Efficient Photocatalytic Reduction of CO 2 Catalyzed by the Metal–Organic Framework MFM-300(Ga)

Open AccessCCS ChemistryCOMMUNICATION5 Aug 2022Efficient Photocatalytic Reduction of CO2 Catalyzed by the Metal–Organic Framework MFM-300(Ga) Tian Luo, Zi Wang, Xue Han, Yinlin Chen, Dinu Iuga, Daniel Lee, Bing An, Shaojun Xu, Xinchen Kang, Floriana Tuna, Eric J. L. McInnes, Lewis Hughes, Ben F. Spencer, Martin Schröder and Sihai Yang Tian Luo Department of Chemistry, University of Manchester, Manchester M13 9PL , Zi Wang Department of Chemistry, University of Manchester, Manchester M13 9PL Photon Science Institute (PSI), University of Manchester, Manchester M13 9PL , Xue Han Department of Chemistry, University of Manchester, Manchester M13 9PL , Yinlin Chen Department of Chemistry, University of Manchester, Manchester M13 9PL , Dinu Iuga Department of Physics, University of Warwick, Coventry CV4 7AL , Daniel Lee Department of Chemical Engineering and Analytical Science, University of Manchester, Manchester M13 9PL , Bing An Department of Chemistry, University of Manchester, Manchester M13 9PL , Shaojun Xu Department of Chemistry, University of Manchester, Manchester M13 9PL , Xinchen Kang Institute of Chemistry, Chinese Academy of Science, Beijing 100190 , Floriana Tuna Department of Chemistry, University of Manchester, Manchester M13 9PL Photon Science Institute (PSI), University of Manchester, Manchester M13 9PL , Eric J. L. McInnes Department of Chemistry, University of Manchester, Manchester M13 9PL Photon Science Institute (PSI), University of Manchester, Manchester M13 9PL , Lewis Hughes Department of Earth and Environmental Sciences, University of Manchester, Manchester M13 9PL , Ben F. Spencer Department of Materials, University of Manchester, Manchester M13 9PL , Martin Schröder *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry, University of Manchester, Manchester M13 9PL and Sihai Yang *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry, University of Manchester, Manchester M13 9PL https://doi.org/10.31635/ccschem.022.202201931 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Photocatalytic reduction of CO2 to carbon fuels is an important target but highly challenging to achieve. Here, we report the efficient photoconversion of CO2 into formic acid over a Ga(III)-based metal–organic framework (MOF) material using triethanolamine as the sacrificial agent. Under light irradiation and at room temperature, photoreduction of CO2 over MFM-300(Ga) yields formic acid with a selectivity of 100%, a high productivity of 502 ± 18 μmol·gcat−1·h−1, and excellent catalytic stability. In situ electron paramagnetic resonance spectroscopy reveals that MFM-300(Ga) promotes the generation of CO2•− radical anions as a reaction intermediate driven by strong binding and activation of CO2 molecules at the bridging –OH sites within the pore. This study represents the first example of a Ga(III)-based MOF catalyst for CO2 reduction. Download figure Download PowerPoint Introduction The development of efficient photocatalysts to convert CO2 into value-added chemicals has attracted much interest.1,2 In addition to TiO2, various Ga-based semiconductors, such as GaN, GaP, Ga2O3, and ZnGa2O4, have been widely investigated as photocatalysts to drive this reaction.3–7 However, their wide optical bandgap (Eg) restricts the use of solar light mainly to the ultraviolet