Critical Assessment of Membrane Technology Integration in a Coal-Fired Power Plant

Abstract

Despite the many technologies for CO₂ capture (e.g., chemical or physical absorption or adsorption), researchers are looking to develop other technologies that can reduce CAPEX and OPEX costs as well as the energy requirements associated with their integration into thermal power plants. The aim of this paper was to analyze the technical and economic integration of spiral wound membranes in a coal-fired power plant with an installed capacity of 330 MW (the case of the Rovinari power plant—in Romania). The study modeled energy processes using CHEMCAD version 8.1 software and polymer membranes developed in the CO₂ Hybrid research project. Thus, different configurations such as a single membrane step with and without the use of a vacuum pump and two membrane steps placed in series were analyzed. In all cases, a compressor placed before the membrane system was considered. The use of two serialized stages allows for both high efficiency (minimum 90%) and CO₂ purity of a minimum of 95%. However, the overall plant efficiency decreased from 45.78 to 23.96% and the LCOE increased from 75.6 to 170 €/kWh. The energy consumption required to capture 1 kg of CO₂ is 2.46 MJ_el and 4.52 MJ_th.

1. Introduction

In the past decade, considerable improvements in polymeric membrane materials for gas separation processes have been observed, where many types of polymer materials have been manufactured. Consequently, their transport properties could provide an energy-efficient path for wide-range gas separations. Due to the urgent demand to mitigate CO₂ emissions from the application of fossil fuels, these planned research potentials have been vastly catalyzed. In general, CCS technologies are believed to be a successful technological solution to reduce the environmental and economic carbon dioxide influences, comprising climate change [1]. The classical process for CO₂ capture, which is based on amine absorption, demands high energy, needs high capital and operating cost, and causes corrosion and environmental troubles in addition to some operational problems. As a result, the membrane is used due to high energy efficiency, operational plainness, system consolidation, and capability to beat thermodynamic solubility restrictions. As a rule, a membrane is a thin interphase layer that plays a significant role in separating two phases [2,3].

Presently, various CO₂ capture processes have been used based on either ejecting CO₂ from flue gas (post-combustion), from syngas (pre-combustion), or injecting pure O₂ instead of air (oxy-fuel combustion) which provides a high CO₂ flow [4,5,6,7].

So far, different membranes have been enhanced and characterized by high permeability and selectivity to achieve a high CO₂ capture process. Various types of membrane materials such as inorganic membranes, common polymers, carbon molecular sieve membranes, mixed matrix membranes, fixed site-carrier (FSC) membranes, and carbon molecular sieve membranes have been utilized for CO₂ separation operation [8]. Nevertheless, to provide a commercially viable membrane to capture CO₂ and contend with the classical amine absorption process, membrane technology must have a relatively low power consumption and low particular capture cost conjointly with an acceptable stability exposure to impurities such as SO₂ and NO_x, which are generally included in the flue gas stream [8].

Over the last decade, preferable polymeric materials with enhanced CO₂ permeability and selectivity have been vastly researched. Generally, more permeable polymers lead to less selectivity and vice versa, similar to a trade-off relationship. In any case, most polymeric materials depend on the solution-diffusion mechanism in which this trade-off is heavily rooted [9,10].

Processability is a basic demand for an effective industrial gas separation membrane. Dense polymers are suitable membrane materials whereas most commercial membranes are manufactured from polymers with low-cost and good scalability, as a result of the credible fabrication of a fine selective layer of 0.1–10 μm and membrane surface area at a range of (1000–1,000,000 m²) [11]. The simplest model utilized to demonstrate and predict the gas permeation process via the dense membrane is the solution-diffusion model, where a molecule can transport from one side of the material to another by a concentration gradient. In typical, the separation can be obtained by the difference in solubility and/or diffusivity. The solubility of specific gases relies on their condensability and affinity to the membrane material. Generally, the molecule condensability expands by increasing critical temperature [12]. On the other hand, the diffusion of a gas molecule depends on the space between two chains, where the chain piece’s random motion permits small kinetic diameter molecules to diffuse through [13]. As a result, the molecule diffusivity increases by the decrease of kinetic diameter.

Table 1. Physical properties of different gases [14].

Gas	Kinetic Diameter/nm	Critical Temperature/°C
CO2	0.330	304.1
N2	0.364	126.2
H2	0.289	33.2
CH4	0.380	190.6

demonstrates the physical properties of different gases.

2. Solution-Diffusion Polymeric Membranes

In this part, the current advances for polymers that are eligible for selective CO₂ separation based on the mechanism of solution diffusion are researched. Five types of polymers are involved: polyethylene oxide (PEO), perfluoro polymers, polymers of intrinsic microporosity (PIMs), thermally rearranged (TR), and iptycene-containing polymers. These types are sorted depending on the discovery period or application in the gas separation process. Polymer string hardness is the main factor that affects this sequence, from the rubbery PEO to the vitreous TR polymers. In addition, the iptycene-containing polymers are derived/adjusted polymers through the integration of iptycene-containing moieties [15].

2.1. PEO-Based Membranes

Despite the CO₂ molecule being non-polar, the different distribution of charges inside gives the molecule a quadrupole moment [16]. In poly (ethylene oxide) (PEO), the polar ether connection (–C–C–O–) is noticed to have a high affinity to CO₂ [17]. Thus, PEO-based polymers offer a major CO₂ solubility, and CO₂ selectivity typically derives from the solubility selectivity. However, one drawback that has been observed is the high degree of crystallinity in pure PEO or in materials related to PEO. Due to the polar ether groups inclination to compose powerful hydrogen bonding, which leads to a compact chain packing [18]. The crystalline area that formed obstructs the diffusion of CO₂ and ultimately restrains the membrane permeability. Weak mechanical strength is also a result of high crystallinity. To compensate for these restrictions, different processes have been applied, including (1) block copolymerization with other rigid pieces, (2) combining with low MW poly (ethylene glycol) (PEG) and its derivatives, and (3) crosslinking to compose a highly branched PEO polymer network [19]. In addition to providing noticeably permeable PEO membrane materials, these efforts also drive a remarkable understanding of the nanostructures of the PEO polymers. Generally, the most accepted process to prevent the high crystallinity of PEO is to block the copolymerization of PEO with rigid pieces [11,20].

Table 2. Transport properties of specific PEO polymers.

Strategy	Material	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
Copolymer	PEO-b-PA6 [14]	35	120	51.4	9.8
	PEO-b-PBT [20]	30	150	51.5	10.3
	PEO-ran-PPO T6T6T [21]	35	470	43	10
	Pent-PI-PEO2000 [22]	35	39	36	4.1
	PEO-b-PBT on PDMS [23]	30	1815 *	50	-
	PEO-b-PS [24]	70	20,400 *	27.7	-
Blending	PEO-PBT/PEG200 [20]	30	208	48.7	11.6
	PEO-PBT/PEG-BE [20]	30	400	50.1	11.8
	PEO-PBT/PEG-DBE [20]	30	750	40	12.4
	PEO-PPO-T6T6T/PDMS-PEG [25]	45	896	36	10.6
Crosslinking	PEO-526/dopamine/PEGDME [26]	50	200	30	6
	PEO-amine/PEO-epoxy [27]	35	376	53	10

* GPU.

demonstrates the permeabilities of CO₂ and CO₂/gas selectivities of the PEO polymers based on various strategies.

The PEO block copolymers commonly demonstrated a CO₂ permeability around 100–200 Barrers with the selectivity of CO₂/N₂ about 50 at 25 °C. Monodisperse tetra-amide (T6T6T) and pentiptycene-based polyimide (pent-PI) are kinds of many new hard segments that were incorporated with PEO to assemble ultra-permeable PEO-based copolymers [22,23]. The self-synthesis trait of the PEO-based block copolymer was also largely studied. A polymer chain rearrangement was noticed by Yave et al. when an ultrathin selective layer was covered onto a hydrophobic PDMS face, resulting in a high CO₂ permeance of 1815 GPU with a CO₂/N₂ selectivity around 50 at 30 °C [23]. Xue et al. assembled a PEO–polystyrene (PS) block copolymer, where cylindrical PEO domains are formed, which provide intense CO₂ permeance of 20,400 GPU with a CO₂/N₂ selectivity around 27.7 at 70 °C [24].

Another strategy that is utilized to increase the ether content in the polymer matrix is known as blending. A 100–2 000 MW short-chain PEG was established in a PEO-based copolymer via polymer chain tangle, which improved the CO₂ solubility and broke the compact packing of the PEO piece in the copolymer. The strong hydrogen bonding among the ether kinds can be reduced due to the tip parts on PEG that supplied another control. As a result, PEG moieties with end parts, such as methyl ether, ally ether, divinyl ether, and butyl ether, were researched [23].

The crosslinking process indicates the bottom-up assembly of extreme branches of PEO or PEO-based copolymers by the polymerization of ethylene oxide monomer or oligomer. In this process, the primary work was based on different methacrylate monomers, where the crosslinking mitigated the crystallinity and enhanced the film-forming capacity [28]. Kline et al. demonstrated that the crosslinking density and heterogeneity could be adjusted via the platform of poly (ethylene glycol) diglycidyl ether and polyether diamine, where the heterogeneity crosslinking enhanced the permeability of CO₂ [27]. In Figure 1, unimodal, bimodal, and clustered PEO networks are assembled.

