Amine-functionalised silica composites

The functionalisation of amine species onto mesoporous silica, a support material, for CO₂ capture has garnered significant attention (Gargiulo et al., Reference Gargiulo, Pepe and Caputo2014). This amine-functionalised silica composite leverages the high CO₂ affinity of amine species and mesoporous silica’s large surface area and pore volume, ensuring uniform distribution of amine species within its porous structure (Chen et al., Reference Chen, Zhang, Row and Ahn2017). Among various porous support materials, mesoporous silica stands out for its suitability in surface functionalisation of amine species due to its large surface area, substantial pore volume, uniform pore size and the ease of surface chemical modification to accommodate the steric hindrance of amine species (Impens et al., Reference Impens, van der Voort and Vansant1999; Asefa and Tao, Reference Asefa and Tao2012).

Mesoporous silica is commonly functionalised with amine species using two primary methods: post-synthetic (wet impregnation or chemical grafting) and direct synthesis (one-pot synthesis) (Chen et al., Reference Chen, Zhang, Row and Ahn2017). Commonly used amines for this purpose include tetraethylenepentamine (TEPA), triethylenetramine (TETA), pentaethylenehexamine (PEHA), branched PEI (BPEI) and linear PEI (LPEI). Wet impregnation is a straightforward and mild synthesis method. In this process, porous silica is first prepared and then soaked in an alcohol solvent containing the desired amount of amine. After soaking, the solvent is dried, resulting in an amine-impregnated silica composite (Xu et al., Reference Xu, Song, Andrésen, Miller and Scaroni2003). Figure 1 illustrates the schematic diagram of the wet impregnation method used to produce an amine-impregnated silica composite. This method is advantageous because it allows a large amount of amine to be incorporated into the silica’s large pore volume (Chen et al., Reference Chen, Zhang, Row and Ahn2017). However, the lack of chemical bonding between the amine and silica support can affect the stability of the composite, potentially hindering its practical application (Chen et al., Reference Chen, Zhang, Row and Ahn2017).

Figure 1.

Schematic diagram of the wet impregnation process for producing amine-impregnated mesoporous silica. Reproduced from brgfx (2024).

Figure 1. Long description

The flowchart consists of three main stages connected by green arrows.

1. Top-left stage: Wet impregnation of amine. A glass beaker contains an amine-containing alcohol solvent. Submerged at the bottom are three hexagonal mesoporous silica particles, each containing a grid of white circular pores.

2. Top-right stage: Drying of residual amine-containing alcohol solvent. The beaker is placed inside a blue heating chamber. Two brown heating elements on the left and right sides emit horizontal arrows labeled Heat toward the beaker. Three vertical arrows point upward from the liquid surface, labeled Evaporation of solvent.

3. Bottom-right stage: Amine-impregnated mesoporous silica. A large-scale view of a single hexagonal mesoporous silica particle. The internal circular pores are now filled with red shapes representing amine functional groups. A circular magnifying glass callout highlights a cluster of these red functional groups within the pores.

Chemical grafting anchors amine species onto the surface of silica support through the formation of covalent bonds, achieved by heating under reflux or using a sonication-assisted method (Knowles et al., Reference Knowles, Chaffee and Chaffee2016; Chen et al., Reference Chen, Zhang, Row and Ahn2017; Hack et al., Reference Hack, Maeda and Meier2022). There are two approaches reported in the literature for preparing amine-grafted silica adsorbents. The first approach involves attaching amine species to the silica surface via a condensation reaction with hydroxyl groups on the silica support (Figure 2) (Chen et al., Reference Chen, Zhang, Row and Ahn2017). The amount of amine incorporated depends on the number of hydroxyl groups on the silica surface (Chen et al., Reference Chen, Zhang, Row and Ahn2017). The second approach involves soaking silica in anhydrous toluene and adding an amino silane to the solution (Huang et al., Reference Huang, Yang, Chinn and Munson2003; Chen et al., Reference Chen, Zhang, Row and Ahn2017). The mixture is then heated under reflux to form an amine-functionalised silica composite (Huang et al., Reference Huang, Yang, Chinn and Munson2003; Chen et al., Reference Chen, Zhang, Row and Ahn2017). The composite is washed several times to remove unreacted amino silane before application (Huang et al., Reference Huang, Yang, Chinn and Munson2003). Unlike the first approach, this method incorporates amine species through siloxane (Si-O-Si) bridges to other grafted amino silane (Chen et al., Reference Chen, Zhang, Row and Ahn2017). In addition to amino silane, other grafting agents such as aziridine (C₂H₅N), PEI, ammonium hydroxide (NH₄OH) and polyacrylic acid (PAA) can be used for amine functionalisation within mesoporous silica (Hicks et al., Reference Hicks, Drese, Fauth, Gray, Qi and Jones2008; Du et al., Reference Du, Du, Zou, Li, Mi and Zhang2013; Ullah et al., Reference Ullah, Atilhan, Aparicio, Canlier and Yavuz2015).

Figure 2.

Schematic diagram of the anhydrous grafting of amine species onto the silica surface through a condensation reaction with hydroxyl groups on the silica support. (R group = triamine chains). Reproduced from Chen et al. (Reference Chen, Zhang, Row and Ahn2017).

Figure 2. Long description

The schematic is organised into three main vertical stages within a rounded frame, with a structural reference to the right.

* At the top left, a box labeled Aminosilane contains a chemical structure with a central S i atom bonded to an R group at the top and three C H sub 3 O groups at the bottom. A thick red L-shaped arrow points from this box down toward the silica surface, with text stating Attached to silica surface via anhydrous grafting.

* In the middle, a horizontal rectangle represents the Pore wall of Hydrated mesoporous silica. The top surface of this wall is lined with twelve O H hydroxyl groups.

* A red downward arrow indicates the transition to the bottom stage, labeled Amine-grafted mesoporous silica. Here, the Pore wall now has four aminosilane molecules grafted to it. The grafting occurs through oxygen atoms. Some sites show the S i atom bonded to two oxygen atoms on the surface, while others show one oxygen bond and remaining C H sub 3 O groups, with occasional residual O H groups on the surface.

* To the right of the main frame, a blue hexagonal icon with white circular pores is labeled Mesoporous silica. A small gray triangle points from one of these pores into the main reaction diagram, indicating that the schematic represents a microscopic view of a single pore wall.

On the other hand, amine-functionalised silica composites (also known as amine-modified silica composites) can be prepared through a one-pot synthesis. The synthesis involves the hydrolysis and co-condensation reaction between an organosilane and an amino silane, using an acid or base catalyst with an organic template (Gunathilake and Jaroniec, Reference Gunathilake and Jaroniec2014; Kong et al., Reference Kong, Jiang, Fan, Shen, Cui and Russell2014; Chen et al., Reference Chen, Zhang, Row and Ahn2017), as illustrated in Figure 3. Furthermore, some studies employ a double-functionalisation method to synthesise amine-modified silica composites. This involves first chemically grafting amine onto the silica surface, followed by incorporating additional amine species via the wet impregnation method to fill the remaining pore volume in the silica support (Fujiki et al., Reference Fujiki, Yamada and Yogo2015; Sanz et al., Reference Sanz, Calleja, Arencibia and Sanz-Pérez2015; Wang et al., Reference Wang, Chen and Guo2015; Jung et al., Reference Jung, Lee, Jeon, Jo, Huh and Kim2016).

Figure 3.

Schematic diagram illustrating the one-pot synthesis of amine-modified silica composite. Reproduced from Xie et al. (Reference Xie, Hill, Xiao, Militz and Mai2010) and Hubbe et al. (Reference Hubbe, Rojas and Lucia2015) [From “Green modification of surface characteristics of cellulosic materials at the molecular or nano scale: A review,” by M A Hubbe, O J Rojas and L A Lucia, 2015, BioResources, 10(3), p. 6095-6206 (10.15376/biores.10.3.Hubbe). Copyright 2015 by the Author(s).].