region, and, more importantly, their nonporous nature limits the mass transport of CO2. This impedes charge transfer between catalyst and substrate and also leads to the undesirable recombination of photogenerated electrons and holes.8 A number of strategies have been exploited to improve the photocatalytic performance of Ga-based semiconductors. For example, doping of metals (e.g., Ge, Zn) or nonmetals (e.g., N, Si) can narrow the bandgap and improve the light-harvesting efficiency of the resultant material.9–12 Fabrication of ultrathin nanosheets, nanowires, or porous structures can increase the surface area and CO2 uptake.13–15 Meanwhile the introduction of a cocatalyst, noble metal nanoparticles, or a second semiconductor to construct Z-scheme or heterojunction-type systems can promote electron transfer.10,16–18 However, the photocatalytic efficiency over state-of-the-art Ga-based semiconductors remains limited, and only gaseous products such as CH4 and CO are produced. For example, a top-performing heterostructure of Au/Al2O3/p-GaN shows a photocatalytic productivity for CO of 230 μmol·g−1·h−1.4 Metal–organic framework (MOF) materials incorporate active sites fixed uniformly in 3D space, thus preventing aggregation of catalytic centers and potentially enhancing charge separation. MOFs are therefore emerging as important photocatalysts for the reduction of CO2, showing potential to overcome the barriers of conventional semiconductors.19,20 For example, the intrinsic microporosity and catalytically active sites confined in MOFs can form unique "microreactors" to promote the adsorption and activation of CO2 via the formation of strong host–guest interactions.21 More importantly, the backbone of MOFs consisting of infinite metal-ligand linkages can facilitate efficient ligand-to-metal-charge-transfer (LMCT), thus prolonging the excitation lifetime by boosting the isolation and utilisation of photoinduced electrons.22 A number of MOF systems have been tested for photoreduction of CO2,19,20 and Ti-based MOFs are particularly attractive.23,24 In contrast, to date, no Ga-based MOF has been shown to exhibit activity for CO2 conversion, and we report here the first example for the efficient photoreduction of CO2. Under light irradiation and at room temperature, MFM-300(Ga) catalyzes the conversion of CO2 into formic acid with a 100% selectivity and an excellent productivity up to 502 ± 18 μmol·gcat−1·h−1, significantly higher than conventional Ga-based semiconductors and among the best-behaving MOF-based photocatalysts for this reaction. Importantly, in situ electron paramagnetic resonance (EPR) spectroscopy confirms that the CO2 radical anion (CO2•−) is generated as an intermediate to the production of formic acid over MFM-300(Ga). Results and Discussion MFM-300(Ga), [Ga2(OH)2(L)] (H4L = biphenyl-3,3ʹ,5,5ʹ-tetracarboxylic acid) was chosen for the photoreduction of CO2 due to its high stability, high adsorption, and strong binding of CO2 molecules via the formation of hydrogen bonds to the bridging –OH groups in the pore.25 MFM-300(Ga) is comprised of chains of [GaO4(OH)2]∞ octahedra linked by cis-μ2-OH groups, and these chains are further bridged by tetracarboxylate ligands to form a 'wine rack' open framework. Desolvated MFM-300(Ga) shows a Brunauer–Emmett–Teller (BET) surface area of 1064 m2·g−1 and an uptake of CO2 of 5.00 mmol·g−1 at 298 K and 1 bar (see Supporting Information Figure S1). The purity of the bulk material has been confirmed by powder X-ray diffraction (PXRD) (see Supporting Information Figure S2) and thermogravimetric analysis (see Supporting Information Figure S3a). Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) show that crystals of MFM-300(Ga) exhibit cuboid-shaped morphology with an average size of 15 μm (Figure 1b). High-resolution X-ray photoelectron spectroscopy (XPS) analysis of MFM-300(Ga) shows the peaks of Ga 2p1/2, 2p3/2, 3d3/2, and 3d5/2 at 1145.5, 1118.6, 21.2, and 20.7 eV, respectively, consistent with the trivalent Ga(III) (Figure 1c, see Supporting Information Figure S4). Solid-state 13C and 71Ga NMR spectroscopy reveal a highly ordered structure consistent with a single repeating octahedral [GaO4(OH)2] environment (see Supporting Information Figure S5).26 The high-field 71Ga{1H} 2D through-space (dipolar) heteronuclear correlation NMR spectrum of MFM-300(Ga) demonstrates this more extensively with strong correlations between the Ga environment and the hydroxyls (at δ{1H} = 2.8 ppm) and a weaker interaction with the more distant aromatic proton between the carboxylates (at δ{1H} = 9.0 ppm), with no other correlations observed (Figure 1d). Moreover, the ratio of the 1H NMR signal intensities of the aromatic protons to the hydroxy protons is ∼3:1 (see Supporting Information Figure S5c), entirely consistent with the structural model shown in Figures 1e–1h. Solid-state UV–vis diffuse reflectance spectroscopy (UV-DRS) of MFM-300(Ga) shows an intensive and broad absorption band in the ultraviolet region, which is assigned to the π→π* transition of the biphenyl ligand (Figure 1a).27 The Tauc plot yields an optical bandgap of 3.30 eV, lower than that of commercial Ga2O3 (Eg = 4.56, 4.70, and 4.67 eV for α-, β-, and γ-Ga2O3, respectively)28 and ZnGa2O4 (Eg = 4.18 eV).9 The photocurrent response of MFM-300(Ga) confirms that the current density increases upon irradiation and decreases upon turning off the light (see Supporting Information Figure S6). Figure 1 | Characterisation and crystal structure of MFM-300(Ga). (a) The UV-DRS spectrum of MFM-300(Ga) with Tauc plot (insert); (b) SEM image of MFM-300(Ga) with TEM image (insert); (c) high-resolution XPS spectrum of Ga 2p, and (d) 71Ga{1H} D-HMQC 2D MAS NMR spectrum of MFM-300(Ga) and corresponding 1D direct excitation 71Ga (top) and 1H (left) MAS NMR spectra, recorded at 20.0 T with a MAS frequency of 60 kHz; crystal structure of MFM-300(Ga): (e) octahedral [GaO4(OH)2]; (f) ligand; (g and h) views of binding sites (bridging –OH groups) for adsorbed CO2 molecules within MFM-300(Ga) studied by In situ synchrotron X-ray single-crystal diffraction.25 Host–guest hydrogen bonds and intermolecular dipole interactions are highlighted in cyan and yellow, respectively. Download figure Download PowerPoint The photocatalytic activity of MFM-300(Ga) toward the reduction of CO2 has been studied in CO2-saturated CH3CN-containing triethanolamine (TEOA) as the sacrificial agent under irradiation at 350–780 nm for 1–12 h. The gaseous products were analysed by gas chromatography and the liquid product by 1H NMR spectroscopy. The photocatalytic efficiency was measured as the moles of product obtained per gram of catalyst per hour (mol·gcat−1·h−1) to afford a direct comparison with reported catalysts. Formic acid was the only product detected in the liquid phase, and