Study on Selected Metal-Organic Framework-Based Catalysts for Cycloaddition Reaction of CO2 with Epoxides: A Highly Economic Solution for Carbon Capture and Utilization

The level of carbon dioxide in the atmosphere is growing rapidly due to fossil fuel combustion processes, heavy oil, coal, oil shelter, and exhausts from automobiles for energy generation, which lead to depletion of the ozone layer and consequently result in global warming. The realization of a carbon-neutral environment is the main focus of science and academic researchers of today. Several processes were employed to minimize carbon dioxide in the air, some of which include the utilization of non-fossil sources of energy like solar, nuclear, and biomass-based fuels. Consequently, these sources were reported to have a relatively high cost of production and maintenance. The applications of both homogeneous and heterogeneous processes in carbon capture and storage were investigated in recent years and the focus now is on the conversion of CO2 into useful chemicals and compounds. It was established that CO2 can undergo cycloaddition reaction with epoxides under the influence of special catalysts to give cyclic carbonates, which can be used as value-added chemicals at a different level of pharmaceutical and industrial applications. Among the various catalysts studied for this reaction, metal-organic frameworks are now on the frontline as a potential catalyst due to their special features and easy synthesis. Several metal-organic framework (MOF)-based catalysts were studied for their application in transforming CO2 to organic carbonates using epoxides. Here, we report some recent studies of porous MOF materials and an in-depth discussion of two repeatedly used metal-organic frameworks as a catalyst in the conversion of CO2 to organic carbonates.


Introduction
The carbon dioxide (CO 2 ) concentration in the atmosphere is becoming alarming due to an increase in anthropogenic activities such as fuel combustion and other energy generation processes (see Figure 1), which may result in global warming [1][2][3][4][5][6]. Carbon capture, storage, and utilization are the most promising alternative among other several processes in reducing CO 2 emission [7][8][9]. The process of carbon capture and sequestration (CCS) was employed initially by using classical solid sorbents such as zeolites, porous carbon, porous silica, and mineral carbonation processes [10][11][12]. In recent years, the attention of the researchers has dwelled on utilizing CO 2 as C 1 feedstock for chemical conversion into various useful products [13][14][15] in the presence of highly selective and task-specific catalysts [8,16,17]. This route has a more beneficial advantage as it allows the utilization of unwanted substances into useful materials [18,19].  [20]. Copyright 2020 Springer Nature.
One of the possible optimistic approaches to climate-relevant carbon capture and sequestration is the catalyst-mediated reaction of CO 2 . Industrial and atmospheric CO 2 are captured and passed through biochemical, photochemical, thermochemical, or electrochemical conversion process to obtained intermediates such as methanol, dimethyl ether (DME), syngas, polyols, and cyclic organic carbonates, which can be further converted into end products as liquid fuels, aromatics among others, or used in the production of cosmetic and personal care products. The search for specific and specially designed catalysts for CO 2 conversion into commercially relevant compounds is indeed a subject of high curiosity these days [16,21,22]. This review focused on the CO 2 conversion by cycloaddition reactions with epoxides and the application of different polymeric metal-organic frameworks (MOFs) materials as potential catalysts that give an intermediate that can be converted into different organic chemicals such as diethyl carbonate (DEC), ethyl methyl carbonate (EMC), and dimethyl carbonate (DMC), among others. Over the years, there has been increasing demand for cyclic carbonate intermediates in various end-use industries such as electrolyte solvents in lithium-ion battery, medical, industrial, and personal care. For instance, according to the Research and Markets data statistic forecast report, the global ethylene carbonate