2. CO2 Capture
2.1. CO2 Capture Technologies
Capture and sequestration of CO
2 (CCS) from aforementioned stationary emission sources has been identified as a paramount option for the issues of global warming and climate change. CCS includes four primary steps known as CO
2 capture, compression, transport, and storage, therefore, developing an efficient and economically feasible technology for the capture and sequestration of CO
2 produced by anthropogenic emissions is critically important. CO
2 capture is the central part of the CCS technology process and gained around 70–80% of the total expensive. However, CSS methods can be classified as, for example, (i) Post-combustion (ii) Pre-combustion, and (iii) Oxy-fuel combustion (Oxygen-fired combustion)
[4][5].
In
post-combustion capture technology, it collects and separates the CO
2 from the emission gases of a combustion system
[6][7][8][9][10]. Firstly, flue gas (mainly consists of CO
2, H
2O, and N
2) passes through denitrification and desulphurization treatments. As the next step, the flue gas is fed to an absorber which contains solvent. Herein, CO
2 regeneration occurs. Then the CO
2-rich absorbent is sent to a CO
2-stripper unit to release the CO
2 gas. Moreover, CO
2-lean absorbent is sent back to the CO
2-absorber unit
[1]. Next, the captured CO
2 is then compressed into supercritical fluid and then transported
[1] as shown in
Figure 1.
Figure 1. Schematic representation of post-combustion technology.
Pre-combustion capture is a technology where CO
2 is captured before the combustion process and CO
2 is generated as an intermediate co-product of conversion process
[11]. The pre-combustion technologies are mainly used in power plants, production of fertilizers and natural gas
[12][13].
In
oxyfuel combustion, the carbon-based fuel consumes in re-circulated flue gas and oxygen (O
2) stream. CSS capture technology is considered expensive due to the high cost of O
2 separation and production. However, the capture and separation of CO
2 are reasonably easy compared to other methods and is considered as an energy-saving method
[14].
Absorption process mainly uses liquids to capture CO
2. During adsorption, once CO
2 is separated from the gas, the sorbent should be regenerated by using a stripper, heating, or depressurization. Moreover, this method is considered as the most established process for CO
2 separation
[15]. In general, adsorbents can be divided into two types, namely, chemical and physical adsorbents.
2.2. Criteria for Selecting CO2 Sorbent Material
Certain economical and technical properties are required in order to select the best solid adsorbent candidate for a particular CO2 capture application. These criteria are listed and described below.
The equilibrium adsorption capacity of a sorbent material is represented by its equilibrium adsorption isotherm. The adsorption capacity is an important parameter when considering the cost. Moreover, which causes reduction in the sorbent quantity, and in the size of the adsorption column. However, to enhance the adsorption capacity of solid sorbents, functionalization has been carried out with existing monoethanolamine (MEA)
[16]. The CO
2 working capacity should be in the range of 2–4 mmol/g of the sorbent
[17].
The adsorption selectivity or selectivity of CO
2 is explained as the sorption uptake ratio of a target gas species compared to another type (as example N
2) contained in a gaseous mixture under given operation conditions. Therefore, it depends on the purity of the adsorbed gas in the effluent
[15]. However, the purity of CO
2 influences transportation and sequestration and, therefore, this criterion plays an important role in CO
2 sequestration
[16].
It is necessary to have fast adsorption/desorption kinetics for CO
2 and it controls the cycle time of a fixed-bed adsorption system. Fast kinetics results in a sharp CO
2 breakthrough curve in which effluent CO
2 concentration changes are measured as a function of time, while slow kinetics provides a distended breakthrough curve. However, both fast and slow adsorption and desorption kinetics impact on the amount of sorbent required. In functionalized solid sorbents, the overall kinetics of CO
2 adsorption mainly depend on the functional groups present, as well as the mass transfer or diffusional resistance of the gas phase through the sorbent structures. The porous support structures of functionalized solid sorbents also can be tailored to minimize the diffusional resistance. The faster an adsorbent can adsorb CO
2 and be desorbed, the less of it will be needed to capture a given volume of flue gas
[16].
