regeneration fraction
closer to the design point, reducing the operational flexibility. The
( ) and temperature are specified parameters, and the model calculates the
pressure ratio PP21 is defined by Lüdtke (Lüdtke, 2004) as being heat requirement and power consumption. See PSE’s gCCS documen
tation (PSE 2016) for additional information on these unit operations.
dependent on the Mach number: For dynamic compressor stability a surge control system is required
Pout [ ] kηp to prevent flow reversal and extreme flow oscillation (McMillan, 2010).
= w(k − 1)M2i + 1 k− 1 (9) The surge control system includes a valve to recycle flow, a
Pin
recycle-breaker that opens when a surge is encountered, and a
Where Pout is the compressor outlet pressure, Pin is the compressor inlet Proportional-Integral (PI) controller that dictates the valve stem position
pressure, and k is the isentropic exponent. based on the volumetric flow through the compressor. The Variable
The process models for compression and liquefaction are developed Frequency Drive (VFD) controller is a Proportional-Integral-Derivative
in gPROMS gCCS, utilising the model library to construct a flowsheet of (PID) controller; it manipulates the speed of the electric drive attached
the conditioning systems, including the compressor sections, inter-stage to the compression stage to deliver a desired output pressure determined
cooling, knock-out drums, surge valves and dehydration unit. Figure 4a by:
shows the model topology for conventional multistage compression,
z = max(min(zcalc , zmax ), zmin ), (10)
based on the base case B0 from IEAGHG (IEAGHG 2011). Figure 4b
shows the model topology for multistage compression and liquefaction, ( Δv− )
based on case D2c from IEAGHG (IEAGHG 2011). More information on zmin = vD 1 − (11)
100
the cases is given in Section 1.1.1. Each compressor section consists of
multiple stages, modelled via polytropic efficiency with negligible ( Δv+ )
zmax = vD 1 − (12)
hold-up and inertia of gas. The thermo-physical properties and phase 100
equilibrium of the fluid is determined through gSAFT (PSE 2016).
As specific outlet conditions for the individual water coolers are not Where z is the controller output (Hz), zmin is the minimum controller
given in IEAGHG (IEAGHG 2011), the two inter-stage water cooling output (Hz), zmax is the maximum controller output (Hz), zcalc is the
calculated controller output based on PID control law (Hz), vD is the
5
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
design speed (Hz), Δv− is the turndown speed (Hz), and Δv+ is the may not be the most energy-efficient systems. Nevertheless, for the
overspeed (Hz) (PSE 2016). purposes of this study they provide a baseline for source to transport
evaluation.
3. Results Figure 5 compares the process simulation for the multistage
compression system against the IEAGHG base case B0, for which the key
3.1. Model validation performance indicators are shown in Table 4. The inlet and outlet
pressures are extremely accurate: less than 0.04% for all streams. The
Using input parameters specified in Table 3, IEAGHG (IEAGHG flowrates into the 2nd, 3rd and 4th compression stages are slightly smaller
2011) and Section 1.1.1, the process simulator calculates several (<2.33%) than presented in IEAGHG (IEAGHG 2011), as less flow is
important design characteristics shown in Table 4. The polytropic effi recycled from the dehydrator. The inlet temperature into the 1st
ciency (ηp ) typically used for compressor evaluation ranges between compression stage is higher than expected due to the surge control
79-87%. Aspelund (Aspelund, 2010) states centrifugal compressors have system, resulting in a higher outlet temperature. These small differences
a polytropic efficiency between 80-85%, and compression stages have a result in the 1st, 2nd, 3rd and 4th compression stage power demand
pressure ratio between 1-5. The simulations show the 1st and 2nd deviating -4.68, 15.41, 12.95, and 8.94%, respectively. Overall, the total
compression stages lie outside these specified bounds, and therefore, power demand (53.42 MW) is 7.10% lower than expected (57.51 MW).
6
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Table 5
Comparison between conventional compression and liquefaction routes.
Source Moisture Overall energy demand (kWh/tCO2)
7
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Table 6
Capture and conventional conditioning energy demand.
Source Capture Energy Demand (GJ/tCO2) Conditioning Energy Demand (GJ/tCO2) Combined Energy demand (GJ/tCO2)
8
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
Fig. 12. Dynamic power consumption and energy demand for the 20 ppm MEA conventional (left) and liquefaction (right) compression cases.