Figure 3. Long description

The diagram is organized into three vertical stages.

1. Top Section: Two chemical equations enclosed in a red box.

- The first equation shows Hydrolysis: R prime S i (O R) sub 3 plus 3 H sub 2 O, with an arrow labeled Acid or Base Catalyst, yields R prime S i (O H) sub 3 plus 3 R O H.

- The second equation shows Self-condensation: R prime Si(OH)₃ plus O H S i R double prime (O H) sub 2 yields R prime S i O S i R double prime (O H) sub 2 plus H sub 2 O.

2. Middle Section: A green arrow points from the equations to a blue hexagonal structure labeled As-synthesized mesoporous silica. This hexagon is filled with dark circular clusters. A circular inset to the right shows an Organic template represented by a dark circle with a wavy tail, with an arrow pointing into the hexagonal structure.

3. Bottom Section: A green arrow points down to a final blue hexagonal structure labeled Mesoporous silica, which contains empty white circular pores. A circular inset to the left, labeled Solvent extraction of organic template, shows the dark wavy-tailed clusters being removed from the structure, indicated by a red arrow pointing away from the hexagon.

Min et al. (Reference Min, Choi and Choi2017) investigated the CO₂ adsorption capabilities of various PEI-functionalised mesoporous silicas. At 40 °C and 0.15 atm, PEI/macroporous silica (MacS) exhibited a higher CO₂ adsorption capacity of 3.84 mmol g⁻¹ compared to PEI/mesocellular silica foam (MCF) (2.96 mmol g⁻¹), PEI/ Santa Barbara Amorphous-15 (SBA-15) (2.53 mmol g⁻¹) and PEI/ Mobil Composition of Matter No. 41 (MCM-41) (0.91 mmol g⁻¹). The extra-large pore diameter (56 nm) and three-dimensional porous structure of MacS minimise pore obstruction by the functionalised PEI, enhancing amine utilisation for CO₂ adsorption. PEI/MacS demonstrated excellent stability, maintaining 90% of its CO₂ adsorption capacity (3.46 mmol g⁻¹) after a 14-day steam treatment at 120 °C.

Fayaz and Sayari (Reference Fayaz and Sayari2017) synthesised 3-[2-(2-aminoethylamino)ethyl amino] propyltrimethoxysilane (triamine)-grafted commercial-grade silica (CARiACT, P10) under optimal grafting conditions using 1.5 cm³ amino silane per g silica and 0.2 cm³ water. This achieved maximum CO₂ adsorption capacities of 1.41 mmol g⁻¹ at 25 °C, 1.93 mmol g⁻¹ at 50 °C and 1.23 mmol g⁻¹ at 75 °C before steam treatment. Increasing steam exposure duration from 3 to 24 h reduced the CO₂ adsorption capacity of triamine-grafted P10 by 56% at 25 °C (0.61 mmol g⁻¹), 21% at 50 °C (1.47 mmol g⁻¹) and 4% at 75 °C (1.18 mmol g⁻¹). Extending steam exposure to 360 h further decreased the CO₂ adsorption capacity by 83% (0.21 mmol g⁻¹), 61% (0.70 mmol g⁻¹) and 26% (0.87 mmol g⁻¹) at 25 °C, 50 °C and 75 °C, respectively. The reduction in CO₂ adsorption capacity with increasing steam exposure time at these temperatures was attributed to decreased average pore width of triamine-grafted P10 and increased diffusional mass transfer resistance.

Supplementary Tables S1(a)–(c) tabulate the CO₂ adsorption performance of amine-impregnated silica composites, amine-grafted silica composites and amine-modified silica composites under various operating conditions. Table 1 provides a detailed comparison of these three groups in terms of their maximum CO₂ adsorption capacity, as summarised in Supplementary Tables S1(a)–(c). As shown in Table 1, the amine-grafted silica composite exhibits a wide adsorption capacity range (0.18–110.64 mmol g⁻¹), while the amine-impregnated silica composite demonstrates a narrower but generally higher overall CO₂ adsorption capacity (1.85–6.10 mmol g⁻¹). The amine-modified silica composite shows a range of 0.80–3.98 mmol g⁻¹. Notably, the reported maximum uptake of 110.64 mmol g⁻¹ for ammonium hydroxide (NH₄OH)-modified SBA-15 (MSBA-15) was measured under 200 atm pressure at 45 °C (Ullah et al., Reference Ullah, Atilhan, Aparicio, Canlier and Yavuz2015). This exceptionally high value reflects a high-pressure special case and should not be interpreted as representative of post-combustion CO₂ capture conditions. The increase from 2.02 to 110.64 mmol g⁻¹ under elevated pressure highlights the strong CO₂ affinity of NH₂ groups and enhanced accessibility of CO₂ to attractive sites within the pores of MSBA-15 (Ullah et al., Reference Ullah, Atilhan, Aparicio, Canlier and Yavuz2015).

Table 1.

Comparative analysis of CO₂ capture performance in hybrid absorption–adsorption systems using amine-impregnated, amine-grafted and amine-modified silica

Table 1. Long description

The table consists of seven columns: Absorbent material matrix, Operating temperature in degrees Celsius, Total pressure P_total in atm, CO₂ partial pressure P_total CO₂ in atm or feed, Minimum CO₂ adsorption capacity in mmol g super minus 1, Maximum CO₂ adsorption capacity in mmol g super minus 1, and References.

* Amine-impregnated silica composite section: Includes 12 materials. Notable entries include TEPA-modified hydrophilic silica aerogel sorbent (S A-I) with a maximum capacity of 6.10 mmol g super minus 1 at 75 degrees Celsius, and T E P A-D E A impregnated M S U-F with a minimum capacity of 4.42 mmol g super minus 1.

* Amine-grafted silica composite section: Includes 12 materials. The highest recorded capacity in the entire table is found here: Ammonium hydroxide modified S B A-15 (M S B A-15) with a maximum capacity of 110.64 mmol g super minus 1 at a pressure range up to 197.4 atm. Other materials like M C M-41-N H sub 2 and A 2-B P M O show lower capacities ranging from 0.18 to 3.86 mmol g super minus 1.

* Amine-modified silica composite section: Includes 5 materials. P E I-functionalised mesoporous silica powders (P E I-M S P-0.32) show a maximum capacity of 3.98 mmol g super minus 1 at temperatures between 35 and 95 degrees Celsius. P S A and S G A adsorbents show lower maximum capacities around 0.80 to 1.02 mmol g super minus 1.

Data marked as n/a indicates values not available or not explicitly reported in the cited literature.

Note : n/a: Data not available or not explicitly reported in the cited literature.

The CO₂ adsorption performance of amine-functionalised silica composite is influenced by multiple factors, including adsorption temperature, moisture content, CO₂ concentration, CO₂ flow rate and adsorption pressure. As shown in Supplementary Tables S1(a)–(c), moisture content, CO₂ concentration and adsorption pressure generally enhance the adsorption capacity of amine-grafted and amine-modified silica composites. In contrast, CO₂ flow rate and adsorption temperature improve adsorption capacity of amine-grafted and amine-modified silica composites only up to a threshold, beyond which further increases in these parameters may lead to a decrease in capacity as contrary to the behaviour observed in amine-impregnated silica composites.