market size alone was estimated to be valued at approximately USD 273 million in 2018 and projected to reach over USD 500 million by 2027, at a compound annual growth rate (CAGR) of 8.1% Figure 2A. The market size for dimethyl carbonate (DMC) was valued over USD 410 million in 2015 and was forecast to have gains exceeding 5% CAGR. In the United States of America (USA) alone, it was estimated to grow over USD 100 million by 2024 in its application as solvents, polycarbonates, and in pharmaceutical, and pesticide industries among others as forecast by Global Market Insights, as shown in Figure 2B. The global polycarbonate business was valued at over USD 14 billion in 2015 and was projected to reach USD 22 billion by 2024. Alkylene carbonates in general, which include, ethylene carbonate, propylene carbonate, glycerin carbonate, etc., are cyclic esters of carbonic acid that possess various properties such as low volatile organic content (VOC), low toxicity, being readily biodegradable, high boiling point, low-odor, and all these properties make them an excellent solvent choice. The global alkylene carbonates market share by end use industry (paints and coatings, agriculture, textiles and fabrics, cosmetics and personal care, and others) is shown in Figure 2C. The revolutionary work in the investigation of metal-organic frameworks (MOFs) as a catalyst for cycloaddition reaction was first discovered in 2009, where the renowned MOF-5 was discovered to have successfully catalyzed CO 2 conversion reaction with epoxides in the presence of quaternary ammonium salts such as tetrabutylammonium bromide (TBABr), see ( Figure 3). In the last decade, different catalysts were designed to facilitate CO 2 conversion reaction into cyclic organic carbonates [24]. The features and composition of post-combustion CO 2 indicated that abilities such as high CO 2 uptake and selectivity, excellent thermal and chemical stability, good reusability, easy synthesis, and operation at ambient conditions should be given by an ideal adsorbent for better CO 2 capture [22].

Reaction Mechanism for the Production of Cyclic Carbonates from CO 2 and Epoxides
CO 2 conversion into cyclic carbonate compounds by cycloaddition reaction to epoxide is regarded as a method with economic advantages to attain a CO 2 -neutral environment and to serve as a source for value-added chemicals. Various researchers reviewed the mechanism for cycloadding CO 2 into epoxides [8,14,17,65,[89][90][91][92].
The process requires a robust acid catalyst to stimulate the epoxide substrate and the highly stable CO 2 double bond and thermodynamically facilitates the opening of the epoxide ring via nucleophilic co-catalyst (TBABr) attack forming an alkoxide as an intermediate, which subsequently combines with the CO 2 -adduct to give the desired carbonates ( Figure 4). The TBABr co-catalyst functions as a nucleophile to motivate the opening of the epoxide rings. The synergistic effect between the MOF catalyst and TBABr is therefore crucial in attaining high catalytic performance [93,94]. The cycloaddition reaction of CO 2 with epoxides was extensively investigated using different potential catalysts [16,[22][23][24][95][96][97][98][99][100][101][102]. The best route for the mechanism was identified as the one that begins with the epoxides ring opening before the addition of carbon dioxide. This therefore proved that the catalytic system in this process strongly depends on the opening of the epoxide ring [19]. The different method for epoxide activation by the MOF catalyst was categorized into four bases on the features of the MOF catalyst ( Figure 5) as follows: (a) MOFs with acidic secondary building units (SBUs) as the only active site, (b) MOFs with acidic linkers as metal active site catalyst, (c) MOFs with Lewis base linkers also acting as a nucleophile and Lewis acidic components, a binary catalytic system (d) MOFs with ionic linkers, where a single-component catalyst is used without the TBABr co-catalyst. The homogenous co-catalyst, tetrabutylammonium bromide (TBABr) would alone promote the epoxide ring-opening in (a and b) and the effort is reinforced in (c). The influence of Lewis-acidic component in SBUs or metal nodes of the MOFs, however, cannot be exempted as indicated (b-d) [102].