Fig. 13. Outlet pressure and temperature profiles for the 20ppm MEA conventional (left) and liquefaction (right) compression cases.
sensitivity analysis looking at various inlet pressures and flowrates. difference is shown between 10 kg/s and 100 kg/s, only reducing 8.09%
Figure 9 shows the energy demand for different inlet stream pres to 105.00 kWh/tCO2. Overall, for the 20 ppm conventional case, there is
sures ranging from 1 bar to 10 bar, for the MEA conventional and a 41.4% reduction in power demand when increasing from 1kg/s to 100
liquefaction scenarios. The mass flowrate is kept at 1kg/s, and the kg/s flow.
pressure range was chosen to investigate the effects of removing the 1st In relation to the power generation technology for the source of CO2,
compression stage. A greater increase is shown during the initial pres Wilkes et al. (Wilkes et al., 2020) shows a Siemens SGT-400 gas turbine
sure increase, and this rate slows down as we approach 7 bar (the outlet operating in open-cycle configuration produces 10.4 MWe of power and
pressure of the first compression stage). At 7 bar and higher, the first 33.8 kg/s of exhaust flow, assuming 4.27 vol.% CO2 concentration and
compression stage is no longer needed; however, the knock-out drum 90% CO2 capture this equates to 2.06 kg/s of CO2 flow into the condi
and cooling stage is still required. Otherwise, there is an increased power tioning system. For a MHPS M701J series producing 478 MWe of power
demand for the second compression stage due to a higher inlet flowrate. in open-cycle and 896 kg/s of exhaust flow (Mitsubishi Power 2020), the
As can be seen for both conventional and liquefaction routes, there is no CO2 output flow is 54.59 kg/s. Based on Figure 11, assuming the CO2 is
difference between 20 and 300 ppm, so moisture content is not an issue captured through an MEA based absorption system, a small-scale 10
as long as the initial purity is >95%: the knock-out drums do most of the MWe plant will require approximately 167 kWh/tCO2 to conventionally
moisture removal. Between 7 to 10 bar there is a greater decrease in the condition the capture CO2 at a 300 ppm moisture control level, whereas
rate of energy drop, which is due to the 2nd compression stage inlet a large-scale 478 MWe plant will require approximately 105 kWh/tCO2.
pressure decreasing. The initial drop in pressure has a greater effect than Therefore, conditioning a CO2 stream from a small-scale power source is
the final drops in pressure (per stage). Overall, the specific energy de 59.05% more energy intensive than the large-scale source.
mand decreases as the starting pressure increases: a similar trend is Table 6 compares the capture and conditioning energy demand for
presented in Aspelund and Jordal (Aspelund and Jordal, 2007) each of the capture technologies, in terms of GJ/tCO2 to enable cross
Figure 10 and Figure 11 show the total power demand and specific comparison with sources in the literature. Although it is not an accurate
energy demand, respectively, for different model input flowrates comparison as the conditioning energy demands are for 1kg/s systems, it
ranging from 1 to 100 kg/s, for both the MEA conventional and lique does highlight the difference between the capture and conditioning re
faction conditioning scenarios. The flowrate range was chosen to quirements. Converting the results in figure to GJ/tCO2, the condition
encompass all possible inputs into large CCUS clusters. This highlights ing energy requirement ranges between 0.31 and 0.65 GJ/tCO2 for inlet
the rate at which the energy demand changes as a function of inlet flowrates ranging from 1 to 100 kg/s. The range of capture energy de
flowrate or plant size. Total power demand increases almost linearly mands far exceeds the range of conditioning energy demands. For large
with increasing inlet flowrate: a greater flow requires a greater amount scale producers (>10kg/s CO2), the conditioning energy requirement is
of power to compress it. At 100 kg/s for the low moisture conventional as low as 0.31 GJ/tCO2 using the liquefaction process. In relation to
and liquefaction routes the power demand is 36.24 and 34.23 MW, CCUS clusters, this poses a unique challenge of balancing quantity and
respectively. Interestingly, a point is reached around 10 kg/s where the efficiency. Certain systems may capture CO2 with a lower specific energy
specific energy demand increases rapidly. A large difference is exhibited requirement (for both capture and conditioning), however, the quantity
between 1kg/s and 10 kg/s. For the 20 ppm MEA conventional case, the captured may be small. Both industrial CO2 capture technologies (SMR-
1 kg/s simulation showed a 179.18 kWh/tCO2 specific energy demand. PSA and Steel-MEM) have lower capture energy demands compared to
At 10 kg/s, this is reduced by 36.24% to 114.24 kWh/tCO2. A smaller the conditioning energy demand, highlighting that these systems may
9
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
benefit from sharing the conditioning duty with neighbouring facilities, capable of preventing surge and reducing the volume of recycled flow
or utilising intermediate storage prior to compression at a centralised (Budinis and Thornhill, 2015). Future work should focus on utilising
hub (Kahlke et al., 2020). MPC during load changes originating from upstream capture units, in
order to optimise the conditioning train during transient operation.
