Adsorption capacities at high pressure
In this report, the adsorption of CO2 on metal organic frameworks (MOFs) is comprehensively reviewed. In Section 1, the problems caused by greenhouse gas emissions are addressed, and different technologies used in CO2 capture are briefly introduced. The aim of this chapter is to provide a comprehensive overview of CO2 adsorption on solid materials with special focus on an emerging class of materials called metal organic frameworks owing to their unique characteristics comprising extraordinary surface areas, high porosity, and the readiness for systematic tailoring of their porous structure. Recent literature on CO2 capture using MOFs is reviewed, and the assessment of CO2 uptake, selectivity, and heat of adsorption of different MOFs is summarized, particularly the performance at low pressures which is relevant to post-combustion capture applications. Different strategies employed to improve the performance of MOFs are summarized along with major challenges facing the application of MOFs in CO2 capture. The last part of this chapter is dedicated to current trends and issues, and new technologies needed to be addressed before MOFs can be used in commercial scales.
- CO2 capture
- solid sorbent
1.1. Environmental problem and CO2 emissions
The increasing level of CO2 emission is considered one of the major environmental challenges that our planet is facing today. The concentration of greenhouse gases in the atmosphere reached a new record in 2013, with CO2 at 396 ppm which represents 142% of the concentration of the pre-industrial era . Findings of a recent global atmosphere watch reported in a greenhouse gas bulletin  revealed that CO2 concentration has increased between 2012 and 2013, more than any other year since 1984, which was attributed to the reduced uptake by the earth’s biosphere. This alarming level of CO2 shows the urgency for taking immediate actions to prevent serious repercussions of climate change. On December 2015, at the Paris Climate Conference (COP21), 195 countries adopted a historical and the first legally binding global climate agreement to keep the increase in global average temperature to well below 2oC above pre-industrial levels. The discovery of new fossil fuel reserves, combined with rising energy demand, led to an increase in the number and capacities of power plants worldwide. This situation is expected to extend into the future due to various factors such as industrial development and economic growth, especially in developing nations, which in turn is expected to further contribute to increasing levels of greenhouse gas emissions in years to come. According to a recent report by the Energy Information Administration, energy consumption is projected to rise by 56% between 2010 and 2040. Fossil fuels will continue to supply about 80% of the world energy through 2040. Industrial energy consumption represents the greatest share of emissions and is projected to consume more than 50% of the energy delivered in 2040. According to currently implemented regulations regarding fossil fuels, CO2 emissions from power plants is projected to increase by 46% compared to emission level in 2010 .
Among several approaches that could be used to overcome the greenhouse gas effect is the utilization of clean energy alternatives which could be the ultimate solution to the climate change problem in terms of reducing CO2 emissions. However, these green technologies still require significant modifications to the current energy framework. The great challenges facing these green technologies lie in the difficulty for implementation at industrial scale, which makes it economically infeasible when compared to fossil fuel-based power plants. This implies that unless green energy alternatives and energy infrastructure for the commercialization and the implementation of these new technologies are attained, the pursuit of new CO2 emission reduction technologies will continue to be the most practical method to address greenhouse gas effects until the advancement in clean energy technologies reaches commercial stages.
There are three different strategies to reduce emissions of CO2 from fossil fuel-based power plants. These include post-combustion capture in which CO2 is separated from the combustion flue gas stream that is mainly composed of nitrogen and some other minor components such as water vapor and oxygen. The separation process in this scheme is a downstream unit which allows for an easy retrofit of a post-combustion capture unit to an existing power plant. However, the limitations of this technique include a low CO2 partial pressure, relatively high flue gas temperature and large quantities of CO2 in the flue gas stream [3, 4]. In the pre-combustion capture scenario, the fossil fuel is treated under certain temperature and pressure to gasify the fuel and produce hydrogen. This method offers streams with high CO2 partial pressure and thus easy separation by utilizing variety of solvents; however, it requires significant modifications to the power generation plant. The last scenario is called the oxy-fuel capture in which the fuel is burned under a pure oxygen environment which requires the separation of oxygen/nitrogen from an air stream. The process produces pure CO2 and water vapor which can be easily recovered through a simple condensation unit. Each separation scenario requires a different capture technology, and therefore the properties, characteristics, and operation of the separation process are also entirely different among the three strategies. The most advanced process for implementation in the field is post-combustion. We will therefore, in this chapter, focus on the post-combustion separation applications.
1.2. Existing technologies for CO2 capture
In order to locate metal organic frameworks (MOFs) on the map of the technologies used for CO2 capture applications, we will briefly describe the major technologies that have been employed and discuss their advantages and limitations. Figure 1 shows the different technologies used for CO2 capture, whereas MOFs are used under the category of membranes and adsorbents.
The most widely investigated technology for CO2 capture from flue gas is absorption using aqueous amine solutions such as monoethanolamine (MEA), diethanolamine (DEA), and methyldiethanolamine (MDEA), as well as blends of different amines [5–7]. Amine scrubbers are considered a well-developed technology and is available in commercial scale for post-combustion capture applications . The major limitations of this technology include the high energy required for solvent regeneration, stability of the amine system at the regeneration conditions, and the negative influence of impurities present in the flue gas that might significantly affect the stability and performance of the solvent [9, 10].
Under the category of absorption technology and in order to overcome the limitations of amine-based technologies, aqueous ammonia as a solvent for CO2 separation has also been widely used to benefit from the low heat of absorption of ammonia-based solvents as compared to amine systems. Besides, the ammonia can also absorb other impurities existing in the gas stream such as NO and SOx. The major drawback of ammonia-based solvents lies in the need to cool the flue gas prior to introducing it to the absorption column to prevent ammonia losses to the gas stream. This adds a huge energy requirement considering the large volume of flue gas stream that typically needs to be treated . The chilled ammonia process faces similar issues in addition to fouling of the heat exchanger by ammonium bicarbonate deposition from saturated solutions .
Great efforts have been made to find new and efficient materials for absorptive CO2 separation. Ionic liquids (ILs) are liquid salts composed of cations and anions, have been proposed as promising solvents to replace the existing amine-based solvents. ILs possess several remarkable properties that make their application in CO2 separation one of the hottest research topics in the last few years [12–14]. These properties include low volatility, high CO2 solubility, good thermal stability, and the possibility of systematically tuning the structure toward certain properties [15–17]. Several review papers reporting experimental data related to CO2 solubility, selectivity, effect of ILs structure on performance, and the stability of ILs are available [12, 18]. Recent developments on the application of the amine-modified ILs, known as task-specific ILs (TSILs), are also widely discussed in the literature [19, 20], including both physical and chemical interactions with CO2. Unfortunately, many ILs and TSILs suffer from a common problem of high viscosity after CO2 absorption. Even though some recent reports mentioned the availability of ILs with low viscosities, it is still evident that much work has to be done to overcome this limitation. Finding cheap routes for the synthesis of these materials is one of the greatest challenges facing researchers working in this area . In this chapter, a great portion will be dedicated to the incorporation of ILs into the pores of MOFs to improve their CO2 capture capabilities.