Recent studies of some metal-organic framework materials as a catalyst for cycloaddition reaction (Table 1) reaffirmed that the cycloaddition reaction in most cases cannot proceed successfully without the presence of a co-catalyst. Tetra-n-butyl ammonium bromide (TBABr) was reported as the most effective co-catalyst among the various nucleophile components, TBAI, TBACl, and KI, and was identified to enhance epoxide ring opening in the reaction [120,[131][132][133]. Some certain MOFs, however, were applied without the presence of a co-catalyst (Table 1, entry [17][18][19][20]. Where this occured, the catalytic activity of MOFs were considered as a single component and were applied without the addition of TBABr. Nonetheless, based on the work studied, this type of reaction can only be successful under harsh reaction conditions of temperature and pressure (Table 1, entry 17-20) [55].  The addition of TBABr co-catalyst in most of the reported MOFs (Table 1, entry [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] further proved that the catalytic function of MOF catalysts in cycloaddition reaction works concurrently with the co-catalyst for successful conversion. Moreover, the catalyst/cocatalyst loading in the data entries (Table 1) was observed to be in relatively low percentage moles. The MOF catalytic materials were found to be active and were reused for at least three consecutive cycles in each case before losing their activity. All the reported MOFs (Table 1) were found to be effective at moderately ambient conditions, except for entry [17][18][19][20], which occured at relatively harsh conditions due to the absence of co-catalyst. Different epoxides such as propylene oxide, styrene oxide, and epichlorohydrin were found to undertake CO 2 cycloaddition under the influence of the MOF catalyst, as shown in Table 1.

MIL-101 Based MOFs in CO 2 Cycloaddition with Epoxides
MIL-101 is one of the repeatedly reported MOF materials with a high potential catalytic activity for the conversion of CO 2 to cyclic carbonates. This was ascribed to its possession of Lewis-acid sites due to Cr 2+ , present at the metal center [56] and structural flexibility, which allows its modifications by substituting different functional groups in the organic ligand but maintains the backbone structures. The synthesis and structural elucidation of MIL-101(Cr) was reported by different researchers [85,86,108,[141][142][143][144][145][146][147][148][149].
MIL-101 is a three-dimensional structure based on chromium terephthalate that was first synthesized by Fėrey et al. [148], having the empirical formula  [110,111,[152][153][154][155][156]. The unique porosity of three-dimensional frameworks forms exclusive channels with large surface areas, which can enhance CO 2 by providing sufficient reaction spaces. It also allows the encapsulation of other catalytic active materials into the large pores to improve the catalytic activity of the MIL-101 by forming a composites with enhanced activity for application in various fields [143,155]. The MOF was also applied as a catalyst in cycloaddition reaction of CO 2 with epoxides as a single component catalyst without the addition of co-catalyst [55].  We report here a series of recently reported MIL-101-based materials used as a catalyst in cycloaddition reaction of CO 2 and epoxides.
Recently, research was reported by Akimana et al. [156], where MIL-101(Cr) was used to catalyzed cycloaddition reaction of epichlorohydrin to CO 2 under mild reaction conditions without any co-catalyst. MIL-101(Cr) was identified as one of the catalytically active MOFs with excellent properties, chemical stability and flexibility, and strong acid sites due to the presence of metal centers in the framework [85,108,148]. It was gathered that special features of MIL-101(Cr) can facilitates the opening of the epoxide ring substrates in combination with the characteristic surface area and large pore size exhibited by the MOF, and that these are good properties for CO 2 adsorption as well are essential features for a superior catalyst that can be applied in the formation of cyclic carbonates from CO 2 .
The MOF was synthesized and applied to catalyze CO 2 cycloaddition with different epoxides. The acidic or basic characteristic was provided by the metal nodes and the high porosity of the MOF attracts the CO 2 gas molecules, which gave the MOF good ability to perform CO 2 cycloaddition and produce cyclic carbonate compounds. The catalytic activity was observed by varying the conditions for the reaction using epichlorohydrin (ECH) as a model substrate. To obtain the optimum reaction condition, the various reaction conditions were alternated. Firstly by using a temperature of 35 • C and 4 bar CO 2 pressure with no solvent or co-catalyst added (catalyst, 50 mg, ECH, 9.2 mmol). The outcome showed a 100% ECH conversion in 48 h. A CO 2 pressure above 1.5 bar while maintaining the condition also exhibited an excellent performance of about 99% conversion in 24 h. The gap between conversion and yield was attributed to the MOF's adsorption for the product and therefore was not detected in the H-NMR, resulting in low percentage yield. The conversion also decreased with a shorter reaction period or by reducing the catalyst loading amount. However, an average catalytic activity was observed with low catalytic loading under mild reaction conditions (35 • C, 1.5 bar) without the addition of co-catalyst or solvent [156]. The MIL-101(Cr) MOF catalytic activity for CO 2 cycloaddition was also investigated with different epoxides and remarkable conversion was obtained Figure 10. The catalyst shows amazing performance in catalytic activity when compared with other reported catalysts and has good reusability property. The application of MIL-101(Cr) as a heterogeneous catalyst in the conversion of CO 2 into organic carbonates revealed high catalytic potentials that are greener and more feasible.