3.4. Dynamic study
4. Conclusion
Another challenge highlighted in Moe et al. (Moe et al., 2020) is the
effect of transient CO2 production on the downstream compression and With the growing interest in CCUS cluster networks and the need to
transportation network. Therefore, to assess each systems capability capture CO2 from a range of sources, the link between capture and
under real-world transient conditions, assuming that load following transportation needs to be clearly identified. Within this paper, vali
capture plant operation (Wilkes et al., 2020) directly relates to fluctu dated process models for two conditioning system are used to evaluate a
ations in the captured CO2 product, the input flowrate in the dynamic range of CO2 sources from power generation and industry, comparing
models is ramped based on real-world gas turbine operation. Using a 1 different end-point pipeline transportation characteristics (moisture
kg/s flowrate basis, this correlates to a 5 MW gas turbine operating in content).
open-cycle configuration. The ramp rate for modern gas turbines is be The study found the use of a liquid pump instead of a final
tween 8-12% max load per minute (Agora Energiewende 2017), so compressor stage offers a 14-19% decrease in energy consumption,
assuming a 10% ramp rate the gas turbine would take 3 minutes to depending on the technology and level of moisture control. The results
decrease to 70% load. Furthermore, assuming constant capture plant also showed the initial pressure of the CO2 stream has a greater effect
performance, the ramp rate is used to simulate a scenario where the gas than the initial moisture content. Increasing initial pressure decreases
turbine, after 2 hours of operation, is forced to reduce to 70% full load the overall specific energy demand; although this is obvious, the results
for 1 hour, then ramps to full load and continues for another 2 hours, show the rate of change and how the initial drop in pressure has a
simulating a typical 5 hour operation (Wilkes et al., 2020). The reporting greater effect than the later drops in pressure. Flowrate has an inversely
interval for the dynamic simulations is 10s. proportional effect on the specific energy demand, especially between 1-
Figure 12 shows the power consumption and energy demand for the 10kg/s, but less so between 10-100 kg/s. Whereas, the power demand is
MEA 20 ppm moisture conventional and liquefaction cases. Considering directly proportional to the size of the flow through the system, high
the total mass of CO2 processed and the total energy supplied to the lighting the possible complications in large scale CCUS clusters, where a
compression stages and dehydration unit, the overall specific energy range of flowrates are attached to one system. A small scale power
demand for the 5 hour transient operation for the conventional and station or industrial CO2 source, producing less than 1 kg/s of CO2, will
liquefaction processes is 182.08 kWh/tCO2 and 153.66 kWh/tCO2 have an effect on the overall performance of the compression and con
respectively. Even though in both scenarios the power demand ditioning aspects of CCUS networks. Increasing the CO2 flowrate from 1
decreased during the partial load stage (t=2 to 3 hours), the lower CO2 to 100 kg/s reduces the power demand by up to 41%.
flowrate means the overall specific energy demand increased. At 2.5 The dynamic simulations showed an interesting relationship be
hours in the conventional case, the power required was 0.4844 MW to tween the transient CO2 flow and specific energy demand. Although at
process 0.7 kg/s of CO2, equating to an instantaneous specific energy lower loads the power demand is smaller (less energy required to
demand of 192.25 kWh/tCO2. This is 7.29% higher than the steady-state compress the smaller flow), the specific energy demand actually in
simulation, highlighting transient CO2 conditioning has a negative effect creases, therefore, costing and requiring more energy per ton of CO2
on specific energy demand. processed. Process control systems ensure the outlet stream conditions
Figure 13 shows that throughout the simulation the individual outlet (pressure) remain constant, as to not disturb downstream usage. Linking
pressures for each compression stages are kept constant, through the use CO2 capture and transportation options provides valuable information
of the surge and variable frequency drive control systems. Implementing on the challenges facing CCUS clusters. Multiple sources have concluded
these control systems enables a realistic response to transient CO2 pro that results are case specific when investigating compression and
duction (Modekurti et al., 2017). The increased outlet compression stage transportation options (MEP 2016, Roussanaly et al., 2013). We
temperature at lower loads should lead to an increased water demand corroborate this conclusion as multiple factors (inlet characteristics,
for interstage cooling, however, there is a 23.38% (conventional) and end-point specifications, and operational transience) affect the design
25.84% (liquefaction) decrease in overall water demand due to the and performance of the conditioning system. This study provides an
lower flowrate through the entire system. There is an increased inlet insight and links various capture sources, conditioning options, and
stream temperature to all the compression stages, however, this is not transportation guidelines.
