Liquid fuel synthesis via CO2 hydrogenation by coupling homogeneous and heterogeneous catalysis

Abstract

•CO2 hydrogenation by coupling homogeneous and heterogeneous catalysis•The reaction is conducted at low temperature with high C5+ selectivity•The activity is comparable to that of the reported Ru0-catalyzed FTS reaction CO2 is a major greenhouse gas, as well as a cheap, nontoxic, and renewable C1 resource. As a clean reductant, H2 can be produced from water by renewable energy, such as wind, hydro, and solar energy. Production of liquid fuel from CO2 and H2 is a promising route to solve the ever-increasing environmental and social challenges of the society. This route usually proceeds through two consecutive reactions accelerated by heterogeneous catalysts. Because of thermodynamic limitations of the cascade reactions, the reported heterogeneous catalysts usually suffer from high temperature and low selectivity. To overcome the barrier of the current technology, we coupled the homogeneous and heterogeneous catalysis in promoting the two reactions, respectively, and outstanding reaction results were achieved. Synthesis of liquid fuel (C5+ hydrocarbons) via CO2 hydrogenation generally involves cascade catalysis of reverse water gas shift (RWGS) and Fischer-Tropsch synthesis (FTS) reactions over heterogeneous catalysts. Because of thermodynamic limitations of the cascade reactions, the reported heterogeneous catalysts usually suffer from high temperature (above 300°C) and low selectivity. Here, we report the liquid fuel production from CO2 hydrogenation by coupling homogeneous RuCl3 and heterogeneous Ru0 catalysts, which proceeded at the lowest temperature reported so far (180°C) and reached the highest C5+ selectivity to date (71.1%). The TOF of the CO2 hydrogenation reached 9.5 h−1, which is comparable with that of the reported Ru0 catalyzed FTS reaction using syngas. Detailed study indicates that synergy of homogeneous and heterogeneous catalysis is the key for the outstanding performance. Synthesis of liquid fuel (C5+ hydrocarbons) via CO2 hydrogenation generally involves cascade catalysis of reverse water gas shift (RWGS) and Fischer-Tropsch synthesis (FTS) reactions over heterogeneous catalysts. Because of thermodynamic limitations of the cascade reactions, the reported heterogeneous catalysts usually suffer from high temperature (above 300°C) and low selectivity. Here, we report the liquid fuel production from CO2 hydrogenation by coupling homogeneous RuCl3 and heterogeneous Ru0 catalysts, which proceeded at the lowest temperature reported so far (180°C) and reached the highest C5+ selectivity to date (71.1%). The TOF of the CO2 hydrogenation reached 9.5 h−1, which is comparable with that of the reported Ru0 catalyzed FTS reaction using syngas. Detailed study indicates that synergy of homogeneous and heterogeneous catalysis is the key for the outstanding performance. Transformation of CO2 into value-added chemicals is of great importance for a sustainable society.1He M. Sun Y. Han B. Green carbon science: scientific basis for integrating carbon resource processing, utilization, and recycling.Angew. Chem. Int. Ed. Engl. 2013; 52: 9620-9633Crossref PubMed Scopus (558) Google Scholar, 2Artz J. Müller T.E. Thenert K. Kleinekorte J. Meys R. Sternberg A. Bardow A. Leitner W. Sustainable conversion of carbon dioxide: an integrated review of catalysis and life cycle assessment.Chem. Rev. 2018; 118: 434-504Crossref PubMed Scopus (882) Google Scholar, 3Liu Q. Wu L. Jackstell R. Beller M. Using carbon dioxide as a building block in organic synthesis.Nat. Commun. 