Dai et al. [157] recently synthesized a new MOF catalyst by implanting quaternary phosphonium salt ionic liquid (IL) on MIL-101(Cr) through a flippant post-synthetic modification to obtain (Cr-MIL-101-[BuPh 3 P]Br). The backbone structure of the parent MIL-101(Cr) was not altered when compared with the Cr-MIL-101-NH 2 , except for the broad Bragg reflections detected. The Cr-MIL-101-NH 2 was then modified further by the addition of phosphonium salt IL through a covalent interaction between the -NH 2 and R-Br group and the crystalline structure of the new hybrid was not affected by the postsynthetic modification (Figure 11). The presence of the phosphonium salt IL in the new compound was confirmed by XRD and FT-IR spectra. The new compound also showed more superior performance in its stability and catalytic activity when compared with the amine-functionalized MIL-101-IL reported earlier [158]. Due to their polarity, structural tenability, thermal stability, and excellent acid-basic property, ILs are considered a strong catalyst for CO 2 cycloaddition reaction with epoxides. The fabrication of heterogeneous material containing active metal sites together with phosphonium salt IL can give rise to a good catalyst with excellent activity in cycloaddition reaction with epoxides [157]. The catalytic function was utilized by the combined ability of two active sites involving Cr 3+ as Lewis acid sites from the MOF and Br-as nucleophile from the IL, which facilitated the epoxide ring-opening by coordination of Cr 3+ sites with an oxygen atom and nucleophilic attack of Br-on the carbon atom of epoxide, respectively. The catalyst performed an outstanding activity for the CO 2 cycloaddition to propylene oxide under a moderate reaction condition of 120 • C and 2.0 Mpa for 2 h using a small amount of catalyst (0.045 mol%) without the addition of any solvent or co-catalyst ( Figure 12  (c) catalyst 0.045 mol%, temperature 120 • C, time 2 h; (d) catalyst 0.045 mol%, temperature 120 • C, CO2 pressure 2.0 MPa. Reprinted with permission from Ref. [157]. Copyright 2020 Elsevier.
An investigation was carried out to optimize the reaction parameters (catalyst loading, temperature, CO 2 pressure, and time) and the outcome is illustrated in Figure 12. A sharp increase in the yield was observed by increasing the concentration of the catalyst, as shown in Figure 12a. The production of propylene carbonate (PC) indicated the need for sufficient energy in the reaction, as shown in Figure 12b, indicating the need for high temperature. The increase in pressure also indicated a significant effect on the reaction up to 2.0 MPa before it begins to have no positive effect on the reaction, as in Figure 12c. Finally, the reaction time showed a significant increase in the formation of PC from 0.5 h to almost 2.5 h (Figure 12d), and after that, extending the time did not affect PC production. Generally, the selectivity of PC was independent in all the parameters involved for the reaction and therefore did not have an effect. The recyclability test for both Cr-MIL-101-[BuPh 3 P]Br and Cr-MIL-101-NH 2 was carried out in the research using similar reaction conditions and Cr-MIL-101-[BuPh 3 P]Br was reused more than four times without any noticeable reduction in PC production. The catalyst was also tested for the conversion of other epoxides and a significant yield was recorded [157].