no carbon-containing gaseous products were detected. The photocatalytic performance as a function of reaction time over MFM-300(Ga) indicates that a high productivity of 502 ± 18 μmol·gcat−1·h−1 for formic acid was achieved at 4 h (Figures 2a, see Supporting Information Figure S7). More importantly, the photocatalytic activity and the crystallinity of MFM-300(Ga) were retained over three cycles of reaction (Figure 2b). The photocatalytic efficiency of MFM-300(Ga) is higher than the majority of reported MOFs (Table 1) for converting CO2 into formic acid, such as amino-functionalised MIL-125(Ti), UiO-66(Zr), and MIL-101(Fe),23,31,44 and is only lower than two cases. One is the mixed metal and mixed ligand systems of NH2-UiO-66(Zr/Ti) and (NH2)2-UiO-66(Zr/Ti),24 which are prepared via post-synthetic modifications to introduce Ti(IV) sites into the framework. The other is a recent report describing a π-conjugated naphthoporphyrin system constructed with Zr metal clusters, which demonstrates the highest value (6630 μmol·gcat−1·h−1) reported in the literature.29 A comparison of state-of-the-art studies of thermal hydrogenation of CO2 into formic acid over MOF-based catalysts is given in Supporting Information Table S1. Figure 2 | Photocatalytic CO2 reduction over MFM-300(Ga). (a) Different reaction time; (b) recycling tests. Reaction conditions: MFM-300(Ga) (10 mg), TEOA/CH3CN (3 mL/15 mL, saturated with CO2), 25 °C, 350–780 nm, light irradiation for 4 h. Download figure Download PowerPoint Table 1 | Comparison of the Photocatalytic Efficiency of MFM-300(Ga) and a Selection of MOFs Reported for this Reaction in the Literature MOF-Based Materials Chemical Formula CO2 Uptake/mmol·g−1 (1 atm) Sacrificial Agent Solvents Photocatalytic Efficiency (μmol·gcat−1·h−1) References TNP-MOF Zr6(OH)4O4(TNP)3 1.69 (273 K) 1.03 (298 K) TEOA CH3CN 6630 29 NH2-UiO-66-Zr(Ti) (NH2)2-UiO-66-Zr(Ti) – Zr4.3Ti1.7O4(OH)4(C8H7O4N)5.17(C8H8O4N2)0.83 – TEOA and BNAH CH3CN 782 1052 24 MFM-300(Ga) Ga2(OH)2(L) 6.77 (273 K) 5.00 (298 K) TEOA CH3CN 502 This work Co-MOF, LIFM-45 Co3(HL0)2·4DMF·4H2O – BNAH DMAc 456 30 NH2-MIL-101(Fe) MIL-101(Fe) NH2-MIL-53(Fe) MIL-53(Fe) NH2-MIL-88(Fe) MIL-88(Fe) Fe3OCl(H2O)3(BDC-NH2)3Fe3OCl(H2O)3(BDC)3FeIII(OH)(BDC-NH2) FeIII(OH)(BDC) FeIII3O(CH3OH)3(BDC-NH2)3FeIII3O(CH3OH)3(BDC)3 1.52 (273 K) 1.18 (273 K) 0.89 (273 K) 0.60 (273 K) 0.64 (273 K) 0.46 (273 K) TEOA CH3CN 445 148 116 74 75 23 31 AD-MOF-2 AD-MOF-1 [Co2(HAD)2(AD)2 (IA)2]·DMF [Co2(HAD)2 (AD)2-(BA)]·DMF·2H2O 1.86 (298 K) 2.33 (298 K) TIPA CH3CN 443 179 32 Fe3-Fe2-NH2Fe3-Fe2 [{Fe2-Tri}{Fe3(μ3-O)(BDC-NH2)3}]·4NO3[{Fe2-Tri}{Fe3(μ3-O)(BDC)3}]·4NO3 1.38 (298 K) 1.23 (298 K) TIPA CH3CN/H2O (1:1) 396 309 33 PCN-138 [Zr6(μ3-O)4(μ3-OH)4][TCPP][TBTB]8/3 2.82 (273 K) 1.82 (298 K) TIPA H2O 168 34 Ir-CP [Y(Ir(ppy)2(dcbpy)2][OH] – TEOA CH3CN 158 35 Eu-Ru(Phen)3 [Eu2(μ2-H2O)(H2O)3(L1)2]·(NO3)2·(2-FBA)2·(H2O)22 – TEOA CH3CN 94 36 Cd/Ru-MOF-1 Cd/Ru-MOF-2 {Cd3[Ru-L2]2·2(Me2NH2)·solvent}n {Cd[Ru-L3]·3(H2O)}n – TEOA CH3CN 67 72 37 PCN-222(Zr) Zr6(μ3-OH)8(OH)8(TCPP)2 2.59 (273 K) 1.56 (298 K) TEOA CH3CN 60 38 NNU-28(Zr) [Zr6O4(OH)4(L4)6]·6DMF 2.83 (273 K) 1.49 (298 K) TEOA CH3CN 53 39 PCN-136 Zr6(μ3-O)4(μ3-OH)4(OH)6(H2O)6(HCHC) 2.72 (273 K) TIPA CH3CN/H2O 44 40 Zr-SDCA-NH2 [Zr6O4(OH)4(L5)6]·8DMF 3.74 (273 K) 1.57 (298 K) TEOA CH3CN 41 41 NNU-31-Zn Fe2Zn(μ3-O)(TCA)2(H2O)3 1.65 (298 K) – H2O 26 42 NH2-MIL-125(Ti) MIL-125(Ti) Ti8O8(OH)4(BDC-NH2)6Ti8O8(OH)4(BDC)6 