PEO-based polymers have the main commercial potential, due to the high CO₂ permeance that has been presented in thin-film-composite membranes under specific testing conditions. Furthermore, the various PEO-based materials that have demonstrated practicable selectivities at an operating temperature of more than 35 °C are of special significance [24,29].

2.2. Perfluoro-Polymers

These are a group of glassy hydrocarbon polymers with added fluorine atoms instead of all or most hydrogen atoms. Due to the powerful C–C and C–F covalent bonds, Perfluoro-polymers are resistant to many chemicals, which leads to these polymers being typical for applications that are submitted to hostile situations [30]. One of the drawbacks is their semi-crystalline nature and low solvent processability, which largely obstruct the development of the polymers in gas separation. In the mid-1980s, the gas permeation data were obtained through the introduction of many amorphous perfluoro-polymers with specific trade names, such as Teflon™ AF, Hyflon™ AD, and Cytop™. In Figure 2, the chemical structures of these perfluoro-polymers, which are commercially obtainable, are demonstrated. They are either cyclic homopolymers or copolymers of tetra-fluoro ethylene and perfluoro odioxole, where they are known for their elevated gas permeability because of the pre-existing micro channels [31]. The properties of these glassy perfluoro-polymers are shown in

Table 3. Transport properties of specific perfluoro-polymers.

Strategy	Material	P (CO2)/atm	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
Commercial	Teflon™ AF2400 [32]	1	35	2200	4.6	0.96
	Teflon™ AF1600 [33]	1	35	520	4.7	1.06
	Hyflon™ AD80 [34]	3	35	473	6.1	1.19
	Hyflon™ AD60 [34]	3	35	124	7.3	1.63
	Cytop™ [34]	1	35	35	7	1.69
	Teflon™ AF2400 [35]	4.4	22	13,000 *	4.8	0.81
	Hyflon™ AD60 [35]	4.4	22	1330 *	7.3	1.28
Homo-polymer	Poly(PFMD) [36]	4.4	35	5.9	8.3	8.47
	Poly(PFMMD) [36]	4.4	35	58	7.5	4.1
Copolymer	Poly(PFMMD-co-PFMD) [37]	4.4	22	403 *	9.1	2.9
	Poly(PFMMD-co-CTFE) [37]	4.4	22	44 *	9	5.7

* GPU.

, listed from the most permeable Teflon™ AF2400 to the least permeable Cytop™.

2.3. Polymers of Intrinsic Microporosity (PIMs)

PIMs are a kind of glassy polymer with hard and twisted macromolecular backbone structures, which were originally reported by Budd and McKeown [38]. Unlike other porous organic polymers, PIMs are solution-treatable [38]. They are generated from the locations of twisting or spiro centers, where the poor molecular packing is induced by the restricted chain turnover of the component macromolecules, performing interconnected holes of smaller than 2 nm [39]. The substantial microporosity of this kind of polymer produces a less than 20% fractional free volume (FFV), resulting in elevated gas permeability [40]. The substantial microporosity of this kind of polymer produces a less than 20% fractional free volume (FFV), resulting in elevated gas permeability. These advantages swiftly stimulated concentrated research potential to assemble various PIMs to enhance the gas permeability and selectivity [38]. Table 4 represents the transport properties of the newly improved PIMs, involving spirobiindane (SBI)-based PIMs, Tröger’s base (TB)-based PIMs, polyimide (PI)-based PIMs, and some other differences. The crosslinking strategy to handle the fast physical aging is also summarized.

The kinks of the PIMs polymer backbone were primarily recognized through inserting the SBI moiety with large pendant sets. SPI can be defined as a molecule with two indanes linked by a spiro carbon center. The SPI piece is generally polymerized with a halogen-including aromatic monomer, which results in the first PIMs used in gas separation membranes, such as PIM-1 and PIM-7 [41,42]. In PIMs, the gas permeation follows the solution-diffusion mechanism. The CO₂/H₂ selectivity is around 1-3, which is due to the size sieving trait that prefers the diffusion of H2. A spiro-fluorene (SBF) element could replace the SBI center, resulting in PIM-SBF offering low chain flexibility and an ultrahigh CO₂ permeability of 13,900 Barrers [43]. Recently, big tetra methyl tetra hydro naphthalene (TMN) units were fused into the SBI units in PIM-1, leading to less conformational flexibility and resulting in higher CO₂ permeability of 17,500 Barrers [44].

Unlike the SBI-based PIMs, Tröger-base (TB), a hard bicyclic unit, has been applied to PIMs. Diamino aromatic polymers with bicyclic rings, such as ethano-anthracene (EA) and triptycene (Trip), are integrated with or without SBI centers to compose an extremely twisted network. The TB group supplies extra Langmuir affinity toward CO₂, which enhances CO₂ selectivity [45]. The contortion sites can be within either the diamine [46] or the dianhydride [47]. Regardless of the utilization of different chemistry for PIM assembly, various enhanced PIMs have been notified by integrating bulky side groups including aromatic rings, such as the hexaphenylbenzene (HPB) unit that is used to reduce physical aging [48]. Carta et al. notified the CO₂ permeability of 333,000 Barrers with a CO₂/N₂ selectivity of 14.9 through fusing bulky TMN and Trip into PIM [44]. Similar to other glassy polymers that have high free volume, PIMs suffer from physical aging, where the repose of the nonequilibrium series leads to the damage of permeability over time [49]. One intensively studied process is crosslinking to provide a more solid polymer network. For PIM-1, different crosslinking methods are used, e.g., thermal, UV, and chemical crosslinking, which are explained in detail in [50,51,52]. Driven by the high cost of the membrane, some polymeric materials have been blended with PIMs, such as carboxylate PIM-1 in Ultem and Matrimid as extremely permeable nanofillers, to mitigate the manufacturing cost as well as raise the gas selectivity [53,54]. Jue et al. found a defect-free HF asymmetric PIM-1 membrane created by phase inversion. A skin layer was provided, and CO₂ permeance of 360 GPU and 27.7 CO₂/N₂ selectivity were gained [55].

Table 4. Transport properties of specific PIMs.

Strategy	Material	P (CO2)/atm	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
SBI-based PIMs	PIM-1 alcohol-treated [41]	1	30	11,200	18.4	3.4
	PIM-7 [42]	0.2	30	1100	26	1.3
	PIM-SBF [43]	1	25	13,900	17.7	2.2
	PIM-TMN-SBI [44]	1	25	17,500	16.2	2.4
TB-based PIMs	PIM-EA-TB [56]	1	25	7140	13.6	0.92
	PIM-SBI-TB [42]	1	25	2900	12.5	1.3
	PIM-Trip-TB [57]	1	25	9709	15.9	1.2
PI-based PIMs	PIM-SBI-PI [47]	1	25	8210	18.7	3.1
	PIM-EA-PI [46]	1	25	7340	19.9	1.7
	6FDA-DAT1-OH [58]	2	35	47	25.9	0.37
PIMs wo SBI	PIM-TMN-Trip [44]	1	25	33,300	14.9	2
	PIM-HPB [48]	1	25	1800	20	7
Crosslinking	TOX-PIM-1 (thermal) [50]	4	22	5100	18.1	1.7
	PIM-1 (UV) [51]	4	22	6374	21.6	2.1
	PAH-PIM-1 (chemical) [52]	1	20	150	22.1	-
Blending	C-PIM-1/Matrimid [54]	3.5	35	2268	18.7	1.4
	PIM-1/Ultem [53]	3.5	35	3276	21.1	-
	PIM-1/POSS-PEG [59]	1	30	1309	31	-
	PIM-1/HCP [60]	2	25	19,086	11.6	-
Membrane	PIM-1 HF [55]	6.9	35	360 *	27.7	1

* GPU.

Table 4. Transport properties of specific PIMs.

Strategy	Material	P (CO2)/atm	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
SBI-based PIMs	PIM-1 alcohol-treated [41]	1	30	11,200	18.4	3.4
	PIM-7 [42]	0.2	30	1100	26	1.3
	PIM-SBF [43]	1	25	13,900	17.7	2.2
	PIM-TMN-SBI [44]	1	25	17,500	16.2	2.4
TB-based PIMs	PIM-EA-TB [56]	1	25	7140	13.6	0.92
	PIM-SBI-TB [42]	1	25	2900	12.5	1.3
	PIM-Trip-TB [57]	1	25	9709	15.9	1.2
PI-based PIMs	PIM-SBI-PI [47]	1	25	8210	18.7	3.1
	PIM-EA-PI [46]	1	25	7340	19.9	1.7
	6FDA-DAT1-OH [58]	2	35	47	25.9	0.37
PIMs wo SBI	PIM-TMN-Trip [44]	1	25	33,300	14.9	2
	PIM-HPB [48]	1	25	1800	20	7
Crosslinking	TOX-PIM-1 (thermal) [50]	4	22	5100	18.1	1.7
	PIM-1 (UV) [51]	4	22	6374	21.6	2.1
	PAH-PIM-1 (chemical) [52]	1	20	150	22.1	-
Blending	C-PIM-1/Matrimid [54]	3.5	35	2268	18.7	1.4
	PIM-1/Ultem [53]	3.5	35	3276	21.1	-
	PIM-1/POSS-PEG [59]	1	30	1309	31	-
	PIM-1/HCP [60]	2	25	19,086	11.6	-
Membrane	PIM-1 HF [55]	6.9	35	360 *	27.7	1

* GPU.