Carbon-based nanoparticles/liquid amine nano-fluids

Nano-fluid is a novel material synthesised through the dispersion of nanometre-sized materials, with length, diameter or thickness ranging from 1–100 nm, into a base fluid, which can be water-soluble or non-water-soluble liquids (Ma and Banerjee, Reference Ma, Banerjee and Balasubramanian2018; Aghel et al., Reference Aghel, Janati, Alobaid, Almoslh and Epple2022). This review focuses on the preparation of carbon-based nanoparticles/liquid amine nano-fluids, as illustrated in Figure 4, for hybrid CO₂ absorption–adsorption. The addition of carbon-based materials such as carbon nanotubes (CNTs) and graphene oxide (GO) as nanoparticles into aqueous amine solvents (e.g., MDEA and diethanolamine [DEA]) has been extensively investigated to increase the liquid absorbents CO₂ absorption efficiency. The unique characteristics of carbon-based nanoparticles – modifiable physical and chemical properties for specialised applications, high specific surface area and high porosity – enhance the CO₂ absorption efficiency of the liquid amine via four mechanisms: grazing effect, hydrodynamic effect, coalescence inhibition and Brownian motion of nanometre adsorbents. These mechanisms facilitate the adsorption of CO₂ and reduce energy consumption in the regeneration process (Pineda et al., Reference Pineda, Choi and Kang2014; Yu et al., Reference Yu, Wang, Park and Fang2019; Aghel et al., Reference Aghel, Janati, Alobaid, Almoslh and Epple2022).

Figure 4.

Schematic diagram of preparation of carbon-based nanoparticles/liquid amine nanofluid. Reproduced from Zunita et al. (Reference Zunita, Makertiharta, Irawanti, Prasetya and Wenten2018) and brgfx (2024). From “Graphene Oxide-Inorganic Composite Membrane: A Review,” by M Zunita, I G B N Makertiharta, R Irawanti, N Prasetya, I G Wenten, 2018, IOP Conference Series : Materials Science & Engineering, 395, p. 1-14 (10.1088/1757-899X/395/1/012005). Copyright 2018 by the Author(s). This work is licensed under CC BY 3.0.

Figure 4. Long description

The diagram is organized into three main stages.

1. Top Section: A rounded rectangular box labeled Carbon-based nanoparticles contains two molecular structures. On the left is a spherical lattice model labeled Carbon nanotubes. On the right is a chemical structure of Graphene oxide G O, showing a hexagonal carbon lattice with various functional groups including O H, O, and C O O H.

2. Middle Section: A red downward-pointing arrow labeled Dispersion process connects the top box to a beaker. The beaker contains a clear blue liquid labeled Base fluid Amine solvent.

3. Bottom-Right Section: A green horizontal arrow points from the first beaker to a second beaker. This final beaker contains the same blue liquid but now features numerous yellow circular dots suspended throughout the fluid. This is labeled Carbon-based nanoparticles forward slash liquid amine nanofluid.

In the absorption of CO₂ using carbon-based nanoparticles (such as CNTs or GO) dispersed in liquid amine nano-fluids, four primary mechanisms have been proposed to explain the observed performance enhancements. To rigorously evaluate these phenomena, it is critical to separate kinetic mass-transfer enhancements (absorption rates) from thermodynamic equilibrium loading (ultimate capacity). Experimental data from the primary literature indicate that the incorporation of carbon-based nanoparticles predominantly drives gas–liquid transport and absorption kinetics rather than changing the baseline chemical equilibrium of the host amine solvent (Cheng et al., Reference Cheng, Liu and Qi2020).

(iv) Brownian motion-induced micro-turbulence (micro-mixing kinetics)

The intrinsic Brownian motion of carbon-based nanoparticles induces micro-scale localised turbulence within the liquid phase (Cheng et al., Reference Cheng, Liu and Qi2020). This random, continuous motion acts as a micro-mixer, promoting the continuous breakup of CO₂ bubbles into smaller fragments, further expanding the gas–liquid interfacial area while reducing bubble collisions (Cheng et al., Reference Cheng, Liu and Qi2020). The kinetic contribution of Brownian motion is most pronounced at smaller particle diameters and moderate operating temperatures. This is indirectly supported by correlations between particle size and mass transfer rate enhancements (Cheng et al., Reference Cheng, Liu and Qi2020), which exert no measurable impact on the final equilibrium uptake capacity.

Among the four mechanisms, bubble coalescence inhibition and Brownian motion are considered the dominant drivers under typical operating conditions. Both explicitly maximise the active gas–liquid interfacial area and lower mass transfer resistance (kinetic parameters). Conversely, boundary-layer disruption and interfacial adsorption function as supplementary kinetic mechanisms, becoming pronounced primarily at elevated nanoparticle loadings.

Characterisation profiles – ranging from dynamic absorption rate measurements and bubble visualisation to size distribution studies – confirm that the observed performance boost is fundamentally a kinetic and transport-rate enhancement (Cheng et al., Reference Cheng, Liu and Qi2020). Because these nanoparticles function as physical transport promoters rather than chemical reactants, they lower the time required to reach saturation without altering the intrinsic equilibrium CO₂ loading of the underlying amine solvent. Given that this systematic framework is currently supported by limited literature (Cheng et al., Reference Cheng, Liu and Qi2020), further independent validation is required to isolate these transport phenomena and map their kinetic boundaries under broader operating limits.

The two-step approach is the most common method used to prepare carbon-based nanoparticles/liquid amine nano-fluids for CO₂ absorption (Tavakoli et al., Reference Tavakoli, Rahimi, Saghandali, Scott and Lovell2022). Dispersion of carbon-based nanoparticles into the liquid amines is performed after the synthesis process using chemical (sol–gel method), physical (vapour phase method) or mechanical (grinding and ball milling) methods (Shen et al., Reference Shen, Shih and Tung2008; Joni et al., Reference Joni, Purwanto, Iskandar and Okuyama2009; Elsalamony et al., Reference Elsalamony, Morsi and Alsabagh2015). Techniques such as ultrasonic vibration, ultrasonication, ultrasonic agitation, microwave irradiation, ultrasonic bath, piezo-actuated and solenoid-actuated baths and a static bath with an immersed sonic horn are employed (Peyravi et al., Reference Peyravi, Keshavarz and Mowla2015; Aghel et al., Reference Aghel, Janati, Alobaid, Almoslh and Epple2022).

Despite the advantages of the two-step method (i.e., low cost, simple preparation process, large-scale production of nanoparticles and suitability for industrial production) (Sarkar et al., Reference Sarkar, Ghosh and Adil2015), this approach has disadvantages due to rapid nano-fluid agglomeration caused by the large and active surface area of the nanoparticles. This agglomeration leads to clumping and subsequent sedimentation, which reduces the stability of the prepared nano-fluids and results in weak physicochemical characteristics (Drzazga et al., Reference Drzazga, Lemanowicz, Dzido and Gierczycki2012). To achieve better dispersion performance of nanoparticles, some works combine dispersing methods such as ultrasonication, magnetic stirring, ultrasound vibration, pH adjustment of the base fluid and high-pressure homogenisers in the two-step preparation method (Zhang et al., Reference Zhang, Cai, Chen, Li, Zhang and Qi2018; Tavakoli et al., Reference Tavakoli, Rahimi, Saghandali, Scott and Lovell2022).