2015; 6: 5933Crossref PubMed Scopus (1202) Google Scholar, 4Aresta M. Carbon Dioxide as Chemical Feedstock. Wiley-VCH Verlag GmbH, 2010Crossref Scopus (17) Google Scholar, 5Qian Q. Zhang J. Cui M. Han B. Synthesis of acetic acid via methanol hydrocarboxylation with CO2 and H2.Nat. Commun. 2016; 7: 11481Crossref PubMed Scopus (91) Google Scholar, 6Wang Y. Qian Q. Zhang J. Bediako B.B.A. Wang Z. Liu H. Han B. Synthesis of higher carboxylic acids from ethers, CO2 and H2.Nat. Commun. 2019; 10: 5395Crossref PubMed Scopus (16) Google Scholar, 7Cui M. Qian Q. He Z. Zhang Z. Ma J. Wu T. Yang G. Han B. Bromide promoted hydrogenation of CO2 to higher alcohols using Ru-Co homogeneous catalyst.Chem. Sci. 2016; 7: 5200-5205Crossref PubMed Google Scholar, 8He Z. Qian Q. Ma J. Meng Q. Zhou H. Song J. Liu Z. Han B. Water- enhanced synthesis of higher alcohols from CO2 hydrogenation over a Pt/Co3O4 catalyst under milder conditions.Angew. Chem. Int. Ed. Engl. 2016; 55: 737-741Crossref PubMed Scopus (130) Google Scholar, 9Asare Bediako B.B.A. Qian Q. Zhang J. Wang Y. Shen X. Shi J. Cui M. Yang G. Wang Z. Tong S. Han B. Ru-catalyzed methanol homologation with CO2 and H2 in an ionic liquid.Green Chem. 2019; 21: 4152-4158Crossref Google Scholar As a clean reductant, H2 can be produced from water by renewable energy, such as wind, hydro, and solar energy.10Higuita Cano M.H. Agbossou K. Kelouwani S. Dubé Y. Experimental evaluation of a power management system for a hybrid renewable energy system with hydrogen production.Renew. Energy. 2017; 113: 1086-1098Crossref Scopus (17) Google Scholar In the past decades, major advances in CO2 hydrogenation have been focused on synthesizing C1 products.11Álvarez A. Bansode A. Urakawa A. Bavykina A.V. Wezendonk T.A. Makkee M. Gascon J. Kapteijn F. Challenges in the greener production of formates/formic acid, methanol, and DME by heterogeneously catalyzed CO2 hydrogenation processes.Chem. Rev. 2017; 117: 9804-9838Crossref PubMed Scopus (590) Google Scholar, 12Klankermayer J. Wesselbaum S. Beydoun K. Leitner W. Selective catalytic synthesis using the combination of carbon dioxide and hydrogen: catalytic chess at the interface of energy and chemistry.Angew. Chem. Int. Ed. Engl. 2016; 55: 7296-7343Crossref PubMed Scopus (501) Google Scholar, 13Jessop P.G. Joó F. Tai C.C. Recent advances in the homogeneous hydrogenation of carbon dioxide.Coord. Chem. Rev. 2004; 248: 2425-2442Crossref Scopus (589) Google Scholar, 14Kattel S. Ramírez P.J. Chen J.G. Rodriguez J.A. Liu P. Active sites for CO2 hydrogenation to methanol on Cu/ZnO catalysts.Science. 2017; 355: 1296-1299Crossref PubMed Scopus (721) Google Scholar, 15Li S. Xu Y. Chen Y. Li W. Lin L. Li M. Deng Y. Wang X. Ge B. Yang C. et al.Tuning the selectivity of catalytic carbon dioxide hydrogenation over iridium/cerium oxide catalysts with a strong metal-support interaction.Angew. Chem. Int. Ed. Engl. 2017; 56: 10761-10765Crossref PubMed Scopus (234) Google Scholar, 16Wang F. He S. Chen H. Wang B. Zheng L. Wei M. Evans D.G. Duan X. Active site dependent reaction mechanism over Ru/CeO2 catalyst toward CO2 methanation.J. Am. Chem. Soc. 2016; 138: 6298-6305Crossref PubMed Scopus (297) Google Scholar Hydrocarbons containing more than five carbons (C5+) are commonly used as liquid fuel, such as gasoline, diesel, and kerosene. Besides the well-established chemistry of liquid hydrocarbons from the syngas, direct synthesis of liquid fuel from CO and H2O was also realized.17Xu Y. Li J. Li W. Li W. Zhang X. Zhao Y. Xie J. Wang X. Liu X. Li Y. et al.Direct conversion of CO and H2O into liquid fuels under mild conditions.Nat. Commun. 