2.4. Thermally Rearranged (TR) Polymers

Thermally rearranged (TR) polymers are harder and planer macromolecules that can be composed by the thermal rearrangement of polyamide (PAs) or ortho-functionalized polyimides (PIs), which were first found by Park et al. [61]. Generally, the TR polymers are characterized by unprecedented polymer chain hardness and a tight bore size distribution, as a result of the microporous nature which occurs due to the high torsional energy fence versus turnover between the phenylene-heterocyclic circles. The main ancestors for the TR conversion are ortho-functionalized PIs or PAs, as shown in Figure 3 below.

When the temperature rises above 350 °C, an intramolecular cyclization is started, and a hard polymer, such as polybenzoxazole (PBO), polybenzimidazole (PBI), or polypyrrolone (PPL), is composed if the ortho-functional set is amino or hydroxyl [62]. The TR polymers obtained from the polycondensation of hydroxy-diamine and diacid chloride are named TR-β, while the ones from the polycondensation of dianhydride and ortho-functional diamine are designated TR-α [62]. The transport properties of various selected TR polymers are shown in

Table 5. The transport properties of various selected TR polymers.

Strategy	Material	P (CO2)/atm	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
TR-α-PBO	6FDA + bisAPAF [61]	10	35	4 045	25.9	1.4
	6FDA + bisAPAF [63]	1	25	4 201	14.8	1
	6FDA + bisAPAF + ADHAB [64]	3	30	151	21	-
	6FDA + HAB [65]	1	35	2.9	29	0
TR-α-PBI	6FDA + DAB [66]	1	25	1 624	26.2	0.91
TR-β-PBO	BPDC + bisAPAF [67]	10	35	532	17.6	1
Crosslinking	6FDA + bisAPAF + DABA/diol [68]	1	25	746	25.2	1.2
	6FDA + bisAPAF + DABA [69]	1	25	491	24.3	1
Copolymer	6FDA + bisAPAF + DAM [70]	0	35	137	21.6	0.78
	6FDA + HAB + 4MPD [71]	10	35	226	10.5	-
TR-labile PI	6FDA + DABA + βCD [72]	10	35	2 707	15.3	0.34
	6FDA + durene + DABA + γCD [73]	2	35	1 024	18.2	0.24
TR w/SBI	TR-PIM-1 [74]	1	35	675	23	1.6
	TR-PIM-2 [74]	1	35	263	24	1
TR HF	6FDA + bisAPAF [75]	1	25	2 326 *	20	1.2
	6FDA + bisAPAF [76]	1	25	2 500 *	16	1.2

* GPU.

.

2.5. Iptycene-Containing Polymers

The two molecules belonging to the iptycene family are triptycene (Trip) and pentiptycene (Pent), where the iptycene family is a group of three-dimensional molecules with arene identities integrated into the (2,2,2)bicyclooctatriene bridgehead method [77]. The iptycene membrane is unique due to the slits (blades) of the benzene creating an interior free volume, which can be compared to the kinetic diameters of the light gases in Table 1. The internal free volume, because of its shape and persisting nature, is not subjected to collapse, and low physical aging is obtained [78]. When the huge iptycenes are integrated into other polymer systems, the polymer chain packing is damaged and the overall free volume is increased [44]. Weidman et al. that proposed iptycene polymers can be classified into three categories: non-ladder, semi-ladder, and ladder, depending on the backbone architecture [79]. The categories are reviewed more systematically in

Table 6. Transport properties of specific iptycene-containing polymers.

Strategy	Material	P (CO2)/atm	T/°C	CO2 Permeability/Barrer	(CO2/N2) Selectivity	(CO2/H2) Selectivity
Non-ladder	6FDA + DATRI [80]	1	35	189	23.3	0.73
	6FDA + DAT2 [49]	2	35	210	23.3	0.74
	6FDA + PPDA(CF3) [81]	1	35	132	19	0.7
	TPDAn + 6FAP [82]	1	35	4.7	25	0
	TPHA-TC [83]	11	35	270	-	-
Semi-ladder	KAUST-PI-1 [84]	2	35	2 389	22.3	0.6
	KAUST-PI-2 [84]	2	35	2 071	21.1	0.87
	6FDA + PAF [85]	2	35	6.8	-	-
	TPDA + APAF [85]	2	35	46	-	-
	PBIBI + PPD [86]	1	25	137.2	27.8	-
Ladder	PIM-Trip-TB [87]	1	25	9 709	15.4	1.2
	PIM-Btrip-TB [87]	1	25	13 200	14.2	1.3

.

3. Membrane Technology for CO₂ Capture

To improve the CO₂ capture system, several standards should be taken into consideration, such as high capture rate and low operating costs. Moreover, the flexibility of the membrane system plays the main role to choose the best configuration for the CO₂ capture process. Presently, several procedures are utilized with different parameters to optimize CO₂ capture technology. The membrane process requires a high separation of acid gases and impurities that are commonly part of the flue gas stream to avert harmful issues and extend its lifetime [88]. Figure 4 shows the membrane system used, with the acid gas separation process, which is integrated with a coal-fired power plant (CFPP).

The low volumetric fraction of CO₂, contrasted with a high volume of the flue gas stream to be treated, is the essential challenge for the post-combustion capture process, which drives a low driving force of CO₂ permeation. To overcome the low motive force in the membrane process fused into post-combustion CO₂ capture technology, either a compressor before the membrane module or a vacuum pump in the permeate flow side, or both together, can be used [89]. As a result, the flue gas stream must be dried before entering the compression station to avoid troubles caused by water droplets.

In this paper, three different schemes with several parameters have been presented, either by using 1-single stage (with and without a vacuum pump) or 2-stages of the membrane to obtain both goals of 90% carbon capture rate and at least 95% purity of CO₂ captured. High purity is required for the transportation purposes and for other goals such as enhance oil recovery (EOR) [90]. As can be noticed, the flue gases must be compressed before any membrane module to produce sufficient driving force for CO₂ separation. As researched further, membrane surface area and energy consumption demonstrate important factors of any membrane module for CO₂ capture. Designing the CO₂ separation system utilizing membrane reveals the size of the membrane technology, such as the required surface area and the suitable configuration to achieve the CO₂ purity goal (min. 95 mole %) with the lowest power consumption. In general, the efficiency of the process relies essentially on the compression ratio of the flue gas, membrane permeability, and membrane surface area. On the other hand, the purity of the CO₂ stream depends on membrane selectivity, capture efficiency, and CO₂ content in the combustion gases. This analysis concentrated on designing the optimal process with minimal energy and surface area needed to reach the proposed project targets.

In the research project (13/2020) a currently developed procedure was utilized to embed CA enzyme into polyacrylamide polymer (PSF 50 K) [91]. Generally, the lifetime of the membrane is 5 years, and after this period the membrane performance will decrease and the material must be replaced [8,92]. The permeability and selectivity for different gases are mentioned in Table 7.

The objective of this article is to evaluate and compare the performance of a 1-single stage and 2-stages membrane unit fused into a conventional CFPP. The flue gas temperature and pressure treated in this study are 50 °C-1.015 bar, respectively (Table 7). Moreover, a high compressor pressure must be utilized in four stages (in this study we assumed it 70 bar) to compress and prepare CO₂ captured for transportation goals with an inter-cooling process to reduce the high temperature generated from the high-pressure compressor [93] in order to accomplish all the demands for CO₂ transport.

Table 7. The main parameters of CFPP and membrane.

Parameters	Main Data
Fuel characteristics [94]	72.30% C, 4.11% H, 1.69% N, 7.45% O,
	0.56% S, 13.89% ash; Moisture: 8%;
	Lower heating value: 28,141 kJ/kg
CFPP parameter
Steam temperature/pressure, [°C/bar]	585/290
Efficiency in high/medium/low-pressure steam turbine, [%]	84.9/91.6/87.8
Condensing pressure, [bar]	0.05
Heating water in the condenser, [°C]	9.5
Coal combustion efficiency in the steam generator, [%]	91
Steam flow rate, [t/h]	914.5
Net power plant efficiency, [%]	45.78
Flue gas Parameters
Temperature/pressure [°C/bar]	50/1.013
Flue gas flow, [kmol/h]	40,320
Flue gas content, [mole%]
CO2	13.12
N2	80.80
O2	6.03
SO2	0.04
Membrane parameters [95]
Membrane material characteristics	Spiral wound in counter current
CO2 permeance, [GPU]	1000
N2 permeance, [GPU]	20
CO2/N2 selectivity	50
Compressor/vacuum pump efficiency, [%]	90
Variations of different membrane parameters simulated
1st Compressor pressure, [bar]	1.5–10
Vacuum pump pressure (case B), [bar]	0.5–0.05
2nd Compressor pressure (case C), [bar]	2–10
1st membrane surface, [m2]	100,000–1,000,000
2nd membrane surface (case C), [m2]	5000–100,000

Table 7. The main parameters of CFPP and membrane.