Irani et al. (Reference Irani, Maleki and Tavasoli2019) reported that dispersing 0.1 and 0.2 mg mL⁻¹ of GO in 40% MDEA via ultrasonication increased CO₂ absorption capacity by 9.4% and 10.1%, respectively, achieving 4.27 and 4.30 mmol g⁻¹ compared to 3.90 mmol g⁻¹ for 40% MDEA without GO at 30 °C. The collision of GO with the gas–liquid interface reduced the gap between the phases, enhancing the diffusion coefficient through hydrodynamic effects. GO nanoparticles adhered to the CO₂ bubble surface, preventing bubble coalescence and promoting Brownian motion collisions with the gas–liquid interface and other GO nanoparticles (Cheng et al., Reference Cheng, Liu and Qi2020). These mechanisms broke CO₂ bubbles into smaller ones, increasing the gas–liquid interface area and promoting CO₂ diffusion into the MDEA solution, thereby enhancing its CO₂ absorption capacity (Cheng et al., Reference Cheng, Liu and Qi2020). During absorption, GO penetrated the mass transfer membrane, adsorbing CO₂ onto its active sites and transporting CO₂ gas into the liquid bulk via the grazing effect, resulting in higher gas absorption (Cheng et al., Reference Cheng, Liu and Qi2020).

Khumalo et al. (Reference Khumalo, Mahlangu, Mamba and Motsa2024) found that the CO₂ absorption rate constant increased by 49% and 34% for nano-fluids containing 0.2 mg mL⁻¹ GO in 30% MEA (0.17 [1/min^0.5]) and 30% EDA (0.18 [1/min^0.5]) solutions, respectively, compared to MEA (0.11 [1/min^0.5]) and EDA (0.13 [1/min^0.5]) solutions without GO. This enhancement was attributed to the Brownian motion of GO nanoparticles, which improved hydrodynamic effects and prevented bubble coalescence, thereby increasing the CO₂ absorption rate (Khumalo et al., Reference Khumalo, Mahlangu, Mamba and Motsa2024). The regeneration of 30% MEA-0.2 mg mL⁻¹ GO nano-fluid showed a minimal decrease in the CO₂ absorption rate due to applied heat, with a reduction of 1.4% in the first and second cycles and 2.8% in the third cycle (Khumalo et al., Reference Khumalo, Mahlangu, Mamba and Motsa2024). The GO nanoparticles in the 30% MEA solution significantly improved the stability of the nano-fluid, maintaining a similar CO₂ absorption rate constant over three regeneration cycles, compared to 30% EDA-0.2 mg mL⁻¹ GO nano-fluid, MEA and EDA solutions. The absorption rate constant for 30% EDA-0.2 mg mL⁻¹ GO nano-fluid, MEA and EDA solutions decreased by 7.7%, 2.2% and 2.9% in the first cycle, 5.7%, 2.7% and 1.7% in the second cycle and 4.6%, 5.8% and 5.6% in the third cycle, which were slightly higher than that of 30% MEA-0.2 mg mL⁻¹ GO nano-fluid (Khumalo et al., Reference Khumalo, Mahlangu, Mamba and Motsa2024).

The CO₂ capture performance of carbon-based nanoparticles/liquid amine nano-fluids via an absorption–adsorption hybrid process is detailed in Supplementary Table S2. Table 4 provides an overview of various types of carbon-based nanoparticles/liquid amine nano-fluids, highlighting their highest CO₂ absorption capacities, recyclability and regeneration costs. Functionalising GO with amine involves the attachment of amine groups (absorption) onto the GO surface (adsorption), followed by their dispersion within the amine matrix (absorption), enhancing the overall functional integration. According to Table 4, functionalising GO with amine resulted in a higher CO₂ absorption capacity (4.50 mmol g⁻¹) compared to pristine GO (4.07 mmol g⁻¹) at a 0.1 wt.% nanoparticle concentration. The amine functional groups chemically interact with CO₂ via adsorption, creating intermediates that produce protonated groups and carbamate on the NH₂-mGO surface (Lashgarinejad et al., Reference Lashgarinejad, Hosseini, Irani, Ghasemi, Mohammadpour and Tavasoli2023). This stabilises CO₂ on the NH₂-mGO surface, facilitating its transport into the liquid bulk (Lashgarinejad et al., Reference Lashgarinejad, Hosseini, Irani, Ghasemi, Mohammadpour and Tavasoli2023). Adsorption and desorption of CO₂ on the NH₂-mGO surface occur in the diffusion layer and liquid bulk, respectively (Lashgarinejad et al., Reference Lashgarinejad, Hosseini, Irani, Ghasemi, Mohammadpour and Tavasoli2023). NH₂-mGO effectively recovers and retains its active sites over repeated absorption cycles. It can be recovered using a magnet, followed by thermal regeneration at 150 °C under vacuum overnight. The CO₂ absorption of 40 wt.% aqueous MDEA nano-fluid was enhanced with the addition of NH₂-mGO by 19.0% in the first cycle, 18.9% in the second cycle, 18.6% in the third cycle, 18.4% in the fourth cycle and 18.3% in the fifth cycle. The 0.1 wt.% NH₂-mGO/40 wt.% aqueous MDEA nano-fluid demonstrated good recyclability, maintaining a CO₂ absorption capacity of 4.47–4.50 mmol g⁻¹ over five absorption cycles, as evidenced in Table 4. Furthermore, the CO₂ absorption capacity of 0.1 wt.% NH₂-mGO/40 wt.% aqueous MDEA nano-fluid decreased with increasing absorption temperature but increased with higher CO₂ partial pressure. Higher temperatures reduced CO₂ solubility, while increased CO₂ partial pressure enhanced CO₂ solubility in the nano-fluid, thereby decreasing the CO₂ absorption rate of the nano-fluid (Irani et al., Reference Irani, Maleki and Tavasoli2019).

Table 4.

Comparative analysis of CO₂ capture performance in hybrid absorption–adsorption systems using carbon nanotube (CNT) nanoparticles/liquid amine nano-fluid and GO nanoparticles/liquid amine nano-fluid

Table 4. Long description

The table consists of 9 columns: Carbon-based nanoparticles/liquid amine nano-fluids, Operating temperature in degrees Celsius, Total pressure P_total total in atm, CO₂ partial pressure P_total CO₂ in atm / feed, Minimum CO₂ absorption capacity in mmol g super minus 1, Maximum CO₂ absorption capacity in mmol g super minus 1, CO₂ absorption capacity over multiple cycles in mmol g super minus 1, and Reference.

Key data points include:

* 0.02 wt. % CNT solution with 5 wt. % D E A solution at 35 degrees Celsius: Max capacity of 0.83 mmol g super minus 1.

* 0.02 wt. % CNT solution with 10 wt. % M D E A solution at 35 degrees Celsius: Max capacity of 0.88 mmol g super minus 1.

* 0.1 wt. % G O with 40 wt. % aqueous M D E A solution at 30 degrees Celsius: Max capacity of 4.07 mmol g super minus 1.

* 0.1 % N H sub 2-m G O with 40 wt. % aqueous M D E A solution at 30 degrees Celsius: Shows a maximum capacity of 4.50 mmol g super minus 1, which remains stable over 5 cycles (4.50, 4.49, 4.48, 4.48, 4.47).

* As temperature increases from 30 to 60 degrees Celsius for the 0.1 % N H sub 2-m G O / 40 wt. % M D E A system, the maximum capacity decreases from 4.50 to 3.70 mmol g super minus 1.

* Pure 40 wt. % M D E A solution at 30 degrees Celsius shows a lower maximum capacity of 3.90 mmol g super minus 1 compared to the nano-fluid versions.

Note: n/a: Data not available or not explicitly reported in the cited literature.

By using GO nanoparticles to prepare GO-based MDEA nano-fluids, it was observed that increasing the concentration of GO nanoparticles to 0.2 wt.% did not significantly enhance the CO₂ absorption capacity compared to using 0.1 wt.% GO nanoparticles. However, a slight increase in CO₂ absorption capacity was noted with 0.2 wt.% GO/40 wt.% aqueous MDEA nano-fluid, the effect was minimal. The increase in GO nanoparticle loading marginally enhanced the mass transfer surface due to the inherent hydrophobic nature of GO nanoparticles (Rahmatmand et al., Reference Rahmatmand, Keshavarz and Ayatollahi2016). Consequently, the stability of the hydrophobic GO nanoparticles in the nano-fluid slightly improved with the increase of the particle concentration up to 0.2 wt.% (Rahmatmand et al., Reference Rahmatmand, Keshavarz and Ayatollahi2016).