Bao et al. [159] reported a green synthesis and cost-effective method for yolk-shell structures of water etching of MOF. The research was carried to fabricate a MOF@mSiO 2 as yolk-shell nanoreactors by silica coating method, which was followed by a wateretching approach. The material was fabricated to curtail the stability challenges of MOF catalyst under hydrothermal conditions as a result of weak metal-ligand bonds in their structures. In this regard, a high-density Lewis acid MIL-101 having unique structures and many catalytic advantages, especially for the acid catalytic reaction, was selected to be encapsulated into SiO 2 and make a yolk-shell arrangement by engraving the MIL-101(Cr) surfaces with water, as shown in Figure 13. The structure of the nanoreactors MIL-101@mSiO 2 -YS was studied and characterized by various spectroscopic techniques, which proved the successful formation of MIL-101(Cr)@mSiO 2 -YS. The study of the new structure revealed that the MIL-101(Cr) nanostructure will have the ability to maintain its Lewis acid sites and the nanoshell structure will deliver effective paths for bulk movement of reactants and products, while the yolk-shell safeguards the MIL-101 and reduces its disintegration during catalytic reaction processes. Thus, the combined special effects of the MIL-101(Cr) framework and that of SiO 2 yolk shells are perfectly brought together to accomplish superior properties and stability that can be used in CO 2 cycloaddition reaction. A cycloaddition of CO 2 with styrene oxide (SO) was tested using the prepared composites and the outcome showed improved catalytic activity by having more cycles compared to the pristine MIL-101, which began to decrease in its catalytic activity after three cycles by the decrease in the yield of styrene carbonate from 98 to 66% due to collapse in its structure, as in Figure 14. The CO 2 reaction was carried in the presence of TBABr co-catalyst using SO in an autoclave charged with 0. 22   Hu et al. [67] reported a DFT study on the cycloaddition of CO 2 to propylene oxide (PO) using binary system MIL-101/TBABr catalyst. A comparative study of noncatalyzed reaction Figure 15, MIL-101-catalyzed reaction as a single component catalyst ( Figure 16A) and TBAB-catalyzed reactions ( Figure 16B) were analyzed using Gibbs free energy surface profile and the transition states in the conversion processes were illustrated. It was discovered that the reaction in the binary system MIL-101/TBABr proceeded in a much simpler manner with a lower energy barrier (18.11 kcal mol −1 ) and 90% conversion. On the other hand, the MIL-101-catalysed alone reaction required a minimum energy of (46.89 kcal mol −1 ) and the TBABr-catalyzed alone required an energy of (26.86 kcal mol −1 ) with the conversion of 27 and 42%, respectively, while the non-catalyzed reaction, which proceeded under ambient conditions, required a minimum energy of (57.67 kcal mol −1 ) [67].  The mechanism study for the reaction showed a three-step path involving (i) the epoxide ring-opening, (ii) the carbon dioxide addition, and (iii) the ring-closing of the cyclic carbonate was preferentially established and was more kinetically positive compared to a two-step path, involving the opening of the epoxide ring and closure of the cyclic carbonate ring only. The Gibbs free energy surface profile for the two different routes is illustrated in Figure 17. The outcomes of the computational studies affirmed that in the presence of a binary catalyst (MIL-101/TBABr) the cycloadding of CO 2 to PO occurs more easily compared to both non-catalyzed and TBABr-Catalyzed and MIL-101(Cr)-catalyzed pathways [67]. Liu et al. [160] carried an investigation for the coupling of propylene oxide (PO) with CO 2 using MIL-101-based MOF composites prepared by post-synthesis and modification of MIL-101(Cr) as Lewis acid site