Parameters	Main Data
Fuel characteristics [94]	72.30% C, 4.11% H, 1.69% N, 7.45% O,
	0.56% S, 13.89% ash; Moisture: 8%;
	Lower heating value: 28,141 kJ/kg
CFPP parameter
Steam temperature/pressure, [°C/bar]	585/290
Efficiency in high/medium/low-pressure steam turbine, [%]	84.9/91.6/87.8
Condensing pressure, [bar]	0.05
Heating water in the condenser, [°C]	9.5
Coal combustion efficiency in the steam generator, [%]	91
Steam flow rate, [t/h]	914.5
Net power plant efficiency, [%]	45.78
Flue gas Parameters
Temperature/pressure [°C/bar]	50/1.013
Flue gas flow, [kmol/h]	40,320
Flue gas content, [mole%]
CO2	13.12
N2	80.80
O2	6.03
SO2	0.04
Membrane parameters [95]
Membrane material characteristics	Spiral wound in counter current
CO2 permeance, [GPU]	1000
N2 permeance, [GPU]	20
CO2/N2 selectivity	50
Compressor/vacuum pump efficiency, [%]	90
Variations of different membrane parameters simulated
1st Compressor pressure, [bar]	1.5–10
Vacuum pump pressure (case B), [bar]	0.5–0.05
2nd Compressor pressure (case C), [bar]	2–10
1st membrane surface, [m2]	100,000–1,000,000
2nd membrane surface (case C), [m2]	5000–100,000

Three configurations for integrating membrane technology in CFPP, where the variations for these configurations are demonstrated in the membrane parameters section of the Table 7, are proposed as follows:

A.

1-single membrane with a compression station before the membrane inlet, see Figure 5;

B.

1-single membrane with a compression station before the membrane and a vacuum pump on the permeate flow side, see Figure 6;

C.

2-stages of a membrane with different compressors and vacuum pumps, check Figure 7.

In general, for a specific CO₂ capture efficiency, the operation at a high-pressure difference across the membrane shows more power consumption required, a smaller membrane surface area, and higher CO₂ purity until specific pressure. Furthermore, the performance at a lower pressure difference implies lower energy consumption, larger membrane surface area, and lower CO₂ purity.

4. Decarbonized Coal-Based Super-Critical Power Plants, Main Design Characteristics, and Assessment Methodology

As a targeted industrial process to be decarbonized by membrane systems, a 330 MW net output coal power plant was investigated, using lignite coal as a fuel. The decarbonization yield is 90% the same as most of the CCS projects [93]. As can be distinguished from Figure 4, the flue gas out of the coal combustion process must be subjected to particulate matter (NO_x and SO_x) removal before the CO₂ capture process.

Table 7 presents the main technical assumptions of investigated CFPP and membrane.

All processes were modeled and simulated by using ChemCAD software. The paper analyzed different parameters of the compressor pressure, membrane surface area, and vacuum pump pressure (if used) in many cases to estimate the techno-economic influence of the analyzed CO₂ capture solutions and achieve the membrane efficiency of 90% and CO₂ purity of 95% with at least the minimum energy consumption required. However, power consumption in membrane technology is only the energy required for the main driving machines such as compressors and vacuum pumps.

To evaluate the techno-economic influence of the analyzed CO₂ capture solutions, the following indicators have been proposed [96]:

$\mathrm{SPECCA}$—specific primary energy consumption for avoiding CO₂ emissions;

$$\mathrm{SPECCA}=\frac{3600·\left(\frac{1}{{\mathsf{\eta }}_{\mathrm{CCS}}}-\frac{1}{{\mathsf{\eta }}_{\mathrm{base}}}\right)}{{\mathrm{P}}_{\mathrm{base}}-{\mathrm{P}}_{\mathrm{CCS}}},$$

(1)

where ${\mathsf{\eta }}_{\mathrm{base}}$: the overall efficiency of energy solutions without CCS technology,

${\mathsf{\eta }}_{\mathrm{CCS}}$: the overall efficiency of energy solutions with CCS technology,

${\mathrm{P}}_{\mathrm{base}}$: the CO₂ pollutant in kg/kWh produced by the CFPP without capture technology,

${\mathrm{P}}_{\mathrm{CCS}}$: the CO₂ pollutant in kg/kWh produced by the CFPP with capture technology.

$\mathrm{SEPCCAs},\mathrm{m}$—takes into account the energy penalty $\left({\mathsf{\eta }}_{\mathrm{base}}-{\mathsf{\eta }}_{\mathrm{CCS}}\right)$ of the energy solution because of the extra heat consumption demanded in the chemical absorption process (${\mathrm{SPECCA}}_{\mathrm{s}}$—Equation (2)), and the electricity consumption required by the capture process using membranes (${\mathrm{SPECCA}}_{\mathrm{m}}$—Equation (3)):

$${\mathrm{SPECCA}}_{\mathrm{s}}=\frac{3600·\left({\mathsf{\eta }}_{\mathrm{base}}-{\mathsf{\eta }}_{\mathrm{CCS}}\right)}{{\mathsf{\eta }}_{\mathrm{base}}·{\mathrm{P}}_{\mathrm{base}}-{\mathsf{\eta }}_{\mathrm{CCS}}·{\mathrm{P}}_{\mathrm{CCS}}},$$

(2)

$${\mathrm{SPECCA}}_{\mathrm{m}}=\frac{3600·\left({\mathrm{W}}_{\mathrm{base},\mathrm{net}}-{\mathrm{W}}_{\mathrm{CCS},\mathrm{net}}\right)}{{\mathrm{W}}_{\mathrm{base},\mathrm{net}}·{\mathrm{P}}_{\mathrm{base}}-{\mathrm{W}}_{\mathrm{CCS},\mathrm{net}}·{\mathrm{P}}_{\mathrm{CCS}}},$$

(3)

where ${\mathrm{W}}_{\mathrm{base},\mathrm{net}}$: the net power generated for the energy solution without capture solution; ${\mathrm{W}}_{\mathrm{CCS}, \mathrm{net}}$: the net power generated for the energy solution with capture solution.

The levelized cost of electricity ($\mathrm{LCOE}$) was calculated by Equation (4), considering the annualized $\mathrm{CAPEX}$ and $\mathrm{OPEX}$ costs, a specified CO₂ recovery (${\mathsf{\phi }}_{\mathrm{CO}2}),$ and the CO₂ recovery flow (${\mathrm{CO}}_{2\mathrm{CCS}})$, where 6570 demonstrates 75% of annual capacity of hours:

$$\mathrm{LCOE}=\left(\frac{\mathrm{CAPEX}+\mathrm{OPEX}}{6570·{\mathsf{\phi }}_{\mathrm{CO}2}·{\mathrm{CO}}_{2\mathrm{CCS}}}\right)$$

(4)

Both recovery and avoided costs of CO₂ indicators are calculated by taking the consideration $\mathrm{LCOE}$, ${\mathrm{CO}}_{2\mathrm{CCS}}$, and specific CO₂ emissions for energy solution with and without CCS solution:

$${\mathrm{CO}}_{2\mathrm{rc}}=\frac{{\mathrm{LCOE}}_{\mathrm{CCS}}-{ \mathrm{LCOE}}_{\mathrm{base}}}{{\mathrm{CO}}_2\hspace{0.33em}\mathrm{recovery}}$$

(5)

$${\mathrm{CO}}_{2\mathrm{ac}}=\frac{{\mathrm{LCOE}}_{\mathrm{CCS}}-{ \mathrm{LCOE}}_{\mathrm{base}}}{{\mathrm{CO}}_{2\mathrm{base}}-{\mathrm{CO}}_{2\mathrm{CCS}}}$$

(6)

In order to set the economic indicators, several data on the unit costs of the components are presented in

Table 8. Main assumptions concerning the economic indicators.

Indicators	Main Data
Availability factor, %	85
Electricity price, €/MWh	160
CO2 tax, €/t [98]	82
Annual hours, h/year	7446 (85/100·8760)
Membrane capture unit
Membrane-specific cost, €/m2	50
Membrane lifetime, years	5
Flue gas, and inter-stage compressor, €/kW	850
Vacuum pump, €/kW	1300
CO2 pump, €/kW	1350
CO2 compressor, €/kW	1800
Membrane replacement cost	20% of the membrane cost
Labor cost, €/h	15
CO2 stream compression
CO2 compressor, M€	11.7
Compressor inter-stage coolers and separators, M€	0.87

. The presented costs are calibrated to the year 2022 based on the index demonstrated on the Chemical Engineering online site [97].