5.90 (273 K) 4.40 (273 K) TEOA CH3CN 16 4 23 NH2-UiO-66-(Zr/Ti) NH2-UiO-66(Zr) (Zr/Ti)6O4(OH)4(BDC-NH2)6Zr6O4(OH)4(BDC-NH2)6 3.79 (273 K) 3.04 (273 K) TEOA CH3CN 12 7 43 NH2-UiO-66(Zr) UiO-66(Zr) Zr6O4(OH)4(BDC-NH2)6Zr6O4(OH)4(BDC)6 3.04 (273 K) 2.37 (273 K) TEOA CH3CN 26 0 44 Notes: TEOA, triethanolamine; BNAH, 1-benzyl-1,4-dihydronicotinamide; H4L, biphenyl-3,3ʹ,5,5ʹ-tetracarboxylic acid; H2BDC, benzene-1,4-dicarboxylic acid; HAD, adenine; BA, butanedioic acid; IA, isobutyric acid; DMF, N,N-dimethylformamide; Tri, 1,2,4-triazole; DMAc, N,N-dimethylacetamide; TIPA, triisopropanolamine; H2TCPP, tetrakis(4-carboxyphenyl)porphyrin; TBTB, 4,4ʹ,4″-(2,4,6-trimethylbenzene-1,3,5-triyl)tribenzoate; ppy, 2-phenylpyridine; dcbpy, 2,2ʹ-bipyridine-4,4ʹ-dicarboxylate; H4L0, 2ʹ-amino-[1,1ʹ:4ʹ,1″-terphenyl]-3,3″,5,5″-tetracarboxylic acid; H3L1, Ru(phen)3-derived tricarboxylate acid metalloligand; 2-FBA, 2-fluorobenzoate; L2=5,5ʹ-dcbpy=2,2ʹ-bipyridine-5,5ʹ-dicarboxylate; L3=(4,4ʹ-dcbpy)2(bpy), bpy=2,2ʹ-bipyridine; H2L4, 4,4ʹ-(anthracene-9,10-diylbis(e-thyne-2,1-diyl))dibenzoic acid; HCHC, hexakis(4-carboxyphenyl)hexabenzocoronene; H2L5, 2,2ʹ-diamino-4,4ʹ-stilbene dicarboxylic acid; TCA, 4,4ʹ,4″-tricarboxytriphenylamine. To gain further insights into this reaction, a series of control experiments were conducted (Table 2). No carbon-containing product was detected from reactions in the absence of (1) MFM-300(Ga), (2) CO2 (where N2 is used instead), or (3) light. These results confirm that the carbon source of formic acid is CO2 and that the reaction proceeds via photocatalytic routes driven by the MOF catalyst. Replacement of TEOA with triethylamine (TEA)45 gives a low productivity of 64 μmol·gcat−1·h−1 for formic acid, which is consistent with recent reports on the important role of TEOA in binding and assisting the transport of CO2 in CH3CN.46,47 A range of different organic solvents have been tested, and CH3CN demonstrates the highest activity due to the optimal efficiency of mass transfer and the enhanced binding of CO2 by TEOA in CH3CN47 (see Supporting Information Figure S8). Interestingly, no product was observed when using Ga2O3 (∼50 mesh, Eg = 4.57), GaN (Eg = 3.04), GaP (Eg = 1.92), or a powdered mixture of Ga(NO3)3 and H4L as the photocatalyst (see Supporting Information Figure S9). This indicates that the framework structure of MFM-300(Ga) is crucial for adsorption of CO2 and the separation and transfer of photoinduced electrons. Table 2 | Summary of Reaction Conditions of Comparison Experiments Entry Catalyst Light (nm) Gas HCOOH 1 MFM-300(Ga) 350–780 TEOA CO2 502 2 n.a. 350–780 TEOA CO2 n.a. 3 MFM-300(Ga) 350–780 TEOA N2 n.a. 4 MFM-300(Ga) 350–780 n.a. CO2 n.a. 5 MFM-300(Ga) n.a. TEOA CO2 n.a. 6 MFM-300(Ga) 350–780 TEA CO2 64 7 Ga(NO3)3·9H2O and H4L 350–780 TEOA CO2 n.a. 8 Ga2O3 350–780 TEOA CO2 n.a. 9 GaN 350–780 TEOA CO2 n.a. 10 GaP 350–780 TEOA CO2 n.a. 11 MIL-53(Ga) 350–780 TEOA CO2 n.a. 12 MIL-68(Ga) 350–780 TEOA CO2 n.a. Reaction conditions: MFM-300(Ga) (10 mg), TEOA/CH3CN (3 mL/15 mL, saturated with CO2), 25 °C, 350–780 nm, light irradiation for 4 h. For entry 6, 3 mL TEA was added to replace TEOA; For entry 7, a powdered mixture of Ga(NO3)3·9H2O (0.04 mmol, 16.7 mg) and H4L (0.02 mmol, 6.6 mg) were used (H4L, biphenyl-3,3,5ʹ,5ʹ-tetracarboxylic acid); For entry 8–12, the catalysts used were 10 mg in each reaction. To investigate