with imidazolium-based ionic liquids to obtained MIL-101-IMBr composite, which fully characterized using different spectroscopic techniques, as in Figure 18. The ionic liquid was functionalized with basic sites that serve as a nucleophile in the reaction by providing halides ions to motivate the opening of the epoxide ring in the cycloaddition reaction. These special features made the composite possess a good catalytic property for cycloaddition reaction at moderately mild condition (80 • C and 0.8 MPa) without the addition of co-catalyst. The PO conversion reached more than 95% with 97% selectivity of propylene carbonate (PC). The catalyst also showed excellent reusability ( Figure 19) and had no visible defect in the catalytic activity throughout the circles [160].  Wang, et al. [158] reported similar research by immobilizing ionic liquid (IL) on an amine-functionalized metal-organic framework in their effort to synthesize bifunctional catalyst with acid-base property for application in CO 2 cycloaddition reaction. The immobilization between the IL and MIL-101-NH 2 was constructed via a post-synthetic procedure to incorporate both amino and carboxylic groups into the material, as shown in Figure 20. The study was supported by density functional theory (DFT) calculations and the hybrid MOF was found to possess acid-base bifunctional futures, and was used as a single catalyst without co-catalyst addition. The synthesized hybrid MOF (IL/MIL-101-NH 2 ) was applied as a catalyst in the formation of propylene carbonate from CO 2 . The reaction was carried out in a low pressure of CO 2 of 1.3 MPa and 1 h duration at a temperature of 120 • C, resulting in 91% yield. The strategic design and synthesis of the catalyst with characteristic acid-base properties opened a new window in the development of new material that can catalyze cycloaddition reaction of epoxides to CO 2 . The composite was found to be very stable by recycling up to five times without a change in its structure [158].

HKUST-1 for CO 2 Cycloaddition with Epoxides
Another group of porous material that attracted the attention of researchers as a potential catalyst for CO 2 cycloaddition with epoxides is HKUST-1 (Hong Kong University of Science and Technology) and also referred as MOF-199, a Cu-based of 1,3,5benzenetricarboxylate (BTC) ligand. HKUST-1 was first synthesized in 1999 using copper(II) ion (Cu 2+ ) as metal center and BTC ligand to give a light blue powder ( Figure 21) [87,88,161], it was also reported using other metals with +2 oxidation state such as Cr 2+ [162], Ni 2+ [163], Zn 2+ [164], Fe 2+ , Co 2+ [165,166], and as bimetallic with two different metals at the center while maintaining the structure [167]. HKUST-1 material has a molecular formula [Cu 3 (BTC) 2 (H 2 O) 3 ] and morphological structure with a sound level of thermal stability ( Figure 22) and a capacity for chemical functionalization of the channel linings with facecentered-cubic crystals comprising a 3D intersecting system with big square-shaped pores (9 Å by 9 Å) of a continuous network through Cu(OAc) 4 paddle wheel SBU, as in Figure 23. The open metal sites in the HKUST-1 structure provides the basis for CO 2 adsorption property [7,168]. The large porous structure can also be used to create effective heterogeneous catalysts as templates to construct dynamic heterogeneous system by encapsulation of other nanomaterials to form composites. The resultant composite material can preserve the original properties of the HKUST-1, while gaining additional unique properties that can be applied in different applications [104,[169][170][171][172].   In recent years, progress was made by various researchers on the modification and application of HKUST-1-based MOFs as a potential catalyst for CO 2 cycloaddition reaction with epoxides. Some of these findings are summarized here.