2019; 10: 1389Crossref PubMed Scopus (15) Google Scholar Hydrogenation of CO2 to liquid fuel is of great importance but is challenging. It usually proceeds through heterogeneously catalyzed reverse water gas shift (RWGS) reaction to generate CO and subsequent CO hydrogenation to hydrocarbons via Fischer-Tropsch synthesis (FTS).18He Z. Cui M. Qian Q. Zhang J. Liu H. Han B. Synthesis of liquid fuel via direct hydrogenation of CO2.Proc. Natl. Acad. Sci. USA. 2019; 116: 12654-12659Crossref PubMed Scopus (60) Google Scholar, 19Prieto G. Carbon dioxide hydrogenation into higher hydrocarbons and oxygenates: thermodynamic and kinetic bounds and progress with heterogeneous and homogeneous catalysis.ChemSusChem. 2017; 10: 1056-1070Crossref PubMed Scopus (93) Google Scholar, 20Marques Mota F.M. Kim D.H. From CO2 methanation to ambitious long-chain hydrocarbons: alternative fuels paving the path to sustainability.Chem. Soc. Rev. 2019; 48: 205-259Crossref PubMed Google Scholar, 21Guo L. Sun J. Ge Q. Tsubaki N. Recent advances in direct catalytic hydrogenation of carbon dioxide to valuable C2+ hydrocarbons.J. Mater. Chem. A. 2018; 6: 23244-23262Crossref Google Scholar, 22Choi Y.H. Jang Y.J. Park H. Kim W.Y. Lee Y.H. Choi S.H. Lee J.S. Carbon dioxide Fischer-Tropsch synthesis: a new path to carbon-neutral fuels.Appl. Catal. B. 2017; 202: 605-610Crossref Scopus (124) Google Scholar, 23Wei J. Ge Q. Yao R. Wen Z. Fang C. Guo L. Xu H. Sun J. Directly converting CO2 into a gasoline fuel.Nat. Commun. 2017; 8: 15174Crossref PubMed Scopus (467) Google Scholar, 24Gao P. Li S. Bu X. Dang S. Liu Z. Wang H. Zhong L. Qiu M. Yang C. Cai J. et al.Direct conversion of CO2 into liquid fuels with high selectivity over a bifunctional catalyst.Nat. Chem. 2017; 9: 1019-1024Crossref PubMed Scopus (458) Google Scholar However, CO2 is a thermodynamically stable and chemically inert molecule, and the RWGS reaction is endothermic and needs a higher temperature,25Su X. Yang X. Zhao B. Huang Y. Designing of highly selective and high-temperature endurable RWGS heterogeneous catalysts: recent advances and the future directions.J. Energy Chem. 2017; 26: 854-867Crossref Scopus (87) Google Scholar whereas the FTS reaction is exothermic and a higher temperature tends to inhibit liquid hydrocarbons and favor secondary products.26Dry M.E. The Fischer-Tropsch process: 1950–2000.Catal. Today. 2002; 71: 227-241Crossref Scopus (1635) Google Scholar,27Maitlis P.M. Zanotti V. The role of electrophilic species in the Fischer-Tropsch reaction.Chem. Commun. 2009; 13: 1619-1634Crossref Scopus (116) Google Scholar Recently, Na-Fe3O4/HZSM-5 multifunctional catalyst23Wei J. Ge Q. Yao R. Wen Z. Fang C. Guo L. Xu H. Sun J. Directly converting CO2 into a gasoline fuel.Nat. Commun. 2017; 8: 15174Crossref PubMed Scopus (467) Google Scholar and In2O3/HZSM-5 bifunctional catalyst24Gao P. Li S. Bu X. Dang S. Liu Z. Wang H. Zhong L. Qiu M. Yang C. Cai J. et al.Direct conversion of CO2 into liquid fuels with high selectivity over a bifunctional catalyst.Nat. Chem. 2017; 9: 1019-1024Crossref PubMed Scopus (458) Google Scholar have been reported as excellent catalysts in converting CO2 into gasoline fuel in flow reactors at above 320°C. It was also shown that Co6/MnOx catalyst could effectively accelerate the reaction.18He Z. Cui M. Qian Q. Zhang J. Liu H. Han B. Synthesis of liquid fuel via direct hydrogenation of CO2.Proc. Natl. Acad. Sci. USA. 2019; 116: 12654-12659Crossref PubMed Scopus (60) Google Scholar Although important advances have been achieved, the above thermodynamic properties still restrict the performance of this route. As a whole, the temperatures of this reaction were usually high (above 300°C), and the C5+ selectivities in total products were generally low (<65%). Moreover, considerable CO (20%∼50%) usually remains in the final product. Without