the effects of ligands and framework structures of Ga-MOFs on the photocatalytic property, MIL-53(Ga) and MIL-68(Ga), both constructed from terephthalic acid, were studied for this reaction.48–50 The phase purity of these MOFs has been confirmed by PXRD (see Supporting Information Figure S10). The bandgaps for MIL-53(Ga) and MIL-68(Ga) are determined to be 3.21 and 3.93 eV, respectively, by UV-DRS analysis (see Supporting Information Figure S11). Under the same reaction conditions as above, neither MOF shows photocatalytic activity toward the reduction of CO2. Although showing similar BET surface areas (1117–1140 m2·g−1) and a structure comprising [GaO4(OH)2] moieties, the CO2 adsorption capacity of MIL-53(Ga) and MIL-68(Ga) (1.46–1.65 mmol·g−1) is much lower than that (5.00 mmol·g−1) of MFM-300(Ga) (see Supporting Information Table S2). This is likely due to the rigid framework and presence of active μ2-OH sites in MFM-300(Ga) that promote the transport and binding of CO2 as confirmed by in situ crystallographic and spectroscopic analyses.25 Furthermore, the biphenyl ligand of MFM-300(Ga) demonstrates increased π-electron delocalisation in the framework backbone and hence enhanced LMCT compared with terephthalate in MIL-53(Ga) and MIL-68(Ga). These collectively result in the drastic difference in their photocatalytic activity. Photophysical and electrochemical experiments were conducted to understand the redox properties of MFM-300(Ga). Mott-Schottky (MS) analysis and valence-band XPS studies were conducted to determine the positions of the conduction band (ECB) and the valence band (EVB) of MFM-300(Ga), respectively.32,51 The EVB is determined to be 2.2 V versus normalised hydrogen electrode (NHE) from the VB-XPS result (see Supporting Information Figure S12a). The positive slope of all MS plots recorded at different frequencies indicates that MFM-300(Ga) is a typical n-type semiconductor, and the intercept is independent of the frequency. The flat band potential (EFB) is determined as −0.99 V versus NHE (see Supporting Information Figure S12b), which is usually ∼0.1 V more positive than the ECB for n-type materials.52 Thus, the ECB is estimated to be −1.09 V versus NHE. The gap between VB and CB is thus 3.29 eV, which is in excellent agreement with the bandgap of 3.30 eV obtained from UV-DRS analysis. To identify the radical species involved in the catalytic process, in situ spin-trapping EPR experiments under photocatalytic conditions were conducted. Since the lifetime of free radicals is several orders of magnitude shorter than the acquisition time of EPR spectra, 5,5-dimethyl-1-pyrroline-n-oxide (DMPO) was used as a spin trap to enable the identification of radicals as long-lived DMPO-radical adducts.53 An intense six-line signal with g = 2.005 and hyperfine coupling constants AN = 15.2 G and AH = 18.9 G was detected and unambiguously assigned to DMPO-CO2•−54,55 (see Figure 3a and Supporting Information Table S3). This indicates that the CO2 radical anion (CO2•−) is the direct intermediate to form formic acid. No radical was captured for the reaction conducted under dark conditions. Significantly, to the best of our knowledge, this is the first time that a direct intermediate radical has been detected in the photoreduction of CO2 over a MOF-based catalyst. Figure 3 | (a) In situ X-band EPR spectra of photocatalytic reactions over MFM-300(Ga) using DMPO