2. CO2 capture using solid sorbents
2.1. Criteria for the evaluation of solid sorbents
In order to evaluate solid materials for their performance in CO2 separation from flue gases, some important performance criteria must be met. These include:
Adsorption capacity: it is a key criteria in evaluating solid sorbent performance. It provides information on the amount of CO2 that could be adsorbed by a given solid material. It can be represented in terms of gravimetric uptake which is the amount of CO2 adsorbed per unit mass sorbent (gram CO2/gram sorbent, or cm3 CO2/gram sorbent). The volumetric uptake is another measure for capacity, and it reports the CO2 uptake per volume of sorbent material (gram CO2/cm3 sorbent, or cm3 CO2/cm3 sorbent). This criterion is of great importance because it represents the amount of sorbent needed for a particular duty and therefore the size of the adsorption bed. It is also considered a crucial factor in determining the energy requirement during the regeneration step.
Selectivity for CO2: it represents the CO2 uptake ratio to the adsorption of any other gas (typically nitrogen for post-combustion capture, and methane for natural gas). It is an essential evaluation criterion, and affects the purity of the adsorbed gas, which will significantly influence the sequestration of CO2. The simplest method to estimate the selectivity factor is to use single-component adsorption isotherms of CO2 and nitrogen.
Enthalpy of adsorption: it represents another critical parameter in the evaluation of the performance of solid sorbents. It is a measure of the energy required to regenerate the solid sorbent, and it therefore significantly influences the cost of the regeneration process. It represents the affinity of the material toward CO2 and measures the strength of the adsorbate–adsorbent interaction.
Physical, thermal, and chemical stability: in order to reduce operating costs, solid sorbents must demonstrate stability under flue gas conditions, adsorption operation conditions, and during the multi-cycle adsorption–regeneration process. In particular, stability in the presence of water vapor is essential for the sustainable performance of the solid sorbent. In addition to thermal properties of the solid sorbent, heat capacity and thermal conductivity are also important in heat transfer operations.
Adsorption/desorption kinetics: the time of the adsorption–regeneration cycle greatly depends on the kinetics of the CO2 adsorption–desorption profile, which is measured in breakthrough experiments. Sorbents that adsorb and desorb CO2 in a shorter time are preferred as these reduce the cycle time as well as the amount of sorbent required, and ultimately the cost of CO2 separation.
Cost of the sorbent material: it is an important factor in the selection of the sorbent material. Materials that exhibit excellent adsorption attributes, and are readily available at low cost, are considered the main targets for researchers working in the field of CO2 capture. Besides, the environmental impact of synthesizing these materials is considered one of the greatest challenges to overcome.
In the following sections, we describe the main solid sorbents used for CO2 capture, their applications, major attributes, and limitations.
Zeolites are porous crystalline aluminosilicate materials available naturally, but can also be prepared synthetically. The zeolite framework is composed of tetrahedral T atoms where T could be Si or Al, connected by oxygen atoms to form rings of different pore structures and sizes. The pore size of the zeolite framework varies between 5 and 12 Å . They are widely used as catalysts in the refining industry [24, 25], fine chemicals synthesis [26, 27], and in gas separation applications [28, 29]. Zeolites are considered promising candidates in CO2 capture application as has been widely reported in the literature [30–32]. CO2 can be adsorbed on zeolites through different mechanisms, such as molecular sieving effect based on the difference in size [33, 34]. Separation can also take place based on polarization interactions between the gas molecule and the electric field on the charged cations in the zeolite framework . Accordingly, CO2 removal with zeolites can be controlled by changing the pore size, polarity, and the nature of the extra framework cation. Among the different zeolites investigated for CO2 capture applications, zeolite 13X is the most widely studied sorbent, and is considered the benchmark technology for solid sorbents [35, 36]. Research on the use of zeolites as sorbents for CO2 capture can be categorized into different areas depending on the approach and the techniques adopted to address the improvement in capture performance. These categories comprise tuning the pore size, designing zeolites with controlled polarities, investigating novel zeolites, optimizing the cation exchange, and most recently incorporating amine moieties and other chemical functions into the zeolite frameworks. Ocean et al.  have studied the selectivity to adsorb CO2 by controlling the pore size of an NaKA zeolite through the synthesis of nanosized NaKA zeolites. Overall, the adsorption kinetics on the nanosized crystals was fast enough for CO2 capture applications; however, the formation of a thin layer on the nanosized NaKA zeolite, due to intergrowth on the surface, did not considerably improve the adsorption kinetics. In contrast, Goj et al.  performed atomistic simulations for silicalite, ITQ-3, and ITQ-7, and reported a positive effect on CO2 uptake and selectivity by tuning the pore apertures. Sravanthi et al.  provided a novel approach to control the pore size and volume by utilizing pore expansion agents and obtained average pore size around 30 nm. The application of the pore-expanded MCM-41 in CO2 separation resulted in the uptake of about 1.2 mmol/g.
Several studies have been conducted to control zeolite affinity toward CO2, which can be realized by tuning the polarity of the zeolite through alteration of the Si/Al ratio and the nature of the cation. Remy et al.  studied the selective separation of CO2 on low-silica KFI zeolite (Si/Al = 1.67) by employing ion exchange with Na, Li, and K. Li-exchanged KFI has shown the highest CO2 uptake which was attributed to the large pore volume as compared to Na and K cations. In comparison with high-silica KFI sample (Si/Al = 3.57–3.67), Li–KFI had the highest capacity at low pressure due to the strong electrostatic field. The overlap between pore size and polarity effects is also strongly observed for amine-supported zeolites, which have gained considerable attention in the last few years [40–45]. For instance, Ahmad et al.  have impregnated melamine into β-zeolite and obtained dynamic CO2 uptake of 3.7 mmol/g at atmospheric pressure and temperature of 25 °C. The major challenge facing amine-modified zeolites is the tradeoff between the increased affinity toward CO2 (strong interaction with the sorbent) and the reduction in pore volume, and consequently the uptake, especially, at low pressures. Factors such as amines loading, distribution, and the nature of the cation can play a vital role to avoid the blockage of the porous structures with the bulky amine moieties [42, 47]. Kim et al.  have performed a rigorous investigation through the simulation of thousands of zeolites to evaluate the adsorption properties of these materials and identify the optimum structures for improved CO2 separation attributes. This study provides a systematic approach to rank and select appropriate zeolites for the required capture objectives. However, important factors such as stability under humid environment, adsorbent and process cost, and the availability of zeolite structures were not taken into consideration.