Comparing the CO₂ absorption performance of 0.02 wt.% CNT nanoparticles dispersed in MDEA and DEA as base fluids, listed in Table 4, the same amine concentration of 10 wt.%, the 0.02 wt.% CNT nanoparticles/10 wt.% DEA nano-fluid exhibited a higher CO₂ absorption capacity of 0.90 mmol g⁻¹ compared to 0.02 wt.% CNT nanoparticles/10 wt.% MDEA, which had an absorption capacity of 0.88 mmol g⁻¹. MDEA solvent is less selective for CO₂ compared to the DEA solvent, which is a very strong chemical absorbent of CO₂. The CO₂ absorption capacity of CNT nanoparticles decreased to 0.53–0.83 mmol g⁻¹ when the concentration of MDEA and DEA solutions was reduced to 5 wt.%. A high concentration of amine solution improves the mass transfer coefficient, thereby increasing the CO₂ absorption rate of the amine solution through the zwitterion mechanism (Irani et al., Reference Irani, Maleki and Tavasoli2019). Also, the CO₂ absorption capacity of both 0.02 wt.% CNT nanoparticles/5 wt.% DEA nano-fluid and 0.02 wt.% CNT nanoparticles/5 wt.% MDEA nano-fluid increased with rising partial pressure of CO₂.

Nonetheless, the use of carbon-based nanoparticles/liquid amine nano-fluids in CO₂ capture applications remains scarce and impractical due to their high regeneration cost and complicated synthesis process. A noticeable effect was observed when CNTs were incorporated into the amine-based nano-fluids, showing an increase of 69.7%–76.9% compared to the regeneration cost of a 40 wt.% MDEA solvent. Future studies are necessary to evaluate the techno-economic feasibility of hybrid CO₂ absorption–adsorption using carbon-based nanoparticles/liquid amine nano-fluids and to develop a simplified, low-cost nano-fluid synthesis method.

Overall, the benefits of using carbon-based nanoparticles/liquid amine nano-fluids in a hybrid absorption–adsorption system compared to amine-based absorbents and functionalised adsorbents, presented in Table 5, are discussed below. Limited research has studied CO₂ adsorption of pristine GO or CNT. Therefore, the adsorption capacities of GO and CNT composites functionalised with amine or metal oxides were used for comparative analysis. As detailed in Table 5, the summary of improvements is as follows:

1. 1. CO ₂ Absorption Performance: The hybrid absorption–adsorption system utilising carbon-based nanoparticles/liquid amine nano-fluids shows a 15.4%–36.4% increase in CO₂ absorption capacity. This surpasses the amine solvents and functionalised adsorption systems due to synergistic effects.

2. 2. Recyclability: The recyclability of carbon-based nanoparticles/liquid amine nano-fluids, such as GO nanoparticles/liquid amine nano-fluids, is enhanced through improved thermal and mass transfer (Lashgarinejad et al., Reference Lashgarinejad, Hosseini, Irani, Ghasemi, Mohammadpour and Tavasoli2023), maintaining performance over five absorption cycles.

3. 3. Nano-fluid Enhancement: Using carbon-based nanoparticles/liquid amine nano-fluids in the hybrid system significantly improves mass transfer by facilitating CO₂ transport through adsorption onto its carbon-based nanoparticles (Lashgarinejad et al., Reference Lashgarinejad, Hosseini, Irani, Ghasemi, Mohammadpour and Tavasoli2023). This addresses the standalone amine absorption system’s mass transfer and absorption limitations while also utilising physisorption on the surface and within the pores of CNT and GO particles to boost overall capture efficiency.

4. 4. Addressing Disadvantages: The hybrid approach enhances CO₂ capture efficiency by improving the recyclability of amine solvents incorporated with carbon-based nanoparticles over multiple absorption cycles. Similarly, adsorbents such as amine/metal oxide functionalised GO composites also exhibit limited adsorption capacity. However, in hybrid systems, energy consumption is lower because both physical and chemical bonding mechanisms govern absorption–adsorption interactions, reducing reliance on thermal regeneration and improving overall efficiency.

   Table 5.

   Comparison of carbon-based nanoparticles/liquid amine nano-fluids as a CO₂ absorbent–adsorbent versus single traditional absorbent

   

   Table 5. Long description

   The table consists of five columns: Criteria, Single absorbent (amine based), Functionalised adsorbents, Carbon-based nanoparticles/liquid amine nano-fluids, and Improvement.

   * CO₂ absorption or adsorption capacity (m mol g super -1): Single absorbents range from 1.75 to 3.90. Functionalised adsorbents range from 0.66 to 3.82 for G O composites and 2.01 to 2.52 for C N T composites. Carbon-based nano-fluids range from 0.90 to 4.50, showing an improvement of +15.4% to 36.4%.

   * Recyclability: Single absorbents have poor recyclability. Functionalised adsorbents are moderate to good. Carbon-based nano-fluids show good recyclability over 5 cycles due to high regenerability of the nanoparticles.

   * Nano-fluid enhancement: Single absorbents provide good mass transfer. Functionalised adsorbents involve moderate to good physisorption. Carbon-based nano-fluids provide enhanced mass transfer, boosting overall capture efficiency.

   * Disadvantages addressed: Single absorbents suffer from poor recyclability and high cost. Functionalised adsorbents suffer from low capacity. Carbon-based nano-fluids, particularly those using G O nanoparticles, mitigate these low absorption capacities.

   Note: ^aLimited research has studied CO₂ adsorption of pristine GO or CNT. Therefore, the adsorption capacities of GO and CNT composites functionalised with amine or metal oxides are used for comparative analysis.

Ionic liquid (IL)-functionalised adsorbent composites

Figure 5 depicts the schematic diagram of functionalising ionic liquids (ILs) onto a porous adsorbent, such as MCM-41 or activated carbon. IL-functionalised adsorbent composites as hybrid absorbent-adsorbents enhance absorption rate by incorporating IL onto a porous material (e.g., mesoporous silica [commonly MCM-41], zeolite, activated carbon and GO) through a simple moisture-controlled impregnation-evaporation method (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014; Xue et al., Reference Xue, Zhu, Du, An, Wang, Duan, Shi, Hao, Xiao and Peng2017; An et al., Reference An, Wang, Ma, Du, Hao, Yang and Guan2023). These supported IL-functionalised adsorbent composites combine the properties of IL and the porous material support, offering lower costs as they require only a small amount of IL for functionalisation (An et al., Reference An, Wang, Ma, Du, Hao, Yang and Guan2023). The porous support not only adsorbs CO₂ but also disperses the ILs effectively, enhancing the absorption and adsorption rates and the mass transfer rate (An et al., Reference An, Wang, Ma, Du, Hao, Yang and Guan2023). Additionally, IL-functionalised adsorbent composites reduce the amount of ILs required for synthesis, thereby lowering overall synthesis costs (An et al., Reference An, Wang, Ma, Du, Hao, Yang and Guan2023). Integrating ILs onto solid porous adsorbents mitigates the slow absorption rate of ILs caused by insufficient gas–liquid contact and slow mass transfer rates (An et al., Reference An, Wang, Ma, Du, Hao, Yang and Guan2023).

Figure 5.