as spin trap, (black) before and (red) after light irradiation, with simulated (blue) spectrum showing a major component, DMPO-CO2•− (green, simulation), and a minor component, DMPO-Ox (cyan, simulation) under photocatalytic conditions over MFM-300(Ga). (b) The proposed mechanism of the photocatalytic reduction of CO2 over MFM-300(Ga). Download figure Download PowerPoint The catalytic cycle for this reaction is proposed (Figure 3b). Upon light irradiation, MFM-300(Ga) is activated, and the electrons in the VB (2.2 V) are promoted to the CB (−1.09 V), and the photoinduced holes at the VB are readily filled up by the electron sacrificial agent TEOA. Surprisingly, the photoinduced electrons with a reductive potential of −1.09 V versus NHE can reduce CO2 to CO2•− radical anions, given the strongly negative redox potential of −1.90 V versus NHE [Eq. (1)] for this process.56 This indicates that the strong binding of CO2 molecules to the bridging μ2-OH sites of MFM-300(Ga) through the formation of hydrogen bonds as determined by in situ synchrotron X-ray diffraction25 not only promotes the transfer of photoinduced electrons from MFM-300(Ga) to the bound CO2 molecules but also activates the adsorbed CO2 molecules. This shifts the reduction to a more anodic potential. Interestingly, the path of photocatalytic reduction of CO2 into formic acid is widely regarded as accepting electrons and protons simultaneously at a redox potential of −0.61 V versus NHE [Eq. (2)].56 By contrast, the adsorbed CO2 molecule in MFM-300(Ga) is able to accept a single electron via host–guest hydrogen bonding to generate a CO2•− radical anion, which is unprecedented in the MOF-driven photoreduction of CO2 and affords new insights into the mechanism of photocatalytic reduction of CO2. Recently, the positive impact of strong host-guest interactions on the activation of the adsorbed substrate has also been demonstrated by the conversion of small molecules over porous catalysts.57,58 Thus, the photocatalytic activity of MOFs is determined by the synergistic effect of bandgap, rate of charge transfer, uptake of CO2, and most importantly, the interaction between the catalyst scaffold and adsorbed CO2 molecules, which will inform the design of new photocatalysts. CO 2 + e − → CO 2 • − E 0 redox = − 1.90 V vs . NHE (1) CO 2 + 2 H + + 2 e − + HCOOH E 0 redox = − 0.61 V vs . NHE (2) Conclusion The porosity and design flexibility of MOFs, coupled with their intrinsic semiconductor and photoelectrical properties, make them promising candidates as efficient photocatalysts. We report the first example of a Ga-MOF-based semiconductor that can promote the photoreduction of CO2 to formic acid with a selectivity of 100% and a high productivity of 502 ± 18 μmol·gcat−1·h−1 under light irradiation and at room temperature using TEOA as an electron sacrificial agent. MFM-300(Ga) shows excellent catalytic stability over three cycles of reactions with full retention of the productivity of formic acid. In situ EPR spectroscopic analysis confirms the generation of the CO2•− radical anion as the reaction intermediate promoted by the strong host–guest interactions between the bridging μ2-OH groups of MFM-300(Ga) and the adsorbed CO2 molecules. Compared with other reported Ga-MOFs in literature, the presence of strong binding sites and efficient LMCT plays an important role in boosting the photocatalytic activity toward CO2 