In order to be able to determine whether an investment project, in this case, the CFPP with and without CO₂ capture, is economically appropriate, an economic and financial analysis is required that considers all cash flows in and out of the established meter. The economic and financial indicators calculated in this analysis are as follows:

Net present value ($\mathrm{NPV}$) calculated with Equation (7):

$$\mathrm{NPV}={{\displaystyle \sum }}_{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{f}}}\frac{{\mathrm{IN}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{i}}-{\mathrm{A}}_{\mathrm{i}}}{{\left(1+\mathrm{r}\right)}^{\mathrm{i}}}-{{\displaystyle \sum }}_{\mathrm{i}=1}^{{\mathrm{n}}_{\mathrm{r}}}{\mathrm{I}}_{\mathrm{i}}·{\left(1+\mathrm{r}\right)}^{\mathrm{i}}, €$$

(7)

where ${ \mathrm{IN}}_{\mathrm{i}}$: the realized revenues for a year $\mathrm{i}$ (€/year); ${\mathrm{C}}_{\mathrm{i}}$: the operating and maintenance expenses for the year $\mathrm{i}$, with taxes and duties but without depreciation (€/year); ${\mathrm{A}}_{\mathrm{i}}$: the annuity for the year $\mathrm{i}$, if a loan was taken (€/year); ${\mathrm{I}}_{\mathrm{i}}$: the realized equity investment for the year $\mathrm{i}$ (€/year); $\mathrm{r}$: the discount rate, which for the energy sector is 8%.

Internal rate of return ($\mathrm{IRR}$) was determined utilizing Equation (8).

$$\mathrm{NPV}={{\displaystyle \sum }}_{\mathrm{i}=1}^{\mathrm{n}}\frac{{\mathrm{IN}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{i}}-{\mathrm{I}}_{\mathrm{i}}}{{\left(1+\mathrm{IRR}\right)}^{\mathrm{i}}}=0,$$

(8)

$\mathrm{IRR}$ for an investment project is equal to the discount rate for which $\mathrm{NPV}$ is 0.

Discounted payback period ($\mathrm{DPP}$) was determined by Equation (9).

$$\mathrm{NPV}={{\displaystyle \sum }}_{\mathrm{i}=1}^{\mathrm{DPP}}\frac{{\mathrm{IN}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{i}}-{\mathrm{I}}_{\mathrm{i}}}{{\left(1+\mathrm{r}\right)}^{\mathrm{i}}},€$$

(9)

$\mathrm{DPP}$ is the period of time after which the initial investment is recovered.

The profitability index ($\mathrm{IP}$) was calculated using Equation (10)

$$\mathrm{IP}=\frac{\mathrm{NPV}+\mathrm{IA}}{\mathrm{IA}}$$

(10)

where $\mathrm{IA}$: the discounted investment. An investment project is economically efficient if $\mathrm{IP}$≥ 1; for $\mathrm{IP}$ < 1 the project is economically inefficient.

5. Results and Discussion

In Case A (no vacuum pump used), various compressor pressures (1.5–10 bar) were used, while all other parameters were fixed. CO₂ capture efficiency rose visibly with the increase of 1st compressor pressure (CP₁), and the power consumption value rose as well. On the other hand, when CP₁ was fixed and 1st membrane surface (SA₁) differed from 100,000 to 1,000,000 m², CO₂ capture efficiency and power consumption increased significantly depending on the raising of CO₂ captured. The 90% efficiency required for the process was obtained at 8.5 bar and 300,000 m² SA₁, while the energy consumption at this point was about 153 MW, and CO₂ purity was 49%. However, the maximum CO₂ purity achieved was 64% at 9.5 bar and 100,000 m², which is quite low. The results show increasing the membrane surface area leads to a decrease in CO₂ purity due to the other particles (e.g., N₂) that pass through the membrane with CO₂ molecules at a higher membrane surface.

Case B demonstrates the utilization of vacuum pump (VP) pressures (0.05–0.5 bar) while the other parameters are constant. The results showed an obvious increase in CO₂ capture efficiency when VP pressure decreased due to the high-pressure difference across the membrane unit. Thus, the power consumption value rose with the decrease of VP pressure. The 90% efficiency required for the process was obtained at 5.5 bar CP₁, 200,000 m² membrane SA₁, and 0.15 bar VP pressure. At the same parameters, the CO₂ purity was 68% and power consumption was about 145 MW. Moreover, the highest CO₂ purity value was achieved at 2 bar CP₁, 0.05 bar VP pressure, and 100,000 m², which is 84%.

Since CO₂ capture efficiency requires a large surface area and CO₂ purity needs a low surface area to be high, 2-stages of membrane units have been recommended to manipulate and increase both the values of CO₂ capture efficiency and CO₂ purity (case C).

In case C, different parameters of CP₁ (2–10 bar) were used where all other parameters were constant. Consequently, the CO₂ capture efficiency increased obviously with increasing CP₁, and the power consumption rose as well. The 2nd compressor pressure (CP₂) rose from 2 to 10 bar while fixing all other components, influenced and increased the 2nd membrane efficiency, and also impacted the CO₂ purity. The moment when SA₁ increased and all other parameters were constant, CO₂ capture efficiency and power consumption rose excessively due to the CO₂ captured rising. On the other hand, the leading factor that influenced the CO₂ purity was the 2nd membrane surface (SA₂), which increased from 5000 to 100,000 m², where CO₂ purity decreased constantly with the increase of the surface area. The efficiency and CO₂ purity required for the process (90%, and 95%, respectively) were achieved at 8 bar CP₁, 4 bar CP₂, 600,000 m² SA₁, and 40,000 m² of SA₂. Moreover, the energy consumed in this case was around 189 MW, which is almost 57% of the total output of energy (330 MW). As shown in Figure 7, the flue gas exits from the second membrane has to be sent back to the mixer as a recirculated flue gas in order to increase CO₂ capture efficiency.

The parameters (such as VP, SA₁, and CP₁) were selected and fixed for all the following figures only to illustrate the variations among the variants.

In Figure 8, the influence of VP and CP₁ on CO₂ capture efficiency was analyzed. In the (no vacuum) line, the maximum capture efficiency value achieved was 87% at high CP₁ (10 bar), while in 0.05 bar VP, the efficiency was much more than other efficiencies, reaching 99.9 % at 10 bar CP₁ due to the high-pressure difference around the membrane. The impact of reducing VP on process efficiency was also clearly apparent. Moreover, it was also demonstrated that CO₂ capture efficiency increases when CP₁ increases successively because of the high CO₂ content passed via the membrane unit.

Figure 9 presents how the CO₂ purity differs based on the compressor and vacuum pump pressure. The necessity of utilizing a vacuum pump is clear to increase CO₂ purity, where the highest amount achieved in no vacuum case is 62%. This value is poor compared with CO₂ purity after using different vacuum pump pressures. On the other hand, increasing compressor pressure value leads to high CO₂ purity till a specific value of the pressure where other molecules (such as N₂) will pass through the membrane decreasing CO₂ purity.

The impact of increasing vacuum pump pressure on CO₂ capture efficiency at various CP₁ is demonstrated below in Figure 10. The efficiency lines go down decreasingly at different CP₁, as a result of the reduction of the pressure difference across the membrane. For example, CO₂ capture efficiency at the point where CP₁ and VP₁ are 10, 0.5 bar, respectively, is higher than what is at 5.5 bar CP₁ and VP₁ of 0.05 bar due to the high pressure difference.

In Figure 11, the CO₂ capture efficiency rises directly when the membrane surface increases in various CP₁ values. At 10 bar CP₁, it is evident that the efficiency over different membrane surface areas is almost steady at 100% due to the flow that is almost fully captured and being passed through the membrane.

Figure 12 demonstrates the influence of the membrane surface at different compressor pressures. The power consumption increases continuously as the membrane surface increases at any CP₁ (1.5–10 bar); it is also noticeable that the power consumption columns are taller when the CP₁ rises due to the high gas flow that passed through the membrane unit. Typically, the power consumption required relies basically on the compressor and vacuum pump energy, and also on the power needed to compress the CO₂ stream at 70 bar which increases constantly with the rise of the membrane surface.

In Figure 13, the impact of the SA₁ with various CP₁ on CO₂ purity has been examined, where the purity reduced slightly at 1.5 bar because of the low compressor pressure value utilized, while at 5 bar the CO₂ purity level was favorable in 100,000 m² of SA₁, almost 70% mole. As discussed before in the results section (case A), when the membrane surface increases the CO₂ purity decreases constantly.

Figure 14 below shows the effect of CP₂ on CO₂ capture efficiency in different CP₁. It is noticeable that CO₂ capture efficiency increases crucially when the CP₁ rises, reaching almost 100 % at 10 bar and CP₂ at 4 bar. The figure also demonstrates that CP₂ has a low impact on CO₂ capture efficiency due to its location after the 1st membrane unit (see Figure 7). The explication of that tiny decrease in CO₂ capture efficiency regarding CP₂ is that the increase of CP₂ leads to a rise in the 2nd membrane CO₂ efficiency which drives a decrease in the recirculated flow to the mixer, finally reducing the flow rate entering the 1st membrane.