The sorbent must show the stable microstructure and morphological structure in adsorption and regeneration steps. Mainly disintegration of the sorbent particles occurs due to the high volumetric flow rate of flue gas, vibration, and temperature. Apart from that, this could also happen due to abrasion or crushing. Therefore, a sufficient mechanical strength of a sorbent particles is required to keep CO
2 capture process cost-effective
[16].
Solid CO
2 capture sorbents such as amine-functionalized sorbents should be stable in an oxidizing environment of flue gas and should be resistant to common flue gas contaminants
[16].
The regeneration of the sorbent is energy saving and is one of the most important parameters required for improving energy efficiency
[18]. Regeneration can be achieved through the adjustment of the thermodynamics of the interaction between CO
2 and the solid adsorbent
[16]. Considering regeneration, physisorption is mostly favored over chemisorption since the latter involves high energy consumption for regeneration.
The production cost is the main key point when considering industrial applications at reasonable gas selectivity and adsorption performance
[16].
2.3. Liquid Amine for CO2 Capture
Development of solvents for CO
2 chemical absorption is a major area of research
[19]. The ideal solvent should have a high CO
2 absorption capacity and react rapidly and reversibly with CO
2 with minimal heat requirement. The solvent should exhibit the following properties such as stability in oxidative and thermal environment, low vapor pressure, toxicity, flammability, and reasonable production cost
[19].
Recently, a most promising CO
2 capture method with chemical absorption is by using liquid amine which can be divided mainly into two groups known as simple alkanolamines and sterically hindered amines
[20]. Examples for simple alkanolamines are monoethanolamine (MEA), diethanolamine (DEA), and triethanolamine (TEA)
[21][22]. Furthermore, alkanolamines are the most widely used sorbents for CO
2 capture. The structures of alkanolamines include primary, secondary, ternary amines containing at least one hydroxyl (-OH) group and amine group-(N-R).
However, these different amine classes have different reaction kinetics with CO
2, CO
2 absorption capacity and equilibria, stability, and corrosion
[20]. As shown in Equations (1) and (2) below, both primary and secondary amines react with CO
2 to form a carbamate and protonated amine, consuming approximately two moles of amine per mole of CO
2 according to the zwitterion mechanism
[23]. According to Equation (3), tertiary amines react with CO
2 gas molecules in the presence of H
2O while forming bicarbonates.
(where R1, R2, and R3 are aryl/alkyl groups).
However, García-Abuín et al.
[24] observed that MEA produced a mixture of carbamate and bicarbonate as the main reaction products during CO
2 absorption. The reaction starts with the reversible reactions between MEA and CO
2 to form carbamate at low CO
2 loading, followed by the CO
2 hydration to form HCO
3−/CO
32− under high CO
2 loading, and accompanied by the hydrolysis of carbamate. The reaction mechanism of CO
2 capture into MEA solution with different CO
2 loadings is shown in
Figure 2.
Figure 2. Reaction mechanism of CO2 capture into MEA solution.
There are three categories of alkanolamines that show increased capital costs due to requirement of specialized and expensive materials for construction
[20]. On the contrary, degradation of alkanolamine causes operational, and environmental problems including high amount of absorbent required, corrosion of equipment, and demanding of energy
[16].
Among three different alkanolamines, MEA is commonly considered as a well-established solvent to separate CO
2 because it can be regenerated easily
[25]. On the other hand, Rinprasertmeechai et al. reported the order of CO
2 absorption capacity of the different alkanolamines as MEA > DEA > TEA
[26]. Moreover, they have further showed the regeneration ability of the amines in the following order: MEEA > > DEA > MEA. MEA exhibits high CO
2 adsorption capacity as it reacts more rapidly with CO
2 compared to MEDA by forming carbamates. However, MEDA shows high regeneration efficiency and requires lower energy
[27]. Moreover, Wang et al. found that, when MEA and MEDA are mixed with the appropriate ratio, the energy consumption for CO
2 regeneration is reduced significantly
[28].
Sterically hindered amines are based on primary or secondary amines with bulky alkyl groups, which is inhibited from reacting with CO
2 through the effect of steric hindrance
[20]. One example of sterically hindered amines is 2-amino-2-methyl-1-propanol (AMP). Steric factor reduces the stability of the formed carbamate due to the weak interaction between the CO
2 molecule and the NH
2 group, promoting fast hydrolysis to form bicarbonate and reducing regeneration energy. Due to the immediate regeneration process of AMP, the NH
2 group can react with CO
2 molecules over and over, increasing CO
2 adsorption. Moreover, Dave et al.