doubt, novel ideas and more efficient catalysts for production of liquid fuel via CO2 hydrogenation are highly desirable. The homogeneous catalysts are well known for their high efficiency at a low reaction temperature. If the RWGS reaction is accelerated effectively at a low temperature by a homogeneous catalyst, the above thermodynamic limitations may be overcome. Enlightened by this idea, we discovered synthesis of liquid fuel via CO2 hydrogenation by coupling a homogeneously catalyzed RWGS reaction and a heterogeneously accelerated FTS reaction in a batch reactor (Figure 1). The reaction could be effectively catalyzed by RuCl3 and Ru0 catalysts in 1-methyl-2-pyrrolidinone (NMP) solvent, where LiCl and LiI were utilized as cocatalyst and promoter, respectively. The reaction was conducted at 180°C, which is much lower than those reported in the literature (Table S1). The selectivity of liquid hydrocarbons (C5-C28 n-paraffins) could reach 71.1%, which is higher than those reported previously. The turnover frequency (TOF) of the reaction was as high as 9.5 h−1, which is comparable with the best level of the FTS reaction using the Ru0 catalyst. Moreover, the products were all n-paraffins, which has not been reported before in liquid fuel synthesis via CO2 hydrogenation. As far as we know, this is the first work to combine homogeneous and heterogeneous catalysis to produce liquid fuel via CO2 hydrogenation. The heterogeneous Ru catalyst (Ru0) was prepared by simple reduction of RuCl3 by NaBH4 in water. The Ru0 particles were well dispersed with size of about 3–5 nm (Figures S1 and S2). The N2 adsorption test indicated that the Brunauer–Emmett–Teller surface area of the Ru0 catalyst was about 5.6 m2/g (Figure S3). The X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure (XAFS) data of the Ru0 catalyst coincided well with those in the literature (Figures S4–S6). To find the suitable combination of homogeneous and heterogeneous catalysts, we screened different catalytic systems in CO2 hydrogenation (Table 1). The reaction could proceed efficiently over the RuCl3/Ru0 catalyst with LiCl as the cocatalyst and LiI as the promoter in 1-methyl-2-pyrrolidinone (NMP) solvent at 180°C (entry 1). The reaction temperature was the lowest to date in liquid fuel synthesis via CO2 hydrogenation. The liquid products were C5-C28 n-paraffins, which were confirmed using a GC-MS (Agilent-7890B-5977A), as well as by comparing the retention times of the standards in the gas chromatography (GC) traces (Figures S7 and S8). The selectivity of C5+ n-paraffins in total product could reach 71.1 C-mol %, which is the highest to date in CO2 hydrogenation. The TOF of the reaction was as high as 9.5 h−1, which is comparable with the best level of the FTS reaction using CO/H2 catalyzed by monometallic Ru0 nanocatalyst.28Xiao C.X. Cai Z.P. Wang T. Kou Y. Yan N. Aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanocluster catalyst.Angew. Chem. Int. Ed. Engl. 2008; 47: 746-749Crossref PubMed Scopus (168) Google Scholar Although heterogeneous Ru-based catalysts are efficient catalysts of the FTS reaction using CO,28Xiao C.X. Cai Z.P. Wang T. Kou Y. Yan N. Aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanocluster catalyst.Angew. Chem. Int. Ed. Engl. 2008; 47: 746-749Crossref PubMed Scopus (168) Google Scholar, 29Li W.-Z. Liu J.-X. Gu J. Zhou W. Yao S.-Y. Si R. Guo Y. Su H.-Y. Yan C.-H. Li W.-X. et al.Chemical insights into the design and development of face-centered cubic ruthenium catalysts for Fischer-Tropsch synthesis.J. Am. Chem. Soc. 2017; 139: 2267-2276Crossref PubMed Scopus (99) Google Scholar, 30Long H.C. Turner M.L. Fornasiero P. Kašpar J. Graziani M. Maitlis P.M. Vinylic initiation of the Fischer-Tropsch reaction over ruthenium on silica catalysts.J. Catal. 