Figure 15 represents the influence of different SA₁ on CO₂ capture efficiency. The CO₂ capture efficiency line goes up obviously with the increase of SA₁ because of the high stream flow passed via the membrane at a higher membrane surface, obtaining almost 95% at 600,000 m².

In Figure 16, the impact of CP₁ on power consumption at different CP₂ has been shown. The main factor that influences the total power consumption rate is CP₁ because of the recirculated flue gas that fuses with the primary flue stream to generate a high flow rate that boosts the power needed to pressure the flow at the 1st compressor. Since the high CP₂ pressures increase the 2nd membrane efficiency, the low recirculated flow decreases constantly, thus providing a lower flow rate entering the 1st compressor, which demonstrates why power consumption reduces at high CP₂.

Figure 17 shows the effect of the 2nd membrane surface area on CO₂ purity at different CP₂. All CO₂ purity lines go down permanently with the increase of membrane surface due to the transit of other components that pass through larger membrane surface (like N₂) producing low CO₂ purity. Furthermore, the CP₂ affects CO₂ purity, where higher CP₂ leads to less CO₂ purity.

Figure 18 below represents the CO₂ purity difference between the 1st and 2nd stages regarding various CP₂. Firstly, it is distinguished that the CO₂ purity of the 2nd membrane is much higher than that of the first because of the low surface area (20,000 m²) used in the 2nd membrane, which shows the significance of utilizing 2-stages of membrane unit. Typically, CO₂ purity is reduced at high pressures. The influence of high CP₂ on the 2nd membrane is more than in the first due to its location (see Figure 7).

On the other hand, Figure 19 demonstrates the variation of CO₂ purity of single-stage (with and without vacuum pump) and 2-stages membrane based on 1st compressor pressure. Low CO₂ purity is remarkable in the 1-stage of the membrane (with and without a vacuum pump) due to the high membrane surface used (400,000 m²). Integrated 2-stages of membrane increase the CO₂ purity with the rise of CP₁ reaching almost 99% at 6 bar. It is observable that all the steam flow contents which passed through the 2nd membrane are CO₂ molecules. However, the usage of the 2nd stage of the membrane is highly recommended to increase CO₂ purity.

Figure 20 exhibits how the CO₂ capture efficiency of single-stage (with and without a vacuum pump) and the 2-stage membrane is affected by 1st compressor pressure. The CO₂ capture efficiency line of single-stage with vacuum is remarkably higher than other lines, where the usage of a vacuum pump increases the pressure difference across the membrane unit, producing high CO₂ capture efficiency.

Figure 21 demonstrates the impact of 1st compressor pressure on power plant efficiency at different 1-single stage membrane surfaces. As shown, high compressor pressure increases the energy consumption required for CCS, which drives to decrease the power plant efficiency. As described before, a high membrane surface such as 700,000 m² increases CO₂ capture efficiency, therefore, generating a significant demand for energy to compress the CO₂ flow. Thus, the power plant efficiency decreases.

Figure 22 shows a comparison of power plant efficiencies regarding 1-stage (no vacuum used), 1-stage (with vacuum), and 2-stages of the membrane. As mentioned above, high compressor pressure decreases the power plant efficiency either for the 1-single stage or 2-stages of the membrane. As shown, the power plant efficiency of 2-stages is higher than the efficiency of one stage due to the lower energy demanded for the 2-stage of membrane case.

Figure 23 represents the difference in capital cost of various membrane surfaces of the 1-single stage based on different 1st compressor pressure. The capital cost lines of the surfaces increase and go up noticeably achieving 10,000 €/kWh at 10 bar of 700,000 m². High surfaces raise the capital cost and also high compressor pressure influences the cost due to the high CO₂ content passed through the membrane, which increases the power consumption required.

The influence of SA₁ on CO₂ capture efficiency of one (with and without VP) and 2-stages of the membrane is shown in Figure 24. The case where the 1-single stage of the membrane with vacuum is used has the highest values of CO₂ capture efficiency due to the high-pressure difference across the membrane unit. At 400,000 m² of SA₁, 1-stage (with vacuum) of membrane achieved almost 100% CO₂ capture efficiency. While, in the same point, CO₂ capture efficiency is less by around 6% and 13% for 1-stage (no vacuum) and 2-stages of the membrane, respectively.

Figure 25 exhibits the variation of power plant efficiency of single-stage (with and without vacuum pump) and 2-stages of membrane regarding SA₁. As noticed from the figure, at 200,000 m², the power plant efficiency for 2-stages and no vacuum case of the membrane are almost the same, which are the highest, while utilization of vacuum pump in the 1-stage reduces the power plant efficiency by around 16% because of the low power consumption needed for capture CO₂ comparing with 1-stage no vacuum case. By increasing SA₁ the power plant efficiency lines of all stages go down constantly due to the increase of CO₂ capture efficiency with SA₁ increase, which leads to high demands of energy.

Figure 26 represents the effect of SA₁ on discounted payback period (DPP) of 1-stage (with and without vacuum pump) and 2-stages of the membrane. The SA₁ of 1-stage (no vacuum) of the membrane has a low influence on DPP, which is the best case due to the low requirements of energy for CO₂ capture compared with other cases. It is observable that increasing SA₁ leads to an increase in DPP because of the high demand for power consumption.

To evaluate the technical and economic assessment for the case A, the different parameters in

Table 9. The specific parameters were chosen for the technical and economical assessment of a single-stage of membrane technology (no vacuum pump used), case (A).

1st Membrane Surface	m2	200,000 (A1)		400,000 (A2)			500,000 (A3)
1st compressor pressure	Bar	8	10	6	8	10	6	8	10
Case abbreviation	-	A12	A13	A21	A22	A23	A31	A32	A33
CO2 efficiency	%	73.14	87.43	75.85	94.16	99.23	84.14	97.82	99.9
CO2 purity	%	56.67	54.63	49.03	45.62	40.49	46.05	41.06	35.41
Power consumption	MW	133	156	126	159	185	135	169	198
CO2 captured/Membrane area	kmol/m2·h	0.019	0.023	0.010	0.012	0.013	0.008	0.010	0.010

are chosen based on the optimum results of CO₂ capture efficiency, CO₂ purity, and power consumption (Figure 27). To abbreviate the surface area and compressor pressure parameters of case A, it was considered that case A₁₂ represents 200,000 m² of 1st membrane surface and 8 bar of 1st compressor pressure, and case A₁₃ represents 200,000 m² of 1st membrane surface and 10 bar of 1st compressor pressure. For 400,000 m² SA₁ and 6 bar of CP₁, the abbreviation is A₂₁ and along with others.

Table 10. The evaluation of the CFPP system with a single-stage of membrane technology (no vacuum pump used).

Parameters	Power Plant	A12	A13	A21	A22	A23	A31	A32	A33
Fuel feedstock, (t/h)	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22
CO2 capture efficiency, (%)	n.a.	73.14	87.43	75.85	94.16	99.23	84.14	97.82	99.90
Net power generated, (MW)	330	202.5	179.3	210	177	150.5	201	166	137.5
Net power plant efficiency, (%)	45.78	31.72	28.51	32.74	28.18	24.51	31.46	26.67	22.71
Capital costs per net electrical capacity, (€/kWh)	2754	5347	6037	5265	6243	7340	5562	67,189	8112
CO2 emission factor, (kg/MWh)	741.15	324.43	171.42	281.51	80.73	12.51	193.39	32.11	1.78
CO2 captured, (kg/MWh)	n.a.	883.41	1192	884.15	1302	1613	1026	1441	1777
Power consumption of membrane plant, (kWe)	n.a.	132,990	156,140	125,660	158,550	184,980	134,900	169,430	197,950
Membrane power consumption, (kWh/tCO2)	n.a.	743.43	730.18	677.36	688.45	762.18	655.52	708.17	810.15
LCOE_tax, (€/kWh)	0.0756	0.1372	0.1404	0.1313	0.1370	0.1555	0.1306	0.1431	0.1708
SPECCAm, (MJth/kg)	n.a.	3.94	3.99	3.54	3.77	4.23	3.49	3.91	4.50
SEPCCAs, (MJel/kg)	n.a.	2.14	2.14	1.90	2.00	2.28	1.85	2.08	2.45
CO2 avoided cost (€/t)	n.a.	147.92	113.79	121.29	92.95	109.64	100.34	95.14	128.82
CO2 captured cost (€/t)	n.a.	69.77	54.37	63.05	47.16	49.54	53.57	46.82	53.61

demonstrates the evaluation and economical estimation with the analyzed solutions considering the indicators presented above of single-stage membrane technology. By fusing the single-stage of the membrane without vacuum pump utilization, the net power plant efficiency decreases by (31–50%) based on the membrane surface and compressor pressure values utilized. Generally, increasing the membrane surface or compressor station unit drives an increase in the power plant efficiency loss. As demonstrated in the paper, the electrical consumption required for a membrane system increases vastly with the increase of CP₁ in addition to membrane surface impact. The LCOE increases with the increase of membrane surface from 200,000 to 500,000 m² of (44–56%) at compressor pressure of 8 bar. The CO₂ avoided cost is higher on the 200,000 m² membrane surface than on other surfaces at the same compressor pressure due to the low power required to capture CO₂.

