Emerging technologies for coal mine methane mitigation with different integration strategies for effective recovery of CH₄

Abstract

Low-concentration CMM (coal mine methane) (CH₄ <30%) is mostly extracted during coal mining, which discharges directly into the air from mining shafts. Herein, recent advances in CH₄ recovery from coal mine gases are summarized. Among them, studies on the use of different adsorbents (activated carbon, zeolites, and metal–organic frameworks (MOFs)) and adsorption processes are extensively reviewed for use with low-concentration CMM. MOFs demonstrate superior performance due to their tunable pore geometries and customizable surface functionalization. These characteristics enable MOFs to achieve higher CH₄ selectivity than traditional activated carbon or zeolite adsorbents. Current research focuses on scaling up these advanced MOF materials and optimizing pressure swing adsorption (PSA) processes for industrial implementation. Compared to alternative separation technologies, such as membrane separation and cryogenic distillation, PSA exhibits distinct advantages for treating low-concentration CH₄ (1–30%). PSA demonstrates better performance in both product purity and recovery rates while maintaining higher technical and economic feasibility. Future research should focus on optimizing the PSA process and integrating it with other technologies. Such developments could provide economic incentives for the widespread adoption of CH₄ recovery systems in coal mining operations.

---

1 Introduction

As the world's largest emitter of anthropogenic methane (CH₄), China faces a critical challenge in its coal mining sector, which contributes to approximately one-third of the total national CH₄ emissions (Fig. 1).^1,2 This substantial CH₄ release stems from geological processes in which CH₄ accumulates in coal seams over millennia, only to be liberated during mining operations.³ Despite China's coal production continuing to rise since 2015, the government has implemented progressive policies to promote the recovery and utilization of coal mine methane (CMM). These initiatives aim to simultaneously address three strategic objectives: meeting growing energy demands, enhancing mining safety, and mitigating climate change impacts through targeted greenhouse gas reduction.⁴ Some studies have shown that anthropogenic CH₄ emissions from coal mining in China reached 55.92 Tg in 2012, equivalent to 1.17 billion tonnes of CO₂ equivalent (Bt CO₂e) when applying a 25 year global warming potential (GWP) factor.^5,6Fig. 2 describes different proportions of abandoned mine methane (AMM) and CMM emissions classified from 2010 to 2100.⁷ There are four different types of coal seam gases that have been classified based on CH₄ concentration. CH₄ concentration in coal seam gases is associated with the extraction schemes. According to its concentration, different utilization routines have been adopted. Ultra-high and mid-high (90–100% vol CH₄ and 30–90% vol CH₄) concentrations of CH₄ can be directly used as a chemical raw material or as a fuel, compressed natural gas (CNG) or Liquefied Natural Gas (LNG). For mid-low (6–30% vol CH₄), it can be used as an auxiliary co-firing fuel to generate energy.^8,9 However, it needs to be concentrated before it can be utilized efficiently. The utilization of ventilation air mine methane (VAM) (0–0.75% vol CH₄) can contribut/e to reducing greenhouse gas emissions.¹⁰

Fig. 1 Methane emission pie charts from different sources in China.

Fig. 2 (a) Methane emission from underground and surface coal mines. (b) Total coal mine methane (CMM) and abandoned mine methane (AMM) emissions from 2010 to 2100.⁴¹ This figure has been adapted/reproduced from ref. 41 with permission from Elsevier, copyright 2020.

(i) Coal bed methane (CBM) has ultra-high (90–100% vol CH₄) or mid-high concentrations of methane (30–90% vol CH₄), which is extracted via surface pre-drainage before coal mining.

(ii) Abandoned mine methane (AAM) typically contains 60–80% CH₄, which refers to CH₄ that escapes from closed or abandoned mines.

(iii) Coal mine methane (CMM) includes mid-low (6–30% vol CH₄) and low (1–6% vol CH₄) concentrations. This range can be explosive.

(iv) Ventilation air mine methane (VAM) contains 0–0.75% vol CH₄. This is vented from large-scale ventilation systems, thereby maintaining safe working conditions.

In response, China has established a comprehensive CMM recovery framework. By 2015, national efforts had achieved 8.4 billion cubic meters of CMM utilization. As a result of national policy initiatives, the Chinese government has established a comprehensive plan for coal mine methane (CMM) recovery.¹¹ And the 14th Five-Year Plan set an ambitious target of 13.2 teragrams (Tg) of CH₄ recovery by 2022.¹² To achieve this objective, mandatory regulations were enacted that require CMM drainage systems must be implemented before coal mining. This threshold balances safety concerns (avoiding explosive 5–15 vol% CH₄/O₂ mixtures) with technical feasibility, as dilute CMM streams require costly purification for industrial use.

CMM primarily consists of methane (CH₄), nitrogen (N₂), oxygen (O₂), and trace lower hydrocarbons (C₂, C₃, C₄).^13–15 The separation of CH₄ from N₂/O₂ presents a significant technical challenge due to their similar molecular diameters (Δd = 0.16 Å for CH₄/N₂) and comparable quadrupole moments. Currently, four primary methods have been used to purify low-concentration CMM, including cryogenic distillation, solvent absorption, membrane separation, and pressure swing adsorption (PSA).¹⁶ Cryogenic distillation achieves high-purity CH₄ but poses explosion risks and requires costly deoxidation. Solvent absorption suffers from poor N₂/CH₄ selectivity and solvent regeneration challenges. Especially, it is unsuitable for removing N₂ from mixed gases containing N₂ concentrations of more than 5%.¹⁷ Membrane separation represents a rapidly advancing technology for gas separation. However, in large-scale industrial applications, the reliability of membranes under long-term operation and the uncertainty of maintenance costs remain the main factors restricting their broader application.¹⁸ In recent years, adsorption-based technologies for capturing medium-to-low concentrations of CH₄ have garnered significant attention. Among these, pressure swing adsorption (PSA) has emerged as a particularly promising approach due to its environmental friendliness, operational flexibility, and process simplicity.¹⁹ For separating coal mine methane (CMM) using the PSA process, the adsorbent performance critically governs the overall system performance. As demands for CMM technology intensify, requiring more economic processes and higher efficiency, operational stability, and environmental sustainability, the development of novel adsorbents and scalable CMM processes has garnered significant attention.

Pressure swing adsorption (PSA) presents a promising alternative to traditional cryogenic distillation for CH₄/N₂ separation, offering superior efficiency and operational simplicity.^20,21 Its viability for this application has been validated experimentally.²² Effective deployment, however, necessitates developing adsorbents with exceptional separation performance and robust regeneration characteristics under low vacuum pressure conditions—a critical prerequisite for process feasibility.^23,24 Diverse adsorbents, including carbon materials,^23,25,26 zeolites,^27,28 aluminosilicates,²² and metal–organic frameworks^29,30 have been explored. MOFs have garnered significant attention due to their high CH₄/N₂ selectivity, tunable pore size, and demonstrated efficiency in PSA/VPSA processes. Their structural versatility allows precise control over pore geometry, aperture configuration, and surface chemistry through strategic modification of organic linkers, secondary building units (SBUs), or host–guest interactions.^31–34 The diffusion and adsorption of gas molecules are dependent on the microscopic properties.^35–37 Therefore, significant efforts have been made to design and synthesize MOF adsorbents with a specific structure to enhance the separation behaviors of CH₄ and N₂.^38–40

This review critically evaluates recent advances in Coal Mine Methane (CMM) capture technologies, with a focus on their technical and economic viability. CMM recovery faces significant challenges driven primarily by: (1) inherently low CH₄ concentrations (typically 5–12 vol% in ventilation air), approaching the lower explosion limit (LEL), and (2) the presence of oxygen, which complicating adsorption-based separation. Among available technologies, PSA demonstrates superior separation efficiency for oxygen-containing CMM streams compared to membrane filtration and cryogenic distillation. This advantage stems from its high CH₄ selectivity and inherent operational safety under moderate pressure conditions. The review systematically examines the developments and ongoing challenges in porous adsorbents (including activated carbons, molecular sieves, ETS molecular sieves, and MOFs) and associated PSA processes for CMM capture.

2 Adsorption separation history

The term “adsorption” was first invented in 1881. It refers to the phenomenon and process in which, when two substances of different phases come into contact, molecules of the less dense substance accumulate on the surface of the denser substance.⁴² The density of the adsorbed phase is much greater than that of a typical gas, potentially approaching that of a liquid. When the gases are a mixture, the composition of the adsorbed phase differs from that of the gas phase due to the pressure differences exerted on the solid surface by the different gas molecules.⁴³ There are three adsorption mechanisms: (1) The adsorbate is transferred from the fluid body through molecular and convective diffusion through the film or boundary layer to the outer surface of the adsorbent, which is called the external diffusion process; (2) the adsorbate is transferred from the outer surface of the adsorbent to the inner surface of the microporous structure through pore diffusion, which is called the internal diffusion process; (3) the adsorbate diffuses along the surface of the pores and is adsorbed on the pore surface.⁴⁴ The evaluation of these mechanisms is crucial for assessing CMM CH₄ recovery using various adsorbents. Consequently, various separation mechanisms are examined in the following sections.

2.1 Adsorption separation mechanism

Adsorption separation has four mechanisms: thermodynamic separation, kinetic separation, molecular sieving and trapdoor separation. The separation of CH₄ from N₂ relies on these mechanisms, contingent upon the types of adsorbents utilized, including activated carbons, zeolites, ETS, and MOFs. Adsorption separation occurs through four different mechanisms in Fig. 3.

Fig. 3 Different separation mechanisms for adsorbent materials for CH₄/N₂ separation.

2.1.1 Kinetic separation. Kinetic separation is achieved based on the difference in diffusion rates of different gas components on the adsorbent. Nitrogen production by air separation on carbon molecular sieve is a typical kinetic separation effect. Also, natural clinoptilolite and magnesium ion-exchanged clinoptilolite exhibit significant kinetic effects in N₂/CH₄ separation, positioning them as potential selective adsorbents for N₂.⁴⁵

2.1.2 Sieving separation. The sieving separation technique relies on the physical size and shape of molecules to separate them. It is mainly applicable to the sieving effect of zeolites and molecular sieves. Typical examples of gas separation using steric effects include 3A zeolite for drying and 5A zeolite for separating carbon dioxide from nitrogen and argon.⁴⁶ For the separation of CH₄ and N₂ molecules, molecular sieve ETS-4 and chabazite CHA type also show similar steric effects, which can selectively allow N₂ molecules to pass through while rejecting CH₄ molecules.⁴⁷

2.1.3 Equilibrium separation. For the equilibrium adsorption separation process, the design/selection basis of the adsorbent are the intrinsic properties of the target molecule, including polarizability, permanent dipole moment, and quadrupole moment. Activated carbon, silica gel, and metal–organic frameworks (MOFs) function as adsorbents by selectively adsorbing target molecules through equilibrium separation effects. This indicates that polar or non-polar molecules with high quadrupole moments are preferentially adsorbed by adsorbents with high electric field gradients.⁴⁸

2.1.4 Trapdoor separation. The trapdoor mechanism is based on the flexible pore-blocking of cations (e.g. strong quadrupole moment), which can suddenly open to permit the entry of guest molecules. This mechanism enhances the separation efficiency for gases with the same molecular diameter, such as nitrogen (N₂) and methane (CH₄). The smaller pressure to trigger gas, like carbon dioxide (CO₂), which has a high quadrupole moment that pushes the gates open, allows CH₄ molecules to enter the pores compared to smaller N₂ molecules. As a result, CH₄ selectivity can be enhanced for various adsorbents, including MOFs and zeolites.

By taking advantage of the differences in adsorption capacity, adsorption rate, and adsorption force for different materials, CH₄ separation from N₂ can be achieved. We have analyzed various materials utilizing distinct mechanisms to achieve enhanced adsorption capacity and selectivity while maintaining low adsorption heat. Various PSA systems were assessed based on their adsorption capacity, employing different processes to enhance purity and recovery. The details of the materials are described below.

2.2 Adsorbent development for CH₄/N₂ separation

2.2.1 Activated carbon materials. Carbon-based materials, including graphite nanofibers, carbon fibers, carbon nanotubes, activated carbon (AC), and carbon molecular sieves, have garnered significant attention in gas separation due to their structural versatility.⁴⁹ Among them, activated carbon stands out as a dominant adsorbent owing to its unique pore architecture, exceptional specific surface area (SSA > 1000 m² g⁻¹), high adsorption capacity, chemical inertness, mechanical robustness, and regenerability.⁵⁰ These properties enable its widespread deployment in industrial gas separation processes, such as CH₄ recovery from coal mine methane (CMM) streams. However, activated carbon materials differ significantly in performance due to variations in raw materials and preparation methods. The operational efficiency of AC is highly dependent on synthesis parameters, including precursor type (e.g., coconut shell, coal, or biomass), activation method (physical vs. chemical), and post-treatment conditions. Consequently, significant differences occur in the pore size distribution (PSD), with most ACs exhibiting broad PSD ranges (0.5–50 nm) due to limited control over PSD during pyrolysis or activation.⁵¹ Despite being a non-polar molecule with a negligible dipole moment, CH₄ exhibits superior adsorption on AC surfaces due to enhanced polarization effects under high surface area microenvironments. This results in higher adsorption capacity and CH₄/N₂ separation selectivity.⁵² The preparation of activated carbons from biomass feedstocks has been extensively investigated through various technological approaches, including physical and chemical activation, physicochemical methods, and microwave-assisted processes (Fig. 4). Adsorption isotherms of N₂, O₂, and CH₄ were collected on five carbon materials: four commercials activated carbon (Centaur, BPL, F30/470, WS42) and one carbon molecular sieve (CMS1) within pressure ranges of 0–4 MPa. Notably, the Centaur-activated carbon demonstrated exceptional adsorption capacities, achieving 5.119 mmol g⁻¹ for CH₄ and 2.968 mmol g⁻¹ for N₂ at 3.7 MPa.⁵³ Significant enhancements in separation performance have been realized through material optimization strategies. Extensive efforts have focused on modulating the inherent separation capabilities of activated carbons, particularly for CH₄/N₂ separation. Among them, there was a substantial improvement in selectivity from 2.42 to 3.54 at 298 K. The introduction of molybdenum oxide (MoO₂) dopants proved to be particularly effective, with an 18.2 wt% MoO₂/AC composite adsorbent achieving a remarkable CH₄/N₂ selectivity of 4.25 at ambient temperature (25 °C).⁵⁴ These findings underscore the potential of dopant-enhanced activated carbons for advanced gas separation applications.

Fig. 4 Different methods for preparing activated carbons for CMM capture: (a) physical activation, (b) chemical treatment, (c) physical activation (heat + chemical) and (d) microwave-assisted radiation methods.