The hydrophilic nature of most zeolite structures is considered a major drawback of zeolites especially for post-combustion CO2 applications [49, 50]. Water competes with CO2 on the available sorption sites and might influence the zeolite structure and framework . As explained earlier, the presence of the exposed cation sites increases CO2 uptake. In a recent study by Serena et al. , the relationship between the water content of the zeolite and the density of the cations was investigated, and a linear relationship was found to describe the decrease of the cation population with increasing water content. This observation highlights the detrimental effect of the presence of water vapor on the adsorption of CO2 on zeolites.
2.3. Carbon-based CO2 capture
Carbon-based adsorbents have been used for CO2 separation in different forms including activated carbons (ACs), carbon nanotubes (CNTs), and graphenes. Activated carbons have an amorphous porous structure with high surface areas that are readily available for CO2 uptake. They have been widely investigated as sorbents for CO2 removal due to their low cost and the availability of raw materials [53–55]. However, there are no active sites to bond with the adsorbed CO2 as cations in zeolite sorbents. This weak interaction results in lower enthalpy and therefore lower energy requirement for regeneration. On the contrary, ACs have very low CO2 uptake at low pressures due to the absence of the electric field on the surface. Kacem et al.  performed a comparison between the performance of ACs and zeolite for CO2/N2 and CO2/CH4 separation based on their capacity, regeneration capacity, and reusability. It was concluded that at higher pressures (above 4 bars), the CO2 uptake for ACs was much higher than zeolites. Also, the recovered CO2 after the regeneration of ACs had higher purity than in the case of zeolites. When compared to zeolites, ACs maintain their adsorption stability even in the presence of water vapor which does not cause any framework failure .
In order to enhance the adsorption capacity on ACs, several studies have been conducted in order to improve the affinity toward CO2 by introducing amine-based functional groups [58–61]. In a recent study, Maria et al.  described a systematic surface modification of microporous ACs through a stepwise chemical treatment. They were successful in grafting amine and amide functional groups on the surface of ACs with only 20% loss of surface area. Gibson et al.  studied the polyamine-impregnated porous carbons and achieved 12 times higher CO2 capacity than bare porous carbon. Chitosan and triethylenetetramine have been successfully impregnated onto the surface of ACs and have shown 60 and 90% increased CO2 uptake at 298 K and 40 bars. In addition to amine functional groups, ammonia-modified ACs, at atmospheric pressure and a temperature range from (303 to 333) K, have been studied . Authors report that an enthalpy of 70.5 kJ/mol was obtained compared to 25.5 kJ/mol for the pristine ACs, suggesting the possibility of chemisorption. Another report has also supported the improved adsorption capacity and selectivity by employing NH3 at high temperature and has considerably improved CO2 uptake from 2.9 mmol/g for the bare AC to 3.22 mmol/g for the modified one at 303 K and 1 bar.
Several studies have been dedicated to the application of amine-modified carbon nano tubes (CNTs) as solid sorbents for CO2 separation [65–69]. Industrial grade CNTs have been functionalized with tetraethylenepentamine (TEPA) by Liu et al. , and the effects of amine loadings on the CO2 uptake, heat of adsorption, and adsorbent regenerability were investigated. TEPA-impregnated CNTs have shown an enhanced capacity of 3.09 mmol/g at 343 K. Similar studies were also reported using different amines such as (3-aminopropyl)triethoxysilane (APTES) , polyethyleneimine (PEI) , and other amines (primary, secondary, tertiary, diamines, and tri-amines) .
Graphene is a planar sheet of carbon atoms extended in two dimensions, and was discovered in 2004 . Graphite-based capture was recently introduced (after 2011) as a promising candidate for CO2 capture applications, and research is growing rapidly in this area [73–77]. A recent review by Najafabadi is available on the current status and research trends of using graphene and its derivatives as solid sorbents for CO2 capture . Research in this area involves grafting various functional groups on graphene such as N-doped graphene composites (surface area = 1336 m2/g), as reported by Kemp et al. , which showed a reversible CO2 capacity of 2.7 mmol/g at 298 K and 1 atm as well as enhanced stability for repeated adsorption cycles. Borane-modified graphene was also reported by Oh et al. , obtaining a CO2 uptake of 1.82 mmol/g at 298 K and 1 atm. Some novel hybrid materials have also been introduced to obtain better improvements in the adsorption properties, including mesoporous graphene oxide (GO)-ZnO nanocomposite , mesoporous TiO2/graphene oxide nanocomposites , Mg–Al layered double hydroxide (LDH), graphene oxide , MOF-5 and aminated graphite oxide (AGO) , UiO-66/graphene oxide composites , and MIL-53(Al) and its hybrid composite with graphene nanoplates (GNP) .
2.4. Metal organic frameworks
A more recent class of porous materials was manufactured and named metal organic frameworks. They represent one of the promising adsorbents and have gained significant attention during recent years for gas separation applications [87, 88]. MOFs are composed of metal ions or clusters (nodes) bridged by organic ligands (connecters) to form various structures and networks. MOFs are well recognized for their extraordinary surface areas, ultrahigh porosity, and most importantly the flexibility to tune the porous structure as well as the surface functionality due to the presence of organic ligands that can easily be chemically modified [89, 90]. One main advantage of MOFs over other solid materials is the possibility to tailor the pore size and functionality by rational selection of the organic ligand, functional group, metal ion, and activation method.
Several review papers are available in the literature for gas separation using MOFs [91–96]; however, great progress has been achieved during the past four years (2012 onward). In order to address the limitations of MOFs and investigate new structures, novel functional groups, in addition to hybrid systems and technologies, more studies are needed to explore the mechanisms involved and to improve the uptake capacity in a humid environment. For these reasons, considerable effort has been observed during the past decade to address gas separation and adsorption using MOFs. Figure 2 shows the number of publications on CO2 capture and separation using MOFs during the past 15 years, which reflects the growing interest of MOFs as efficient solid sorbents.