Schematic diagram of functionalising ionic liquids (ILs) onto porous adsorbents, such as MCM-41 or activated carbon. Reproduced from Fellinger (Reference Fellinger2017). From “Sol–gel carbons from ionothermal syntheses,” by T-P Fellinger, 2017, BioResources, 81(1), p. 52–58 (10.1007/s10971-016-4115-z). Copyright 2017 by the Author(s).

Figure 5. Long description

At the bottom is a rounded rectangular box labeled Porous solid adsorbents containing two circular icons. The left icon shows a hexagonal honeycomb lattice labeled M C M 41. The right icon shows a chaotic porous network labeled Activated carbon. Attached to the top surface of this box are five blue circles labeled I L. A red circle highlights two of these I L spheres, with lines connecting them to two large upper panels separated by the word O R.

The top-left panel is labeled Cation I L s and contains six chemical structures.

- Imidazolium: a five-membered ring with two nitrogen atoms.

- Pyridinium: a six-membered aromatic ring with one nitrogen atom.

- Pyrrolidinium: a five-membered saturated ring with one nitrogen atom.

- Phosphonium: a central phosphorus atom bonded to four R groups.

- Ammonium: a central nitrogen atom bonded to four R groups.

- Sulfonium: a central sulfur atom bonded to three R groups.

The top-right panel is labeled Anion I L s and contains seven chemical structures.

- Alkylsulfate: an SO₄ group with one R chain.

- Hexafluorophosphate: a central phosphorus atom surrounded by six fluorine atoms.

- Bis(trifluoromethylsulfonyl)imide: a central nitrogen atom between two sulfonyl groups with CF₃ ends.

- Tetrafluoroborate: a central boron atom with four fluorine atoms.

- Amino acid: a structure with an NH₂ group and a carboxylate group.

- Carboxylate: an R group attached to a COO minus group.

- Halide: a circle labeled X with a negative charge.

Avila et al. (Reference Avila, Lepre, Santini, Tiano, Denis-Quanquin, Szeto, Padua and Costa Gomes2021) synthesised a porous IL on the ZIF-8 MOF support using phosphonium acetate or levulinate salts. At 30 °C and 5 bar, the absorbent prepared with pure tetrabutylphosphonium levulinate, [P₄₄₄₄][Lev], on ZIF-8 MOF (PoIL3-Z5) and pure trihexyltetradecylphosphonium bis(trifluoromethyl sulfonyl amide), [P₆₆₆₁₄][NTf₂], on ZIF-8 MOF (PoIL1-Z5) exhibited higher CO₂ adsorption capacities of 0.14 and 0.10 mmol g⁻¹, respectively (Avila et al., Reference Avila, Lepre, Santini, Tiano, Denis-Quanquin, Szeto, Padua and Costa Gomes2021). These values are higher compared to adsorbents prepared by mixing ILs [P₆₆₆₁₄][NTf₂] and tetrabutylphosphonium acetate, [P₄₄₄₄][OAc], at varying composition ratios (0.25, 0.50 and 0.75) on ZIF-8 MOF, which showed adsorption capacities ranging from 0.017 to 0.064 mmol g⁻¹ (Avila et al., Reference Avila, Lepre, Santini, Tiano, Denis-Quanquin, Szeto, Padua and Costa Gomes2021).

Cheng et al. (Reference Cheng, Li, Hu, Zhou and Cen2016) found that an anion-functionalised IL, trihexy(tetradecyl)phosphonium 2-hydroxypyridinate ([P₆₆₆₁₄][2-Op]), loaded onto MCM-41 (MCM-41–50% IL) exhibited a higher CO₂ adsorption rate of 1.17 mmol CO₂ g⁻¹ MCM-41–50% IL min⁻¹, which is 2.1 times higher than that of pure [P₆₆₆₁₄][2-Op] (0.56 mmol CO₂ g⁻¹ IL min⁻¹) during the initial 1 minute under a pure CO₂ atmosphere. MCM-41-50% IL has more exposed IL compared to pure [P₆₆₆₁₄][2-Op] (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). CO₂ molecules rapidly diffused throughout the reactor and contacted the surface of [P₆₆₆₁₄][2-Op] after CO₂ was introduced into the reactor (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). This resulted in a higher CO₂ adsorption rate by MCM-41-50% IL than pure [P₆₆₆₁₄][2-Op] in the initial 1 minute (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016).

The CO₂ capture performance of the IL-functionalised adsorbent composites using the absorption–adsorption hybrid process is tabulated in Supplementary Table S3. Table 6 highlights IL-functionalised adsorbent composite’s highest CO₂ adsorption capacities, operating conditions and recyclability. According to the data in Supplementary Table S3, increasing the adsorption temperature from 120 °C to 180 °C reduces the CO₂ adsorption capacity by 50 wt.% PAP (Tetrabutylphosphonium bromide ([P₄₄₄₄]Br) prepared using an anion-exchange resin and neutralised with 3-aminopyrazole) modified MCM-41 (50PAP/M41) from 1.45 to 0.23 mmol g⁻¹, 50 wt.% PAP modified mesoporous alumina (50PAP/pA) from 1.20 to 0.25 mmol g⁻¹, 50 wt.% PI ([P₄₄₄₄]Br prepared using an anion-exchange resin and neutralised with imidazole) modified MCM-41 (50PI/M41) from 0.64 to 0.098 mmol g⁻¹ and 50PI/pA from 1.16 to 0.17 mmol g⁻¹ (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014). Higher temperatures induce the deactivation of IL for CO₂ capture (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014).

Table 6.

Comparative analysis of CO₂ capture performance in hybrid absorption–adsorption systems using IL-functionalised adsorbent composites

Table 6. Long description

The table contains 10 columns: I L-functionalised adsorbent composites, Operating temperature in degrees Celsius, Total pressure P_total in atm, CO₂ partial pressure P_total CO₂ in atm or feed, Minimum CO₂ adsorption capacity in mmol g sub negative 1, Maximum CO₂ adsorption capacity in mmol g sub negative 1, Relative adsorption capacity over multiple cycles in percent, CO₂ adsorption capacity over multiple cycles in mmol g sub negative 1, and Reference.

Key data rows include:

* 67 wt. percent P A P modified M C M 41: At 120 degrees Celsius and 1 atm, capacity drops from 100 percent (1.91 mmol g sub negative 1) in the 1st cycle to 34 percent (0.65 mmol g sub negative 1) by the 7th cycle.

* 50 wt. percent P I modified mesoporous alumina: At 120 to 180 degrees Celsius, capacity drops from 100 percent (1.35 mmol g sub negative 1) in the 1st cycle to 83 percent (1.12 mmol g sub negative 1) by the 7th cycle.

* M C M 41 50 percent [P sub 6 6 6 14] [2 O p]: Operating at 19 degrees Celsius with a maximum capacity of 2.03 mmol g sub negative 1.

* Activated carbon supported composites (v b t m a with glycine, L-arginine, or L-lysine): Show high maximum capacities ranging from 24.70 to 30.90 mmol g sub negative 1 at 25 to 60 degrees Celsius.

* Activated carbon impregnated with 20 wt. percent [d m e d a h] [propanoate]: Maintains 90 percent capacity (15.15 mmol g sub negative 1) by the 4th cycle.

Note: n/a: Data not available or not explicitly reported in the cited literature.

As the IL loading on MCM-41 increases from 15 to 67 wt.%, the CO₂ adsorption capacity at 120 °C rises, ranging from 0.17 to 0.91 mmol g⁻¹ for nPAP/M41 and 0.21–1.84 mmol g⁻¹ for nPI/M41 (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014). Similarly, at 120 °C, the CO₂ adsorption capacity of PAP and PI-functionalised mesoporous alumina (p-Al₂O₃) increases with IL loading from 15 to 50 wt.%, ranging from 0.49–1.20 mmol g⁻¹ for nPAP/pA and 0.65–1.16 mmol g⁻¹ for nPI/pA (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014). The same amount of PAP or PI functionalised onto p-Al₂O₃ adsorbs more CO₂ than MCM-41, except for 50PAP/pA, due to the less reduced surface area in nPAP/pA and nPI/pA (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014). This provides more surface area for CO₂ molecules to interact with PAP or PI during adsorption.