Recent advancements in CH₄ enrichment through coconut shell-based activated carbon modification have been reported by Zhang et al.,⁵⁵ who systematically investigated pore structure engineering via chemical activation. Their comprehensive study established a quantitative relationship between micropore characteristics and separation efficiency. The optimized coconut shell-based carbon material (GAC-3) demonstrated a remarkable equilibrium selectivity of 3.95 alongside a CH₄ adsorption capacity of 3.28 mmol g⁻¹ at 298 K, confirming the dominant role of micropore networks in separation performance. In parallel developments, Liu et al.⁵⁶ engineered KOH-activated carbons from low-rank bituminous coal (DF-AC and SM-AC, Fig. 5a–e) with their adsorption behaviors systematically characterized through isotherm analysis (Fig. 5f and g). The study revealed distinct adsorption equilibria for CH₄ and N₂, with a rapid increase in separation factors from 273 to 373 K. Notably, DF-AC achieved maximum selectivity (S = 2.82) under low-pressure conditions (Fig. 5h and i). KOH-activated carbons derived from low-rank bituminous coal exhibited ultra-high specific surface areas (∼3000 m² g⁻¹). An exceptional CH₄ adsorption capacity (13 mmol g⁻¹ at 6 MPa) and CH₄/N₂ (30/70 vol%) IAST selectivity (4.8) were demonstrated. Zhao et al.⁵⁷ improved the CH₄/N₂ separation of coal-based porous carbons. Subsequent investigations using Shanxi coal feedstock also revealed that KOH activation via the hydrothermal method generates microporous structures with enhanced surface area and reduced surface functional groups. The CH₄ adsorption capacity and CH₄/N₂ selectivity are as high as 1.72 mmol g⁻¹ and 5.5, respectively. Micropores from 0.5–1.0 nm play a vital role in the CH₄/N₂ separation. Innovative synthetic strategies were further developed by Li et al.,⁵⁸ who introduced an in situ growth method for fabricating carbon nanofibers (CNFs) with selective ultra-microporous coatings (Fig. 6a). The approach leverages controlled interactions between bridging metal ions (Ni²⁺, Co²⁺) and polymer coatings (PC) to modulate porosity. As evidenced by SEM analysis (Fig. 6b–d), all samples exhibited uniform fiber morphology, with Ni²⁺ and Co²⁺ (ionic radii 0.63–0.67 Å) demonstrating superior interfacial bonding compared to Mn²⁺ (0.75 Å), as illustrated in Fig. 6e and f. The enhanced interaction correlates with improved structural regularity in the PF-M-1 composites. The optimized CNFs exhibited remarkable gas separation performance with a rapid CH₄ diffusion rate exceeding those of conventional materials by more than 2-fold, coupled with a CH₄/N₂ (30/70 vol%) IAST selectivity of 6.8 under a gas mixture. Quantitative analysis confirmed substantial CH₄ uptake (0.97 mmol g⁻¹) with preferential adsorption kinetics (Fig. 6h–k). Morphological analysis further revealed shortened diffusion pathways and enhanced site accessibility (Fig. 6l and m), which synergistically improve dynamic separation efficiency through accelerated mass transfer.

Fig. 5 KOH-activated carbon (DF-AC) was synthesized through a controlled activation process for coal mine methane capture, involving the following key steps: (a) coal pulverization, (b) carbonization followed by chemical activation, (c) acid washing to remove impurities, and (d and e) vacuum drying to obtain the final product. (f and g) Comprehensive characterization included temperature-programmed CH₄ and N₂ adsorption isotherms. (h and i) Adsorption enthalpy analysis demonstrating the selective CH₄ capture properties of DF-AC under varying conditions.⁵⁶ This figure has been adapted/reproduced from ref. 56 with permission from Elsevier, copyright 2019.

Fig. 6 (a) Schematic of the facile in situ growth process for polymer fiber fabrication. (b–d) Field-emission SEM micrographs revealing morphological evolution of Ni-, Co-, and Mn-based PFs, demonstrating uniform fiber diameters and hierarchical microstructures. (e and f) Zeta potential profiles and pH titration curves elucidating the synthesis mechanism through metal–polymer interactions. (g) Fiber thickness of different metals (Mn, Co, and Ni). (h) Comparative nitrogen adsorption isotherms (77 K) of all PCF samples. (i and j) Adsorption uptakes of CH₄ and N₂ on these samples at 298 K. (k) IAST selectivity of CH₄/N₂ under 30 : 70 (v/v) CH₄/N₂ mixtures. (l and m) Dynamic breakthrough curves for PCF-Co-0.5 and PCF-Co-0.25, with thickness-dependent separation efficiency.⁵⁸ This figure has been adapted/reproduced from ref. 58 with permission from Elsevier, copyright 2023.

2.2.2 Carbon molecular sieve. Due to the lower kinetic diameters of O₂ and N₂ compared to CH₄, carbon molecular sieves (CMSs) with more uniform and smaller pore sizes than activated carbon (AC) preferentially adsorb O₂ and N₂ over CH₄.⁵⁹ CMS exhibits a substantial adsorption capacity for CH₄ in the CH₄ and N₂ systems, with equilibrium separation coefficients ranging from 1.35 to 4.74. Meanwhile, N₂ demonstrates fast diffusion rates and preferential adsorption kinetics.^60,61 The potential of utilizing the kinetic effect of CMS for CH₄/N₂ separation has been validated. Despite exhibiting a 133-fold superior adsorption rate constant for N₂ compared to CH₄ on CMS, the enhanced equilibrium adsorption capacity of CH₄ adversely affects the total kinetic selectivity of 1.9 for the N₂/CH₄.⁶² Therefore, enhancing the separation efficacy of CMSs relies on suppressing CH₄ equilibrium adsorption and augmenting the kinetic separation impact, both of which are achievable via the regulation of porosity characteristics. In the kinetic separation, micropores with sizes of 0.4–0.7 nm predominantly absorb N₂ rather than CH₄, the principal component.⁶³ CMS with heteroatom doping was examined, obtaining a CH₄/N₂ selectivity of up to 6.50.⁶⁴ As compared to zeolites and MOFs, CMS shows low selctivity and adorption capacity. Recently, Yang et al.²⁵ modified CMSs made from coal by using organic substances such as polyethylene imine, lauryl sodium sulfate, and tetracosane based on the plasma modification technique. Additionally, CMS samples were altered in CH₄ and N₂ environments, using a low-temperature plasma treatment, and named as CMS-P-C and CMS-P-N, respectively (Fig. 7a). Fig. 7b and c shows no discernible change in the textural characteristics of the CMS samples after alteration. After low-temperature plasma alteration, the smallest change in micropore volume was observed. This verifies that the modification encouraged the creation of additional micropores, which caused the selective adsorption of CH₄ on the changed samples. Also, SEM images show that the samples have tiny pores on the surface (Fig. 7d–g). The CMS-P-N sample was the most effective, with corresponding saturation adsorption uptakes (q_m) of 5.56 mmol g⁻¹ for N₂ and 6.76 mmol g⁻¹ for CH₄. For CMS-P-N, its CH₄/N₂ separation coefficient was 3.32. Additionally, the extended breakthrough time demonstrated the high CH₄ adsorption for the plasma treatment samples CMS-P-N in a N₂ atmosphere (Fig. 7h and i). Although significant progress has been achieved on activated carbon materials, they still exhibit fundamental limitations in large-scale CH₄ recovery operations, particularly in achieving the requisite separation selectivity for industrial applications. It is necessary to develop advanced adsorbent systems, with molecular sieves and MOFs emerging as promising candidates owing to their demonstrated potential for enhanced gas-phase separation efficiency.

Fig. 7 (a) CMS samples were altered in CH₄ and N₂ settings using a low-temperature plasma treatment, and named CMS-P-C and CMS-P-N. (b and c) Adsorbent surface area and pore size distribution. SEM images of anthracite precursor material: (d) CMS-SDS, (e) CMS-PEI, (f) CMS-P-C, and (g) CMS-P-N. (h and i) Breakthrough curves of CH₄ on samples: (a) CMS, CMS-C24, and CMS-SDS; (b) CMS-PEI, CMS-P-C, and CMS-P-N.²⁵ This figure has been adapted/reproduced from ref. 25 with permission from Elsevier, copyright 2020.

2.2.3 Zeolite molecular sieves. Zeolite molecular sieves refer to natural or artificially synthesized crystalline aluminosilicates containing alkali or alkaline earth metal oxides. Its microporous sieve-like structure is formed by connecting silica and alumina tetrahedra as basic units. Compared with activated carbon materials, the surface of zeolite molecular sieves is exposed to a large amount of anionic oxygen and isolated cations, which have a stronger surface electric field gradient.^63,65 Therefore, these frameworks suit polar molecules with permanent dipole moments, unsaturated molecules, and higher quadrupole distances. Non-polar molecules have a strong polarization effect, resulting in a preferential adsorption capacity.^66,67

Over the past decade, zeolites have become pivotal materials for CH₄/N₂ separation, with small-pore frameworks demonstrating exceptional performance through distinct mechanisms. Recently, copper-doped zeolite have achieved a high CH₄/N₂ selectivity (N₂/CH₄) of 6.8 at 298 K,²² highlighting their potential for industrial CH₄ purification. However, scalability challenges warrant systematic resolution, particularly synthesis reproducibility and cost-efficiency. Li et al.⁶⁸ synthesized K-chabazite with controlled crystallite size (∼500 nm), optimized porosity (BET surface: 68.68 m² g⁻¹; micropore volume: 0.31 cm³ g⁻¹), and superior separation performance (SN₂/CH₄ ≈ 5.0) over two different chabazite-based zeolites (NCHA and MCHA), as evidenced by CH₄/N₂ adsorption isotherms (Fig. 8a–c). Dynamic breakthrough tests (Fig. 8d–f) confirmed sustained CH₄ retention (>120 min at 1 bar) and 90% product purity, outperforming clinoptilolite and titanium silicalite ETS benchmarks. P. A. Webley et al.⁶⁹ revealed temperature-dependent selectivity inversion in K-chabazite: preferential CH₄ adsorption dominates above 293 K via thermodynamic equilibrium, while molecular sieving favors N₂ at 253 K. This tunability underscores the potential of adaptive adsorbents for dynamic separation processes. Y. Xu and Z. Liu et al.⁷⁰ demonstrated that while CH₄/N₂ separation holds industrial promise, its practical implementation remains constrained by insufficient adsorbent efficiency. Their work employed amine-functionalized Y zeolites (Fig. 9a), where ion exchange with tetramethylammonium cations (TMA⁺) in NaY frameworks selectively enhanced CH₄ adsorption capacity (0.5 mmol g⁻¹) while suppressing N₂ uptake (0.12 mmol g⁻¹) under equivalent conditions (Fig. 9b and c). In addition, IM-5@100 zeolite (IMF type) uses a simple method to control the aluminium quantity. Nano-IM-5@100 reaches adsorption equilibrium faster and performs better in mass-transfer tests. In addition, IM-5@100 exhibits strong CH₄ adsorption and high IAST selectivity of 0.79 mmol g⁻¹ and 4.7, respectively.⁷¹ Here, β-cyclodextrin was used to manufacture nano-sized ZK-5 for the first time. It is possible to reduce the crystal size of ZK-5 from micron-size (3 µm) to nano-size (50–100 nm). The nano-ZK-5 has an adsorption capacity and IAST selectivity of 1.34 mmol g⁻¹ and 4.2 (298 K), respectively.⁷² This selectivity enhancement occurred concomitantly with reduced specific surface areas, a phenomenon attributable to preferential CH₄-framework interactions overriding textural effects. Critical characterization validates the modification strategy: (i) SEM imaging confirmed the unaltered crystal morphology between TMAY and pristine NaY (Fig. 9d and e); (ii) adsorption calorimetry and breakthrough experiments substantiated TMAY's superior separation performance versus NaY, as evidenced by enhanced CH₄ affinity and prolonged retention dynamics (Fig. 9f and g). These results confirm that targeted cation exchange effectively optimizes zeolitic separation potential without structural compromise. Lederman et al.⁷³ examined the adsorption isotherms for CH₄ and N₂ on the molecular sieve at 2 bar and 295 K. The selectivity rises with decreasing temperatures from 1.8 to 2. The adsorption kinetics and equilibrium on the molecular sieve align well with the findings obtained from gravimetric methods. Two different zeolites 13X while NaX exhibited separation selectivity values of 2.0 and 1.5, respectively. But eight-membered ring molecular sieves, such as SAPO-34, DDR, ERI, CHA, and LTA with small micropore windows of approximately 4 Å, showed better separation performance.^74,75 These sieves have been extensively studied because their unique pore and channel structures are suitable for adsorption and separation applications. Studies have shown that SAPO-34 exhibits an average CH₄/N₂ separation factor of around 3.0.⁷⁶ Notwithstanding these advances, the modest adsorption capacities (≤0.5 mmol g⁻¹) underscore fundamental material limitations, directing attention toward metal–organic frameworks (MOFs) as next-generation adsorbents requiring urgent development.

Fig. 8 (a) CH₄ and N₂ sorption isotherms at 298 K and 1 bar. (b) CH₄/N₂ selectivities of NCHA and MCHA. (c) Selectivities of commercial zeolites compared to MCHA and NCHA. (d) CH₄ and N₂ kinetic adsorption patterns on MCHA and NCHA. (e) Breakthrough curve of CH₄/N₂ (50/50, V/V) mixtures on different zeolites. (f) CH₄ gas purity of MCHA and NCHA zeolites.⁶⁸ This figure has been adapted/reproduced from ref. 68 with permission from Wiley, copyright 2022.

Fig. 9 (a) Crystal images of NaY ion exchange to TAMY zeolite. (b and c) The adsorption capacity of CH₄ and N₂ on different zeolites such as NaY, TMAY, ChY, and TEAY. SEM images show no difference for (d) NaY and (e) TMAY. (f and g) Isosteric adsorption heat of CH₄ on different samples. (g) Comparison of breakthrough for the working capacity of zeolite samples.⁷⁰ This figure has been adapted/reproduced from ref. 70 with permission from Elsevier, copyright 2021.

2.2.4 Titanium silicate molecular sieve ETS-4. Titanosilicate ETS-4 (Engelhard Titanosilicate-4) features a three-dimensional framework comprising interconnected SiO₄ tetrahedra and TiO₆ octahedra,⁷⁷ forming one-dimensional channels with 8-membered ring apertures (3–4 Å). This material exhibits inherent thermal instability, where Na-ETS-4 undergoes structural collapse via chain dissociation along channel axes upon hydration.^78,79 Cation exchange with Sr²⁺ or Ba²⁺ significantly enhances thermal stability, with Sr-ETS-4 maintaining framework integrity up to 350 °C. Controlled dehydration at 473 K induces lattice contraction in Sr-ETS-4, reducing the effective pore size from 4.0 Å to 3.7 Å. The precisely tuned aperture—intermediate between N₂ (3.64 Å) and CH₄ (3.8 Å) kinetic diameters—activates a molecular gate mechanism, achieving N₂/CH₄ kinetic selectivity of 26 at 298 K through steric exclusion of CH₄.^80,81 Parallel studies on Ba-ETS-4 demonstrate temperature-dependent dehydration kinetics directly governing adsorption equilibria, with optimal performance achieved at 270 °C treatment. Engelhard's prototype system utilizing ETS-4 adsorbents successfully reduced nitrogen content in CH₄/N₂ mixtures from 15–18% to 3–5%, attaining 90% CH₄ recovery.⁸²Fig. 10a–c showed different ETS-4 against DDR and silicalite-1 frameworks: silicalite-1 demonstrates superior CH₄/N₂ adsorption capacity (0.6–1.0 mmol g⁻¹, Fig. 10d and e) and dynamic separation performance (Fig. 10f and g), though ETS-4 maintains advantages in selectivity tuning.⁸³

Fig. 10 (a–c) DDR: 8 ring, silicalite-1: 10 ring, and beta: 12 ring show primary zeolite channels. (d and e) The adsorption isotherm of DDR and silicalite-1 for CH₄ and N₂ is illustrated. (f and g) The breakthrough of silicalite-1 on the CH₄/N₂ mixture also shows the selectivity on all three ETS-4.⁸³ This figure has been adapted/reproduced from ref. 83 with permission from the American Chemical Society, copyright 2013.

2.2.5 Metal–organic framework (MOF). Metal–organic frameworks (MOFs) offer distinct advantages over conventional adsorbents (activated carbon, molecular sieves, and zeolites), including ultralow density, tunable pore size, exceptionally large surface areas (>6000 m² g⁻¹), and programmable surface morphologies. These crystalline porous materials are constructed via coordination bonds between metal ions/clusters and multifunctional organic linkers. The MOF research paradigm has progressively transitioned from novel material discovery to targeted structural modifications and application-driven performance optimization.^84–86 Numerous MOF architectures demonstrate exceptional CH₄/N₂ separation performance through tailored host–guest interactions, as systematically compared in Table 1. Representative systems include: zirconium-based MOF-801-Zr–FA,⁸⁷ aluminum-carboxylate CAU-21-BPDC,⁸⁸ nickel-coordinated Ni-INA⁸⁹ and Ni–MA–BPY,⁹⁰ zinc-imidazolate ZIF-94,⁹¹ copper-formate frameworks (Cu-MOF,⁹² ATC–Cu⁹³), and specialized architectures such as ZSTU-1,⁹⁴ Al-CDC,⁹⁵ and [Ni₃(HCOO)₆].⁹⁶ Each system leverages unique pore geometries and surface chemistries to achieve competitive adsorption selectivity and capacity.