3. Adsorption of CO2 on metal organic frameworks
CO2 capture performance of different MOFs will be comprehensively reviewed in terms of their capacity, selectivity, heat of reaction, and major challenges facing researchers, and some ideas to approach these challenges will also be provided. The next section is dedicated to review the most recent studies of CO2 capture and separation on MOFs, and we will mainly target the works published in the last four years.
3.1. Evaluation of MOFs in CO2 Capture
As introduced earlier, capacity, selectivity, and heat of adsorption are considered the main criteria for the evaluation of MOFs for CO2 separation. CO2 uptake is a proportional function of pressure in the gas phase, where the low pressure corresponds to post-combustion applications. The gravimetric uptake of CO2 is indicative of the ability of MOFs to adsorb CO2 and, therefore, we have reported CO2 uptake along with MOF surface area, and other properties for MOFs published after 2012 which could be added to the published reviews that have listed these data in a table format. Table 1 represents the properties of MOFs at high-pressure applications, while Table 2 presents the low-pressure data.
|Surface Area (m2/g)|
|Common name||BET||Langmuir||Capacity (wt%)||Pressure|
|Basolite® C 300||1706.42||41.9||224.99||318||18|||
|a IAST selectivity|
|Surface Area (m2/g)|
|Common name||BET||Langmuir||Capacity (wt%)||Pressure|
|Temp. (K)||Selectivity||Qst (kJ/mol)||Ref.|
|Basolite® C 300||1706.42||9.4||0.95||318||18|||
|a IAST selectivity|
3.2. Strategies to Improve the CO2 Capture Performance on MOFs
Several strategies have been adopted to improve the performance of MOFs in CO2 capture applications. The ability to precisely tune the MOF structures has led to versatile approaches that can be utilized to enhance CO2 uptake, selectivity, and the affinity toward CO2. These methods could be classified into effects of open metal sites, pre-synthetic modifications of the organic ligand, and post-synthetic functionalization schemes.
3.2.1. Open Metal Sites
Open metal sites in MOFs are formed by the removal of a solvent molecule coordinated to the metal nodes by applying vacuum and/or heat after the synthesis of framework in a process called “activation.” The presence of open metal sites on the MOF framework has a great impact on the selectivity toward CO2 as well as on the binding energy between the adsorbed CO2 molecules and the surface of MOF sorbents. These coordinately open metal centers act as binding sites where CO2 molecules can attach and bind to the pore surface by the induction of dipole–quadrupole interactions. Allison et al.  have developed a systematic procedure to precisely understand the interactions between the CO2 molecule and the force field generated by the open metal sites in MOF-74. The developed method allows for accurate estimation of adsorption isotherms using computational approach which enables the evaluation of different hypothetical open metal sites. These observations confirm previous findings of Kong et al. on understanding CO2 dynamics in MOFs with open metal centers . Among the MOF family, HKUST-1, M-MIL-100, M-MIL-101, and M-MOF-74 are the most widely studied frameworks with open metal sites (M represents the metal site). However, to precisely investigate the influence of the open metal sites, we need to isolate the effects of the nature of organic ligands, the synthesis route, and functional groups present in the framework. It was observed that utilizing light metal sites provides higher surface areas, and therefore improve CO2 uptake at low pressures for MOF-74 . Several studies have reported the effects of metal centers using computational approach as reported for M-MOF-74 [138–140] where noble metals such as Rh, Pd, Os, Ir, and Pt are considered promising candidates for CO2 capture (see Figure 3).
Casey et al.  studied the isostructural series of HKUST-1 for various metal centers (Mo, Ni, Zn, Fe, Cu, and Cr) to get insights into the adsorption mechanism and the force field created by different metal types. It was found that the presence of divalent metals such as Mg2+ significantly increased CO2 binding strength and resulted in higher selectivity toward CO2. In addition to the nature of the metal nodes, it was found that the activation method plays a vital role in determining CO2 uptake and affinity toward CO2 which was in agreement with Llewellyn et al.  for MIL-100 and MIL-101, where various activation methods resulted in different CO2 loadings and heat of adsorption.
In a recent study, Cabelo and coworkers  investigated the interaction between CO2 and the unsaturated Cr(III), V(III), and Sc(III) metal sites in MIL-100 framework using variable temperature infrared spectroscopy. The enthalpy of adsorption for Cr(III), V(III), and Sc(III) were amounted to be (−63, −54, and −48) kJ/mol, respectively, which are considered among the highest values for CO2 adsorption on MOFs with open metal centers to date. The synthesis and characterization of an M-DABCO series (M = Ni, Co, Cu, Zn) were described by Sumboon et al.  to systematically evaluate the effect of the metal identity on surface area, pore volume, and CO2 uptake. It was concluded that Ni-DABCO has shown the highest pore volume and specific surface area due to the high charge density present at the metal center. Comparison of the M-DABCO with activated carbons and MIL-100(Cr) revealed that the unsaturated cations possess exceptional CO2 uptake of 180 cm3/g at 1 bar and 298 K [as compared to 30 cm3/g for ACs and 60 cm3/g for MIL-100(Cr),144].
3.2.2. Pre-synthetic Modifications of MOFs
Organic ligands are the linkers that connect the metal nodes together and therefore determine the framework structure, pore volume, pore window, and surface area which are very crucial characteristics in CO2 separation process. Ligand functionalization is considered to be a powerful tool to improve the adsorption of CO2 on MOFs due to the wide range of functional groups and the ease of modifying the organic ligand through strong covalent interactions. In a recent computational work by Torrissi et al., the impacts of various functional groups attached to the ligand part were investigated by density functional theory (DFT) . The incorporation of amine functional moieties to the organic ligands has witnessed much attention in recent years, due to the proven positive effect of the presence of open nitrogen sites on the MOF frameworks . Keceli et al.  studied four biphenyl ligands modified with amide groups of different chain lengths. Varying the length of the alkylamide group has shown a great impact on the porosity, surface area, and CO2 capacity. It was also evident that the activation procedure has great influence on the surface area of the resulting material which is attributed to the different mechanisms of solvent removal from the MOF framework. Three amino-functionalized MOFs have been prepared from 2-aminoterephthalate (ABDC) and three different metals (Mg, Co, and Sr). Despite a low surface area (63, 71, and 2.5 for Mg, Co, and Sr, respectively) and a relatively low CO2 uptake (1.4 mmol/g at 1 bar and 298 K), the prepared MOFs had exceptional selectivity toward CO2 (396 was recorded for Mg-ABDC) and exhibited high heat of adsorption . Shimizu and coworkers  used 3-amino-1,2,4-triazole ligands to design a 3D structure MOF with 782 m2/g surface area and 0.19 cm3/g pore volume that is capable to achieve CO2 uptake of 4.35 mmol/g at 1.2 bar and 273 K. Moreover, the as-synthesized MOF has shown enthalpy of adsorption of 40.8 kJ/mol at zero coverage which was comparable to the commercial zeolite NaX (48.2 kJ/mol). In a similar study, Xiong et al.  used triazole ligands to prepare a new framework called UTSA-49 by incorporating nitrogen atoms and methyl functional groups on 5-methyl-1H-tetrazole ligands which recorded 13.6 wt% CO2 uptake at 1 bar and 298 K and 27 kJ/mol enthalpy (Figure 4). These observations were in agreement with work reported by Gao et al. for the influence of triazolate linkers . It is essential to understand the synergistic effect between the multiple functional groups on the pore surface and their size exclusion effects which are considered potential approaches to optimize the performance of functionalized MOFs. Table 3 summarizes CO2 capture properties of MOFs modified with different amino functional groups.