1997; 167: 172-179Crossref Scopus (48) Google Scholar C1 products (mostly methane) were usually produced when they are used in CO2 hydrogenation.16Wang F. He S. Chen H. Wang B. Zheng L. Wei M. Evans D.G. Duan X. Active site dependent reaction mechanism over Ru/CeO2 catalyst toward CO2 methanation.J. Am. Chem. Soc. 2016; 138: 6298-6305Crossref PubMed Scopus (297) Google Scholar,31Kim A. Debecker D.P. Devred F. Dubois V. Sanchez C. Sassoye C. CO2 methanation on Ru/TiO2 catalysts: on the effect of mixing anatase and rutile TiO2 supports.Appl. Catal. B. 2018; 220: 615-625Crossref Scopus (96) Google Scholar, 32Roldán L. Marco Y. García-Bordejé E. Function of the support and metal loading on catalytic carbon dioxide reduction using ruthenium nanoparticles supported on carbon nanofibers.ChemCatChem. 2015; 7: 1347-1356Crossref Scopus (20) Google Scholar, 33Kwak J.H. Kovarik L. Szanyi J. CO2 reduction on supported Ru/Al2O3 catalysts: cluster size dependence of product selectivity.ACS Catal. 2013; 3: 2449-2455Crossref Scopus (253) Google Scholar In fact, this is the first report of efficient synthesis of liquid fuel via CO2 hydrogenation using Ru catalysts. Figure 2 shows the detailed n-paraffin distribution and the Anderson-Schulz-Flory plot. Interestingly, the final products in this work were totally n-paraffins and no CO or other byproduct were observed. In the reported synthesis of liquid fuel via CO2 hydrogenation, the product usually contained various kinds of compounds, such as olefins, aromatics, CO, n-paraffins, branched paraffins, naphthene, and/or oxygenates.18He Z. Cui M. Qian Q. Zhang J. Liu H. Han B. Synthesis of liquid fuel via direct hydrogenation of CO2.Proc. Natl. Acad. Sci. USA. 2019; 116: 12654-12659Crossref PubMed Scopus (60) Google Scholar, 19Prieto G. Carbon dioxide hydrogenation into higher hydrocarbons and oxygenates: thermodynamic and kinetic bounds and progress with heterogeneous and homogeneous catalysis.ChemSusChem. 2017; 10: 1056-1070Crossref PubMed Scopus (93) Google Scholar, 20Marques Mota F.M. Kim D.H. From CO2 methanation to ambitious long-chain hydrocarbons: alternative fuels paving the path to sustainability.Chem. Soc. Rev. 2019; 48: 205-259Crossref PubMed Google Scholar, 21Guo L. Sun J. Ge Q. Tsubaki N. Recent advances in direct catalytic hydrogenation of carbon dioxide to valuable C2+ hydrocarbons.J. Mater. Chem. A. 2018; 6: 23244-23262Crossref Google Scholar, 22Choi Y.H. Jang Y.J. Park H. Kim W.Y. Lee Y.H. Choi S.H. Lee J.S. Carbon dioxide Fischer-Tropsch synthesis: a new path to carbon-neutral fuels.Appl. Catal. B. 2017; 202: 605-610Crossref Scopus (124) Google Scholar, 23Wei J. Ge Q. Yao R. Wen Z. Fang C. Guo L. Xu H. Sun J. Directly converting CO2 into a gasoline fuel.Nat. Commun. 2017; 8: 15174Crossref PubMed Scopus (467) Google Scholar, 24Gao P. Li S. Bu X. Dang S. Liu Z. Wang H. Zhong L. Qiu M. Yang C. Cai J. et al.Direct conversion of CO2 into liquid fuels with high selectivity over a bifunctional catalyst.Nat. Chem. 2017; 9: 1019-1024Crossref PubMed Scopus (458) Google Scholar The chain length distribution of the paraffins followed the Anderson-Schulz-Flory statistics, and the chain growth probability (α) was as high as 0.83, which agreed with the fact that considerable products of longer chains were produced in this work.Table 1Hydrogenation of CO2 to produce C5+ n-paraffins using different catalytic systemsEntryCatalystCo-catalystPromoterSolventSelectivity (C-mol %)TOFaTOF denotes moles of CO2 consumption per mole of total Ru catalysts per hour. (h-1)COC1-4C5+CH3OH1bThe selectivity of CH4 was 17.5 C-mol %, and the C2-4 selectivity was 11.4 C-mol %.RuCl3, Ru0LiClLiINMP028.971.109.52Ru0--NMP0100.0003.63Ru0LiClLiINMP0100.0004.34RuCl3--NMP0100.00025.35cThe solution after reaction was transparent and clear, indicating a homogeneous reaction.RuCl3LiClLiINMP84.810.205.0122.86RuCl3, Ru0--NMP0100.0003.77RuCl3, Ru0LiCl-NMP091.98.107.98RuCl3, Ru0-LiINMP0100.0006.69RuBr3, Ru0LiClLiINMP043.156.906.910RuI3, Ru0LiClLiINMP076.923.106.811Ru2(CO)6Cl4, Ru0LiClLiINMP034.066.007.912Ru3(CO)12, Ru0LiClLiINMP052.847.206.413RuCl3, Ru0NaClLiINMP066.833.206.014RuCl3, Ru0KClLiINMP075.324.701.915RuCl3, Ru0AlCl3LiINMP084.016.001.316RuCl3, Ru0[Bmim]ClLiINMP58.321.020.707.517RuCl3, Ru0LiBrLiINMP042.957.108.318RuCl3, Ru0LiBF4LiINMP092.17.901.519RuCl3, Ru0LiClNaINMP047.352.707.820RuCl3, Ru0LiClKINMP062.337.706.721RuCl3, Ru0LiClLiBrNMP065.334.709.822RuCl3, Ru0LiClLiIDMI040.659.405.223RuCl3, Ru0LiClLiIToluene0100.0000.224RuCl3, Ru0LiClLiISqualane084.915.100.425RuCl3, Ru0LiClLiIH2O0100.0000.226dCO2 was replaced by 2 MPa CO.Ru0--NMP-23.476.603.127cThe solution after reaction was transparent and clear, indicating a homogeneous reaction.RuCl3LiCl-NMP61.814.6023.6110.728cThe solution after reaction was transparent and clear, indicating a homogeneous reaction.RuCl3-LiINMP81.914.103.980.029dCO2 was replaced by 2 MPa CO.Ru0LiCl-NMP-7.292.802.230dCO2 was replaced by 2 MPa CO.Ru0-LiINMP-27.572.503.431dCO2 was replaced by 2 MPa CO.Ru0LiClLiINMP-19.480.602.432cThe solution after reaction was transparent and clear, indicating a homogeneous reaction.,dCO2 was replaced by 2 MPa CO.RuCl3LiClLiINMP-25.0075.014.733dCO2 was replaced by 2 MPa CO.,e50 μL methanol was added into the catalytic system.Ru0LiClLiINMP-67.432.6-1.0Reaction conditions: 3 μmol Ru homogeneous catalyst and 37 μmol Ru0 (based on the metal), 0.75 mmol cocatalyst, 0.4 mmol promoter, 2 mL solvent, 5 MPa CO2 and 5 MPa H2 (at room temperature), 180 oC, 12 h.a TOF denotes moles of CO2 consumption per mole of total Ru catalysts per hour.b The selectivity of CH4 was 17.5 C-mol %, and the C2-4 selectivity was 11.4 C-mol %.c The solution after reaction was transparent and clear, indicating a homogeneous reaction.d CO2 was replaced by 2 MPa CO.e 50 μL methanol was added into the catalytic system. Open table in a new tab Reaction conditions: 3 μmol Ru homogeneous catalyst and 37 μmol Ru0 (based on the metal), 0.75 mmol cocatalyst, 0.4 mmol promoter, 2 mL solvent, 5 MPa CO2 and 5 MPa H2 (at room temperature), 180 oC, 12 h. Both RuCl3 and Ru0 are necessary for efficient synthesis of C5+ paraffins via CO2 hydrogenation. When it was catalyzed by the Ru0 catalyst, only methane was produced (entries 2, 3). When only RuCl3 was used as catalyst, it was in situ reduced to Ru0 by H2, and methane was the sole product formed (entry 4). When LiCl and LiI were used together with RuCl3, CO was produced at a very high rate (entry 5). The solution after the reaction was transparent and clear, suggesting a homogeneous reaction. The photos of the solutions after the reactions accelerated by RuCl3 catalyst and RuCl3/Ru0 catalyst, respectively, are given in Figure S9. As for the binary RuCl3/Ru0 catalyst, the lithium halides were also required in the reaction. Without lithium halide, only methane was detected after the reaction (entry 6). When LiCl was added to the reaction, the catalytic rate increased and minor C5+ paraffins formed (entry 7). This indicated that LiCl acted as cocatalyst of RuCl3/Ru0. LiI could not accelerate the synthesis of C5+ paraffins over RuCl3/Ru0 (entries 6, 8), but when it was added to RuCl3/Ru0/LiCl, the production of C5+ paraffins increased considerably (entries 1, 7). Hence LiI was a promoter in the target reaction. Based on