Table 1 Adsorption capacities and selectivities of MOF adsorbents for methane (CH₄) and nitrogen (N₂) separation

MOF adsorbent	Adsorption capacity (mmol g^−1)	STP (298 K and 1 bar)		Selectivity	Ref.
		Q ^0st (kJ mol^−1) CH4	Q ^0st (kJ mol^−1) N2
MOF-801-Zr–FA	1.30	25.3	20.4	10.5	87
CAU-21-BPDC	0.99	32.4	29.9	11.9	88
Ni-INA	1.82	28.0	17.0	15.8	89
ZSTU-1	1.37	25.9	12.0	12	94
Ni-Qc-5	1.26	31.5	22.5	7.7	104
Al-CDC	1.43	27.5	18.6	13.1	95
Co3(C4O4)2(OH)2	0.40	25.1	18.1	8.5–12	105
Cu-MOF	0.63	20.0	15.0	12	92
SBMOF-1	0.92	23.2	16.1	11.5	106
ATC–Cu	2.90	26.8	16.0	8.9	93
MOF-890	0.45	26.0	22.0	—	107
Cu(INA)2	0.8	22.5	16.5	8.4	108
ZIF-94	1.50	23.9	20.0	—	91
Ni–MA–BPY	1.01	23.5	19.6	7.2	90
Co–MA–BPY	0.91	22.8	18.4	7.4	90
Zn2(5-aip)2(bpy)	0.34	25.5	13.3	7.1	30
Ni(OAc)2L	1.151	26.7	20.0	—	109
Cu(hfipbb)(H2hfipbb)0.5	0.46	24.0	20.0	—	110
[Ni3(HCOO)6]	0.79	24.8	19.3	6.3	96

---

Early MOF syntheses employed simple ligands to construct prototype frameworks. Cu–BTC (HKUST-1) forms via Cu²⁺ coordination with benzene-1,3,5-tricarboxylate (BTC), yielding a 3D network with 9 Å cubic pores interconnected by 4.6 Å triangular channels.^97,98 Activation exposes unsaturated Cu²⁺ sites that preferentially adsorb polar molecules (CO) and quadrupolar species (CO₂) through electrostatic interactions. Al-BDC (MIL-53-Al) adopts a flexible octahedral framework with Al³⁺-terephthalate coordination, featuring 8.5 Å × 8.5 Å 1D channels.^99,100 This framework exhibits a “breathing effect”, reversible pore contraction/expansion during gas adsorption, achieving a CH₄/N₂ selectivity of 2.7 through steric discrimination.¹⁰¹ Subsequent studies on Basolite® A100 revealed temperature-dependent CH₄/N₂ separation factors (3.4–4.4), with adsorption strength hierarchy: CO₂ > CH₄ > N₂ > O₂.¹⁰² Despite structural diversity, ZIFs generally exhibit moderate CH₄/N₂ selectivity due to small window apertures and large pore volumes. At 303 K, IAST selectivity values for ZIF-8, ZIF-69, and ZIF-76 are 2.8, 3.08, and 1.19, respectively.¹⁰³ ZIF-8 demonstrates exceptional chemical stability and achieves a separation factor of 3.4 at 298 K. The SOD-type ZIF series (Fig. 11a) combines minimal surface area with high pore volume; namely, ZIF-8, ZIF-90, SIM-1 (ZIF-94), and ZIF-93. Among these, SIM-1 (ZIF-94) exhibits superior CH₄ uptake (1.5 mmol g⁻¹vs. N₂: 0.28 mmol g⁻¹; Fig. 11b and c), attributed to its optimized SOD cage geometry (0.84 nm) generating strong adsorption potentials. Comparative working capacity analyses (Fig. 11d–g) confirm the dominance of ZIF-94 over other ZIFs (e.g., ZIF-93, RHO-type), with IAST selectivity reaching 7.0 for CH₄/N₂ mixtures (30 : 70 and 50 : 50 v/v; Fig. 11h and i).

Fig. 11 (a) Crystal structures of ZIF-8, ZIF-90, SIM-1 (ZIF-94), and ZIF-93. (b and c) Adsorption isotherms for CH₄ and N₂ on all samples at 298 K. Breakthrough curves for the experimental column and the simulated system for a 30%/70% mixture at STP with a 5 mL min⁻¹ flow rate for (d) ZIF-8, (e) ZIF-90, (f) SIM-1 (ZIF-94), and (g) ZIF-93. (h and i) CH₄/N₂ IAST selectivities for ratios of 30%/70% and 50%/50%.⁹¹ This figure has been adapted/reproduced from ref. 91 with permission from Elsevier, copyright 2020.

MOF design strategies focus on engineering pore geometry and polar adsorption sites can enhance the separation performance and selectivity. Wang et al.¹⁰⁸ systematically compared four ultra-microporous MOFs (Fig. 12a–d) involving (a) [Ni₃(HCOO)₆] MOF, (b) [Cu(INA)₂] MOF, (c) Al-BDC MOF, and (d) Ni-MOF-74 MOF. [Ni₃(HCOO)₆] and [Cu(INA)₂] own weak/apolar frameworks lacking functional groups, where selectivity arises from pore confinement effects. The low-polarity sites of Al-BDC are derived from the terephthalate linker electronegativity. Ni-MOF-74 exhibits highly polar sites via coordinatively unsaturated metal sites (CMS), enabling dipole-induced CH₄ interactions. Notably, [Cu(INA)₂] achieves CH₄/N₂ selectivity > 7 through synergistic pore size optimization (3.8 Å) and CH₄ quadrupole-CMS electrostatic interactions (Fig. 12e–i). This framework demonstrates superior CH₄ capacity (1.1 mmol g⁻¹ at 1 bar) versus comparators, attributed to its balanced surface area (1250 m² g⁻¹) and micropore volume (0.45 cm³ g⁻¹). Deng et al.¹¹⁰ developed ultra-microporous [Cu(hfipbb)(H₂hfipbb)_0.5] via microwave-assisted synthesis, achieving a CH₄/N₂ selectivity of 6.9 through precise pore geometry control (3.7 Å windows). Zhou et al.¹¹¹ designed a pyrene-based 3D MOF (Rod-8) with an interlayer spacing of 4.35 Å, which combines molecular sieving and π-complexation effects to reach selectivity of 9.0 and CH₄ capacity of 0.77 mmol g⁻¹.

Fig. 12 Crystal structures of MOF samples: (a) [Ni₃(HCOO)₆] MOF; (b) [Cu(INA)₂] MOF; (c) Al-BDC MOF and (d) Ni-MOF-74 MOF. (e) The adsorption isotherms of all samples at 77 K for surface area and pore size distribution were calculated using the NLDFT model. Gas adsorption experiments at three different temperatures for CH₄ and N₂ (f) [Ni₃(HCOO)₆] MOF; (g) [Cu(INA)₂] MOF; (h) Al-BDC MOF and (i) Ni-MOF-74 MOF.¹⁰⁸ This figure has been adapted/reproduced from ref. 108 with permission from the Royal Society of Chemistry, copyright 2016.

The near-identical kinetic diameters of CH₄ (3.8 Å) and N₂ (3.64 Å) necessitate ultramicroporous MOFs (<4 Å) with tailored adsorption sites for effective separation. Zhou et al.¹¹² developed a Ni-based MOF featuring open metal sites (OMS) via 2,5-dihydroxyterephthalate ligands. Density functional theory (DFT) calculations revealed strong CH₄ binding energies (−28.6 kJ mol⁻¹) at the OMS, facilitated by 1D channels (3.9 Å) that optimize guest-OMS interactions. In parallel developments, Liu et al.⁹² synthesized a Cu–BTC variant with dual micropores (7 Å and 5 Å; Fig. 13a and b). Structural analysis shows µ₂-coordinated carboxylate oxygen atoms in paddlewheel [Cu₂(COO)₄] clusters create hydrophilic pore regions (Fig. 13c and d). CH₄ molecules form multiple C–H⋯O van der Waals contacts (bond length: 2.7–3.1 Å) with these oxygen-rich sites, while N₂ experiences weaker interactions due to limited access to Cu OMS. This mechanism yields exceptional CH₄/N₂ selectivity (10–12.6) and adsorption capacity (1.8 mmol g⁻¹ at 1 bar), validated by breakthrough tests (Fig. 13e and f). Stability analyses confirm retained performance after 15 hydration/dehydration cycles (Fig. 13g and h). Further innovation emerged with the [Co₃(C₄O₄)₂(OH)₂] framework,¹⁰⁵ where hydroxide groups provide polarized adsorption sites. The material achieves CH₄/N₂ selectivities of 8.5–12.5 with 0.89 mmol per g CH₄ uptake, demonstrating how anionic oxygen centers enhance quadrupole-mediated adsorption.

Fig. 13 (a and b) Cu-MOF crystal with hydrophobic and hydrophilic pores along the b-axis and c-axis. (c and d) Cu-MOF has CH₄ and N₂ adsorption binding sites. (e and f) Single-component adsorption isotherms of Cu-MOF and IAST selectivity at 298 K. (g and h) Adsorption capability after water treatment of the sample (inset: PXRD comparison) and cyclic regeneration performance.⁹² This figure has been adapted/reproduced from ref. 92 with permission from the American Chemical Society, copyright 2019.

Recent advances in MOF design for coalbed methane purification emphasize strategic integration of polar adsorption sites and pore confinement effects. The Cu₂(ATC) framework, constructed from 1,3,5,7-adamantane tetracarboxylate ligands, features rectangular channels (4.8 × 5.2 Å) lined with alkyl groups and coordinatively unsaturated Cu²⁺ sites. This architecture achieves exceptional CH₄/N₂ selectivity (9.7) and capacity (2.90 mmol g⁻¹ at 1 bar), with density functional theory (DFT) calculations confirming preferential CH₄ adsorption at low-polarity, aliphatic cavities via weak C–H⋯π interactions (binding energy: −24.3 kJ mol⁻¹).⁹³ Al-based CAU-21-BPDC leverages symmetric polar sites within its 6.5 Å pores to achieve a CH₄/N₂ selectivity of 11.9 and CH₄ uptake of 0.99 mmol g⁻¹ at standard temperature and pressure (STP). The enhanced CH₄ affinity originates from quadrupole-mediated interactions between CH₄ tetrahedral electron cloud and electron-deficient aromatic linkers.⁸⁸ Similarly, SBMOF-1, a Ca²⁺-based framework with 7.3 Å pores lined with low-polarity diphenyl sulfone ligands, exhibits a selectivity of 11.5 and a CH₄ capacity of 0.93 mmol g⁻¹, demonstrating the efficacy of π-surface engineering.¹⁰⁶ Al-CDC's 1D ultramicroporous channels (5.4 Å diameter) combine saturated C–H bonds and ligand trans-conformations to achieve a record CH₄/N₂ selectivity (16.7) with 1.42 mmol g⁻¹ CH₄ uptake. This performance underscores the critical role of pore geometry in amplifying dispersion forces for nonpolar gas separation.⁹⁵ The Ni(INA)₂ MOF (Fig. 14a and b) exemplifies coordination engineering, where Ni²⁺ centers adopt octahedral geometry with N/O donor atoms, forming 5.8 Å channels achieving high CH₄/N₂ selectivity¹⁵ and capacity (1.42 mmol g⁻¹ at 298 K), attributed to strong electrostatic interactions (3.16–3.47 Å) between CH₄ and pyridinic N-sites, as validated by DFT (Fig. 14c–f). Dynamic breakthrough tests confirm sustained performance (Fig. 14g and h), highlighting its industrial viability.⁸⁹

Fig. 14 (a and b) Crystal structure of Ni(INA)₂ and pore size of Ni(INA)₂ MOF. (c and d) Adsorption isotherms of Ni(INA)₂ for CH₄ and N₂ and the adsorption–desorption profile. (e and f) DFT calculations show an electrostatic interaction between CH₄ and N-pyridine. (g and h) Dynamic breakthrough and IAST selectivity as compared to all adsorbents for CH₄.⁸⁹ This figure has been adapted/reproduced from ref. 89 with permission from Wiley, copyright 2022.

Additionally, it is highly desirable to examine the effect of trace-level impurities on the adsorption performance of CMM for the promising MOF candidates with high CH₄ adsorption capacity and high selectivity. Therefore, establishing the water stability of MOFs due to high relative humidity in CMM is also important for practical applications through PSA. In this regard, several studies show that different factors influence the water stability of CH₄ capture. These factors include textural properties, ligand functionality (hydrophobic or hydrophilic), and bond strength of the metal oxide and linker. Thus, some MOFs have been shown to have excellent water resistance, including zirconium-based MOF-801-Zr–FA,⁸⁷ aluminum-carboxylate CAU-21-BPDC,⁸⁸ Al-CDC,⁹⁵ Co–MA–BPY and Ni–MA–BPY,⁹⁰ and copper-formate frameworks (Cu-MOF).⁹² Additionally, Wang et al.⁸⁷ synthesized MOF-801-M–FA (M = Zr or Hf) using a mixed-linker approach (Fig. 15a and b). MOF chemical and water stability were confirmed at different pH levels from 1 to 10 by recording the PXRD patterns (Fig. 15c). The fine-tuning of MOF-801-M–FA (M = Zr or Hf) structures improved their gas adsorption capabilities for CH₄ (2.4 mmol g⁻¹) and N₂ (0.48 mmol g⁻¹) at 298 K (Fig. 15d–g). The findings further show that MOF-801-M–FA (M = Zr or Hf) are excellent candidates for collecting CH₄ from N₂ in industry due to their high capacities, excellent chemical and water resistance, and ease of regeneration (Fig. 15h and i). Recently, Liu et al.¹¹³ prepared ultramicroporous MOF CoNi(pyz-NH₂) with a small pore size and high OMS density to effectively separate CH₄/N₂ mixtures. Compared to alternative adsorbents, it shows high adsorption capacity (1.6 mmol g⁻¹) and selectivity (16.5) at 273 K and 1 bar. In addition, CoNi(pyz-NH₂) shows high hydrothermal and chemical stabilities based on different tests with aqueous solutions at pH 12 for seven days. PXRD characterization confirmed that the crystal structure remained integrated under harsh conditions. On this basis, MOFs are stable and tolerant to trace impurities for capturing coal mine methane. A critical analysis of all three adsorbents is presented below.

Fig. 15 (a and b) MOF synthesis strategy using dual mix-linker ligands fumarate and formate. (c) Powder X-ray diffraction patterns of MOF-801-FA after immersion in aqueous solutions with different pH. Adsorption isotherm of CH₄ and N₂ on both materials (d and e) MOF-801-Zr–FA and (f and g) MOF-801-Hf–FA. (h and i) Water stability analysis based on 1–10 cycles using breakthrough curve analysis.¹¹⁴ This figure has been adapted/reproduced from ref. 114 with permission from Elsevier, copyright 2022.