A series of benzene tricarboxylic acid (btc)-based metal-organic frameworks were synthesized by Rani, et al. [58]. The MOFs were solvothermally synthesized under the same experimental conditions and fully characterized by different analytical methods. The synthesized MOFs included; Zn, Co, Ni, and Cu btc-based MOFs, which were described to exhibit diverse structural and catalytic behaviors [58]. A study on the application of MOFs for CO 2 cycloaddition with PO under mild conditions was carried out successfully. For the first time, Zn, Co, and Ni btc-based frameworks were exploited as a catalyst for CO 2 cycloaddition reaction with epoxides and displayed remarkable yield under certain reaction conditions. An attempt was made previously with Cu-based (HKUST-1) MOF on this catalytic application, but interestingly, in this study, the Cu-btc was also reported to exhibit higher yield than in the previous reports [54,169]. The research further revealed that among the four synthesized MOF catalysts, Zn-btc was identified to perform higher catalytic performance in the CO 2 fixation reaction as compared with other MOFs having a large turnover number (TON) of 3785 mol −1 and a turnover frequency (TOF) of 946 h −1 with more than 99% yield (Figure 24) [58]. and two molecules of lattice vapor. In the asymmetric unit, there were two types of Zn atoms; each was present in a distorted octahedral geometry and linked in the planar position to three separate btc ligands and two coordinated water molecules in the equatorial position. Two among the three btc ligands acted as bidentate ligands. The Zn 2 -coordinated bidentate ligand was tentatively coordinated with additional Zn 1 and Zn 2 . The overall coordination of Zn and btc ligands contributed to the creation of 2D sheet structure, as shown in Figure 25 [58].
The Co-btc and Ni-btc crystals were obtained using the same experimental conditions by reacting cobalt nitrate and nickel nitrate with btc ligand, respectively. SXRD was used to interpret the structures of the two MOFs, which contain two distinct metal centers that are isostructural, both in a distorted octahedral geometry. The structure consisted of coordinated metal centers with two molecules in the central position and two water molecules and btc acting as a bidentate ligand in a planer position. Another metal center was coordinated to two monodentate btc ligands in the central position and four water molecules in a planar position. The btc was bidentatively connected on both sides to one metal center and monodentatively to two separate metal centers and vice-versa, forming a zig-zag ID chain through H-bonding, which resulted in forming a 2D-sheet structure with large pores (Figure 26   A single crystal structure study for Cu-btc showed the Cu metal in a square pyramidal geometry coordinated with four planar btc carboxylic oxygen molecules and water molecules in an equatorial position. Via four btc ligands, two Cu centers were connected forming a structure resembling paddle wheels, which were interconnected to give a 3D system structure with large pores of~32 Å as earlier described, see Figure 23. The recyclability test for all the synthesized MOF catalysts was investigated using the optimum conditions at 120 • C and CO 2 pressure of 0.7 MPa for the length of 4 h to explore the catalyst recycling ability of the MOFs. The outcome reveals that Zn-btc had the highest recycling ability, followed by Cu-btc. [58]  Another related research was carried out by Ding et al. [133] who conducted a computational study on a series of HKUST-1(Cu-btc) structures with different metal centers to design a special catalyst for CO 2 fixation with epoxides. The catalytic activity of M-HKUST-1(M = Sc, Ti, V, Cr, Mo, Cd, Ru, W, Cu, Fe, Zn) under the influence of TBA salts in a series of four different halogen anions denoted as TBAX (X = F, Cl, Br and I) as co-catalysts in the formation of propylene carbonate from CO 2 and PO was investigated. The recent development in computing technology has driven great advancement in designing materials with high accuracy and efficiency and this technology was employed in the development of new catalytic materials for the fixation of CO 2 , which mainly depends on trials of a different experiment.