related to the cooling system, but rather the activation of the surge
system recycling hot gases. The temperature and enthalpy change across Credit author statement
the compression stages is related to the polytropic efficiency (see
equation 7). The ratio between the inlet and outlet temperatures, also Mathew Dennis Wilkes: Conceptualization, Methodology, Investi
known as polytropic ratio (Guo et al., 2007), increases for the first gation Software. Sanjay Mukherjee: Writing- Reviewing and Editing.
compression stage resulting in the polytropic efficiency decreasing by Solomon Brown: Supervision, Writing- Reviewing and Editing
1.34%, whereas the polytropic ratio in the 2nd and 3rd stages is smaller at
lower loads, resulting in the polytropic efficiency increasing by 4.62% Declaration of Competing Interest
and 12.72%, respectively. During the 4th stage the ratio remains con
stant, hence no change in efficiency, thus highlighting the importance of The authors declare that they have no known competing financial
controlling the inlet temperature during load changes. interests or personal relationships that could have appeared to influence
It is possible to incorporate a cooling step within the surge recycle. the work reported in this paper.
However, Budinis and Thornhill (Budinis and Thornhill, 2016) showed
this increases the power consumption due to a greater volume of flow Acknowledgements
and cooling requirement of the recycle stream. The study focused on the
compression stage configuration. A possible alternative configuration This work is funded by the Engineering and Physical Sciences
for CO2 systems could be recycling the surge system prior to the heat Research Council (EPSRC) Centre for Doctoral Training in Carbon
exchanger unit, ensuring a constant inlet compression stage tempera Capture and Storage and Cleaner Fossil Energy (reference: EP/L016362/
ture. Another possible solution is model predictive control (MPC), 1), and Drax Group Plc. The author would like to acknowledge the
10
s)
h)
(%)
(◦ C)
(Hz)
(m/s)
(MW)
(MW)
Figure 6.
Figure 5.
Table A2
Table A1
(kg/h)
(m3/s)
Parameter
Parameter
Appendix
Power (%)
Power (%)
M.D. Wilkes et al.
Diameter (m)
Diameter (m)
Percentage of Total
Percentage of Total
Surge Volume Flow
Inlet Mass Flowrate
Outlet Temperature
Power Requirement
Power Requirement
Electric Drive Speed
dynamic stability.
Inlet Temperature (◦ C)
Inlet Temperature (◦ C)
Outlet Temperature (◦ C)
Inlet Mass Flowrate (kg/
CS1
7.00
1.22
1.60
39.38
22.63
79.90
44.90
28.20
53.47
7.00
1.22
1.60
202.03
306.00
556484
41.59
22.65
79.90
45.00
28.20
53.47
4400.00
202.12
306.00
556484
4395.00
Fellowship (reference: IF\192046).
CS2
CS2
1.00
6.60
35.84
20.60
33.99
80.10
10.00
25.40
19.21
1.00
6.60
165.93
251.00
594440
4800.00
37.81
20.59
34.00
80.10
10.00
25.40
19.21
165.91
251.00
594440
4800.00
-
-
-
-
-
-
DH1
-
-
-
-
-
-
DH1
7.03
4.04
32.70
24.87
33.60
24.00
7.42
4.04
592956
32.70
24.87
33.60
24.00
592956
CS3
CS3
6.81
1.56
0.53
11.85
70.03
89.47
81.30
26.30
32.69
19.48
134
6.23
1.54
0.54
136.00
4850
11.44
66.04
83.69
79.60
24.70
32.70
19.46
535447
4900.00
535323
CS4
LP1
-
-
-
-
-
5.89
3.39
0.74
0.38
1.74
0.95
82.20
80.60
36.20
97.10
69.63
40.07
26.37
80.00
65.64
20.00
111.23
111.00
536761
4800.00
533400
Process model outputs for the multistage compression and liquefaction route
performance, as this impacts the output pressure, power demand and
respectively. Also included in A3 and A4 is the surge control system
(compressor rotational speed) required to ensure the output pressure is
compression route, and Table 4 and Table A2 for the multistage
characteristics shown in Table 4 and Table A1 for the conventional
based on the IEAGHG case D2c (IEAGHG 2011), data within is also presented in
on the IEAGHG base case B0 (IEAGHG 2011), data within is also presented in
support of the Royal Academy of Engineering through the Industrial
Process model outputs for the conventional multistage compression route based
shown in Section 3.1, the model calculates several important design
11
Table A3
Process model outputs during each case study for the conventional compression route, including operational and design characteristics of the compression stages and surge control systems.