2.3 Critical analysis of adsorbents for CMM capture

Based on practical applications of the PSA process, water-resistance, thermal and chemical stability should be considered for an adsorbent. The activated carbon typically shows hydrophobic nature, and water preferentially adsorbs in the micropore channels, forming clusters. Due to presence of water vapor can reduce the efficiency of CH₄ based on co-adsorption in the micropores of carbon, thus CH₄ adsorption capacity and selectivity decrease. Porous carbon possesses a wider pore size distribution than zeolites and MOFs, making them less effective for CMM. As an adsorbent, zeolites are often limited during CH₄ separation owing to their hydrophilic surface, making them inferior in terms of CH₄ adsorption capacity and selectivity in comparison to porous carbons and MOFs. MOFs have gained much attention because they have the highest CH₄ adsorption capacity and selectivity among the three types of adsorbents. Similar kinetic diameter of CH₄ and N₂, MOFs are highly suitable for controlling pore size, metal interaction, and the linker with N-group makes them favorable for CH₄ capture. Additionally, the low cost of MOFs favors their use with low-concentration CMM. A deep study should be conducted based on PSA performance and energy efficiency of these adsorbents for critical analysis.

The evaluation of these adsorbents is primarily based on separation mechanisms. Different mechanisms—such as kinetic separation, molecular sieving, and the trap-door effect—govern the selectivity of adsorbents in CH₄/N₂ separation. As summarized in Table 2, several methods (denoted as methods a–g) have been employed to calculate selectivity, taking into account factors such as adsorption capacity, isosteric heat of adsorption, and uptake ratio. Among the adsorbents evaluated, MOFs demonstrate particularly high performance, with a selectivity of approximately 12 and an uptake of 2.9 mmol g⁻¹.

Table 2 Comparison of adsorbents (activated carbon, zeolites, and MOF) for CH₄/N₂ separation based on their selectivity, adsorption capacity, uptake ratio, and isosteric adsorption heat

Adsorbents	CH4/N2 selectivity	Uptake, (mmol g^−1) (298 and 1 bar)		Uptake ratio	Q ^0st, (kJ mol^−1)		ΔQ^0st, (kJ mol^−1)	Ref.
		CH4	N2		CH4	N2
a Predicted by IAST theory. b Obtained from mixture-based adsorption measurement. c Calculated from the ratio of Henry constants. d Calculated by theoretical calculations. e Obtained from binary gas breakthrough experiments.
Carbon materials
GAC	3.95	3.28	1.6	2.05	—	—	—	55
DF-AC	7.2^a	2.6	1.2	2.16	18.25	13.5	3–4	56
PCF-Mn	6.1^a	1.34	0.45	2.9	—	—	—	58
PCF-Co	6.8^a	0.96	0.30	3.2	—	—	—	58
CGUC-0.5-6	6.7^a	0.90	0.25	3.61	—	—	—	115
PRC-850	5.7^a	1.12	0.40	2.79	23.00	18.40	4.60	26
SC-1	5.7^a	1.33	0.50	2.66				116
AC beads	5.5^a	1.10	0.30	3.67	20.60	14.60	6.00	117
Norit R1 extra	5.0^a	0.17	0.04	4.10	20.60			118
SA-1-600	5.0^a	1.30	0.40	3.23	22.00	18.00	4.00	119
sOMC	3.5^c	0.98	0.31	3.19	15.00	14.80	0.20	64
AC (F30-470 Degussa)	3.0–4.0^c	0.25			—	—	—	120
Linde 4A molecular	3.4^a	0.49	0.13	3.81	—	—	—	75
FCNS (288 K)	3.3^c	1.12	0.36	3.08	25.10	24.00	1.10	121
Maxsorb	3.3^a	0.87	0.25	3.48	—	—	—	122
	2.4^c
CMS	1.9^a	0.75	0.40	1.87	—	—	—	25
Zeolites
ZK-5	4.3^a	0.82	0.34	2.42	—	—	—	72
	5.4^c
NaY	3.9^c	1.01	0.30	3.39	—	—	—	123
Ca-SAPO-34	3.5^e	0.61	0.28	2.18	—	—	—	124
H-ZSM-5 (313 K)	3.3^c	0.71	0.26	2.74	—	—	—	125
Na-SAPO-34	2.4–3.0^e	0.60	0.28	2.13	12.20	22.00	−9.80	124
Na-BETA-25 (313 K)	2.0^c	0.54	0.22	2.47	11.20	6.80	4.40	27
Zeolite 13X	1.9^b	0.46	0.13	3.68	14.70	15.00	−0.30	28
Zeolite 5A	0.9^c	1.01	1.15	0.88	—	—	—	126
MOF materials
Al-CDC	13.1^a	1.43	0.23	6.29	27.50	18.60	8.90	95
	13.1^c
Co3(C4O4)2(OH)2	12.5^a	0.40	0.18	2.20	25.10	18.10	7.00	105
	7.6^c
SBMOF-1	11.5^a	0.92	0.18	5.16	23.20	16.10	7.10	106
	11.1^c
ATC–Cu	9.7^a	2.90	0.75	3.87	26.80	16.00	10.80	93
	5.2^c
STAM-1	11.1^a	0.63	0.11	5.93	20.00	15.00	5.00	92
	10.9^c
NKMOF-8-Me	9.0^a	1.76	0.31	5.63	28.00	18.90	9.10	127
	9.4^c
MOF-888	8.4^a	0.45	0.08	5.61	26.00	22.00	4.00	107
Cu(INA)2	8.3^c	0.80	0.12	6.60	17.50			21
MOF-891	7.8^a	1.34	0.29	4.68	22.00	21.00	1.00	107
ZIF-94	7.4^a	1.51	0.37	4.12	23.90	20.00	3.90	91
	8.0^c
Ni-Qc-5-Dia	7.4^a	1.31	0.28	4.72	19.50	14.00	5.50	104
Ni–MA–BPY	7.4^a	1.01	0.21	4.80	23.50	19.60	3.90	90
Co–MA–BPY	7.2^a	0.92	0.20	4.69	22.80	18.40	4.40
Ni(OAc)2L	7.0^a	1.15	0.47	2.46	26.70	20.00	6.70	109
MOF-890	7.0^a	1.07	0.27	4.00	23.00	19.00	4.00	107
Cu(hfipbb)(H2hfipbb)0.5	6.9^a	0.47	0.13	3.64	24.00	20.00	4.00	110
MOF-889	6.4^a	1.16	0.24	4.90	22.00	19.00	3.00	107
[Ni3(HCOO)6]	6.2^c	0.82	0.18	4.56				29, 96 and 128
(Alias: Ni–FA)	6.1^c	0.79	0.17	4.54	24.80	19.30	5.50
	6.0^a	0.80	0.17	4.59	22.20	18.00	4.20
Ni–BTC	5.1^a	1.67	0.41	4.07				129
UiO-66–Br2	5.1^a	0.72	0.20	3.67	—	—	—	130
UTSA-30a	5.0^a	0.63	0.17	3.81	—	—	—	131
Cu(OTf)2	4.8^a	0.25	0.07	3.56	19.60	16.00	3.60	132
[Cu(Me-4py-trz-ia)]	4.2^a	1.12			18.00	12.00	6.00	102
MOF-177	4.0^c	0.56	0.13	4.20	11.70			126
Ni-MOF-74	3.8^a	1.84	0.91	2.03	—	—	—	133
	3.5^c
MIL-53(Al)	3.7^a	0.74	0.22	3.32	19.00	15.90	3.10	132
Cu–BTC	3.7^a	0.90	0.35	2.58				134
Al-BDC	3.6^c	0.73	0.22	3.26	18.70			21
ZIF-69	3.0^a	0.50	0.14	3.62	—	—	—
MIL-100(Cr)	3.0^a	0.60	0.17	3.55	—	—	—	133
MIL-101(Cr) (293 K)	2.7^a	0.65	0.13	5.05	15.70	12.00	3.70	135
ZIF-8	2.5^a	0.27	0.09	3.00	17.00	10.80	6.20	91
	3.2^c	0.20	0.07	2.82	12.44	9.83	2.60	103
	2.8^d
NU-1000	2.5^a	0.48	0.20	2.46	—	—	—	136
ZIF-90	2.3^a	0.47	0.20	2.35	15.90	12.50	3.40	91
UiO-67	2.3^a	0.40	0.13	3.08				136

---

3 The evolution of different technologies for CMM

3.1 CMM separation performance through practically applicable adsorbents

Adsorption separation exploits differential equilibrium adsorption capacities of gas components on porous materials to achieve selective separation. PSA regenerates adsorbents by modulating pressure, offering rapid cycling and low energy consumption (0.5–1 kWh per kg CH₄ for 15% CH₄ enrichment). In contrast, Temperature Swing Adsorption (TSA) employs thermal desorption, suited for low-concentration impurity removal (5–20% CH₄ enrichment) at higher energy costs (2.7–12.8 kWh per kg CH₄).^137,138

The Skarstrom cycle (two-bed, two-step configuration) established continuous PSA operation,¹³⁹ enhanced separation through pressure equalization steps, laying the groundwork for Vacuum PSA (VPSA).¹⁴⁰ UOP's five-bed PSA system using activated carbon achieved 96.4% CH₄ purity (70% feed) with 85% recovery.¹⁴¹ The subsequent three-bed PSA units improved nitrogen rejection from natural gas (30% N₂ feed) via optimized cycle timing.¹⁴² Yang et al.^143,144 demonstrated 90% CH₄ purity (85% feed) using titanium silicate/mg-clinoptilolite in a six-step PSA cycle. Olajossy's activated carbon system elevated CH₄ purity from 55.2% to 98% with 91% recovery.¹⁴⁵ Delgado's simulated two-bed PSA enriched an 85% CH₄ feed to 96% purity (74% recovery, 0.048 kg CH₄/kg adsorbent).¹⁴⁶ A three-bed PSA system was simulated based on silicalite adsorbent with 50 vol% CH₄, showing high purity (96.3%) and recovery (95.6%), but lower energy consumption (0.140 kWh per kg CH₄).¹⁴⁷ Zhang's VPSA system processed 25% CH₄ feed (65% N₂, 10% O₂) to 50.4% purity with 86.3% recovery at 0.25 kWh per kg CH₄ energy consumption.¹⁴⁸ Deng's two-bed, six-step PSA using silicalite-1 (Fig. 16a) achieved 50.5% CH₄ purity (20% feed) with 84.4% recovery, demonstrating the critical role of pressure equalization in enhancing purity with low energy consumption (0.1667 kWh per kg CH₄) (Fig. 16b–f).²²

Fig. 16 (a) Cycle steps and bed sequence for the two-bed VPSA process. (b) Bed pressure profiles according to the different steps. CH₄ purity and recovery on different cyclic steps for two different CH₄ ratios: (c) 20% CH₄/80% CH₄ and (d) different cyclic steps, such as pressure equalization. (e) Purity and recovery of CH₄ from a 30% mixture at different flow rates and (f) different cyclic steps, with the addition of pressurized equalization.²² This figure has been adapted/reproduced from ref. 22 with permission from Elsevier, copyright 2020.

The two-bed VPSA system with a six-step cyclic configuration (Fig. 17a) demonstrates enhanced CH₄ purity (40–45%) at 2 L min⁻¹ flow rates for 20–35% feed concentrations. Janfang et al.¹⁴⁹ pioneered skid-mounted two-bed VPSA prototypes for coal mine applications, achieving CH₄ enrichment from 16–30% to 43.3–70.7% using TUTJ-1 zeolite, with a peak productivity of 43.6 m³ (kg h)⁻¹, albeit with high energy consumption (1.393 kWh per kg CH₄). While MOFs exhibit high CH₄ capacities, their industrial deployment remains limited by moderate selectivity (<8) and concentration ratios. Dual Reflux VSA (DR-VSA) employing ionic liquid zeolites (ILZ) attained 80.2% CH₄ purity and 95.5% recovery from 20% feed gas, albeit with elevated energy demand (3.13 kWh per kg CH₄).¹⁵⁰ A dual-stage VPSA process using activated carbon adsorbent resulted in 90% CH₄ purity and 98% recovery from a 20% feed. The results also show that process energy consumption at lower adsorption pressure using an improved adsorbent was more energy-efficient, only 0.704 kWh per kg CH₄.¹⁵¹ Based on the measured equilibrium and kinetic parameters, PSASIM® simulations of [Ni₃(HCOO)₆] MOF-based dual PSA yielded 96% purity (50/50 CH₄/N₂ feed) with 1.7 × 10⁻⁴ mol (kg s)⁻¹ productivity and 1.111 kWh per kg CH₄ thermal energy.¹⁵² Comparative studies of 4-column VSA configurations have revealed that pressure equalization and purge optimization have a critical impact on product purity and recovery in activated carbon systems.¹⁵³ Recent four-bed VPSA trials with [Cu(INA)₂] MOF (Fig. 17a) employed LDF-modeled breakthrough curves to optimize mass transfer coefficients and operational parameters, including input gas flow rate, adsorption period, pressure, and evacuation pressure. Under the optimal operational conditions, 90% CH₄ recovery with 50% purity was achieved with low energy consumption (0.861 kWh per kg CH₄), as shown in Fig. 17b–f.²⁴ These advances highlight the critical interplay between adsorbent properties (activated carbons, zeolites, molecular sieves and MOFs) and process engineering in CMM enrichment.

Fig. 17 (a) Crystal structure of the Cu(INA)₂ MOF. (b) Four-bed six-step VPSA process for 15% CMM capture. (c and d) Effect on the purity and recovery of the vacuum pressure and product flow rate. (e and f) Productivity and energy consumption variation based on vacuum pressure and product flow rate.²⁴ This figure has been adapted/reproduced from ref. 24 with permission from Elsevier, copyright 2022.

3.2 VAM separation performance evaluation

VAM is currently being investigated using various separation technologies, including temperature swing adsorption (TSA), PSA, and different hybrid approaches. Innovations in these processes are crucial for enhancing CH₄ recovery efficiency, operational safety, and scalability. Several adsorbents have demonstrated promising performance, with high CH₄ adsorption capacities ranging from 1.0 to 2.75 mmol g⁻¹ and selectivities of up to 12, as summarized in Table 2. The use of such efficient adsorbents in PSA and TSA processes has the potential to significantly enhance VAM treatment performance.

Early studies on VAM enrichment employed PSA or vacuum pressure swing adsorption (VPSA) methods. Liu et al.¹⁵⁴ evaluated a VPSA process using coconut shell-derived activated carbon, observing that the CH₄ content increased from 0.3% to 0.74% with rising adsorption pressure. Although the enriched VAM concentration remained below 1%, these initial findings stimulated further investigation into PSA-based VAM upgrading. Subsequently, Ouyang et al.¹⁵⁵ investigated VAM enrichment using the same type of coconut shell-based activated carbon, achieving a selectivity of up to 6.18 via a VPSA process. The CH₄ content was raised from 0.42% to 1.09%, meeting the requirement for use in lean-burn gas turbines. More recently, Yang et al.¹⁵⁶ conducted a VPSA study incorporating an additional vacuum step. Using the same adsorbent with an equilibrium selectivity of 5, they enriched CH₄ from 0.2–0.4% to 3% by increasing the vacuum exhaust ratio. Their laboratory-scale three-bed, two-step unit successfully increased CH₄ concentration from 0.2% to 1.2%, satisfying the feed specifications for flow-reversal reactors or lean-burn turbines in power generation applications.

A prototype two-bed VAM enrichment unit was designed and constructed in compliance with Australian Standards and local mining regulations (Fig. 18a). In 2014, Bae et al.¹⁵⁷ employed a honeycomb monolithic carbon fiber composite (HMFC) for VAM enrichment. They highlighted the importance of an initial vacuum step to enrich VAM streams containing less than 1% CH₄. A unique vacuum and temperature swing desorption strategy was applied during adsorbent regeneration: first a vacuum swing, followed by a temperature-vacuum swing. The initial vacuum swing (VS) proved critical in determining the final CH₄ concentration. While one adsorption step could enrich VAM with less than 1% CH₄ by factors of 5 or 11, a two-step adsorption process achieved CH₄ concentrations as high as 25%. The two-step VS process—both short and full cycles—not only enhanced CH₄ enrichment but also maintained operational safety by ensuring oxygen levels remained within permissible limits. Additionally, one-step adsorption of VAM with less than 5% CH₄ was sufficient to fuel lean-burn turbines. The conventional temperature and vacuum swing methods were modified to include an initial vacuum process. The adsorption process was specifically designed to prevent CH₄ concentrations from reaching explosive limits during enrichment. Using VAM feed streams with three different CH₄ ratios (0.3–0.98 vol%), the system achieved enriched CH₄ levels of 19.28–36.92 vol% CH₄. Fig. 19(b–e) illustrates the desorption CH₄ concentrations for the first and second beds at initial and final stages for these three feed ratios.