In addition, the surface roughness and types of materials affect IL immobilisation. For example, functionalising 15 wt.% IL on SBA-15 leads to micropore plugging and mesopore surface smoothing, preventing some active sites from adsorbing CO₂, resulting in a lower adsorption capacity of 0.12 mmol g⁻¹ compared to 15PAP/M41 (0.21 mmol g⁻¹) (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014). Zeolite NaY, GO and activated carbon also failed to enhance the CO₂ adsorption capability of PAP, achieving lower adsorption capacities of 0.21–0.73 mmol g⁻¹ (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014).

CO₂ concentration, flow rate, weight percentage of IL and adsorption time influence the adsorption performance of molecular sieve MCM-41 loaded with 50 wt.% [P₆₆₆₁₄][2-Op] (MCM-41-50% [P₆₆₆₁₄][2-Op]). As evidenced in Supplementary Table S3, under a constant 100% CO₂ concentration, increasing the adsorption time from 1 to 400 min increases the adsorption capacity of MCM-41-50% [P₆₆₆₁₄][2-Op] from 1.17 to 2.03 mmol g⁻¹ at 10 mL min⁻¹ CO₂ flow rate, 19 °C and 1 atm (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). Similarly, in a 40% CO₂/CH₄/H₂ mixture, the CO₂ adsorption capacity of MCM-41-50% [P₆₆₆₁₄][2-Op] was enhanced from 0.56 to 1.43 mmol g⁻¹ as the adsorption time increased from 2 to 240 min (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). The CO₂ adsorption at 240 min also showed comparable adsorption capacities of 1.43 mmol g⁻¹ in a 40% CO₂/30% CH₄/30% H₂ mixture and 1.49 mmol g⁻¹ in a 40% CO₂/60% CH₄ mixture at 19 °C and 1 atm (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). MCM-41-50% [P₆₆₆₁₄][2-Op] exhibits higher CO₂ adsorption capacity in a pure CO₂ (100%) atmosphere than in biohythane (40% CO₂/30% CH₄/30% H₂ mixture) and biogas (40% CO₂/60% CH₄ mixture) atmospheres. This indicates that H₂ and CH₄ have slight effects on CO₂ adsorption of MCM-41-50% [P₆₆₆₁₄][2-Op] in biohythane and biogas atmospheres (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016).

On top of that, at 19 °C and 1 atm, the CO₂ adsorption capacity increases from 0.98 to 1.16 mmol g⁻¹ for adsorption at 30 min and from 1.43 to 1.70 mmol g⁻¹ at 240 min when increasing the biohythane flow rate from 10 to 40 mL min⁻¹ (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). The CO₂ adsorption rate of MCM-41-50% [P₆₆₆₁₄][2-Op] was enhanced 1.1 times with increasing flow rates. Under a biohythane atmosphere using a 10 mL min⁻¹ flow rate, increasing the weight percentage of [P₆₆₆₁₄][2-Op] from 30 to 50 wt.% enhances the CO₂ adsorption capacity of MCM-41-w[P₆₆₆₁₄][2-Op] from 0.32 to 0.56 mmol g⁻¹ at 2 min of adsorption, from 0.73 to 0.96 mmol g⁻¹ at 25 min of adsorption and from 1.37 to 1.43 mmol g⁻¹ at 240 min of adsorption at 19 °C and 1 atm (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). More exposed IL within MCM-41-50% [P₆₆₆₁₄][2-Op] facilitates faster diffusion of CO₂ molecules and interaction with the IL surface during the initial 2 min of adsorption compared to MCM-41-30% [P₆₆₆₁₄][2-Op] and MCM-41-40% [P₆₆₆₁₄][2-Op] (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). As the adsorption continues for 25 and 240 min, the viscosity of IL rises and generates a protective film, slowing CO₂ diffusion into the IL located within inner pores (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016). This resulted in the slowly increasing MCM-41-w[P₆₆₆₁₄][2-Op] CO₂ adsorption rate after 25 min (Cheng et al., Reference Cheng, Li, Hu, Zhou and Cen2016).

The loading percentage influences the CO₂ adsorption performance of IL-impregnated activated carbon at 25 °C and pressure of 0.99, 4.93, 9.87, 14.80 and 19.74 atm. Supplementary Table S3 shows that the CO₂ adsorption capacity for all IL-impregnated activated carbons initially increases with IL loading and then decreases, indicating the presence of an optimal condition (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). Each IL exhibits a different optimal loading, influenced by the amine functionalities supported on the activated carbon (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). At the optimal loading, ILs are well impregnated into the activated carbon, achieving higher CO₂ adsorption capacity (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). However, the CO₂ adsorption capacity significantly reduces when ILs oversaturate the activated carbon, blocking its pores (Md Nurul et al., Reference Md Nurul, Bustam Mohammad, Yunus and Man2014; Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). This reduction occurs regardless of the type of IL used, as the blocked pores hinder the physical adsorption of CO₂ (Md Nurul et al., Reference Md Nurul, Bustam Mohammad, Yunus and Man2014; Gholidoust et al., Reference Gholidoust, Atkinson and Hashisho2017).

For all activated carbons impregnated with mono-functionalised amine-based ILs, including unsym-dimethylethylenediamine formate ([dmedah][formate]), unsym-dimethylethylenediamine acetate ([dmedah][acetate]), unsym-dimethylethylenediamine propanoate ([dmedah][propanoate]) and (vinyl benzyl)trimethylammonium chloride glycine ([vbtma][gly]), at 25 °C and 19.74 atm, they achieved optimum CO₂ adsorption capacities ranging from 25.44 to 30.90 mmol g⁻¹ at 20 wt.% IL loading (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). Higher loading of these ILs onto the activated carbon surfaces enhances the CO₂ adsorption capacity due to the acidic role in CO₂ adsorption (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). However, higher IL loadings of 30 and 40 wt.% are required to achieve the optimum CO₂ adsorption capacity when activated carbon is impregnated with multifunctionalised amines such as (vinyl benzyl)trimethylammonium chloride L-arginine ([vbtma][arg]) and (vinyl benzyl)trimethylammonium chloride L-lysine ([vbtma][lys]) (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). At 25 °C and 19.74 atm, the optimum adsorption capacities of these two IL-impregnated activated carbons are 24.70 mmol g⁻¹ at 30 wt.% [vbtma][arg] loading and 25.81 mmol g⁻¹ at 40 wt.% [vbtma][lys] loading (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). The higher number of active sites on the basic amino acids, argininate ([arg]) and lysinate ([lys]), provides more CO₂ attraction to the activated carbon, thereby enhancing adsorption (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019).