Source Rotational Speed (RPM) Tip Speed (m/s) Average Impeller Diameter (m) Section Surge Margin (%) Surge Volume Flow (m3/s)
CS1 CS2 CS3 CS4 CS1 CS2 CS3 CS4 CS1 CS2 CS3 CS4 CS1 CS2 CS3 CS4 CS1 CS2 CS3 CS4
Post-MEA-20 4385 4150 3450 3050 77.20 64.50 63.90 64.90 0.31 0.25 0.25 0.25 28.60 37.00 45.00 50.80 0.2920 0.0759 0.0137 0.0057
Post-MEA-300 4385 4150 3450 3050 77.20 64.50 63.80 64.80 0.31 0.25 0.25 0.25 28.60 37.00 45.00 50.80 0.2920 0.0759 0.0136 0.0057
Post-MEA-600 4385 4150 3450 3050 77.20 62.80 61.90 63.60 0.31 0.25 0.25 0.25 28.70 33.90 44.70 50.70 0.2920 0.0740 0.0132 0.0056
Post-NH3-20 4700 4150 3400 3050 63.90 64.90 63.90 65.00 0.25 0.25 0.25 0.25 28.20 38.20 45.50 51.20 0.0669 0.0765 0.0138 0.0057
Post-NH3-300 4700 4150 3400 3050 63.90 64.90 63.70 64.90 0.25 0.25 0.25 0.25 28.20 38.20 45.50 51.20 0.0669 0.0765 0.0138 0.0057
Post-NH3-600 4700 4150 3400 3050 63.80 63.00 61.80 63.70 0.25 0.25 0.25 0.25 28.20 35.20 45.20 51.00 0.0669 0.0744 0.0134 0.0056
Pre- MDEA-20 4385 4150 3450 3050 87.00 64.60 64.00 65.10 0.35 0.25 0.25 0.25 28.30 37.10 45.10 50.90 0.4160 0.0760 0.0137 0.0057
Pre- MDEA-300 4385 4150 3450 3050 87.00 64.60 63.80 64.90 0.35 0.25 0.25 0.25 28.30 37.10 45.00 50.90 0.4160 0.0760 0.0136 0.0057
Pre- MDEA-600 4385 4150 3450 3050 87.00 62.90 62.00 63.80 0.35 0.25 0.25 0.25 28.30 34.00 44.70 50.70 0.4160 0.0741 0.0132 0.0056
SMR-PSA-20 4875 4000 3315 3000 86.70 63.80 63.80 64.70 0.35 0.25 0.25 0.25 26.70 38.80 45.90 51.30 0.4130 0.0708 0.0133 0.0058
SMR-PSA-300 4875 4000 3315 3000 86.70 63.80 63.80 64.70 0.35 0.25 0.25 0.25 26.70 38.80 45.90 51.30 0.4130 0.0708 0.0133 0.0058
SMR-PSA-600 4875 4000 3315 3000 86.70 63.80 63.80 64.70 0.35 0.25 0.25 0.25 26.70 38.80 45.90 51.30 0.4130 0.0708 0.0133 0.0058
Steel-MEM-20 4385 4150 3450 3050 84.10 61.80 62.20 63.60 0.36 0.25 0.25 0.25 27.20 34.50 45.10 50.80 0.4410 0.0736 0.0136 0.0059
Steel-MEM-300 4385 4150 3450 3050 84.10 61.80 62.20 63.60 0.36 0.25 0.25 0.25 27.20 34.50 45.10 50.80 0.4410 0.0736 0.0136 0.0059
Steel-MEM-600 4385 4150 3450 3050 84.10 61.80 62.20 63.60 0.36 0.25 0.25 0.25 27.20 34.50 45.10 50.80 0.4410 0.0736 0.0136 0.0059
IGCC-Selexol-20 4385 4150 3450 3050 61.30 64.10 64.20 65.30 0.25 0.25 0.25 0.25 30.50 37.20 45.30 51.10 0.0870 0.0755 0.0137 0.0057
IGCC-Selexol-300 4385 4150 3450 3030 63.70 64.80 64.20 64.90 0.25 0.25 0.25 0.25 30.60 37.50 45.30 51.10 0.0904 0.0763 0.0137 0.0058
IGCC-Selexol-600 4385 4150 3450 3030 63.70 63.10 62.30 63.70 0.25 0.25 0.25 0.25 30.60 34.50 45.00 50.90 0.0904 0.0742 0.0133 0.0056
IGCC-MCFC-20 4775 4050 3415 3050 87.10 63.50 63.90 65.10 0.35 0.25 0.25 0.25 27.70 38.30 45.50 51.30 0.4190 0.0766 0.0138 0.0057
IGCC-MCFC-300 4775 4050 3415 3050 87.10 63.40 63.70 65.00 0.35 0.25 0.25 0.25 27.70 38.30 45.50 51.20 0.4190 0.0766 0.0138 0.0057
IGCC-MCFC-600 4775 4175 3540 3050 87.10 63.40 63.80 65.00 0.35 0.25 0.25 0.25 27.70 35.40 45.30 51.20 0.4190 0.0743 0.0133 0.0057
IGCC-SOFC-CLC-20 4760 4000 3335 3000 86.80 63.50 63.80 64.30 0.35 0.25 0.25 0.25 27.50 38.70 45.80 51.20 0.4200 0.0705 0.0132 0.0058
IGCC-SOFC-CLC-300 4760 4000 3335 3000 86.80 63.50 63.80 64.30 0.35 0.25 0.25 0.25 27.50 38.70 45.80 51.20 0.4200 0.0705 0.0132 0.0058
IGCC-SOFC-CLC-600 4760 4000 3335 3000 86.80 63.50 63.80 64.30 0.35 0.25 0.25 0.25 27.50 38.70 45.80 51.20 0.4200 0.0705 0.0132 0.0058
International Journal of Greenhouse Gas Control 111 (2021) 103449
References
LP1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Agora Energiewende, 2017. Flexibility in thermal power plants. Agora Energiewende,
Berlin.