Fig. 18 (a) Schematic of a two-bed large-scale VAM enrichment process. (b and c) CH₄ desorption concentration, initial and final stage of column 1. (d and e) CH₄ desorption concentration, initial and final stage of column 2.¹⁵⁹ This figure has been adapted/reproduced from ref. 159 with permission from the American Chemical Society, copyright 2019.

Fig. 19 Schematic of the MOF and MMM preparation. (a) Precursors and structure of the CoNi–DABCO MOF. (b) CoNi–DABCO@PDMS/PVDF MMM preparation. (c and d) CH₄ and N₂ single-component permeation and binary mixture performance at different filler loadings. (e and f) Feed pressure affects gas permeability and selectivity of the 20 wt%-CoNi–DABCO@PDMS/PVDF MMM and long-term operational stability.¹⁷⁵ This figure has been adapted/reproduced from ref. 175 with permission from Elsevier, copyright 2025.

Recently, Bae et al. developed a prototype VAM capture system based on a two-stage vacuum-temperature-vacuum swing adsorption (VTVSA) process using a honeycomb monolithic structure, deployed at an Australian coal mine. This technology exceeded conventional PSA enrichment limits (21×) by producing 49.6 mol% CH₄ from a 2.4 mol% feed at 3.6 bar operating pressure. Field trials of the VTVSA system with carbon fiber composites demonstrated CH₄ enrichment factors of 44–63 from feed concentrations of 0.54–0.73 vol% in Australian coal mines.^158,159 The study found that using temperature swing (TS) alone was insufficient for high CH₄ recovery, as a significant amount of CH₄ remained in the gas phase within the adsorption column. To improve recovery, a vacuum swing (VS) step was introduced following the TS phase.

3.3 Membranes for CH₄ enrichment

Membrane separation stands as an energy-efficient and environmentally sustainable alternative to conventional separation methods, distinguished by its operational simplicity and scalability. Since its industrial emergence in the late 1970s, this technology has been extensively adopted in critical sectors, including petroleum refining, natural gas processing, and chemical manufacturing. The separation mechanism relies on the differential permeation rates of gas components through a semi-permeable membrane, governed by their distinct diffusion coefficients and solubility within the membrane material. This selectivity arises from the interplay between molecular size, condensability, and membrane–pore interactions, enabling preferential transport of specific components under applied pressure gradients.¹⁶⁰

3.3.1 Inorganic membranes. Membranes are categorized into inorganic and organic types, based on material composition. Molecular sieve membranes represent an emerging class of inorganic membranes that are distinguished by their monodispersed pore architecture (<1 nm), robust hydrothermal stability (>500 °C), and cation exchange capacity. These attributes enable precision separations in gas mixtures (e.g., CH₄/N₂) and liquid pervaporation processes through size-exclusion and surface diffusion mechanisms.^161–163 Microporous molecular sieve membranes, with pore dimensions commensurate with gas kinetic diameters (3.0–4.0 Å), demonstrate exceptional mixed-gas separation performance.^164,165 Krishna et al.¹⁶⁶ reported SAPO-34 membranes achieving a N₂/CH₄ selectivity of 5.0 via combined molecular dynamics simulations and permeation experiments. Morooka et al.¹⁶⁷ synthesized titanium silicate membranes on α-Al₂O₃ supports via hydrothermal crystallization, demonstrating a N₂ permeance of 1.2 × 10⁻⁷ mol (m² s Pa)⁻¹ with a N₂/CH₄ selectivity of 4.1, attributed to preferential N₂ surface diffusion. Seyed et al.¹⁶⁸ engineered PBI/Matrimid composite membranes (1 : 3 ratio) exhibiting a N₂ permeability of 7.99 barrer and a N₂/CH₄ selectivity of 3.8, outperforming pure polymeric analogs. CVD-fabricated carbon nanotube (CNT) membranes on alumina substrates displayed tunable CH₄/N₂ selectivity (1.8–3.85) dependent on the CNT inner diameter (24–34 nm) and feed composition. Notably, increasing the CH₄ concentration from 20% to 60% elevated the CH₄ permeability by 38%, while suppressing N₂ transport by 22%, indicative of competitive adsorption effects.¹⁶⁹

3.3.2 MOF-based membranes. MOFs have emerged as transformative materials in gas separation due to their ultralow density (>6000 m² per g surface area), monodispersed pore architectures, and tunable surface functionalities. The correlation between molecular kinetic diameters and diffusivity in ZIF-8 at 35 °C highlights its molecular sieving potential.^170,171 Among MOFs, zeolitic imidazolate frameworks (ZIFs), notably ZIF-8, exhibit exceptional thermal stability (>400 °C) and chemically resistant microporous structures (<3.4 Å pores), making them ideal for membrane applications. Song et al.¹⁷² engineered hybrid matrices via the incorporation of ZIF-8 nanoparticles into Matrimid® 5218, achieving unprecedented H₂/CH₄ selectivity (156) at 150 °C under 20 wt% ZIF-8 loading. This performance persisted post 18 hour vacuum annealing, demonstrating structural integrity under thermal stress.^173,174 A salt-etching strategy utilizing AgNO₃ as a porogen yielded Ag-ZIF-8-2 membranes with hierarchical porosity. The optimized bilayer architecture exhibited N₂ permeance of 4457 GPU (1 GPU = 3.35 × 10⁻¹⁰ mol (m² s Pa)⁻¹) and N₂/CH₄ selectivity of 10.5, outperforming conventional polymeric membranes more than 5–10 times in permeance (Table 3). Recently, MMMs have been constructed by adding CoNi–DABCO to polydimethylsiloxane, which has several oppositely adjacent open metal sites, tiny pores, and significant CH₄ binding affinity (Fig. 19a and b). MMM with 20 wt% CoNi–DABCO exhibits 1285 barrer CH₄ permeability and 3.7 CH₄/N₂ mixed-gas selectivity (Fig. 19c and d). Additionally, MMMs have excellent pressure resistance (up to 10 bar) and 30 day stability (Fig. 19e and f).¹⁷⁵

Table 3 Permeability and selectivity for CH₄ and N₂ of ZIF-related membranes

Membrane material	Permeability [barrer]		Selective		Ref.
	N2	CH4	N2/CH4	CH4/N2
a Matrimid®-ZIF-8 nanocomposite membranes with 20 wt% loading of ZIF-8 nanoparticles. Membrane samples were annealed at various temperatures for 18 h under vacuum. b Matrimid® membrane and Matrimid®-ZIF-8 composite membranes. All membrane samples were annealed under vacuum at 230 °C for 18 h. c Pure gas permeation. d Mixed gases permeation.
Matrimid®-ZIF-8, 20 wt%, 60 °C^a^,^c	1.77	1.06	1.67		176
Matrimid®-ZIF-8, 20 wt%, 150 °C^a^,^c	0.42	0.23	1.83		176
Matrimid®-ZIF-8, 20 wt%, 180 °C^a^,^c	0.61	0.31	1.97		176
Matrimid®-ZIF-8, 20 wt%, 200 °C^a^,^c	0.61	0.36	1.96		176
Matrimid®-ZIF-8, 20 wt%, 230 °C^a^,^c	0.88	0.46	1.91		176
Matrimid®c	0.36	0.23	1.57		176
Matrimid®-ZIF-8, 5 wt%, 230 °C^b^,^c	0.47	0.26	1.81		176
Matrimid®-ZIF-8, 10 wt%, 230 °C^b^,^c	0.63	0.45	1.40		176
Matrimid®-ZIF-8, 30 wt%, 230 °C^b^,^c	1.68	1.16	1.45		176
ZIF-7^c	49	53		1.08	177
ZIF-7^c	13	19		1.46	178
ZIF-8^c	466	430	1.08		179
ZIF-8^c	890	794	1.12		173
ZIF-8^c	412	376	1.10		180
ZIF-22^c	3393	3608		1.06	181
ZIF-22^d	3500	3954		1.13	181
ZIF-90^c	1183	938	1.26		182
ZIF-90^c	1266	980	1.29		182
ZIF-90^c	765	645	1.19		183
ZIF-90^d	806	615	1.31		183

---

Membrane-based gas separation systems are operationally advantageous due to their modular scalability and low maintenance demands. However, CH₄/N₂ separation poses inherent challenges stemming from the near-identical kinetic diameters of the gases (CH₄: 3.8 Å vs. N₂: 3.64 Å) and their nonpolar nature, resulting in suboptimal selectivity (<4) and permeability (<50 barrer) for conventional polymeric membranes. Economic viability requires a separation factor of more than 5, a threshold balancing energy costs against CH₄ recovery efficiency, to justify industrial deployment. Current research prioritizes the rational design of high-flux membranes with engineered selectivity through advanced materials (e.g., mixed-matrix membranes, MOF hybrids) coupled with process optimization (e.g., multistage cascades, sweep gas configurations) to overcome these intrinsic limitations in coal mine methane (CMM) capture applications.

3.4 Cryogenic processes for CMM enrichment

Cryogenic separation is a promising technology for CH₄ and N₂ separation. The fundamental separation mechanism is based on differential boiling points, with the adiabatic expansion of compressed gas achieving a cryogenic (boiling) point. Cryogenic technology was first utilized for air separation and then for dry gas recovery by a commercial-scale refinery invented by Linde in 1985.¹⁸⁴ For nitrogen and methane separation, N₂ and CH₄ have boiling points of 77.35 K and 111.7 K, respectively, at 101.3 kPa, which is a 34 K boiling point differential; therefore, the gases can be separated by cryogenic-based rectification technology. Low-temperature cryogenic separation technology was used to remove nitrogen from natural gas in 1982. The gas mix with 50% N₂ and 41.1% CH₄ at a gauge pressure of 1000 psi can be enriched to obtain 95.7% CH₄.¹⁸⁵ Southwest Research Institute of Chemical Industry has accomplished a 99.4% CH₄ recovery rate even for a gas mix with 2.3% CH₄ concentration. Auxiliary circulation equipment was included during the separation process to enrich CH₄ purity from 45% to 95–99%, along with recovery rates ranging from 95 to 99% in the CBM. The liquefaction separation technology for CH₄ purification with air-bearing coal-bed methane in the center of the fractionation tower is being cold through evaporation in the evaporator at the bottom. The fractionation tower provides nitrogen and LNG with high purity at the top and bottom, respectively.^186,187 The Shanxi Yangquan Coal Group accomplished a commercial application of the approach in 2007 with ∼98% CH₄ purity in the product gas. A small NGL process was designed, and the system's heat and cooling parameters were also analyzed. The results showed that the N₂/CH₄ mixture, lacking propane cooling, and the compressor's power consumption affect the per-unit LNG cost.¹⁸⁶ However, disadvantages, such as higher energy consumption, higher primary investment, and operating costs, limit its wide application. It is preferred for commercial-scale separation equipment with a capacity of more than 10⁶ m³ per day. In contrast, oxygen with CMM is generally separated in small-scale operations.

3.4.1 Assessing CMM separation methods: performance, and energy consumption. Table 4 presents a comparative evaluation of adsorption, cryogenic, and membrane-based separation technologies across four critical performance metrics: feedstock composition tolerance, product purity thresholds, and specific energy consumption.

Table 4 Comparison of CMM capture technologies for different process performance

Process	Technology	Materials design	Feed flow rate (L min^−1)	Feed pressure (bar)	Feed temperature (°C)	Feed conditions (%)	CH4 product purity (%)	CH4 product recovery (%)	CH4 energy consumption (kWh kg^−1)	Reference
Adsorption	VSA four-step sixteen-step	Activated carbon	1.98–4.6	1.0	23	CH4: 17–20	42	80	1.939	153
						N2: 78.15
						O2: 17.85
	PSA three-bed six-step	Activated carbon	N/A	1.0	35	CH4: 50	89	75	0.624	188
						N2: 50
	VPSA two-step eight-step	Activated carbon	2	3.0	25	CH4: 25	50.40	86.3	0.251	148
						N2: 65
						O2:10
	VPSA four-bed twelve-step	Zeolite TUTJ-1	50	3.0	N/A	CH4: 15–30	43–70.7	78–90	1.393	149
						N2: 78.15
	PSA two-bed six-step	Silicalite-1	1.5	1.7	25	CH4: 20–40	40–50.1	84.4	0.1667	22
						N2: 60–80
	Dual-VPSA six-bed	Activated carbon	150	3.0	23	CH4: 20–30	90	98.71	0.704	151
						N2: 70–80
	VPSA one bed four-step	Activated carbon	1.0	2.5	25	CH4: 10	75	89		189
						CO2: 78.15
						N2: 12.85
	DR-PSA	Ionic liquid zeolite	15	5.5	25	CH4: 1–15	88.3	88.3	0.426	190
						N2: 85–99
	HP-PVSA	Ionic liquid zeolite	15	1.21	25	CH4: 1–15	87.5	80.3	3.139	190
						N2: 85–99
	VPSA six-bed twelve-step	Ionic liquid zeolite	30–40	1.20	18	CH4: 5–16.1	31	90	0.31–0.43	191
						N2: 84.9–95
	PSA two-bed six-step	Crystalline-shaped silicalite-1	0.075–0.1	—	25	CH4: 20	42	85	—	192
						N2: 80
	PSA two-bed three-step	Zeolite/activated carbon monolith	0.0296	—	25	CH4: 20–50	40–80	82–85	—	193
						N2: 50–80
	Dual-PSA two-bed five-step	[Ni3(HCOO)6] MOF	15.42	6–13	25	CH4: 50	96	97	0.153	152
						N2: 50
	VPSA four-bed six-step	Cu(INA)2	N/A	3	25	CH4: 15	52	91	0.2278	24
						N2: 78.15
						O2: 17.85
Cryogenic	Cryogenic distillation		N/A	1.0	35	CH4: 50	63	91	0.90	188
						N2: 50
	Distillation column		6.572 × 10^6	1.1	40	CH4: 40	99.91	97.12	1.131	194
						N2: 47.4
						O2: 12.6
	Distillation column		6.676 × 10^6	5.0	40	CH4: 30	98.0	98	1.06–1.61	195
						N2: 70
	Distillation column		4.01 × 10^3	3.2	N/A	CH4: 40	99.30	90	N/A	196
						N2: 47.53
						O2: 12.47
Membrane	Membrane system	PDMS membrane	N/A	1.0	35	CH4: 50	72	64	N/A	188
						N2: 50
	Membrane process selectivity (CH4/N2): 4.0	Polyamide–polyether block copolymer, Pebax® 2533	1.966 × 10^5	2.7	25	CH4: 75–90	96	86	0.136	197
						N2: 10–25
	Membrane process selectivity (N2/CH4): 8.0	SAPO-34 membranes	3.933 × 10^5	6.4	23	CH4: 85	96	N/A	0.28	198
						N2: 15

---

Table 4 compares PSA, cryogenic, and membrane technologies for coal-mine methane (CMM) purification. Adsorption-based systems demonstrate notable performance variations. Kevin Gang Li et al.¹⁵³ attained 42% CH₄ purity with 80% recovery using activated carbon in vacuum swing adsorption (VSA), while Salman et al.²⁴ improved these metrics to 52% purity and 91% recovery via vacuum-pressure swing adsorption (VPSA) with low energy consumption (0.2278 kWh per kg CH₄). Advanced configurations, such as those developed by Goafei Chen et al.,¹⁵¹ achieved 90% purity with 98.71% recovery using a six-bed dual VPSA system based on 0.704 kWh per kg CH₄ energy consumption. The highest reported adsorption performance is attributed to Martinez et al.,¹⁵² who utilized [Ni₃(HCOO)₆] MOF pellets in a dual PSA cycle to reach 96% purity and 99.9% recovery, albeit requiring high energy 1.112 kWh per kg CH₄ for the product. Cryogenic methods exhibit comparable efficacy, with simulation studies¹⁶³ reporting 98% purity and matching recovery rates. Subsequent research enhanced these figures to 99.91% purity and 97.12% recovery, although both implementations required intensive refrigeration cycle compressor energy. Experimental validation studies confirmed 99.3% CH₄ purity in a pilot-scale cryogenic distillation. Membrane technology has shown progressive development, with carbon nanotube membranes achieving CH₄/N₂ selectivity of 1.8–3.8 under varied feed conditions.^164,165 Carbon molecular sieve membranes demonstrated 5.2 barrer permeability with 6.0N₂/CH₄ selectivity,¹⁶⁶ while modified membranes¹⁶⁷ attained exceptional CO₂/CH₄ (45.2) and O₂/N₂ (7.3) selectivity. Simulation results¹⁶⁸ indicate 96% CH₄ recovery potential for membrane systems. Cross-technology benchmarking (Fig. 21) reveals PSA as the optimum option with the highest CH₄ recovery (99.9%), followed by cryogenic (97.1%) and membrane (96.0%) processes, establishing a hierarchy of technical feasibility for CMM upgrading applications.