|MOF name||Type of functional group||CO2 uptake (wt. %)||Enthalpy of adsorption (kJ/mol)||Pressure||Temperature||Surface area (m2/g)||Ref.|
|Zn4(bpta)2-1||Bipyridine pillar ligands||8.2||34.82||1.2||298||413|||
|Zn4(bpta)2-1||Bipyridine pillar ligands||3.1||27.69||1.2||298||51|||
|Zn(ad)(ain)||2-Aminoisonicotinate and adeninate||9.2||40||1||298||399|||
Apart from amine groups, there are other functional moieties that are proven to be effective in enhancing the performance of MOFs in CO2 capture. Phosphonate and sulfonate organic ligands have gained tremendous attention recently due to their significant improvements in MOF stability toward water . Several studies are reported based on the use of phosphonate and sulfonate ligands, for instance, the selective CO2/N2 separation over nitrogen-containing phosphonate MOFs was studied by Marco et al. , and the synthesis, stability, porosity of the phosphonate MOFs , and their major applications were reported for water stability studies [159–161]. The shielding effect exhibited by phosphonate groups were responsible for the improved stability under humid conditions up to 90% relative humidity at 353 K as observed for CALF-30 . The enhanced water stability of these MOFs was attributed to the kinetic blocking effect which makes the framework completely hydrophobic .
MOFs containing nitrogen-donor building blocks were also widely investigated, particularly adenine group which was extensively used due to framework robustness, richness in nitrogen sites, and framework diversity . Song et al.  reported the preparation of three new adenine-based MOFs by controlling the adenine coordination with Cd metal sites. This study has provided insight into the controlled synthesis of MOFs by controlling the structure building units (SBU) which can be utilized to extend the idea to include multiple building units within the same framework. Similar studies are also available based on adenine groups as building units, where the effect of the adenine functionalization on framework topology, porosity, and adsorption behavior was investigated . The use of Zn-adeninate SBU led to the discovery of highly porous Bio-MOF-11 to 14 series  and Bio-MOF-100  with exceptional surface area (4300 m2/g) and very large pore volume (4.3cm3/g); however, the framework stability of these materials still needs to be addressed as the material tends to lose its porosity under harsh activation environment. This issue has been tackled by Zhang et al.  to prepare more stable adenine-based PCN-530 structure. Lin et al. have observed high density of open nitrogen-donor sites on 1,3,5-tris(2H-tetrazol-5-yl)benzene (H3BTT) which was responsible for the enhanced CO2 capacity  through the improvement of the framework porosity and the utilization of nitrogen sites readily available to adsorb CO2. However, the richness of nitrogen atoms in the framework does not necessarily favor CO2 adsorption, as reported by Gao et al.  for the case of tetrazolate-based rth-MOF that has more exposed nitrogen sites as compared to pyrazolate-based rht-MOF and yet was showing less CO2 uptake attributed to the strong electric field observed on the pyrazolate-based rht-MOF.
Other ligand modifications are also reported in literature by deploying several types of functional groups such as hydroxyl groups (OH) on Zn(BDC) , (CH3)2, (OH), and (COOH) on MIL-53(Al) , NO2 on IRMOF-8  as well as alkyl and nitro groups grafted on DUT-5 . Based on the contribution by Yaghi’s group , several studies were dedicated to understand the effects of ligand extension on the pore size, surface area, and the sorption behavior of MOFs [98, 109, 133, 171–173]. Recently, zeolite-like MOFs denoted as ZTIFs have attracted great interest due to their unique characteristics for tuning the structure toward various applications [174, 175]. New frameworks (ZTIF-1 and ZTIF-2) were recently reported based on the incorporation of tetrazolates into Zn-Imidazolate structures , with similar structures. UTSA-49 was also reported by Chen and coworkers for the selective separation of CO2/N2 mixture .
Lately, the idea of mixed ligand approach for the synthesis of MOFs with tunable properties has gained much attention which allows for incorporating several functionalities within each ligand to target certain properties such as improving the stability and the capacity for CO2 simultaneously . For instance, the water stability issue was tackled by Marco and coworkers  by utilizing two heterocyclic N-donor-mixed phosphonate-based organic ligands. The designed MOF has shown great water stability and achieved CO2 uptake of 77 cm3/g at low pressure and 195 K. By deploying the mixed-ligand approach, Liu et al.  have successfully prepared Co-based MOFs containing both benzenetricarboxylic- and triazole-based ligands by using a solvo-thermal synthesis technique. The synthesized MOFs displayed CO2 uptake of up to 15.2 wt. % at 1 bar and 295 K as well as remarkable selectivity toward nitrogen. A detailed investigation of mixed ligand approach in the design of MOFs is available in literature [180, 181]; however, further work is still needed to optimize the synthesis conditions and correlate the observed performance to the appropriate constituents on the organic ligands. Recent work by Yaghi et al. provided tools to quantitatively map different functional groups incorporated into the same MOF structure .
3.2.3. Post-synthetic Functionalization of MOFs
As mentioned previously, tuning the affinity of the framework functionalities toward CO2 is crucial for improving adsorptive capacities. The aim is to decorate the pore surface in order to have high adsorption selectivity and capacity and yet minimize the regeneration energy. In addition to the pre-synthetic modification of the organic linker, post-synthetic functionalization of MOFs (PSM) is considered a viable route to insert functionalities into the MOF structure after the formation of the basic framework. This approach can overcome the limitations observed in pre-synthetic functionalization, for instance precise control of the synthesis conditions is needed to preserve the functional groups during the solvo-thermal synthesis conditions. Note that some functional groups are not stable under synthesis conditions which require a narrow range of conditions to prepare the MOFs. Others, however, cannot be introduced to the synthesis mixture due to solubility issues, hindrance effects, and they might participate in the crystallization process and yield unwanted materials. Besides, inserting functional groups on the metal sites prior to the synthesis of the framework might intervene in the formation of the building units which can result in the deterioration of the crystal structure [183–185]. PSM is therefore considered an attractive pathway to tailor the properties of MOFs toward better CO2 capture performance.