Alabdulkarem, A., Hwang, Y., Radermacher, R., 2012. Development of CO2 liquefaction
0.0128
0.0128
0.0139
0.0138
0.0127
0.0133
0.0133
0.0128
0.0128
0.0128
0.0132
0.0131
0.0126
0.0134
0.0133
0.0129
0.0132
0.0131
0.0126
0.0129
0.0129
0.0129
0.0128
0.0128
cycles for CO2 sequestration. Appl. Therm. Eng. 33-34, 144–156.
Process model outputs during each case study for the multistage compression and liquefaction route, including operational and design characteristics of the compression stages and surge control systems.
CS3
Surge Volume Flow (m3/s)
Aspelund, A., Jordal, K., 2007. Gas conditioning—The interface between CO2 capture
and transport. Int. J. Greenhouse Gas Control 1 (3), 343–354.
A. Aspelund, “12 - Gas purification, compression and liquefaction processes and
0.0766
0.0743
0.0742
0.0742
0.0742
0.0744
0.0744
0.0744
0.0742
0.0742
0.0742
0.0755
0.0755
0.0735
0.0766
0.0759
0.0759
0.0739
0.0770
0.0770
0.0710
0.0759
0.0759
0.0740
technology for carbon dioxide (CO2) transport,” in Developments and Innovation in
CS2
Carbon Dioxide (CO2) Capture and Storage Technology, Woodhead Publishing, 2010,
pp. 383-407.
Aungier, R.H., 2000. Centrifugal Compressors: A Strategy for Aerodynamic Design and
Analysis. American Society of Mechanical Engineers.
0.4330
0.0870
0.0870
0.0870
0.4190
0.4190
0.4190
0.4190
0.4190
0.4190
0.2900
0.2920
0.2920
0.0671
0.0671
0.0671
0.4160
0.4160
0.4160
0.4200
0.4200
0.4200
0.4330
0.4330
Brunsvold, A., Jakobsen, J.P., Mazzetti, M.J., Skaugen, G., Hammer, M., Eickhoff, C.,
CS1
Neele, F., 2016. Key findings and recommendations from the IMPACTS project. Int.
J. Greenhouse Gas Control 54 (2), 588–598.
Budinis, S., Thornhill, N.F., 2015. Control of centrifugal compressors via model
LP1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
9–14.
Budinis, S., Thornhill, N.F., 2016. Supercritical fluid recycle for surge control of CO2
44.40
44.30
44.40
44.90
44.90
44.70
44.90
44.90
44.90
44.40
44.40
44.30
44.90
44.90
44.90
44.40
44.40
44.30
44.90
44.90
44.90
44.90
44.90
44.90
Bui, M., Adjiman, C.S., Bardow, A., Anthony, E.J., Boston, A., Brown, S., Fennell, P.S.,
Fuss, S., Galindo, A., Hackett, L.A., Hallett, J.P., Harzog, H.J., Jackson, G.,
35.40 Kemper, J., Krevor, S., Maitland, G.C., Matuszewski, M., Metcalfe, I.S., Petit, C.,
35.40
35.40
35.40
35.40
35.40
35.40
35.40
37.20
37.20
34.10
38.30
38.30
37.00
37.00
33.90
38.00
38.00
38.50
37.00
37.00
34.00
35.40
35.40
Puxty, G., Reimer, J., Reiner, D., Rubin, E., Scott, S., Shah, N., Smit, B., Trusler, J.P.,
CS2
Webley, P., Wilcox, J., Dowell, N.Mac, 2018. Carbon capture and storage (CCS): the
way forward. Energy Environ. Sci. 11 (5), 1062–1176.