Fig. 20 establishes PSA as the optimum with respect to CH₄ purity (96%), and the recovery was high (97.9% CH₄) while requiring substantially lower energy intensity (0.135 kWh per kg CH₄). Table 4 reveals a direct correlation between CH₄ recovery efficiency and energy demand among the proven technologies. Cryogenic separation demonstrates intermediate energy consumption at 1.06 kWh per kg CH₄, while membrane technology emerges as the most energy-efficient option (0.136 kWh per kg CH₄). This efficiency hierarchy presents operational trade-offs: membrane systems face intrinsic recovery limitations, only achieving 86% due to permeability–selectivity constraints. Enhancing product recovery in membrane processes necessitates either expanded membrane surface areas or multi-stage module configurations, thereby elevating both energy expenditure and capital costs. In contrast, PSA achieves superior purity and recovery metrics through less energy-intensive phase-change operations and cyclic pressurization–depressurization mechanisms, respectively.

Fig. 20 CH₄ purity, recovery, and energy consumption of pressure swing adsorption, membrane, and cryogenic technologies.

While cryogenic separation yields high purity (>95% CH₄), its liquefaction requires an enormous amount of energy (1.5–3.0 kWh per kg CH₄) and hence is not economical for large-scale use. While less efficient with lower concentrations, PSA (Pressure Swing Adsorption) is adjustable for medium concentration gas (20–80% CH₄) with moderate energy inputs (0.5–1.5 kWh per kg CH₄). Membranes are energy-efficient and cost-effective (0.3–1.0 kWh per kg CH₄) for high-concentration streams (<80% CH₄) but yield only 90% purity. The compressor energy is a significant cost component for all three technologies. With the use of vacuum pumps for desorption, VPSA (Vacuum PSA) reduces energy consumption (∼0.1–1.2 kWh per kg CH₄), but the complex pretreatment and cycling require a greater initial expenditure ($6 to 20 M). Standard PSA (0.5–1.5 kWh per kg CH₄) is less efficient for low-concentration CH₄ (<20%) but costs less ($5 to 15 M). Dual PSA recovers more material (∼0.3–1.0 kWh per kg CH₄) by cascading adsorption levels, but the enhanced hardware increases the cost ($8 to 25 M). For dilute streams (<30% CH₄), single-stage membrane units are most economical ($2 to 10 M, 0.3–1.0 kWh per kg CH₄). Conversely, hybrid membrane–PSA units (e.g., for >90% purity) raise costs ($10 to 18 M) while lowering energy use (∼0.5–1.2 kWh per kg CH₄). The CH₄ concentration dictates the trade-offs: membranes are optimal for low-grade gas, while PSA modifications achieve the higher purity requirements at higher costs for complex configurations.

3.5 Critical comparison of emerging technologies

PSA has emerged as a superior technology for CMM capture compared to membrane separation and cryogenic distillation due to its ambient pressure operation, modular design, and energy efficiency. PSA can operate effectively at 1–3 bar, eliminating the energy-intensive compression required for membrane and cryogenic processes. It can compete for fluctuating feed gas composition (e.g., variable O₂ and N₂ levels). The non-cryogenic nature of PSA reduces explosion risks and enables deployment in distributed, small-to-medium-scale CMM recovery projects.^199,200 Advances in porous materials (e.g., zeolites, MOFs) have enhanced CH₄ selectivity and cycle stability, mitigating historical limitations in single-cycle recovery efficiency. Membrane separation has high energy consumption due to compressed air requirements. Additionally, poor selectivity and limited industrial scalability restrict its deployment application in CMM capture. Meanwhile, the cryogenic process is economically viable only for large-scale CMM streams targeting LNG production.²⁰¹ Energy-intensive compression refrigeration and inherent O₂ co-condensation risks are unavoidable. From the perspective of technology readiness level (TRL), PSA is currently the most commercialized technology for CMM capture. Table 5 summarizes the pros and cons of all coal mine methane enrichment technologies. For PSA and membrane separation, separation performance is strongly dependent on adsorbents and membrane materials. In Tables 2 and 4, classical materials were considered, including AC, zeolites, and MOFs adsorbents, as well as inorganic and organic membranes. For PSA and membrane separation, the separation performance is strongly dependent on adsorbents and membrane materials. Hyrbrid system is only suitable for higher concentrations of CH₄ in the feed gas.

Table 5 Pros and cons of coal mine methane for different separation technologies

Separation processes	PSA^a	Cryo-distillation^b	Mem-separation^c	Hybrid systems
a PSA (pressure awing adsorption). b Cryo-dist (cryogenic distillation). c Mem-sep (membrane separation).
Process	Pressure swing adsorption	Cryogenic process	Membrane process	PSA–membrane system
Mechanism	Selectivity	The difference in boiling points	Permeability difference	Selectivity
Material consideration	Pellets form (activated carbon, zeolites, MOFs)	No material	Membrane (inorganic and organic MOF-based membranes)	Used a combination of adsorbent and membrane
Change in phase	No	Yes	No	No
Pressure requirement	Atmospheric in PSA or ambient in vacuum-PSA 1–3 bar	Higher pressure 10–15 bar	Very high pressure 5–10 bar	Medium pressure 0–5 bar
Pros	Better safety, life, and operational flexibility with low energy consumption	Good enrichment performance, used for large-scale processes, is beneficial for LNG	High operation flexibility, low energy consumption	Higher purity and recovery with low energy consumption
Cons	Limited enrichment for one cycle, but increases with higher cycles	Higher consumption of energy and process complexity, and higher risk	High methane enrichment but low CH4/N2 selectivity, short membrane life	Limited membrane life can potentially be loss, but achieved 99% pure CH4
Development status	Used for commercial applications	Site trials in several coal mines	Laboratory development stage	Needs to be considered for this application

---

4 Roadmap for designing an efficient system for CMM

Addressing coal mine methane (CMM) purification challenges requires enhanced separation efficiency through technological integration. Hybrid systems combining cryogenic, membrane, and PSA processes could synergistically optimize energy efficiency while achieving higher product purity. Recent PSA innovations in cycle optimization and advanced adsorbents (e.g., MOF-based materials) demonstrate potential for energy reduction per recovery percentage point. Cross-disciplinary collaboration between academia and industry remains critical to commercializing emerging concepts such as thermally coupled PSA cycles or mixed-matrix membranes. Strategic government subsidies targeting R&D tax credits and carbon pricing mechanisms could improve project Net Present Value (NPV), simultaneously advancing CH₄ emission reduction targets. PSA systems for CMM capture demonstrate configuration-dependent performance across dual-bed, three-bed, and four-bed architectures (Fig. 21). Dual-bed PSA systems constitute the baseline configuration for decentralized CH₄ capture, prioritizing capital efficiency and operational simplicity through minimized valve actuation sequences. Their compact footprint suits space-constrained installations with a small-scale capacity, though the absence of pressure equalization protocols induces lower CH₄recovery efficiency with elevated specific energy consumption. Three-bed systems mitigate these inefficiencies through staged pressure equalization functionality, improving adsorbent working capacity. The intermediate bed facilitates inter-vessel pressure-gradient optimization during phase transitions, achieving higher recovery and lower energy intensity. This configuration will increase capital cost and maintenance frequency due to increased valve cycling requirements. However, it maintains operational viability for medium-scale deployments through adaptive cycle algorithms that balance CH₄ purity against compressor duty cycle constraints.

Fig. 21 Different configurations for coal mine methane capture are (a) three-bed PSA, (b) four-bed PSA, (c) dual PSA with two beds, and (d) dual PSA with three-bed configurations.²⁰² This figure has been adapted/reproduced from ref. 202 with permission from the American Chemical Society, copyright 2024.

Four-bed PSA systems represent the pinnacle of CH₄ capture technology for high-volume applications, employing multi-stage pressure gradient management and phase-synchronized bed cycling to achieve higher recovery efficiency at moderate specific energy consumption. Their continuous adsorption–desorption sequencing minimizes product purity fluctuations through real-time pressure compensation algorithms. Thus, the product can sustain high CH₄ molar content and volumetric flow rates. However, this operational sophistication demands greater capital investment and increased maintenance frequency due to complex valve actuation matrices.

The selection of PSA configurations for coal mine methane (CMM) recovery follows a techno-economic hierarchy dictated by operational scale and feed gas characteristics. Dual-bed systems represent the entry-level solution for low-flow operations, achieving lower CH₄ recovery at relatively high energy intensity within compact footprints. Their operational simplicity is enabled by minimal valve cycling and basic control architecture. As a result, it leads to adsorbent underutilization due to the absence of pressure equalization protocols. Three-bed configurations optimize mid-scale applications through partial pressure equalization, enhanced CH₄ recovery, and reduced specific energy consumption. The intermediate bed enables inter-vessel pressure-gradient optimization during phase transitions, although it increases capital expenditure and maintenance frequency through enhanced valve actuation requirements. Four-bed systems dominate large-scale deployments via multi-stage pressure management, achieving higher recovery efficiency at lower energy consumption through synchronized bed cycling. Their continuous operation protocol maintains product purity within a lower variance but demands larger installation areas and higher capital costs.

5 Conclusions and future outlooks

China's coal mining sector emits 55.92 teragrams of CH₄ annually (equivalent to 28 billion cubic meters), predominantly released untreated into the atmosphere. This unmitigated emission stream presents threefold imperatives: securing clean energy reserves, reducing environmental impact, and enhancing mine safety. Our systematic review evaluates CH₄ recovery technologies across molecular-scale mechanisms to industrial process design, yielding four critical insights:

(1) Pressure swing adsorption demonstrates particular efficacy for low-concentration coal mine methane enrichment. Commercial viability has advanced through process optimizations, including reduced adsorption pressures and integrated deoxygenation phases, achieving higher CH₄ purity coupled with higher recovery rates. Adsorbent innovation remains pivotal—recent breakthroughs in metal–organic frameworks (MOFs) enable higher CH₄ working capacities than conventional activated carbons at equivalent selectivity thresholds.

(2) Molecular-scale separation criteria are being established. CH₄ and N₂ differentiation hinges on three molecular properties: polarizability differential, kinetic diameter variance, and quadrupole moment contrast. MOFs surpass conventional adsorbents through tunable pore geometries and surface functionalization, achieving higher CH₄/N₂ selectivity.

(3) The material performance hierarchy is summarized. On activated carbons, a maximum CH₄/N₂ selectivity of 7.0 was achieved via ultramicropore (<8 Å) optimization. For zeolites, Clinoptilolite demonstrates 6.53 kinetic selectivity through size-sieving effects, while 9.8 mmol per g CH₄ capacity was obtained at 298 K/100 kPa via coordinatively unsaturated metal sites on MOFs.

(4) Process economics comparisons were conducted. While cryogenic separation achieves 99% purity, its energy intensity and inflexibility limit scalability. Membrane systems require higher feed pressures for viable flux, in contrast to PSA's operational advantages. Large-scale PSA implementations currently provide the most viable pathway for CMM valorization, particularly when integrated with CH₄-to-liquid infrastructure for transportation fuel synthesis.

This review summarizes various aspects of capturing CMM to address global environmental issues. This study provides a deeper understanding of how coal mine methane emissions can be reduced, while also offering additional economic benefits to the mining industry and ultimately contributing to the emerging low-carbon economy.

Abbreviations

CMM	Coal mine methane
CBM	Coal bed methane
VAM	Ventilation air mine methane
AMM	Abounded mine methane
LEL	Lower explosive limit
CH4	Methane
N2	Nitrogen
O2	Oxygen
CO2	Carbon dioxide
PSD	Pore size distribution
GWP	Global warming potential
Bt CO2e	Billion tonnes CO2-equivalent
TRL	Technology readiness level
Tg	Teragrams
MOF	Metal organic framework
AC	Activated carbon
GAC-3	Shell-based carbon material
CMS	Carbon molecular sieve
PCFs	Polymer-derived carbon fibers
TMA^+	Tetramethylammonium cations
ETS	Engelhard Titanosilicate
ILZ	Ionic liquid zeolites
ZIFs	Zeolitic imidazolate frameworks
OMS	Open metal sites
BTC	Benzene-1,3,5-tricarboxylate
DHT	2,5-Dihydroxyterephthalate
ATC	1,3,5,7-Adamantane tetracarboxylate
PSA	Pressure swing adsorption system
TSA	Temperature swing adsorption
VPSA	Vacuum pressure swing adsorption system
DR-VSA	Dual reflux vacuum pressure swing adsorption system
VT-VSA	Vacuum-temperature-vacuum pressure swing adsorption system
VSA	Vacuum swing adsorption system
LDF	Linear driving force
LNG	Liquefied natural gas

Author contributions

Salman Qadir: conceptualization, investigation, writing – original draft, writing review & editing. Muhammad Kamran, Muhammad Sajjad, Sivadasan Dharani: methodology, visualization. Ahmad Naquash, Muhammad Islam: software. Shao-Tao Bai: funding acquisition, project administration, and visualization. Wang Sheng: writing – review & editing.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

No primary research results, software or code have been included, and no new data were generated or analysed as part of this review.

Acknowledgements

This work was financially supported by the Shenzhen Public Service Platform for Carbon Capture, Utilization and Storage (CCUS) Technology (XMHT20230108018), Research Projects of the Department of Education of Guangdong Province (2023ZDZX2086), the Technology Innovation Center for Carbon Sequestration and Geological Energy Storage (grant number MNRCCUS022302). S. Qadir thanks the financial support from Shenzhen Polytechnic University's Startup Fund (6025331002K).