[29] compared the CO
2 absorption of different liquid amine classes and showed a lower regeneration energy requirement for 30 wt% AMP over 30% MEA, 30% MEDA, 2.5% NH
3, and 5% NH
3 [29].
Recently, ionic liquids (IL) have also been investigated as liquid solvents for CO
2 capture due to their low vapor pressure, thermal stability, non-toxicity, and adsorption capacity
[30][31][32]. The widely studied ILs include bis(trifluoromethylsulfonyl)imide (TF
2N), tetrafluoroborate (BF
4), and hexafluorophosphate (PF
6)
[30][31][32]. However, the main drawbacks of the ILs are high viscosity and production high cost.
2.4. Comparison between Major Non-Carbonaceous Solid Sorbents for CO2 Capture and Importance of Silica Materials
Due to the low contact area between gas and liquid, low CO
2 loading, and absorbent corrosion associated with liquid amine-based sorbents, solid sorbents for CO
2 capture have attracted significant attention in recent years
[33][34]. Various solid adsorbents have been proposed according to their structures and compositions, adsorption mechanisms, and regeneration process
[34]. Many solid sorbents are cheap and readily available and show low heat capacities, fast adsorption kinetics, high CO
2 adsorption capacities and selectivity, and high thermal, chemical, and mechanical stabilities
[34].
Commercially available solid adsorbents for CO
2 capture include carbonaceous materials such as activated carbons, nanofibrillated cellulose (CFCs), carbon nanotubes (CNTs), and non-carbonaceous materials, including silica, zeolites, hollow fibers, and alumina
[5]. These materials show different surface morphologies, pore structures, specific surface areas, and functional groups.
Carbonaceous adsorbents are widely used for CO
2 capture due to their relative abundance, low cost, renewability, and high thermal stability. However, the weak CO
2 adsorption capacities of carbonaceous materials at 50–120 °C make it challenging to use in industrial CO
2 capture
[35]. Therefore, much research focus has been given to non-carbonaceous materials.
Zeolites are aluminosilicates with ordered three-dimensional (3D) microporous structures with high crystallinity and surface area
[35]. The adsorption efficiencies of zeolites are primarily affected by their size, charge density, and chemical composition of cations in their porous structures
[28]. It has been reported that the CO
2 adsorption of zeolites increases as the Si/Al ratio increases and is exchanged with alkali and alkaline-earth cations in the structure of zeolites
[36]. However, zeolites present several drawbacks, such as relatively low CO
2/N
2 selectivity and high hydrophilicity
[37]. Apart from the above, zeolites show reduced CO
2 adsorption capacity when CO
2/N
2 mixtures contain moisture, and zeolites require high temperatures (>300 °C) for regeneration
[38].
Recently, metal-organic frameworks (MOFs) have gained much attention owing to their unique properties, such as tunable pore structure and high surface area
[39]. However, when exposed to gas mixtures, the MOFs show decreased adsorption capacities
[37]. Moreover, previous reports indicate that MOFs are promising materials for CO
2 capture in laboratory settings; however, further research is required to confirm their practical applicability
[40]. Water vapor also negatively affects the application of these sorbents by competing and adsorbing them onto physisorbents, thus decreasing their CO
2 adsorption capacity
[41].
Ordered mesoporous silica materials are good candidates because of their high surface area, high pore volume, tunable pore size, and good thermal and mechanical stability. So far, mesoporous silica includes the families of MCM (Mobil Company Matter: M41S, Santa Barbara Amorphous type material (SBA-n), anionic surfactant-template mesoporous silica (AMS)
[35]. However, the CO
2 adsorption capacities of them observed at atmospheric pressure are not high. Therefore, many studies have been recently reported on the functionalized mesoporous and nanoporous silica for efficient CO
2 capture
[42][43].