Chen, S., Lior, N., Xiang, W., 2015. Coal gasification integration with solid oxide fuel cell
30.50
30.50
30.50
27.70
27.70
27.70
27.70
27.70
27.70
28.30
28.30
28.30
27.70
27.70
27.70
27.70
27.70
27.70
28.10
28.60
28.70
28.20
28.20
28.20
Cooper, R., Bove, D., Audasso, E., Ferrari, M.C., Bosio, B., 2020. A feasibility assessment
Average Impeller Diameter (m)
of a retrofit Molten Carbonate Fuel Cell coal-fired plant for flue gas CO2 segregation.
Int. J. Hydrogen Energy p. in proof.
Cormos, C.-C., 2011. Evaluation of power generation schemes based on hydrogen-fuelled
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
CS3
combined cycle with carbon capture and storage (CCS. Int. J. Hydrogen Energy 36
(5), 3726–3738.
Duan, L., Sun, S., Yue, L., Qu, W., Yang, Y., 2015. Study on a new IGCC (Integrated
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
CS2
IEAGHG, “Rotating equipment for carbon dioxide capture and storage,” 2011.
Institute, G.C., 2020. Global Status of CCS: 2020. Global CCS Institute, Australia.
IPCC, 2005. Carbon Dioxide Capture and Storage. Cambridge University Press,
Cambridge.
63.80
63.40
63.40
63.40
63.60
63.40
63.40
63.40
64.10
64.10
62.40
63.50
63.40
64.50
64.50
63.50
63.80
63.80
64.00
64.50
64.50
63.20
63.60
63.60
Tip Speed (m/s)
CS2
Jensen, M.D., Steven, J.A.S., Schlasner, M., Hamling, J.A., 2014. Operational flexibility
of CO2 transport and storage. Energy Procedia 63, 2715–2722.
Kahlke, S.-L., Pumpa, M., Schütz, S., Kather, A., Rütters, H., 2020. Potential Dynamics of
CO2 Stream Composition and Mass Flow Rates in CCS Clusters. Processes 8, 1188.
61.30
61.30
87.10
87.10
87.10
87.10
87.10
87.10
86.90
86.90
86.90
87.20
87.20
87.20
88.00
88.00
88.00
61.30
76.70
77.20
77.20
63.90
63.90
63.80
CS1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
California.
McMillan, G.K., 2010. Centrifugal and Axial Compressor Control, 10th ed. Momentum
Press, New York.
4250
4050
4050
4200
4175
4175
4175
4150
4175
4175
4175
4175
4175
4175
4175
4150
4150
4150
4150
4200
4050
4050
4000
4150
CS2
MEP, 2016. Feasibility study for full-scale CCS in Norway. Ministry of Petroleum and
Energy.
Mitsubishi Power, “M701J Series,” 2020. [Online]. Available: https://power.mhi.
com/products/gasturbines/lineup/m701j. [Accessed 09 December 2020].
4775
4775
4775
4775
4775
4385
4385
4385
4775
4775
4775
4775
4385
4385
4385
4700
4700
4700
4375
4375
4375
4775
4775
4775
CS1
Modekurti, S., Eslick, J., Omell, B., Bhattacharyya, D., Miller, D.C., Zitney, S.E., 2017.
Design, dynamic modeling, and control of a multistage CO2 compression system. Int.
J. Greenhouse Gas Control 62, 31–45.
IGCC-SOFC-CLC-300
IGCC-SOFC-CLC-600
Moe, A.M., Dugstad, A., Benrath, D., Jukes, E., Anderson, E., Catalanotti, E., Durusut, E.,
IGCC-SOFC-CLC-20
Neele, F., Grunert, F., Mahgerefteh, H., Gazendam, J., Barnett, J., Hammer, M.,
IGCC-Selexol-300
IGCC-Selexol-600
IGCC-MCFC-300
IGCC-MCFC-600
IGCC-Selexol-20
Pre- MDEA-300
Pre- MDEA-600
Steel-MEM-300
Steel-MEM-600
IGCC-MCFC-20
Span, R., Brown, S., Munkejod, S.T., Weber, V., 2020. A Trans-European CO2
Pre- MDEA-20
Post-MEA-300
Post-MEA-600
Steel-MEM-20
Post-NH3-300
Post-NH3-600
SMR-PSA-300
SMR-PSA-600
Post-MEA-20
Post-NH3-20
SMR-PSA-20
Source
Muhammad, H.A., Roh, C., Cho, J., Rehman, Z., Sultan, H., Baik, Y.-J., Lee, B., 2020.
A comprehensive thermodynamic performance assessment of CO2 liquefaction and
12
M.D. Wilkes et al. International Journal of Greenhouse Gas Control 111 (2021) 103449
pressurization system using a heat pump for carbon capture and storage (CCS) Stewart, M., 2018. Surface Production Operations: Volume IV - Pump and Compressor
process. Energy Convers. Manage. 206, 112489. Systems: Mechanical Design and Specification. Gulf Professional Publishing.