In order to make use of high amine affinity toward CO2, several amine moieties were selected for the modification of various solid sorbents [186–189] including MOFs [190, 191]. Ethylene diamine (en) is considered the most commonly used type of amine for PSM of MOFs for CO2 capture application. In 2014, Lee et al.  reported grafting the diamine into the expanded MOF-74 or Mg(dopbdc) structure at amine loadings of 16.7 wt. % at room temperature which exhibited very high CO2 uptake of 13.7 wt% at 0.15 bar higher than the 12.1 wt. % capacity reported by McDonald et al. for N,N'-dimethylethylenediamine (mmen) grafted on Mg(dopbdc) . The isosteric heat of adsorption was recorded to be 49 to 51 kJ/mol indicating chemisorption of CO2 molecules which was further confirmed by the formation of carbamic acid probed by the in situ Fourier transform infrared spectroscopy (FTIR) experiments. The en grafted Mg(dopbdc) was further evaluated for the multicycle adsorption, and it has only lost 3% of its CO2 uptake after five cycles. Moreover, en-Mg(dopbdc) has also shown stable structure and capacity after exposure to different moisture contents, and therefore this material has a potential for large-scale CO2 capture (see Figure 5). NH2, CH2NH2, CH2NHMe along with other functional groups were recently grafted on IRMOF-74, and it was found that IRMOF-74-III-CH2NH2 displayed CO2 capacity of 3.2 mmol/g at 1 bar . The sodalite-type structure Cu-BTTri was also grafted by en functional group  which showed chemisorption interaction with the adsorbed CO2 molecule as can be observed from the high isosteric heat of adsorption (90 kJ/mol). However, the en-Cu-BTTri has only shown improved capacity at low pressure while the unmodified MOF shows higher uptakes at high pressure which is attributed to the significant reduction in Brunauer–Emmett–Teller (BET) surface area from 1770 to 345 m2/g due to the pore blocking effect of the en group. In an attempt to address this issue, McDonald at al. functionalized mmen group on Cu-BTTri and preserved a BET surface area of 870 m2/g with 96 kJ/mol isosteric heat of adsorption, nitrogen selectivity of 327, and CO2 uptake of 9.5 wt.% under 0.15 bar CO2/0.75 bar N2 mixture at 25 °C. The negative impact of alkylamine functional groups on reducing the surface area was evident, and one approach to overcome this issue is to introduce ligand extension prior to the introduction of the amine group so as to increase the MOF porosity and avoid the pore-blocking problem during PSMs . Also, a deep insight into the mechanism of CO2 adsorption on alkylamine-grafted MOFs is crucial to further understand the interactions for improved structural design and amine loadings . Other amine functionalities such as piperazine were also grafted into Cu-BTTri  and exhibited 2.5 times higher CO2 uptake as compared to bare Cu-BTTri, while the heat of adsorption confirms the chemisorption interactions. The area reduction was also evident as it was reduced from 1700 m2/g to 380 m2/g (similar to ethylendiamin, (en)- Cu-BTTri ). Pyridine was also grafted on Ni-DOBDC to improve the water stability and increase the hydrophobicity of the material . Experimental observations supported by simulations results confirmed the enhanced water stability for the Pyridine-Ni-DOBDC samples while maintaining the CO2 uptake at atmospheric conditions and low pressures. It was also concluded that the amine moiety was grafted on the unsaturated metal sites of the framework, which makes this approach desirable for amine functionalization. From a combined experimental and simulation study, it was found that pyridine modification of an MOF can reduce H2O adsorption while retaining considerable CO2 capacity at conditions of interest for flue gas separation. This indicates that post-synthesis modification of MOFs by coordinating hydrophobic ligands to unsaturated metal sites may be a powerful method to generate new sorbents for gas separation under humid conditions. Amine functionalization to target the water stability of MOFs will be further discussed in the next section.
It is evident from the previous discussion that amine impregnation into MOFs always sacrifices the surface area of the final product. Therefore, the choice of the amine that can improve the affinity toward CO2 and attain high surface area simultaneously is a trade-off issue. MIL-101 materials were reported to have the highest pore volume and surface area among MOFs to date (BET = 3125 m2/g and 1.63 cm3/g). Hence they allow the incorporation of amines with longer alkyl chains such as polyethyleneimine while at the same time maintaining relatively high surface area (1112.6 m2/g after 75 wt% amine loading). PEI-loaded MIL-101 prepared by Lin et al.  exhibited remarkably high CO2 uptake of 4.2 mmol/g at 0.15 bar and 298 K with exceptional CO2/N2 selectivity of 770 at 25 °C.
Optimization of amine loadings and distribution within the MOF structure is a detrimental factor for the impact of these functionalities on the performance in CO2 capture process. Precise control of the different factors during the grafting process is crucial to append these groups exactly on the unsaturated metal centers, while avoid blocking the pores and hindering access to the interior volume. Improving the PSM methods is considered one of the means to achieve the ideal grafting and amine distribution .
4. Recent Advances and Current Trends
4.1. Hybrid Systems Based on MOFs
For more efficient utilization of MOFs sorbents, several hybrid systems based on MOFs with other solid sorbents have been investigated in the literature. The objective of having hybrid materials is to utilize the synergism between the two sorbents and therefore ultimately improve the overall performance in CO2 separation. Moreover, sorbents such as activated carbons, graphenes, and CNTs provide the added feature of high surface area and easily functionalized sites which contribute to the tuning of the final properties of the composite material. CNTs represent one of the effective candidates that can improve the properties of MOFs for gas adsorption applications. Zhu et al.  incorporated HKUST-1 in the interspace of CNTs. The designed composite exhibited superior selectivity and a CO2 saturation capacity of 7.83 mmol/g at 298 K, which was attributed to the high porosity and surface area. In a similar study, multiwall CNTs, well dispersed in MIL-101 (Cr), were successfully prepared and maintained the same framework and crystal structure as MIL-101. An increase of 60% in CO2 uptake was observed for the MWCNT-MIL-101 composite which was attributed to the increased porosity as a result of incorporating CNTs , as was confirmed by similar work on MWCNT-MIL-53(Cu) composite .