Table 6 indicates the optimum activated carbon supported with 20 wt.% (vinyl benzyl)trimethylammonium chloride glycine (AC-20 wt.%[vbtma][gly]) achieved a higher CO₂ adsorption capacity compared to activated carbon supported with 30 wt.% (vinyl benzyl)trimethylammonium chloride L-arginine (AC-30 wt.%[vbtma][arg]) and 40 wt.% (vinyl benzyl)trimethylammonium chloride L-lysine (AC-40 wt.%[vbtma][lys]). This is due to the steric hindrance effect of the four functionalised amine molecules in [arg] and the two primary amines in [lys] structure (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). The isolation of amine groups by secondary and tertiary carbons in their respective structures’ limits CO₂ diffusion to the amine molecules in activated carbon, thereby reducing CO₂ adsorption (Gholidoust et al., Reference Gholidoust, Atkinson and Hashisho2017; Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). For protic ILs (e.g., [dmedah][formate], [dmedah][acetate] and [dmedah][propanoate]), the CO₂ adsorption capacity increased with the increasing alkyl chain length in the anion, which enhances the hydrophobicity of the protic IL molecules (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). The increase in hydrophobicity raises the number of amine active sites on the activated carbon, leading to higher CO₂ adsorption (Hassan et al., Reference Hassan, Duclaux, Lévêque, Reinert, Farooq and Yasin2014). Furthermore, increasing the alkyl chain length also increases the IL polarity, resulting in a stronger CO₂ interaction with the impregnated activated carbon (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). On top of that, Supplementary Table S3 data illustrate that the CO₂ adsorption capacity of all IL-impregnated activated carbons increased with rising pressure. This is because the increasing pressure forces more CO₂ gas into the impregnated activated carbon (Md Nurul et al., Reference Md Nurul, Bustam Mohammad, Yunus and Man2014).

The effect of temperature on CO₂ adsorption by AC-20 wt.%[vbtma][gly] and activated carbon impregnated with 20 wt.% unsym-dimethylethylenediamine propanoate (AC-20 wt.%[dmedah][propanoate]) is observed in Supplementary Table S3. The CO₂ adsorption capacity of these two IL-impregnated activated carbons is higher at lower temperatures (25 °C) and decreases at higher temperatures (40 °C and 60 °C). As demonstrated in Supplementary Table S3, a significant reduction in CO₂ adsorption capacity is noted for AC-20 wt.%[vbtma][gly], from 1.51–30.90 to 0.82–20.22 mmol g⁻¹, and for AC-20 wt.%[dmedah][propanoate], from 1.06–28.31 mmol g⁻¹ to 0.99–12.85 mmol g⁻¹, when temperature increase from 25 °C to 40 °C (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). At 60 °C, the CO₂ adsorption capacity is slightly lower than at 40 °C, ranging from 0.93 to 9.77 mmol g⁻¹ for AC-20 wt.%[vbtma][gly] and 0.91–11.75 mmol g⁻¹ for AC-20 wt.%[dmedah][propanoate] (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019). The higher CO₂ adsorption capacity at 25 °C is due to increased CO₂ diffusion to the reactive amine sites, facilitated by the exothermic physisorption process (Lee et al., Reference Lee, Jin, Jung, Lee, Lee, Jeong and Jeon2011; Gholidoust et al., Reference Gholidoust, Atkinson and Hashisho2017). The lower CO₂ adsorption capacity at 60 °C was due to the adsorption shifts towards the endothermic desorption process, and full desorption has not been achieved at this temperature (Gholidoust et al., Reference Gholidoust, Atkinson and Hashisho2017; Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019).

The integration of ionic liquids (ILs) with porous adsorbents, such as MCM-41 and activated carbon, in hybrid absorption–adsorption systems delivers significant performance improvements over standalone ILs or adsorbents. These improvements can be understood in terms of both measured outcomes (Table 7) and key design criteria for material selection, including mass transfer efficiency, thermodynamic regeneration properties and material stability across adsorption cycles, as summarised below. By explicitly linking adsorption capacity to mass transfer, regeneration costs to thermodynamics and recyclability to stability, Section “Ionic liquid (IL)-functionalised adsorbent composites” now provides a framework for evaluating IL-functionalised adsorbent composites not only by their measured outcomes but also by the underlying design principles that guide material selection. Building on these design criteria, the performance outcomes of IL-functionalised adsorbent composites can be summarised as follows:

1. 1. CO₂ Adsorption Performance and Mass Transfer Considerations:

• • The hybrid IL-functionalised adsorbent composites achieve a 4.50%–99.70% increase in CO₂ adsorption capacity. This surpasses standalone ILs absorption and single MCM-41 or activated carbon adsorption.

• • This improvement arises from synergistic mass transfer effects: the porous adsorbent provides high surface area and rapid diffusion pathways, while ILs contribute reactive amine groups for chemical absorption (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019).

• • Design criterion: Materials should balance pore size distribution [6–15 nm] (to minimise diffusion resistance) with IL loading [10–25 wt.%] (to maximise reactive site density without blocking pores).

1. 2. Recyclability and Material Stability:

• • IL-functionalised adsorbent composites maintain performance over four to seven adsorption cycles, outperforming standalone ILs, which degrade rapidly due to chemical instability under repeated regeneration (Wan et al., Reference Wan, Zhu, Li, Ma, Liu and Zhu2014; Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019).

• • This stability is comparable to the recyclability of traditional adsorbents (MCM-41 and activated carbon), which show variable recyclability depending on the structural stability and surface chemistry.

• • Design criterion: Selection should prioritise adsorbents with high mechanical and thermal stability to resist oxidative or thermal degradation at ~600 °C and > 900 °C during regeneration, hence maximising functional operation time.

1. 3. Enhancement of Functionalised Ionic Liquids (ILs) and Physisorption Limitations:

• • The use of IL-functionalised adsorbent composites in the hybrid system significantly improves the physical adsorption of CO₂ onto the adsorbent surface, in addition to interacting with the reactive amine groups of IL, complementing the chemical absorption of ILs (Raja Shahrom et al., Reference Raja Shahrom, Nordin and Wilfred2019).

• • This dual mechanism addresses the physisorption limitations of CO₂ in the standalone absorption and adsorption systems, particularly at higher operating temperatures.

• • Design criterion: Functionalisation strategies should optimise IL-adsorbent interactions to maximise both physisorption (surface affinity) and chemisorption (reactive binding).

1. 4. Addressing Disadvantages and Overall Efficiency:

• • The hybrid approach enhances CO₂ capture efficiency by improving the physisorption of adsorbents functionalised with IL at higher temperatures and with moderate to good recyclability.

• • This efficiency is particularly notable in comparison to the typically high costs and the low adsorption capacity of both the ILs and MCM-41 and activated carbon adsorbents.

• • Design criterion: Materials must be selected to balance adsorption kinetics, regeneration energy and long-term stability, ensuring that operational efficiency is maintained across multiple cycles.

  Table 7.

  Comparison of IL-functionalised adsorbent composites as a CO₂ absorbent-adsorbent versus single traditional adsorbent

  

  Table 7. Long description

  The table consists of five columns: Criteria, Single absorbent (pure ionic liquid I L), Single adsorbent (M C M 41 and activated carbon), I L functionalised adsorbent composites, and Improvement.

  * Row 1: CO₂ adsorption capacity (m mol g super minus 1). Single absorbent: 0.25 to 2.20. Single adsorbent: Activated carbon 0.078 to 15.47; M C M 41 0.017 to 0.63. I L functionalised composites: 0.43 to 30.90. Improvement: plus 4.50 percent to 99.7 percent.

  * Row 2: Recyclability. Single absorbent: Poor, as I L degrades. Single adsorbent: Moderate to good. I L functionalised composites: Moderate to good over 4 to 7 cycles. Improvement: Maintains capacity over repeated cycles comparable to traditional adsorbents.

  * Row 3: Enhancement of functionalised I L s. Single absorbent: Good mass transfer. Single adsorbent: Good CO₂ physisorption. I L functionalised composites: Enhanced physisorption. Improvement: Enhances interaction between support and CO₂ while facilitating reaction with amine groups.

  * Row 4: Disadvantages addressed. Single absorbent: Low reaction surface area. Single adsorbent: Low adsorption capacity. I L functionalised composites: Potential pore oversaturation at high loadings and slow adsorption rates over long durations. Improvement: Optimal loading enhances uptake and mitigates pore blockage.