Peletiri, S.P., Mujtaba, I.M., Rahmanian, N., 2019. Process simulation of impurity Streb, A., Mazzotti, M., 2020. Novel Adsorption Process for Co-Production of Hydrogen
impacts on CO2 fluids flowing in pipelines. J. Cleaner Prod. 240, 118145. and CO2 from a Multicomponent Stream—Part 2: Application to Steam Methane
Pipitone, G., Bolland, O., 2009. Power generation with CO2 capture: Technology for CO2 Reforming and Autothermal Reforming Gases. Ind. Eng. Chem. Res. 59 (21),
purification. Int. J. Greenhouse Gas Control 3 (5), 528–534. 10093–10109.
Porter, R.T.J., Mahgerefteh, H., Brown, S., Martynov, S., Collard, A., Woolley, R.M., Tebodin, 2011. KNOWLEDGE SHARING REPORT – CO2 Liquid Logistics Shipping
Fairweather, M., Falle, S.A.E.G., Wareing, C.J., Nikolaidis, I.K., Boulougouris, G.C., Concept (LLSC) Overall Supply Chain Optimization. Global Carbon Capture and
Peristeras, L.D., Tsangaris, D.M., Economou, J.G., Salvador, C., Krevor, S., 2016. Storage Institute, Canberra.
Techno-economic assessment of CO2 quality effect on its storage and transport: Visser, E.d., Hendriks, C., Barrio, M., Mølnvik, M.J., Koeijer, G.d., Liljemark, S., Gallo, Y.
CO2QUEST: An overview of aims, objectives and main findings. Int. J. Greenhouse L., 2008. Dynamis CO2 quality recommendations. Int. J. Greenhouse Gas Control 2
Gas Control 54 (2), 662–681. (4), 478–484.
Porter, R.T.J., Fairweather, M., Kolster, C., Dowell, N.M., Shah, N., Woolley, R.M., 2017. Wetenhall, B., 2014. Impact of CO2 impurity on CO2 compression, liquefaction and
Cost and performance of some carbon capture technology options for producing transportation. Energy Procedia 63, 2764–2778.
different quality CO2 product streams. Int. J. Greenhouse Gas Control 57, 185–195. Wilkes, M.D., Mukherjee, S., Brown, S., 2020. Transient CO2 capture for open-cycle gas
PSE, 2016. gCCS Documentation. Process Systems Enterprise Limited, London. turbines in future energy systems. Energy, 119258.
Romeo, L.M., Bolea, I., Lara, Y., Escosa, J.M., 2009. Optimization of intercooling Wilkes, M.D., Mukherjee, S., Brown, S., 2021. Compression system power requirements
compression in CO2 capture systems. Appl. Therm. Eng. 29 (8), 1744–1751. for various CO2 sources and transportation options. Comp. Aid. Chem. Eng. 49.
Romano, M.C., Chiesa, P., Lozza, G., 2010. Pre-combustion CO2 capture from natural gas Accepted Manuscript.
power plants, with ATR and MDEA processes. Int. J. Greenhouse Gas Control 4 (5), Witkowski, A.S., Majkut, M., 2012. The impact of CO2 compression systems on the
785–797. compressor power required for a pulverized coal-fired power plant in post-
Roussanaly, S., Bureau-Cauchois, G., Husebye, J., 2013. Costs benchmark of CO2 combustion carbon dioxide sequestration. The J. Committee Mach. Build. Polish
transport technologies for a group of various size industries. Int. J. Greenhouse Gas Acad. Sci. 59 (3), 343–360.
Control 12, 341–350. Yan, J., Zhang, Z., 2019. Carbon Capture, Utilization and Storage (CCUS). Appl. Energy
Sircar, S., Golden, T.C., 2000. Purification of Hydrogen by Pressure Swing. Seper. Sci. 235, 1289–1299.
Tech. 35 (5), 667–687. Yu, H., Morgan, S., Allport, A., Cottrell, A., Do, T., McGregor, J., Wardhaugh, L.,
Spallina, V., Nocerino, P., Romano, M.C., Annaland, M.v.S., Campanari, S., Gallucci, F., Feron, P., 2011. Results from trialling aqueous NH3 based post-combustion capture
2018. Integration of solid oxide fuel cell (SOFC) and chemical looping combustion in a pilot plant at Munmorah power station: Absorption. Chem. Eng. Res. Des. 89 (8),
(CLC) for ultra-high efficiency power generation and CO2 production. Int. J. 1204–1215.
Greenhouse Gas Control 71, 9–19.
13