Graphene oxide composites with different MOFs are extensively reported in the literature such as HKUST-1 , MOF-5 , and Cu-BTC . Graphite oxide (GO) is considered a stabilizing agent for MOFs under humid environment, and it has shown remarkable CO2 capacity of 3.3 mmol/g and great stability under simulated flue gas conditions for GO/Cu-BTC composite . The synthesis of Cu-based MOFs composite with aminated graphite oxides (GO) was carried out and fully characterized by Zhao et al. . The composite exhibited 50% enhanced porosity as compared to the parent MOF and displayed unique structure and pore sizes effective for size exclusion separation of CO2 from the flue gas. Silica aerogel (SA) was also investigated as a promising candidate for hybrid systems with ZIF-8 . The detailed characterizations of the SA/ZIF-8 confirmed the presence of the two phases in the composite after sol–gel synthesis procedure with different ZIF-8 loadings and mild BET surface area .
Several composite materials have been reported for various applications; however, utilizing these hybrid systems in CO2 adsorption might be a promising route for improving the CO2 capture process. Ahmed et al.  published a review of information related to the synthesis and adsorption applications of MOF composite materials.
4.2. Ionic Liquids/MOF Composites
Ionic liquids as solvents for the absorption separation of CO2 from flue gas are discussed in Section 1.2 in order to overcome the limitations related to the poor dynamics of CO2 separation in ILs due to their high viscosity. MOFs can act as an ideal support material for the incorporation of ILs into their porous structure while preserving their unique properties. The concept of immobilization of ILs into solid sorbents has been reported for various applications. For instance, ILs immobilization on mesoporous silica was reported for the catalytic esterification reaction , ILs addition into polymer gels for ionic conductivity applications , ILs/Zeolite composites , in addition to several review papers available on this topic indicating the widespread use of this new approach over the past years [210, 211]. Computational investigation of the theoretical possibility of incorporating ILs into MOFs was studied by Jiang’s group for (BMIM)PF6 IL supported on IRMOF-1 for CO2 capture applications . The confinement effects of the narrow pore on the ILs and the ionic interactions between [BMIM]+ favor the open pore while the anion, [PF6]−, was attached to the open metal sites, was observed in a simulation study. It was ascertained that CO2 was favorably attached to the [PF6]− anions sites. The study demonstrated that IL/MOF composites are a potential candidate for CO2 adsorption and have displayed significantly high CO2/N2 selectivity. To the best of our knowledge, the first report on an experimental attempt to immobilize ILs into MOF structures was published by Liu et al.  for the insertion of Bronsted acidic ILs (BAIL) into the pores of MIL-101 using post-synthetic approach with triethylene diamine (TEDA) or imidazole (IMIZ) as a solvent assisting during the functionalization process. Nitrogen adsorption isotherms of bare MIL-101, TEDA-BAIL/MIL-101, and all (IMIZ-BAIL/MIL-101) samples showed type-I isotherm indicating the microporous nature of the composite. BET surface area was 1873 m2/g for the bare MIL-101 which was slightly decreased to 1728 m2/g and 1148 m2/g for IMIZ-BAIL/MIL-101 and TEDA-BAIL/MIL-101, respectively. Following this leading report, Jhung’s group  successfully grafted up to 50 wt.% acidic chloroaluminate IL on MIL-101 which reduced the BET surface area of the bare MIL-101 by around 60%. The incorporation of ILs with basic nature which is favorable for CO2 adsorption was for the first time reported by Kitagawa et al. . A detailed characterization and investigation of the phase behavior of the immobilized 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonylamide) denoted as EMI-TFSA ILs into ZIF-8 was presented in this study. A reduction of 29% in pore volume was measured in N2 adsorption isotherm experiments and computational calculations. The EMI-TFSA/ZIF-8 composite has shown distinctive ion conductivity at low temperature as reported in a second paper by the same group . The prospect of IL/MOF composite for gas separation is still under computational investigation with no reported experimental studies of CO2 adsorption on these composites. Recently, Vincent-Luna et al.  investigated the effects of adding room temperature ILs (RTILs) into the pores of Cu-BTC structures. The adsorption of CO2, N2, CH4, and their mixtures were studied by utilizing various RTILs having the same cation 1-ethyl-3-methylimidazolium [EMIM]+ and different anions such as bis[(trifluoromethyl)-sulfonyl]imide [Tf2N]−, thiocyanate [SCN]−, nitrate [NO3]−, tetrafluoroborate [BF4]−, and hexafluorophosphate [PF6]−. The RTIL/Cu-BTC composite has shown enhanced CO2 uptakes at low pressures with high CO2/N2 selectivity due to the polarization driving force rendering these materials as a promising system for post-combustion CO2 capture. Another application of IL/MOF composite as a precursor for the preparation of nitrogen and boron–nitrogen (N- and BN-)-decorated porous carbons was recently reported by Aijaz et al.  as a novel synthesis strategy.
4.3. Ab/Adsorption in Ionic Liquids/MOF Slurry System
Another approach to utilize the combined synergistic advantages of MOF and IL composites is through a novel hybrid adsorption–absorption technology. This novel technology can provide an efficient approach to utilize the high capacity, selectivity, and low heat of adsorption of the solid sorbents along with the advantages of having a continuous flow process that allows for better heat integration and separation rates in contrast to the conventional batch process used in adsorption-only process. Mass transfer enhancement due to the dispersion of fine solids in liquid solvents was studied and insight into the mechanism and the analysis of different mass transfer resistances were described by Zhang and coworkers  which was in agreement with previous findings [219–221]. As far as enhancement of CO2 capture in slurry systems is concerned, a study dealing with AC particles dispersed in K2CO3 aqueous solution was reported by Sumin et al.  to investigate the influence of the hydrodynamics on the mass transfer improvements. In a similar work by Rosu et al. , AC particles were also found to improve the absorptive CO2 capture process. The unique characteristics of MOFs in CO2 adsorption and their recent applications in aqueous solution environment  have opened the door toward the possibility of immersing MOF particles in various physical and chemical solvents for CO2 separation application. This novel unit operation process can overcome the limitations reported for conventional adsorption on MOFs such as high pressure drop and the necessity for formulating the powders into different shapes and sizes which affects their structural stability and reduces the active surface area. Liu et al.  reported, for the first time, the preparation of ZIF-8/glycol and ZIF-8/glycol/2-methylimidazole slurries (Figure 6). CO2 uptake of 1.25 mmol/L was recorded for the slurry system with CO2/N2 selectivity of 394 at 1 bar, and most importantly a very low enthalpy of 29 kJ/mol. In a similar work by Lei et al. , ILs [EMIM, TF2N] and [OMIM, PF6] were used to prepare slurry systems with ZIF-8 and ZIF-7. CO2 adsorption in the slurry system has shown a promising performance with isosteric heat of adsorption less than 26 kJ/mol. Following these two studies, the solubility of CO2 in physical solvents such as methanol mixed with ZIF-8 was also investigated . The study revealed that ZIF-8 can significantly improve the low pressure-CO2 uptake in physical solvents and can dramatically reduce the solvent losses by evaporation to the gas phase at the top of the absorber. Increasing ZIF-8 loadings has shown further enhancement of the CO2 capacity, as observed previously ; however, it is worth noting that a high solid loading in the slurry system was not recommended from process engineering point of view as it might cause some problems during the pumping of the slurry mixture and increases the solid losses in the multicycle separation process .