reduction, which sheds light on the design of future MOF-based photocatalysts with improved activity. Supporting Information Supporting Information is available and includes detailed experimental procedures and characterization data. Conflict of Interest There is no conflict of interest to report. Acknowledgments This research was supported by EPSRC (EP/I011870, EP/V056409), the Royal Society and the University of Manchester with funding, and EPSRC for funding of the EPSRC National EPR Facility at the University of Manchester. This project has also received funding from the European Research Council under the European Union's Horizon 2020 research and innovation programme (grant agreement No 742401, NANOCHEM). The UK High-Field Solid-State NMR Facility used in this research was funded by EPSRC and BBSRC (EP/T015063/1) as well as the University of Warwick including via partial funding through Birmingham Science City Advanced Materials Projects 1 and 2 supported by Advantage West Midlands and the European Regional Development Fund. The authors wish to acknowledge Dr. Marek Nikiel for the help with XPS measurement. References 1. Ran J.; Jaroniec M.; Qiao S. Z.Cocatalysts in Semiconductor-Based Photocatalytic CO2 Reduction: Achievements, Challenges, and Opportunities.Adv. Mater.2018, 30, 1704649. Google Scholar 2. Zhang W.; Mohamed A. R.; Ong W. J.Z-Scheme Photocatalytic Systems for Carbon Dioxide Reduction:Where Are We Now?Angew. Chem. Int. Ed.2020, 59, 22894–22915. Google Scholar 3. Dhakshinamoorthy A.; Navalon S.; Corma A.; Garcia H.Photocatalytic CO2 Reduction by TiO2 and Related Titanium Containing Solids.Energy Environ. Sci.2012, 5, 9217–9233. Google Scholar 4. Li R.; Cheng W.-H.; Richter M. H.; DuChene J. S.; Tian W.; Li C.; Atwater H. A.Unassisted Highly Selective Gas-Phase CO2 Reduction with a Plasmonic Au/p-GaN Photocatalyst Using H2O as an Electron Donor.ACS Energy Lett.2021, 6, 1849–1856. Google Scholar 5. Barton E. E.; Rampulla D. M.; Bocarsly A. B.Selective Solar-Driven Reduction of CO2 to Methanol Using a Catalyzed p-GaP Based Photoelectrochemical Cell.J. Am. Chem. Soc.2008, 130, 6342–6344. Google Scholar 6. Akatsuka M.; Kawaguchi Y.; Itoh R.; Ozawa A.; Yamamoto M.; Tanabe T.; Yoshida T.Preparation of Ga2O3 Photocatalyst Highly Active for CO2 Reduction with Water Without Cocatalyst.Appl. Catal. B Environ.2020, 262, 118247. Google Scholar 7. Yan S. C.; Ouyang S. X.; Gao J.; Yang M.; Feng J. Y.; Fan X. X.; Wan L. J.; Li Z. S.; Ye J. H.; Zhou Y.; Zou Z. G.A Room-Temperature Reactive-Template Route to Mesoporous ZnGa2O4 with Improved Photocatalytic Activity in Reduction of CO2.Angew. Chem. Int. Ed.2010, 122, 6544–6548. Google Scholar 8. Marszewski M.; Cao S.; Yu J.; Jaroniec M.Semiconductor-Based Photocatalytic CO2 Conversion.Mater.Horizons2015, 2, 261–278. Google Scholar 9. Yan S.; Wang J.; Gao H.; Wang N.; Yu H.; Li Z.; Zhou Y.; Zou Z.Zinc Gallogermanate Solid Solution: A Novel Photocatalyst for Efficiently Converting CO2 into Solar Fuels.Adv. Funct. Mater.2013, 23, 1839–1845. Google Scholar 10. Tatsumi H.; Teramura K.; Huang Z.; Wang Z.; Asakura H.; Hosokawa S.; Tanaka T.Enhancement of CO Evolution by Modification of Ga2O3 with Rare-Earth Elements for the Photocatalytic Conversion of CO2 by H2O.Langmuir2017, 33, 13929–13935. Google Scholar 11. Yan S.; Yu H.; Wang N.; Li Z.; Zou Z.Efficient Conversion of CO2 and H2O into Hydrocarbon Fuel over ZnAl2O4-Mod