3. CO2 Adsorption Using Mesoporous Silica Materials (Physisorbents)
3.1. Mesoporous Silica Materials
Mesoporous silica materials are used for various applications, including catalysis and wastewater treatment
[44]. Mesoporous silica has unique properties such as uniformity of pore distribution (with size between 0.7 and 50 nm), high surface area (around 1000 m
2/g), and good thermal stability
[45]. The first synthesized mesoporous silica material was M41S in the 1990s
[46]. However, the development of surfactants and synthesis protocols have been able to prepare many types of mesoporous silicas such as MCM-41, SBA-15, SBA-16, FDU-2, MCM-50, and KIT-5 with a diverse range of pore geometries such as cubic, and hexagonal, and morphologies such as rods, spheres, and discs
[47].
In 1990, Mobil Oil Corporation discovered molecular sieves of the M41S family consisting of silicate/aluminosilicate
[48]. Typically, these materials are prepared via the sol-gel method. Three well-defined structural arrangements have been identified after studying the effect of surfactant concentration, and those are hexagonal (MCM-41), cubic (MCM-48), and lamellar (MCM-50) structures. Therefore, these materials (M41S family) exhibit mesoporous arrays with amorphous walls of about 10 Å (1 nm)
[48]. Moreover, the structural ordering of these M41S family materials can be changed with increasing hydrothermal synthesis temperature and time
[48]. These M41S molecular sieves are mainly applied in catalysis
[49], adsorption
[48], and controlled release of drugs
[50]. The main advantage of this mesoporous silica is its unique chemical structure consisting of the high density of functional silanol groups (Si–OH), pore size and shape can be molded during the synthesis process, and the internal surface can be easily modified with organic and inorganic groups
[48][51][52].
Santa Barbara Amorphous family (SBA) first prepared silica-based materials with well-ordered mesoporous in 1998
[48]. This material group consists of SBA-2 (hexagonal close-packed array), SBA-12 (three-dimensional hexagonal network), SBA-14 (cubic structure), SBA-15 (two-dimensional hexagonal), and SBA-16 (structured in a cubic cage)
[48][53]. These nanostructured mesoporous materials comprise a silica-based framework with uniform and well-ordered mesopores, large pores, thick and porous walls, high surface area, and high thermal stability
[52][54]. The most widely investigated members of the SBA-n family in the literature are SBA-15 and SBA-16. The SBA-15- and SBA-16-based mesoporous arrays are commonly utilized as adsorbents
[52], catalysts or catalytic
[55], and drug deliveries
[56].
The Fudan University synthesized mesoporous materials family (FDU-n)-based mesoporous silica arrays with well-ordered mesostructures and pore arrangements, high surface area, large and uniform distribution of pore diameter, amorphous pore-wall structures, and thermal and mechanical stability
[57]. FDU-1-based mesoporous materials have a 3D face-centered cubic (FCC) structure with large cage-like mesopores, while FDU-2 mesoporous array possesses a mesostructured FCC unit cell and well-ordered 3D architecture
[52].
On the contrary, the mesoporous material series of the KIT-n family, where n = 1, 5, or 6, are mainly represented by the KIT-1, KIT-5, and KIT-6. However, KIT-1-based mesoporous silicas exhibit a 3D architecture in a disordered framework with high surface area, large pore volume and pore diameter, and thermal and hydrothermal stability
[58]. KIT-5-based nanostructured mesoporous materials have well-ordered 3D cage-like mesopores in a face-centered close-packed cubic lattice architecture
[52]. In addition, KIT-6 shows 3D mesoporous amorphous walls with large pore size, uniform pore distribution, high surface area, and thermal stability
[52].
Moreover, mesoporous silica materials of the M41S, SBA-n, FDU-n, and KIT-n families are used in a wide range of applications such as separation, catalysis, drug release adsorption, sensors, matrix solid-phase dispersion (MSPD) and solid-phase extraction
[52].
3.2. Synthesis Procedures of Mesoporous Silica
Initially, Stöber et al.