The models for Cr-HKUST-1, Cu-HKUST-1, and Fe-HKUST-1 obtained from the Crystallographic Data Centre (CCDC) of Cambridge University and others through isostructural substitution of the Cu atom having identical crystal environment from the known HKUST-1 MOF. A cluster model with a narrow dimension was utilized for the single-site catalytic reaction, which usually happens inside the HKUST-1 cavities. The study of the mechanism was carried out by using condensed cluster models and the effect of the metal substitution was further studied. The simple cluster was developed to imitate the HKUST-1 framework ( Figure 28). The most generally accepted mechanism for a MOF-based catalyst in CO 2 cycloaddition with epoxides can be summarized into four steps, starting with the adsorption of epoxide, epoxide ring opening, CO 2 insertion, and finally cyclic carbonate formation by closing the ring. The predicted mechanism for HKUST-1 is illustrated in Figure 29. The study also revealed a similar mechanism to that of regular Cu-HKUST-1 with the ring-opening playing the most critical part, not the insertion of CO 2 . This was calculated to require moderately lower barrier as defected in the Gibbs free energy barrier ( Figure 30) [133].  The M-HKUST-1 computational screening can be depicted by studying the ratedetermining step in the reaction. This can be obtained by assessing the activity of the whole M-HKUST-1 systems, thereby discovering the most effective catalyst. The energy barriers for the key step were prescribed as presented in Figure 30. For the M-HKUST-1/Br − catalyst, the rate-determining energy barrier relied primarily on the metal centers. According to the studies, W-HKUST-1 would accomplish a high potential to attain superior activity among the M-HKUST-1 system. Furthermore, TBABr was forecast as the most effective co-catalyst among the TBAX studied for the better catalytic function of M-HKUST-1 [133]. Wu 3.0 MPa CO 2 pressure for 9 h. Mn-BTC catalyst also showed remarkable stability and could be reused with a small reduction in its catalytic ability over three times. Moreover, CO 2 was added to other epoxides and DFT calculations were performed, and the yields revealed in the order ECH > propene oxide > 1,2-epoxybutane > allyl glycidyl ether [17].
Previously, research was carried out by Kurisingal et al. on two forms of metal-organic frameworks (Cu-BTC and UiO-66) as a binary system, which were synthesized solvothermally for the first time and applied as catalysts for CO 2 -epoxide cycloaddition without the addition of solvent. The research focused on binary MOF with Cu and Zr metal centers and assessed their catalytic potentials for CO 2 conversion into cyclic carbonate compounds with epoxides in the absence of any solvent. The study of the effects of some reaction parameters such as catalyst amount, temperature, reaction time, and CO 2 pressure was examined. The synthesized binary MOFs were used for the CO 2 -epoxide cycloaddition in the presence of TBABr co-catalyst. The catalyst (UiO-66/Cu-BTC) accomplished good conversion for epichlorohydrin (ECH) with over 99% selectivity. The remarkable conversion of ECH by the UiO-66/Cu-BTC/TBAB binary system was promoted by the combined effect of the two metal centers (Cu and Zr) and the bromide ion (Br − ) from TBAB. The catalyst reusability was explored by reusing the catalyst six times without losing its catalytic properties [64].

Summary and Outlooks
The application of metal-organic framework-based catalytic materials for the CO 2 cycloaddition with epoxides in some recent studies was summarized and reported in this review. MIL-101(Cr) and HKUST-1 based MOFs were identified as a high potential MOF catalysts for the conversion of CO 2 into cyclic organic carbonates. The two forms of MOF materials, MIL-101(Cr) and HKUST-1, were both identified to possess special features that gave the advantage in catalyzing the reaction of CO 2 with epoxides. Both MIL-101(Cr) and HKUST-1 frame works were characterized with Lewis-acid property due to the presence of metal (II) ions at their metal centers and their structural flexibilities. Thus, allowing their modifications by replacing different functional groups in the organic ligand of MIL-101 and substituting the metal centers in HKUST-1 respectively, while maintaining their backbone structures. The large porosity in the structures of the two materials contributed to their adsorption ability in CO 2 and can also be utilized in making effective heterogeneous system by encapsulation of other nanomaterials to form composites. The resultant composites would preserve the original properties of the MOFs, while gaining additional unique properties that can be applied in different applications. There is currently on-going research to improve the catalytic properties of the two MOFs (MIL-101(Cr) and HKUST-1) for better performance. The review also shown that MOF-based catalysts operate effectively when coupled with a co-catalyst. The presence of a co-catalyst helps in performing the crucial role of the activation of the epoxide ring-opening in the second step of the reaction chain. Among the several co-catalyst materials, the one with high electrophilic ability served the best in catalyzing the reaction and for this process, tetrabutylammonium bromide (TBABr) was proven to show the highest performance. The continuous modification of the porous MOFs revealed positive results towards increasing their catalytic activity for CO 2 cycloaddition with epoxides.

Conflicts of Interest:
The authors declare no conflict of interest.