For future studies on MOF-based slurry systems, there is basic selection of criteria that needs to be satisfied by both MOF and the liquid solution. The selection of the MOF possessing the appropriate pore size for the preparation of the slurry system is very important to guarantee that the size of the liquid is large enough and does not occupy the pores which leaves no space for CO2 to adsorb. Moreover, the structural stability of the MOF in the aqueous solution is essential so that it does not lose its porous framework nor its surface area. The selection of the liquid candidate is crucial, as it should not provide any extra mass transfer resistance for CO2 molecules. Further, experimental and computational investigations are still required to understand the separation mechanism in slurry mixtures and to have insight into the different types of interactions between the gas, liquid, and solid materials.
5. Challenges and Outlook
In conclusion, MOFs are considered the largest growing research area in CO2 capture, with great achievements and developments. Due to their versatile structures and possibilities for various functionalization approaches, the door is still open for further improvements and advancements of their performance under real flue gas conditions, and in large-scale applications. Although we have reported MOFs with distinguished properties and exceptional CO2 capacity, selectivity, and stability, there are still some concerns that need to be addressed before reaching commercial scale level. The lack of information about the performance of MOFs under real gas mixture conditions is one of the key issues to understand the actual working uptakes and identify any possible limitations. Further experimental testing of MOFs using, for example, a gas mixture containing all the impurities that might be present in an actual flue gas is needed high-throughput technique. Computational gas mixture studies can provide essential information in this regard; however, experimental investigation is still considered the most reliable approach. MOF stabilities in humid conditions, high temperature, and harsh mechanical stress situations must be given much attention. Several studies were performed to target MOF stability, and great achievements were recorded in this field [228, 229], as reviewed in references [160, 230]. Finally, in the following section, we focus on water stability studies as it is one of the main drawbacks of MOFs.
5.1. Water Stability of MOFs
Water stability is a major challenge that has to be overcome before metal organic framework can be used in removing carbon dioxide from flue gas. The core structure of MOF reacts with water vapor content in the flue gas leading to severe distortion of the structure and even failure. As a consequence, the physical structure of MOF is changed, e.g., reduction of porosity and surface area, etc. that decreases the capacity and selectivity for CO2. Complete dehydration of flue gas increases the cost of separation. It is therefore essential for MOFs to exhibit stability in the presence of water up to certain extent .
Metal–ligand coordination bond, which is the most significant part of MOF, is hydrolyzed with water, resulting in the displacement of ligand bond; and as a consequence, the whole structure usually collapses . The stability of MOF in the presence of water depends on the strength of metal ligand bond. pKa values of the ligand atom can be considered as the strength of this metal- ligand bond. Since the hydrolysis reaction between MOF and water molecule is governed by Gibbs free energy and activation energy of the reactant and product molecules, thermodynamics and kinetics factors have great influence on the water stability of MOF . Insight into the molecular structure, more specifically the metal–ligand strength, the weakest part of MOF, and thermodynamics as well as kinetics study of hydrolysis reaction are very important to improve water stability. Several strategies based on these two important aspects have been taken into consideration.
Jasuja et al.  performed a study on the effects of functionalization of the organic ligand in a series of isostructural MOFs in the Zn(BDC-X)-(DABCO)0.5 family on water stability. In this experiment, they cyclically stabilized an unstable parent structure in humid conditions through the incorporation of tetramethyl-BDC ligand. The results of molecular simulation disclosed that the kinetic stability is improved due to the carboxylate oxygen in the DMOF-TM2 structure which acted as a shield to prevent hydrogen-bonding interactions and subsequent structural transformations. Hence, electrophilic zinc atoms in this structure became inaccessible to the nucleophilic oxygen atoms in water, resulting in prevention of the hydrolysis reactions for the displacement ligand. They also performed another study to evaluate the effect of strength of metal–ligand coordination bond and catenation in the framework on water stability . According to their results, the non-interpenetrated MOFs constructed from a pillar ligand of higher pKa exhibited higher stability; however, interpenetrated MOFs constructed from a pillar ligand of lower pKa values exhibited less stability. The interpenetration in MOF with incorporation of ligands of relatively high basicity exhibited good water stability. By considering the results of previous experiment, they synthesized cobalt-, nickel-, copper-, and zinc-based, new pillared MOFs of similar topologies which exhibited good water stability . The grafted methyl group on the benzene dicarboxlate (BDC) ligand introduced steric factors around the metal centers; consequently, water stability of MOF drastically improved. The basicity of BTTB-based MOFs synthesized with bipyridyl pillar ligands had lower basicity than DABCO; however, they exhibited better stability in the presence of humid condition.
Bae et al.  performed a study to modify Ni-DOBDC with pyridine molecules. The study showed that pyridine molecule made the normally hydrophilic internal surface more hydrophobic; as a result, water absorption was reduced, while substantial CO2 capture capacity was retained to a certain level. Fracaroli et al.  improved the interior of IRMOF-74-III by covalently functionalizing it with a primary amine, and used a MOF, IRMOF-74-IIICH2NH2, for the selective capture of CO2 in 65% relative humidity.
Zhang et al.  performed a study to modify the surface of the MOF hydrophilic to hydrophobic to improve water stability. They demonstrated a new strategy to modify hydrophobic polydimethysiloxane (PDMS) on the surface to significantly enhance their water resistance by a facile vapor deposition technique. In this study, they successfully coated three vulnerable MOFs according to the water stability (MOF-5, HKUST-1, and ZnBT), while the porosity, crystalline characteristics, and surface area were unchanged.
All these studies demonstrated that water stability of MOFs can be improved by incorporating specific factors (e.g., metal–ligand strength, thermodynamic and kinetic factors, etc.) which govern the structural stability of the framework.