[59] discovered an effective method for synthesizing monodispersed silica particles. This process consists of hydrolysis of tetraethyl orthosilicate (TEOS) using ammonia as a catalyst in water and ethanol solution. This method leads to the synthesis of silica particles
[60]. In this reaction, TEOS undergoes hydrolysis in an ethanol/ammonia solution. As a result, it produces silanol monomer (-Si-OH) with the epoxy groups (-Si-OEt), as shown in Equation (4). Then silanol groups undergo condensation to produce branched siloxane clusters, which causes to initiate the nucleation and growth of silica particles, see Equation (5). Simultaneously, silanol monomers react with the unhydrolyzed TEOS via condensation (see Equation (6)) and participate in the nucleation and growth of silica particles
[22]. Moreover, the particle size of Stöber silica depends on the concentration of the aqueous ammonia solution and water in the ethanol reaction
[22].
Many experimental factors control hydrolysis, silica condensation rate, assembly kinetics, nucleation, and growth rates
[48][61]. The pH is an essential factor that influences the charges of silica species. Rates of hydrolysis of silane and condensation of the siloxane bond depend strongly on the charge states. Hydrolysis of the Si–OR bond in silanes could be catalyzed by acid and base conditions, but its rate is prolonged near the neutral conditions
[61].
Sakamoto et al.
[62] prepared silica nanoparticles (NPs) via the evaporation and self-assembly of silicate and quaternarytrialkylmethylammonium as a surfactant. This study shows that the size of NPs depends on the ratio between the surfactant and silica precursor. Apart from that, Sihler et al.
[63] used dye-stabilized emulsion to synthesize SiO
2 NPs. Moreover, this synthesis method provides silica capsules and sub-particles with precise size control. Monodispersed colloidal silica NPs (diameter of 15–25 nm) were prepared by Murray et al.
[64]. In this study, as the silica source, octadecyltrimethoxysilane (OTMS) was used.
Simple synthesis methods called soft and hard templating are also applied to increase the pore volume and loading capacity of prepared hollow mesoporous SiO
2 [65]. Template synthesis of mesoporous materials typically enrolls in three steps: template preparation, template-directed synthesis of the target materials using sol-gel, precipitation, hydrothermal synthesis, and template removal
[66][67].
The hard-templating method involves nano-casting using pre-synthesized mesoporous solids
[68]. Hard templating is a facile synthesis method for fabricating porous materials with a stable porous structure. The structure replication is very straightforward
[66]. This approach utilizes porous hard templates such as mesoporous silica. The pores of these templates are impregnated with a precursor compound for the desired product, which is then thermally converted into the product. The template is finally removed to yield the desired mesoporous material as a negative structural replica of the hard template
[66]. However, the method is costly and time-consuming. Moreover, the mesoporous parameters, such as mesostructure and pore sizes, are difficult to change
[67].
In contrast, soft templating methods use cationic and anionic surfactants or block copolymers as templates
[61]. During the synthesis, surfactant or block copolymers are used as a soft template. Moreover, the increase in surfactant micelle concentration causes the formation of a large assembly or self-assembly of 3D mesoporous
[22]. Different 3D micelle structures can be obtained by varying the solvent ratio between the aqueous and non-aqueous and adding co-solvents. Moreover, the silica source interacts with the structure-directing agent (SDA) without any phase separation. The interactions between ions or charged molecules are vital in forming well-defined porous nanostructures
[68].
The soft templating method mainly depends on the self-assembly of the surfactant
[66]. The process is based on the interactions between inorganics. The mesoporous structure of the final material is obtained after the removal of the pore-templating surfactant or block copolymers by low-temperature calcination (up to 600 °C) or by different washing techniques (extraction)
[66].
Figure 3 represents the synthesis mechanism of mesoporous silica in the presence of a cationic surfactant. The synthesis process of mesoporous silica is carried out using TEOS as the silica source
[22]. In this process, surfactant plays a significant role in defining the pore size and volume of silica
[22]. Cationic surfactant forms micelle structures with water, which arranges the cationic “heads” of the surfactant molecules to the outer side. It resulted in the hydrophobic “tails” collected in the center of each micelle. As the next step, silica molecules cover the micelle surface. Finally, the surfactant is removed via calcination or extraction, and it results in porous silica
[22][69][70].