References

1. S. M. Miller, A. M. Michalak, R. G. Detmers, O. P. Hasekamp, L. M. Bruhwiler and S. Schwietzke, Nat. Commun., 2019, 10, 303.
2. G. Janssens-Maenhout, M. Crippa, D. Guizzardi, M. Muntean, E. Schaaf, F. Dentener, P. Bergamaschi, V. Pagliari, J. G. Olivier and J. A. Peters, Earth Syst. Sci. Data, 2019, 11, 959–1002.
3. S. Peng, S. Piao, P. Bousquet, P. Ciais, B. Li, X. Lin, S. Tao, Z. Wang, Y. Zhang and F. Zhou, Atmos. Chem. Phys., 2016, 16, 14545–14562.
4. F. Barbi, L. d. C. Ferreira and S. Guo, J. Chin. Gov., 2016, 1, 324–339.
5. E. Ng and C. Ren, Urban Clim., 2018, 23, 352–372.
6. L. Liu, Q. Zhuang, Y. Oh, N. J. Shurpali, S. Kim and B. Poulter, J. Geophys. Res.:Biogeosci., 2020, 125, e2019JG005428.
7. S. Tao, S. Chen and Z. Pan, Energy Sci. Eng., 2019, 7, 1059–1074.
8. Y. Lu, H. Zhang, Z. Zhou, Z. Ge, C. Chen, Y. Hou and M. Ye, Energy Fuels, 2021, 35, 9102–9123.
9. L. Qiu, S. Wang, B. Hu, J. Yan, S. Deng, J. Shen, C. Shi, X. Ge and B. Chen, Energy Convers. Manage., 2024, 319, 118980.
10. S. Bai, Y. Zhang, F. Li, Y. Yan, H. Chen, S. Feng, F. Jiang, S. Sun, Z. Wang and C. Zhou, Sci. Total Environ., 2024, 950, 175446.
11. T. R. Scarpelli, D. J. Jacob, J. D. Maasakkers, M. P. Sulprizio, J.-X. Sheng, K. Rose, L. Romeo, J. R. Worden and G. Janssens-Maenhout, Earth Syst. Sci. Data, 2020, 12(1), 563–575.
12. International energy data and analysis–China, US EIA, 2015.
13. K. L. Cashdollar, I. A. Zlochower, G. M. Green, R. A. Thomas and M. Hertzberg, J. Loss Prev. Process. Ind., 2000, 13, 327–340.
14. K. Lee, W. C. Isley III, A. L. Dzubak, P. Verma, S. J. Stoneburner, L.-C. Lin, J. D. Howe, E. D. Bloch, D. A. Reed and M. R. Hudson, J. Am. Chem. Soc., 2014, 136, 698–704.
15. M. Tagliabue, D. Farrusseng, S. Valencia, S. Aguado, U. Ravon, C. Rizzo, A. Corma and C. Mirodatos, Chem. Eng. J., 2009, 155, 553–566.
16. K. Wei, H. Liu, P. Qiu and Z. Guo, Chem. Eng. J., 2025, 512, 162352.
17. J. K. Adewole, F. B. Owoyale, H. B. Oladipo and A. L. Ahmed, Discov. Mater., 2025, 5, 1–53.
18. A. Imtiaz, M. H. D. Othman, A. Jilani, I. U. Khan, R. Kamaludin, J. Iqbal and A. G. Al-Sehemi, Membranes, 2022, 12, 646.
19. T. Kopac, Int. J. Energy Res., 2021, 45, 15839–15856.
20. J. Hu, T. Sun, X. Liu, S. Zhao and S. Wang, Microporous Mesoporous Mater., 2016, 225, 456–464.
21. J. Hu, T. Sun, X. Liu, Y. Guo and S. Wang, RSC Adv., 2016, 6, 64039–64046.
22. J. Yang, H. Bai, H. Shang, J. Wang, J. Li and S. Deng, Chem. Eng. J., 2020, 388, 124222.
23. X. Yang, Z. Li, C. Zhang, H. Wang, E. Zhang, Y. Xing, P. Xiao, R. T. Yang, Y. Liu and P. A. Webley, Chem. Eng. J., 2019, 367, 295–303.
24. S. Qadir, D. Li, Y. Gu, Z. Yuan, Y. Zhao, S. Wang and S. Wang, Chem. Eng. J., 2021, 408, 127238.
25. Z. Yang, D. Wang, Z. Meng and Y. Li, Sep. Purif. Technol., 2019, 218, 130–137.
26. R. Tang, Q. Dai, W. Liang, Y. Wu, X. Zhou, H. Pan and Z. Li, Chem. Eng. J., 2020, 384, 123388.
27. A. Henrique, M. Karimi, J. A. Silva and A. E. Rodrigues, Chem. Eng. Technol., 2019, 42, 327–342.
28. M. Mofarahi and A. Bakhtyari, J. Chem. Eng. Data, 2015, 60, 683–696.
29. Y. Guo, J. Hu, X. Liu, T. Sun, S. Zhao and S. Wang, Chem. Eng. J., 2017, 327, 564–572.
30. Y. Chen, H. Wu, Y. Yuan, D. Lv, Z. Qiao, D. An, X. Wu, H. Liang, Z. Li and Q. Xia, Chem. Eng. J., 2020, 385, 123836.
31. W. Lu, Z. Wei, Z. Y. Gu, T. F. Liu, J. Park, J. Park, J. Tian, M. Zhang, Q. Zhang, T. Gentle 3rd, M. Bosch and H. C. Zhou, Chem. Soc. Rev., 2014, 43, 5561–5593.
32. G. Ferey, C. Serre, T. Devic, G. Maurin, H. Jobic, P. L. Llewellyn, G. De Weireld, A. Vimont, M. Daturi and J. S. Chang, Chem. Soc. Rev., 2011, 40, 550–562.
33. H. Furukawa, K. E. Cordova, M. O'Keeffe and O. M. Yaghi, Science, 2013, 341, 1230444.
34. S. Kitagawa, Angew Chem. Int. Ed. Engl., 2015, 54, 10686–10687.
35. H.-P. Li, S.-N. Li, H.-M. Sun, M.-C. Hu, Y.-C. Jiang and Q.-G. Zhai, Cryst. Growth Des., 2018, 18, 3229–3235.
36. C. Gu, N. Hosono, J. J. Zheng, Y. Sato, S. Kusaka, S. Sakaki and S. Kitagawa, Science, 2019, 363, 387–391.
37. J. D. Humby, O. Benson, G. L. Smith, S. P. Argent, I. da Silva, Y. Cheng, S. Rudic, P. Manuel, M. D. Frogley, G. Cinque, L. K. Saunders, I. J. Vitorica-Yrezabal, G. F. S. Whitehead, T. L. Easun, W. Lewis, A. J. Blake, A. J. Ramirez-Cuesta, S. Yang and M. Schroder, Chem. Sci., 2019, 10, 1098–1106.
38. S. Cavenati, C. A. Grande and A. E. Rodrigues, Chem. Eng. Sci., 2006, 61, 3893–3906.
39. D. S. Sholl and R. P. Lively, Nature, 2016, 532, 435–437.
40. Z. Zhang, Z.-Z. Yao, S. Xiang and B. Chen, Energy Environ. Sci., 2014, 7(9), 2868–2899.
41. N. Kholod, M. Evans, R. C. Pilcher, V. Roshchanka, F. Ruiz, M. Coté and R. Collings, J. Clean. Prod., 2020, 256, 120489.
42. G. Raj, Surface Chemistry, Krishna Prakashan Media, 2002.
43. C. Tien, Introduction to Adsorption: Basics, Analysis, and Applications, Elsevier, 2018.
44. K. S. Knaebel, Adsorbent selection, accessed on 6.8.2011.
45. M. W. Ackley, R. Giese and R. T. Yang, Zeolites, 1992, 12, 780–788.
46. A. Ertan, CO₂, N₂, and Ar adsorption on zeolites, Izmir Institute of Technology, Turkey, 2004.
47. S. U. Nandanwar, D. R. Corbin and M. B. Shiflett, Ind. Eng. Chem. Res., 2020, 59, 13355–13369.
48. J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504.
49. Z. Kong, H. Zhang, T. Zhou, L. Xie, B. Wang and X. Jiang, Sep. Purif. Technol., 2024, 129099.
50. S. Kundu, T. Khandaker, M. A.-A. M. Anik, M. K. Hasan, P. K. Dhar, S. K. Dutta, M. A. Latif and M. S. Hossain, RSC Adv., 2024, 14, 29693–29736.
51. J.-T. Liu, Y.-C. Zheng, X. Hou, X.-R. Feng, K. Jiang and M. Wang, Chem. Eng. J., 2024, 154013.
52. D. Su, P. Chen, C. Li, Y. Yan, R. Zhao, Q. Yue and Y. Qiao, Molecules, 2024, 29, 4404.
53. Y. Belmabkhout, G. D. Weireld and M. Frere, J. Chem. Eng. Data, 2004, 49, 1379–1391.
54. M. Baksh, A. Kapoor and R. Yang, Sep. Sci. Technol., 1990, 25, 845–868.
55. J. Zhang, L. Li and Q. Qin, Clean Energy, 2021, 5, 329–338.
56. Y. Zheng, Q. Li, C. Yuan, Q. Tao, Y. Zhao, G. Zhang and J. Liu, Powder Technol., 2019, 347, 42–49.
57. Y. Lei, Y. Zhao, H. Lian, Y. Wang, X. Xue, G. Zhang and Y. Zhang, J. Environ. Chem. Eng., 2024, 12, 112477.
58. T.-Y. Li, Y.-S. Wang, M.-Y. Zhang, B.-Y. Liu, Z. Zheng, G.-P. Hao and A.-H. Lu, Chem. Eng. J., 2023, 462, 142118.
59. H. Qinglin, S. Farooq and I. Karimi, Langmuir, 2003, 19, 5722–5734.
60. X. Peng, D. Cao and W. Wang, Chem. Eng. Sci., 2011, 66, 2266–2276.
61. S. Cavenati, C. A. Grande and A. E. Rodrigues, Sep. Sci. Technol., 2005, 40, 2721–2743.
62. M. W. Ackley and R. T. Yang, AIChE J., 1990, 36, 1229–1238.
63. J. Zhang, Coal Sci. Technol., 2015, 43(6), 141–145.
64. K. X. Yao, Y. Chen, Y. Lu, Y. Zhao and Y. Ding, Carbon, 2017, 122, 258–265.
65. X. Tang, X. Bai, Y. Wang, X. Yang, X. Li, J. Li and J. Yang, Chem. Eng. J., 2024, 495, 153630.
66. C. Han, J. Yang, S. Dong, L. Ma, Q. Dai and J. Guo, Sep. Purif. Technol., 2024, 128957.
67. R. Maceiras, J. Feijoo, L. Perez-Rial, V. Alfonsin and P. Falcon, Fuel, 2024, 376, 132732.
68. J. Yang, J. Liu, P. Liu, L. Li, X. Tang, H. Shang, J. Li and B. Chen, Angew. Chem., Int. Ed., 2022, 61, e202116850.
69. J. Shang, G. Li, Q. Gu, R. Singh, P. Xiao, J. Z. Liu and P. A. Webley, Chem. Commun., 2014, 50, 4544–4546.
70. Y. Wu, D. Yuan, S. Zeng, L. Yang, X. Dong, Q. Zhang, Y. Xu and Z. Liu, Fuel, 2021, 301, 121077.
71. X. Tang, Y. Wang, M. Wei, X. Zhang, Y. Li, X. Li, J. Li and J. Yang, Sep. Purif. Technol., 2023, 318, 124003.
72. J. Yang, X. Tang, J. Liu, J. Wang, H. Shang, L. Wu, J. Li and S. Deng, Chem. Eng. J., 2021, 406, 126599.
73. P. B. Lederman and B. Williams, The adsorption of nitrogen-methane on molecular sieves, 1964.
74. J. Yang, Q. Zhao, H. Xu, L. Li, J. Dong and J. Li, J. Chem. Eng. Data, 2012, 57, 3701–3709.
75. N. K. Jensen, T. E. Rufford, G. Watson, D. K. Zhang, K. I. Chan and E. F. May, J. Chem. Eng. Data, 2012, 57, 106–113.
76. G. C. Watson, N. K. Jensen, T. E. Rufford, K. I. Chan and E. F. May, J. Chem. Eng. Data, 2012, 57, 93–101.
77. J. Rocha and M. W. Anderson, Eur. J. Inorg. Chem., 2000, 2000, 801–818.
78. Y. Zhou, Y. Yuan, S. Cong, X. Liu and Z. Wang, Sep. Purif. Technol., 2022, 300, 121808.
79. S. M. Kuznicki, V. A. Bell, S. Nair, H. W. Hillhouse, R. M. Jacubinas, C. M. Braunbarth, B. H. Toby and M. Tsapatsis, Nature, 2001, 412, 720–724.
80. J. A. Delgado, M. A. Uguina, V. I. Águeda and A. García-Sanz, Langmuir, 2008, 24, 6107–6115.
81. B. Majumdar, S. Bhadra, R. Marathe and S. Farooq, Ind. Eng. Chem. Res., 2011, 50, 3021–3034.
82. S. J. Bhadra and S. Farooq, Ind. Eng. Chem. Res., 2011, 50, 14030–14045.
83. J. Yang, J. Li, W. Wang, L. Li and J. Li, Ind. Eng. Chem. Res., 2013, 52, 17856–17864.
84. S. M. Cohen, Chem. Rev., 2012, 112, 970–1000.
85. A. U. Czaja, N. Trukhan and U. Müller, Chem. Soc. Rev., 2009, 38, 1284–1293.
86. H. Furukawa, K. E. Cordova, M. O'Keeffe and O. M. Yaghi, Science, 2013, 341, 1230444.
87. Y.-M. Gu, Y.-Y. Yuan, S. Qadir, Z.-S. Yuan, S.-S. Zhao, T.-J. Sun, X.-W. Liu and S.-D. Wang, Chem. Eng. J., 2022, 433, 134447.
88. D. Lv, Y. Wu, J. Chen, Y. Tu, Y. Yuan, H. Wu, Y. Chen, B. Liu, H. Xi and Z. Li, AIChE J., 2020, 66, e16287.
89. S. M. Wang, M. Shivanna and Q. Y. Yang, Angew. Chem., Int. Ed., 2022, 61, e202201017.
90. X.-W. Liu, Y.-M. Gu, T.-J. Sun, Y. Guo, X.-L. Wei, S.-S. Zhao and S.-D. Wang, Ind. Eng. Chem. Res., 2019, 58, 20392–20400.
91. Q. Shi, J. Wang, H. Shang, H. Bai, Y. Zhao, J. Yang, J. Dong and J. Li, Sep. Purif. Technol., 2020, 230, 115850.
92. M. Chang, Y. Zhao, Q. Yang and D. Liu, ACS Omega, 2019, 4, 14511–14516.
93. Z. Niu, X. Cui, T. Pham, P. C. Lan, H. Xing, K. A. Forrest, L. Wojtas, B. Space and S. Ma, Angew. Chem., Int. Ed., 2019, 58, 10138–10141.
94. P. Liu, L. Fan, J. Li, Z. Wu, Y. Wang, Y. Chen, J. Gao, J. Yang, J. Li and B. Chen, Chem. Mater., 2024, 36, 2925–2932.
95. M. Chang, Y. Zhao, D. Liu, J. Yang, J. Li and C. Zhong, Sustain. Energy Fuels, 2020, 4, 138–142.
96. X.-W. Liu, Y. Guo, A. Tao, M. Fischer, T.-J. Sun, P. Z. Moghadam, D. Fairen-Jimenez and S.-D. Wang, Chem. Commun., 2017, 53, 11437–11440.
97. S. S.-Y. Chui, S. M.-F. Lo, J. P. Charmant, A. G. Orpen and I. D. Williams, Science, 1999, 283, 1148–1150.
98. L. Hamon, E. Jolimaître and G. D. Pirngruber, Ind. Eng. Chem. Res., 2010, 49, 7497–7503.
99. I. J. Kang, N. A. Khan, E. Haque and S. H. Jhung, Chem.–Eur. J., 2011, 17, 6437–6442.
100. M. Gaab, N. Trukhan, S. Maurer, R. Gummaraju and U. Müller, Microporous Mesoporous Mater., 2012, 157, 131–136.
101. P. Rallapalli, K. Prasanth, D. Patil, R. S. Somani, R. Jasra and H. Bajaj, J. Porous Mater., 2011, 18, 205–210.