By presuming the CO₂ tax is 82 €/ton and retail electricity cost is 160 €/MWh and considering that all carbon certificates are sold, the assumptions for all cases are in the Tables below. The economic evaluation of the 1-single stage of the membrane (no vacuum used) was analyzed in

Table 11. The economical assessment of the CFPP system with a single-stage of membrane technology (no vacuum pump used).

Parameters	Units	A12	A13	A21	A22	A23	A31	A32	A33
NPV	[M€]	813	963	922	1087	985	1035	1073	880
VNA	[M€]	753	892	853	1006	912	958	994	815
IRR	[%]	0.16	0.17	0.17	0.18	0.17	0.18	0.18	0.16
UPP	[year]	7.60	7.18	7.34	6.93	7.18	7.08	6.99	7.48
DPP	[year]	10.16	9.34	9.65	8.87	9.33	9.14	8.97	9.91
PI	[-]	1.75	1.89	1.83	1.98	1.89	1.93	1.96	1.79

. As summarized, by increasing the membrane surface from 200,000 to 500,000 m² at the same compressor pressure, the net present value increases at 400,000 m² by around 25%, then reduces by 1% at 500,000 m² membrane surface due to the increase of energy consumption needed for the CO₂ capture process.

To evaluate the technical and economic assessment for the case B, the different parameters in

Table 12. The specific parameters were chosen for the technical and economical assessment of a single- stage of membrane technology (with a vacuum pump), case (B).

1st Membrane Surface	m2	200,000 (B1)			400,000 (B2)			500,000 (B3)
1st compressor pressure	Bar	6	8	10	6	8	10	6	8	10
Case abbreviation	-	B11	B12	B13	B21	B22	B23	B31	B32	B33
CO2 efficiency	%	88.27	96.17	98.93	98.78	99.9	100	99.70	99.99	100
CO2 purity	%	55.45	56.67	54.63	51.88	44.35	38.6	46.48	38.92	33.47
Power consumption	MW	142	171	195	173	209	242	187	228	265
CO2 captured/Membrane area	kmol/m2·h	0.023	0.025	0.026	0.013	0.013	0.013	0.010	0.010	0.010

are chosen based on the optimum results of CO₂ capture efficiency, CO₂ purity, and power consumption.

Table 13. The evaluation and economical assessment of the CFPP system with a single-stage of membrane technology (vacuum pump used).

Parameters	Power Plant	B11	B12	B13	B21	B22	B23	B31	B32	B33
Fuel feedstock, (t/h)	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22
CO2 capture efficiency, (%)	n.a.	88.27	96.17	98.93	98.78	99.90	100	99.70	99.99	100
Net power generated, (MW)	330	194	164.5	140.5	162	126	93.5	148	107.7	70
Net power plant efficiency, (%)	45.78	30.49	26.46	23.13	26.14	21.16	16.60	24.21	18.57	13.35
Capital costs per net electrical capacity, (€/kWh)	2754	5592	6581	7705	6809	8744	11,819	7521	10,360	15,932
CO2 emission factor, (kg/MWh)	741.15	148.19	56.94	18.62	18.39	1.94	0.00	4.95	0.23	0.00
CO2 captured, (kg/MWh)	n.a.	1115	1429.7	1722	1489	1934	2617	1644	2271	2493
Power consumption of membrane plant, (kWe)	n.a.	141,890	170,960	194,970	173,240	209,150	242,020	187,150	227,790	265,460
Membrane power consumption, (kWh/tCO2)	n.a.	657.22	726.82	805.77	717.06	855.98	989.52	767.48	931.43	1085.4
LCOE_tax, (€/kWh)	0.0756	0.13	0.1434	0.1650	0.1444	0.1852	0.2518	0.1583	0.22	0.34
SPECCAm, (MJth/kg)	n.a.	3.54	4.02	4.47	3.97	4.73	5.28	4.26	5.07	5.47
SEPCCAs, (MJel/kg)	n.a.	1.87	2.15	2.43	2.11	2.62	3.10	2.30	2.89	3.44
CO2 avoided cost (€/t)	n.a.	89.91	99.12	123.75	95.21	148.21	237.70	112.35	194.04	356.69
CO2 captured cost (€/t)	n.a.	47.81	47.44	51.93	46.21	56.65	67.32	50.32	63.31	75.69

summarizes the assessment with the examined solutions based on the indicators shown above at a 1-stage of membrane technology with vacuum pump usage. In all analyses, the vacuum pump pressure was assumed to be 0.25 bar. After the integration, the net power plant efficiency decreased by 42% at 200,000 m² and 8 bar, then reduced constantly, reaching mitigation of 59% of the main net power plant efficiency at 500,000 m². As explained in the paper, the high pressure difference across the membrane module guides an increase in the power plant efficiency loss. The table shows that LCOE increases by 23% with the increase of membrane surface from 200,000 to 400,000 m² and by 16% with the increase from 400,000 to 500,000 m² due to the high demands of energy CO₂ capture. In terms of CO₂ avoided cost, for the 200,000 m² membrane surface the cost is less than 400,000 and 500,000 m² by around 33% and 49%, respectively, at compressor pressure 8 bar because of the low energy required for the CO₂ capture process.

Table 14. The economic assessment of the CFPP system with a single-stage of membrane technology (vacuum pump used).

Parameters	Units	B11	B12	B13	B21	B22	B23	B31	B32	B33
NPV	[M€]	1106	1050	912	1074	794	516	968	627	306
VNA	[M€]	1024	972	844	994	736	478	896	581	283
IRR	[%]	0.19	0.18	0.17	0.18	0.16	0.13	0.17	0.14	0.11
UPP	[year]	6.84	6.97	7.32	6.96	7.71	8.72	7.25	8.30	9.76
DPP	[year]	8.70	8.93	9.60	8.92	10.38	12.64	9.46	11.65	15.5
PI	[-]	2.02	1.97	1.84	1.97	1.72	1.47	1.87	1.56	1.27

demonstrates the economic evaluation for different parameters which were examined. The net present value in the economic assessment decreases by 24% when the membrane surface increases from 200,000 to 400,000 m², and also decreases if the surface differs from 400,000 to 500,000 m² by around 21% at 8 bar CP₁. On the other hand, the DDP scale increases with the membrane surface increase reaching 15.5 years at 500,000 m² due to the high power consumption that increases with the rise of SA₁.

To estimate the technical and economic assessment for the case C, the different parameters in

Table 15. The specific parameters were chosen for the technical and economical assessment of 2-stages of membrane technology, case (C).

1st Membrane Surface	m2	200,000 (C1)			400,000 (C2)			600,000 (C3)
1st compressor pressure	Bar	6	8	10	6	8	10	6	8	10
Case abbreviation	-	C11	C12	C13	C21	C22	C23	C31	C32	C33
CO2 efficiency	%	43.21	65.78	80.57	64.52	86.08	95.56	78.00	94.85	99.26
CO2 purity	%	91.34	94.74	95.99	94.61	96.36	96.89	95.81	96.85	97.07
Power consumption	MW	102	133	160	121	162	198	140	189	233
CO2 captured/Membrane area	kmol/m2·h	0.011	0.017	0.021	0.009	0.011	0.012	0.007	0.008	0.009

are chosen based on the optimum results of CO₂ capture efficiency, CO₂ purity, and power consumption.

Table 16. The evaluation and economical assessment of the CFPP system with 2-stages of membrane technology.

Parameters	Power Plant	C11	C12	C13	C21	C22	C23	C31	C32	C33
Fuel feedstock, (t/h)	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22	92.22
CO2 capture efficiency, (%)	n.a.	43.21	65.78	80.57	64.52	86.08	95.56	78.00	94.85	99.26
Net power generated, (MW)	330	233	203	176	214	174	138	196	147	102
Net power plant efficiency, (%)	45.78	36.00	31.74	28.00	33.36	27.75	22.71	30.81	23.96	17.79
Capital costs per net electrical capacity, (€/kWh)	2754	4659	5365	6188	5176	6380	8066	5774	7717	11,085
CO2 emission factor, (kg/MWh)	741.15	595.25	413.08	270.53	404.98	195.85	78.97	274.67	85.94	17.74
CO2 captured, (kg/MWh)	n.a.	452.91	794.05	1,121.79	736.46	1 211	1 700	973.83	1 583	2 379
Power consumption of membrane plant, (MWe)	n.a.	102	133	160	121	162	198	140	189	233
Membrane power consumption, (kWh/tCO2)	n.a.	966.46	825.86	811.02	768.10	767.80	847.07	731.67	814.36	961.61
LCOE_tax, (€/kWh)	0.0756	0.14	0.14	0.15	0.14	0.15	0.18	0.14	0.17	0.24
SPECCAm, (MJth/kg)	n.a.	4.71	4.38	4.45	3.98	4.22	4.70	3.92	4.52	5.20
SEPCCAs, (MJel/kg)	n.a.	2.82	2.43	2.43	2.19	2.28	2.58	2.12	2.46	3.00
CO2 avoided cost (€/t)	n.a.	462.77	209.65	160.77	188.22	134.10	151.91	139.99	141.19	220.86
CO2 captured cost (€/t)	n.a.	149.07	86.62	67.45	85.91	60.38	59.18	67.06	58.44	67.15

shows the technical estimation of 2-stages of membrane integrated into CFPP. As demonstrated, increasing SA₁ drives the decrease of the net power plant efficiency by 14% when the surface is increased from 200,000 to 600,000 m² at 6 bar of CP₁. Integrating 2-stages of the membrane increases the LCOE by about 50% at 8 bar compressor pressure and 400,000 m² membrane surface. By considering the impact of CP₁ on CO₂ avoided cost, the values of the cost decrease with CP₁ increasing (6–10 bar) by around 66% at 200,000 m² of SA₁.