Figure 4 shows the schematic diagram for synthesizing mesoporous silica using block copolymer. As can be seen from
Figure 4, titania-incorporated organosilica-mesostructures (Ti-MO) are synthesized via condensation method using silica precursors ([3-(trimethoxysilyl) propyl] isocyanurate and tetraethylorthosilicate) and titanium precursor (titanium isopropoxide) in the presence of the triblock copolymer, Pluronic P123
[71]. This method consists of template removal using two independent steps (i) extraction with a 95% ethanol solution and (ii) calcination of the sample at 350 °C. This method improves the adsorption capacity and enhances the structural properties such as specific surface area, micro-porosity, and pore volume.
Figure 3. Mechanism for the synthesis of mesoporous silica in the presence of a cationic surfactant.
Figure 4. Mechanism for the synthesis of mesoporous silica using block copolymer.
The synthesis of MCM-41 and SBA-15 is performed using cetrimoniumbromide (CTAB) and Pluronic P123 surfactant. The CTAB is an ionic surfactant and acts as stearidonic acid (SDA) and which causes the formation of a hexagonal array of mesostructured composites
[11]. However, as the final step, surfactants are removed by heating in air at high temperatures or by solvent extraction to obtain MCM-41 and SBA-15
[22]. Wu et al.
[62] and Hao et al.
[71] reported a detailed description of the mechanism. Paneka and co-workers have reported the synthesis of MCM-41 from fly ash using a hydrothermal process. However, the synthesis of MCM-41 shows reduced BET surface area, increased pore volume, and pore size
[72].
Recently, Singh and Polshettiwar
[73] reported the synthesis of silica nano-sheets using ammonium hydroxide. They have developed a method to synthesize silica nano-sheets using lamellar micelles as soft templates in a water-cyclohexane solvent mixture. Zhang et al.
[74] also reported the large-scale synthesis of mesoporous silica nanoparticles. Reported data show that various morphologies and particle sizes have been obtained during the synthesis. For synthesis process, the reaction occurred at atmospheric pressure with a sol–gel technique using CTAB as a template.
3.3. Importance of Micro-Porosity and CO2 Adsorption Capacity of Mesoporous Silica Materials
The textural properties, including surface area, pore diameter and volume of mesoporous materials, are usually measured by studying nitrogen adsorption–desorption isotherms. The specific surface area is calculated using the volume adsorbed at different relative pressure data by the Brunauer–Emmett–Teller (BET) method
[48]. Apart from that, the pore volume and pore size distribution are determined using the Barrett–Joyner–Halenda (BJH) method
[48].
Furthermore, the textural properties are important parameters when considering CO
2 adsorption using physisorbents. Moreover, microporosity plays a major role in CO
2 gas adsorption because it involves the diffusion of CO
2 molecules into the physisorbent
[75][76][77].
The pore size of mesoporous materials varied in the descending order of KIT-6 > SBA-15 > SBA-16 > MCM-48 > MCM-41. The KIT-6 exhibited the largest pore volume among the other sorbents. These combined features of large pore size and large pore volume would enable KIT-6 to better accommodate the bulky polyethyleneimine (PEI) with little hindrance, allowing higher loadings inside silica particles than other silica-supported materials. Moreover, Zelěnák and co-workers prepared three mesoporous silica materials with different pore sizes (3.3 nm MCM-41; 3.8 nm SBA-12; 7.1 nm SBA-15)
[78]. During their studies, amine functionalization was investigated with the effect of pore size and architecture on CO
2 sorption. According to the data, SBA-15 showed the highest CO
2 adsorption of 1.5 mmol/g due to the highest amine surface density in SBA-15
[78].
Lashaki and Sayari
[79] also investigated the impact of the support pore structure on the CO
2 adsorption performance of SBA-15 silica. In this study, SBA-15 silica supports were used to obtain different pore sizes and intra-wall pore volumes. These materials were functionalized further with triamine through dry and wet grafting. CO
2 sorption measurements showed the positive impact of support with large pore size and high intra-wall pore volume on adsorptive properties, with the former being dominant. Large pore volume influenced the load of more amine groups, CO
2 uptakes, and CO
2/N
2 ratios and faster kinetics. When the intra-wall pore volume decreased by 53%, it caused a reduction in CO
2 uptake capacity by up to 63% and CO
2/N
2 ratios by up to 62% and slower adsorption kinetics. Moreover, it was inferred that large pore size and high intra-wall pore volume of the support improved the adsorptive properties via enhanced amine accessibility
[79].