102. J. Möllmer, M. Lange, A. Möller, C. Patzschke, K. Stein, D. Lässig, J. Lincke, R. Gläser, H. Krautscheid and R. Staudt, J. Mater. Chem., 2012, 22, 10274–10286.
103. J. Pérez-Pellitero, H. Amrouche, F. R. Siperstein, G. Pirngruber, C. Nieto-Draghi, G. Chaplais, A. Simon-Masseron, D. Bazer-Bachi, D. Peralta and N. Bats, Chem. - Eur. J., 2010, 16, 1560–1571.
104. S. Qadir, Y. Gu, S. Ali, D. Li, S. Zhao, S. Wang, H. Xu and S. Wang, Chem. Eng. J., 2022, 428, 131136.
105. L. Li, L. Yang, J. Wang, Z. Zhang, Q. Yang, Y. Yang, Q. Ren and Z. Bao, AIChE J., 2018, 64, 3681–3689.
106. M. Chang, J. Ren, Q. Yang and D. Liu, Chem. Eng. J., 2021, 408, 127294.
107. P. T. Nguyen, H. T. Nguyen, H. Q. Pham, J. Kim, K. E. Cordova and H. Furukawa, Inorg. Chem., 2015, 54, 10065–10072.
108. J. Hu, T. Sun, X. Liu, Y. Guo and S. Wang, RSC Adv., 2016, 6, 64039–64046.
109. C. E. Kivi, B. S. Gelfand, H. Dureckova, H. T. Ho, C. Ma, G. K. Shimizu, T. K. Woo and D. Song, Chem. Commun., 2018, 54, 14104–14107.
110. X. Wu, B. Yuan, Z. Bao and S. Deng, J. Colloid Interface Sci., 2014, 430, 78–84.
111. R.-J. Li, M. Li, X.-P. Zhou, S. W. Ng, M. O'Keeffe and D. Li, CrystEngComm, 2014, 16, 6291–6295.
112. H. Wu, W. Zhou and T. Yildirim, J. Am. Chem. Soc., 2009, 131, 4995–5000.
113. P. Guo, Y. Ying and D. Liu, ACS Appl. Mater. Interfaces, 2024, 16, 7338–7344.
114. Y.-M. Gu, Y.-Y. Yuan, S. Qadir, Z.-S. Yuan, S.-S. Zhao, T.-J. Sun, X.-W. Liu and S.-D. Wang, Chem. Eng. J., 2022, 433, 134447.
115. P. Zhang, J. Wang, W. Fan, Y. Zhong, Y. Zhang, Q. Deng, Z. Zeng and S. Deng, Chem. Eng. J., 2019, 375, 121931.
116. S. Du, Y. Wu, X. Wang, Q. Xia, J. Xiao, X. Zhou and Z. Li, AIChE J., 2020, 66, e16231.
117. Y.-J. Wu, Y. Yang, X.-M. Kong, P. Li, J.-G. Yu, A. M. Ribeiro and A. E. Rodrigues, J. Chem. Eng. Data, 2015, 60, 2684–2693.
118. F. Dreisbach, R. Staudt and J. Keller, Adsorption, 1999, 5, 215–227.
119. F. Yang, J. Wang, L. Liu, P. Zhang, W. Yu, Q. Deng, Z. Zeng and S. Deng, ACS Sustain. Chem. Eng., 2018, 6, 15550–15559.
120. W. Sievers and A. Mersmann, Chem. Eng. Technol., 1994, 17, 325–337.
121. S. Y. Sawant, R. S. Somani, S. S. Sharma and H. C. Bajaj, Carbon, 2014, 68, 210–220.
122. M. A. Sheikh, M. M. Hassan and K. F. Loughlin, Gas Sep. Purif., 1996, 10, 161–168.
123. Y. Hasegawa, T. Tanaka, K. Watanabe, B.-H. Jeong, K. Kusakabe and S. Morooka, Korean J. Chem. Eng., 2002, 19, 309–313.
124. M. E. Rivera-Ramos and A. J. Hernández-Maldonado, Ind. Eng. Chem. Res., 2007, 46, 4991–5002.
125. P. Harlick and F. Tezel, Sep. Purif. Technol., 2003, 33, 199–210.
126. D. Saha, Z. Bao, F. Jia and S. Deng, Environ. Sci. Technol., 2010, 44, 1820–1826.
127. M. Chang, F. Wang, Y. Wei, Q. Yang, J. X. Wang, D. Liu and J. F. Chen, AIChE J., 2022, 68, e17794.
128. X. Ren, T. Sun, J. Hu and S. Wang, Microporous Mesoporous Mater., 2014, 186, 137–145.
129. X. Jia, N. Yuan, L. Wang, J. Yang and J. Li, Eur. J. Inorg. Chem., 2018, 2018, 1047–1052.
130. T.-H. Kim, S.-Y. Kim, T.-U. Yoon, M.-B. Kim, W. Park, H. H. Han, C.-i. Kong, C.-Y. Park, J.-H. Kim and Y.-S. Bae, Chem. Eng. J., 2020, 399, 125717.
131. Y. He, S. Xiang, Z. Zhang, S. Xiong, F. R. Fronczek, R. Krishna, M. O'Keeffe and B. Chen, Chem. Commun., 2012, 48, 10856–10858.
132. X. Wang, L. Li, J. Yang and J. Li, Chin. J. Chem. Eng., 2016, 24, 1687–1694.
133. L. Li, J. Yang, J. Li, Y. Chen and J. Li, Microporous Mesoporous Mater., 2014, 198, 236–246.
134. Q. M. Wang, D. Shen, M. Bülow, M. L. Lau, S. Deng, F. R. Fitch, N. O. Lemcoff and J. Semanscin, Microporous Mesoporous Mater., 2002, 55, 217–230.
135. Y. Zhang, W. Su, Y. Sun, J. Liu, X. Liu and X. Wang, J. Chem. Eng. Data, 2015, 60, 2951–2957.
136. M.-B. Kim and P. K. Thallapally, J. Coord. Chem., 2021, 74, 216–225.
137. R. T. Yang, Adsorbents: Fundamentals and Applications, John Wiley & Sons, 2003.
138. J. A. Delgado, V. I. Águeda, J. García and S. Álvarez-Torrellas, Sep. Purif. Technol., 2019, 209, 550–559.
139. C. W. Skarstrom, Method and apparatus for fractionating gaseous mixtures by adsorption, US Pat., 2944627, 1960.
140. G. D. M. Pierre and D. Daniel, Process for separating a binary gaseous mixture by adsorption, US Pat., 3155468, 1964.
141. A. L. Kohl and R. Nielsen, Gas Purification, Elsevier, 1997.
142. H. E. Reinhold III, K. S. Knaebel, M. Huber and D. R. King, Separation of gases by pressure swing adsorption, US Pat., 5792239, 1998.
143. A. Jayaraman, A. J. Hernandez-Maldonado, R. T. Yang, D. Chinn, C. L. Munson and D. H. Mohr, Chem. Eng. Sci., 2004, 59, 2407–2417.
144. A. Jayaraman, New Nano-Structured Sorbent Materials for Gas Separation, Purification and Storage, University of Michigan, 2004.
145. A. Olajossy, A. Gawdzik, Z. Budner and J. Dula, Chem. Eng. Res. Des., 2003, 81, 474–482.
146. J. A. Delgado, M. A. Uguina, J. L. Sotelo and B. Ruiz, Sep. Purif. Technol., 2006, 50, 192–203.
147. J. A. Delgado, M. A. Uguina, J. L. Sotelo, V. I. Águeda and P. Gómez, Sep. Purif. Technol., 2011, 77, 7–17.
148. Y. Zhou, Q. Fu, Y. Shen, W. Sun, D. Zhang, D. Li and H. Yan, Energy Fuels, 2016, 30, 1496–1509.
149. H. Shang, H. Bai, X. Li, J. Li and J. Yang, Sep. Purif. Technol., 2022, 295, 121271.
150. Y. Guo, X. Gu, G. Hu, P. A. Webley and G. K. Li, Sep. Purif. Technol., 2022, 301, 121907.
151. B. Lu, Y. Shen, Z. Tang, D. Zhang and G. Chen, Chin. J. Chem. Eng., 2021, 32, 264–280.
152. V. I. Á. Maté, J. A. D. Dobladez, S. Álvarez-Torrellas, M. Larriba and Á. M. Rodríguez, Chem. Eng. J., 2019, 361, 1007–1018.
153. G. Hu, Q. Zhao, L. Tao, P. Xiao, P. A. Webley and K. G. Li, Chem. Eng. Sci., 2021, 229, 116152.
154. Y. Liu, X. Yang, Y. Li, H. Yang, C. Zhang and Y. Meng, Energy Educ. Sci. Technol., Part A, 2012, 29, 217–226.
155. S. Ouyang, S. Xu, N. Song and S. Jiao, Fuel, 2013, 113, 420–425.
156. X. Yang, Y. Liu, Z. Li, C. Zhang and Y. Xing, Energies, 2018, 11, 1030.
157. J.-S. Bae, S. Su and X. X. Yu, Environ. Sci. Technol., 2014, 48, 6043–6049.
158. J.-S. Bae, S. Su, X. X. Yu, J. Yin, A. Villella, M. Jara and M. Loney, Ind. Eng. Chem. Res., 2020, 59, 15732–15741.
159. J.-S. Bae, S. Su and X. X. Yu, Ind. Eng. Chem. Res., 2019, 58, 21700–21707.
160. M. A. Carreon, J. Mater. Res., 2018, 33, 32–43.
161. T. C. Bowen, R. D. Noble and J. L. Falconer, J. Membr. Sci., 2004, 245, 1–33.
162. B. Smitha, D. Suhanya, S. Sridhar and M. Ramakrishna, J. Membr. Sci., 2004, 241, 1–21.
163. D. Q. Vu, W. J. Koros and S. J. Miller, J. Membr. Sci., 2003, 211, 311–334.
164. T. Remy, S. A. Peter, L. Van Tendeloo, S. Van der Perre, Y. Lorgouilloux, C. E. Kirschhock, G. V. Baron and J. F. Denayer, Langmuir, 2013, 29, 4998–5012.
165. C. A. Scholes, G. W. Stevens and S. E. Kentish, Fuel, 2012, 96, 15–28.
166. S. Li, J. L. Falconer, R. D. Noble and R. Krishna, Ind. Eng. Chem. Res., 2007, 46, 3904–3911.
167. G. Guan, K. Kusakabe and S. Morooka, Microporous Mesoporous Mater., 2001, 50, 109–120.
168. S. S. Hosseini and T. S. Chung, J. Membr. Sci., 2009, 328, 174–185.
169. N. Gilani, J. T. Daryan, A. Rashidi and M. R. Omidkhah, Appl. Surf. Sci., 2012, 258, 4819–4825.
170. O. Shekhah, J. Liu, R. Fischer and C. Wöll, Chem. Soc. Rev., 2011, 40, 1081–1106.
171. D. Zacher, O. Shekhah, C. Wöll and R. A. Fischer, Chem. Soc. Rev., 2009, 38, 1418–1429.
172. Q. Song, S. Nataraj, M. V. Roussenova, J. C. Tan, D. J. Hughes, W. Li, P. Bourgoin, M. A. Alam, A. K. Cheetham and S. A. Al-Muhtaseb, Energy Environ. Sci., 2012, 5, 8359–8369.
173. M. C. McCarthy, V. Varela-Guerrero, G. V. Barnett and H.-K. Jeong, Langmuir, 2010, 26, 14636–14641.
174. H. Bux, F. Liang, Y. Li, J. Cravillon, M. Wiebcke and J. r. Caro, J. Am. Chem. Soc., 2009, 131, 16000–16001.
175. J. Wang, Y. Ying, P. Guo, D. Liu, W. Zhang and Q. Yang, J. Membr. Sci., 2025, 713, 123387.
176. Q. Song, S. K. Nataraj, M. V. Roussenova, J. C. Tan, D. J. Hughes, W. Li, P. Bourgoin, M. A. Alam, A. K. Cheetham, S. A. Al-Muhtaseb and E. Sivaniah, Energy Environ. Sci., 2012, 5, 8359.
177. Y.-S. Li, F.-Y. Liang, H. Bux, A. Feldhoff, W.-S. Yang and J. Caro, Angew. Chem., Int. Ed., 2010, 49, 548–551.
178. Y. Li, F. Liang, H. Bux, W. Yang and J. Caro, J. Membr. Sci., 2010, 354, 48–54.
179. H. Bux, F. Liang, Y. Li, J. Cravillon, M. Wiebcke and J. Caro, J. Am. Chem. Soc., 2009, 131, 16000–16001.
180. H. Bux, A. Feldhoff, J. Cravillon, M. Wiebcke, Y.-S. Li and J. Caro, Chem. Mater., 2011, 23, 2262–2269.
181. A. Huang, H. Bux, F. Steinbach and J. Caro, Angew. Chem., Int. Ed., 2010, 49, 4958–4961.
182. A. Huang, W. Dou and J. r. Caro, J. Am. Chem. Soc., 2010, 132, 15562–15564.
183. A. Huang and J. Caro, Angew. Chem., Int. Ed., 2011, 50, 4979–4982.
184. J. Brennan and I. L. Spain, in High Pressure Technology, Routledge, 2017, pp. 253–291.
185. R. F. Pahade, J. B. Saunders and J. J. Maloney, Process for separating methane and nitrogen, US Pat., 4592767, 1986.
186. S. Sanaye and S. M. S. Ghoreishi, Cryogenics, 2019, 102, 66–76.
187. M. Waqar, H. Ishaq and S. R. Jamil, Energy Convers. Manage.:X, 2022, 14, 100200.
188. G. B. Zhang, S. S. Fan, X. M. Lang and Y. H. Wang, Adv. Mater. Res., 2012, 403, 585–592.
189. D. Qu, Y. Yang, Z. Qian, P. Li, J. Yu, A. M. Ribeiro and A. E. Rodrigues, Chem. Eng. J., 2020, 380, 122509.
190. G. Hu, Y. Guo, J. Luo, G. Xiao, R. Weh, K. G. Li, T. Qi, P. A. Webley and E. F. May, AIChE J., 2025, 71, e18573.
191. G. Hu, G. Xiao, Y. Guo, M. Manning, L. Chen, L. Yu, K. G. Li and E. F. May, AIChE J., 2022, 68, e17668.
192. Y. Yang, X. Wang, X. Li, Y. Zhou, Y. Li, X. Bai, J. Li and J. Yang, Chem. Eng. J., 2025, 507, 160858.
193. J. Liu, H. Shang, J. Yang, J. Wang, J. Li and S. Deng, Chem. Eng. J., 2021, 426, 130163.
194. G. Cui, Z.-L. Li and Y.-L. Zhao, RSC Adv., 2015, 5, 68218–68226.
195. T. Gao, W. Lin, W. Liu, A. Gu and Y. Qi, J. Chem. Eng. Jpn., 2011, 44, 555–560.
196. L. Xiao and J. Chen, Processes, 2021, 9, 606.
197. K. A. Lokhandwala, I. Pinnau, Z. He, K. D. Amo, A. R. DaCosta, J. G. Wijmans and R. W. Baker, J. Membr. Sci., 2010, 346, 270–279.
198. S. Li, Z. Zong, S. J. Zhou, Y. Huang, Z. Song, X. Feng, R. Zhou, H. S. Meyer, M. Yu and M. A. Carreon, J. Membr. Sci., 2015, 487, 141–151.
199. H. Zentou, M. Aliyu, M. A. Abdalla, O. Y. Abdelaziz, B. Hoque, A. M. Alloush, I. M. Tayeb, K. Patchigolla and M. M. Abdelnaby, Chem. Rec., 2025, 25, e202400188.
200. A. Risco-Bravo, C. Varela, J. Bartels and E. Zondervan, Renew. Sustain. Energy Rev., 2024, 189, 113930.
201. B. Shingan, M. Pujari, A. K. Arya and V. Singh, Chem. Eng. Technol., 2024, 47, e202400085.
202. S. Ganesan, R. L. Canevesi and C. A. Grande, Ind. Eng. Chem. Res., 2024, 63, 18497–18512.

---