In terms of economic assessment of 2-stages of the membrane,

Table 17. The economic assessment of the CFPP system with 2-stages of membrane technology.

Parameters	Units	C11	C 12	C 13	C 21	C 22	C 23	C 31	C 32	C 33
NPV	[M€]	343	630	761	679	859	780	831	820	547
IRR	[%]	0.12	0.14	0.15	0.15	0.16	0.16	0.16	0.16	0.13
UPP	[year]	9.51	8.22	7.78	8.09	7.53	7.76	7.66	7.69	8.64
DPP	[year]	14.76	11.49	10.52	11.20	10.00	10.49	10.27	10.34	12.45
PI	[-]	1.32	1.58	1.70	1.61	1.77	1.70	1.73	1.73	1.48

summarizes it, where the net present value decreases with the increase of SA₁ from 200,000 to 600,000 m² by 59%. It is noticeable that increasing SA₁ impacts directly on the DPP, where DPP decreases from almost 15 to 11 years with the rise of SA₁ 200,000–600,000 m².

The sensitivity assessment for 1-stage, no vacuum, SA₁ of 500,000 m², CP₁ 6 bar is given (as an example). For all cases, the variation is approximately the same. The influence of the CO₂ tax and the CO₂ capture efficiency on the levelized cost of electricity is remarkable. Taking into account that CO₂ avoided cost is over 100 €/t for almost all cases, we calculated a possible 140 €/t of CO₂ tax which allows us to increase the profitability index from 1.93 to 2.44.

Different economic parameters of 1- and 2-stages of the membrane are compared in

Table 18. A comparison 1- and 2-stages of membrane regarding different parameters.

Parameters	Units	1-Stage (without Vacuum)	1-Stage (with Vacuum)	2-Stages
CO2 capture efficiency	%	94.16	96.17	94.85
CO2 purity	%	45.62	56.67	96.85
Power consumption	MW	159	171	189
LCOE_tax	€/kWh	0.1370	0.1434	0.17
CO2 avoided cost	€/t	92.95	99.12	141.19
CO2 captured cost	€/t	47.16	47.44	58.44
NPV	M€	1087	1050	820

considering the optimal CO₂ capture efficiency and purity. According to the table, increasing the membrane stages leads to an increase in CO₂ purity, but at the same time increases the power consumption and investment cost of the project, which is also due to the usage of more equipment that requires energy to function.

In order to provide a clear vision regarding modeling using the CHEMCAD process with membrane, a comparison between the current and other papers studied from technical and economical points of view has been presented in

Table 19. The comparison of the current paper and different papers regarding technical and economical parameters.

Parameters	Current Study	Study 1	Study 2	Study 3	Study 4
Fuel feedstock, [t/h]	92.22	n.a.	n.a.	n.a.	n.a.
CO2 capture efficiency, [%]	94.85	90.01	67	90	90
CO2 purity, [%]	96.85	95.67	88	95	95
CO2 permeance, [GPU]	1000	740	270	100	1000
CO2/N2 selectivity	50	135	34	43	50
1st, 2nd Compressor pressure, [bar]	8, 4	2.5	5	1	n.a.
1st, 2nd vacuum pump pressure, [bar]	No utilization	0.25	n.a.	n.a.	n.a.
Flue gas flow, [kmol/h]	40,320	9580	6690	n.a.	500 *
Net power plant efficiency, [%]	23.96	35.5	n.a.	n.a.	n.a.
Capital costs per net electrical capacity, [€/kWh]	7717	n.a.	n.a.	n.a.	n.a.
CO2 emission factor, [kg/MWh]	85.94	n.a.	n.a.	n.a.	n.a.
Power consumption of membrane plant, [MWe]	189	165	137	n.a.	145
LCOE_tax, [€/kWh]	0.17	n.a.	n.a.	n.a.	n.a.
SPECCAm, [MJth/kg]	4.52	n.a.	n.a.	n.a.	n.a.
SEPCCAs, [MJel/kg]	2.46	1.09	n.a.	n.a.	n.a.
CO2 avoided cost [€/t]	141.19	47.40	197	n.a.	n.a.
CO2 captured cost [€/t]	58.44	47.11	n.a.	72	39

* m³/s.

below.

The current paper focused on using membrane technology integrated into CFPP in the coming years (at least 5 years); in that time, regarding the improvements of pieces of equipment and performance, we believe the efficiency of equipment (such as compressors and vacuum pumps) could be enhanced. However, the difference between using a compressor or pump efficiency of 90% instead of 85% leads to 6% decrease in the energy consumption at the CO₂ capture efficiency of 90%.

Study₁ represents a paper studied by Xuezhong He and May-Britt Hägg [8], while study₂ demonstrates a research article authored by Xuezhong He; Jon Arvid Lie; Edel Sheridan; and May-Britt Häg [99], study₃ exhibits an article studied by Van der Sluus; Hendriks; and K. Blok [100], and study₄ shows a paper studied by Merkel; Lin H; Wei X; and Baker [101]. The current study shows an increase regarding the economic side compared with study₁ due to the elevated flue rate, which is almost 76% higher, which also explains why the power consumption of the present paper is larger as well. On the other hand, the capture efficiency and purity of CO₂ is more than the others because of the high CO₂ permeance and the 1st compressor pressure used.

6. Conclusions

The aim of this paper was to assess and compare 1-single stage (with and without usage of vacuum pump) and 2-stages of membrane performance, from a technical and economical point of view, integrated into a 330 MW coal-based super-critical power plant with different configurations and parameters to achieve 90% CO₂ capture efficiency and CO₂ purity of at least 95%.

The application of a 1-single stage of the membrane with and without a vacuum pump station is remarkably undesirable due to neither the low CO₂ capture efficiency and CO₂ purity when no vacuum is used nor the poor CO₂ purity when a vacuum station is harnessed. The required CO₂ capture efficiency (90%) is obtained at 8.5 bar CP₁ and 300,000 m² of SA₁ when no vacuum is utilized. At this point, CO₂ purity attained was 50%, which is extremely low. The moment a vacuum pump is utilized, CO₂ capture efficiency and CO₂ purity can be improved. A 90% CO₂ capture efficiency can be obtained at 4.5 bar CP₁ and 200,000 m² of SA₁. At the same point, the power consumption is 10% less than no vacuum station used at 90% capture efficiency. The fundamental obstacle that has been examined is the low value of CO₂ purity at any parameters utilized (84% mole max); therefore, a 2nd stage of the membrane with lower surface areas has been suggested to enhance the CO₂ purity. The results show that CO₂ purity has increased by almost 16% reaching 97% at 6 bar CP₁, 4 bar CP₂, SA₁ of 600,000 m², and SA₂ of 40,000 m². On the other hand, CO₂ capture efficiency at this point is around 93%, which is the required level. However, high CP₁ is a considerable factor in boosting the carbon capture rate. By increasing the CP₁ from 4 to 6 bar, CO₂ capture efficiency rises by around 51%, and this value has been obtained at 600,000 m² of SA₁. The SA₁ impacts the CO₂ capture rate as well, for example increasing the SA₁ from 200,000 m² to 600,000 m² drives a rise in CO₂ capture efficiency by approximately 31%. At the suitable parameters to achieve 93% and 97% of CO₂ capture efficiency and CO₂ purity, respectively, the power consumption required for the process is around 57% of the total plant energy (330 MW).

One of the main influences on the economic section is SA₁, where increasing the surface drives an increase in CO₂ captured constantly which means higher electrical energy required. Moreover, CP₁ has a senior role in affecting the economic side, where high CP₁ leads to an increase in investment cost. The LCOE is highly influenced by increasing CP₁, where raising CP₁ 2–6 bar leads to an increase of 42% at the 2-stages of the membrane and 600,000 m² SA₁. However, at the optimum point where 93% and 97% of CO₂ capture efficiency and CO₂ purity, respectively, has achieved, the economic assessments are credible from DPP, CO₂ avoided, and LCOE points of view. In applicable future growth methods, the increase of CO₂ permeability is a significant factor to consider.