Ru0/LiCl/LiI, we tested other precursors of the homogeneous catalyst, i.e., RuBr3, RuI3, Ru2(CO)6Cl4, and Ru3(CO)12. The catalytic performance of the Ru halides followed the order: RuCl3> RuBr3> RuI3 (entries 1, 9, 10). This showed that Cl− was the best ligand for the catalysis of the target reaction. The eminent ligand effect of Cl− was further verified by the control test using Ru2(CO)6Cl4 and Ru3(CO)12, respectively (entries 11, 12). We also tried the cocatalyst with other cations (Na+, K+, Al3+, [Bmim]+) and anions (Br-, BF4-), and the results indicated that LiCl was most appropriate for the reaction (entries 1, 13–18). We further conducted reactions with other promoters (NaI, KI, and LiBr), and the results demonstrated that LiI was the best promoter (entries 1, 19–21). In the tested cocatalysts and promoters, Li+ was proved to be the best cationic species. When the metal cation became larger, both the catalytic activity and C5+ selectivity decreased evidently (entries 1, 13–15, 19, 20). The solvent was also crucial to the reaction. We also conducted the reaction in 1,3-dimethyl-2-imidazolidinone (DMI), squalane, toluene, and water, which possess different molecular structure, product solubility, and physico-chemical properties. For example, DMI has similar molecular structure to NMP, squalane has similar structure and property to the products, toluene is an aromatic compound and can dissolve the products, and water is a common proton solvent. After screening the reaction solvents, we found that NMP was the most effective for the reaction (entries 22–25). As a weak Lewis base, NMP may not only stabilize the catalyst but also absorb the acidic CO2.34Gholami F. Azizi S. Peyghambarzadeh S.M. Bohloul M.R. The modelling and experimental study on molecular diffusion coefficient of CO2 in N-methyl pyrolidone.Sep. Sci. Technol. 2017; 52: 2435-2442Crossref Scopus (3) Google Scholar We would like to mention that no liquid hydrocarbons were observed when other Lewis bases, such as triethylamine, dimethyl formamide (DMF), or pyridine, were used as solvents. In the reactions accelerated by the organometallic catalysts, even a small modification of the solvent structure may cause a large change in the rate and in the pattern of the processes in solution.35Gutmann V. Solvent effects on the reactivities of organometallic compounds.Coord. Chem. Rev. 1976; 18: 225-255Crossref Scopus (772) Google Scholar The solvent with similar structure to cyclic amide, such as NMP or DMI, may simultaneously facilitate the RWGS reaction and the C-C bond formation.36Wang Y. Zhang J. Qian Q. Asare Bediako B.B.A. Cui M. Yang G. Yan J. Han B. Efficient synthesis of ethanol by methanol homologation using CO2 at lower temperature.Green Chem. 2019; 21: 589-596Crossref Google Scholar In a word, the catalytic system that consisted of RuCl3/Ru0/LiCl/LiI and NMP was the best for the target reaction. Using the optimized catalytic system, we studied the impact of gas pressure and RuCl3/Ru0 dosage on the reaction (Table S2). At a fixed ratio of CO2 and H2 (1/1), the catalytic activity and C5+ selectivity increased evidently with elevating total pressure from 4 to 10 MPa (entries 1–4). The reaction was less sensitive to the pressure when the pressure was further increased (entry 5). The ratio of CO2/H2 also affected the reaction, and the optimal ratio of CO2 and H2 was 1/1 (entries 4, 6, 7). Without CO2 or H2, the reaction did not take place, indicating that both CO2 and H2 took part in the reaction (entries 8, 9). The dosage of the Ru catalysts also affected the catalytic results. When the total dosage of RuCl3 and Ru0 catalysts was fixed at 40 μmol, the best ratio of RuCl3/Ru0 was 3/37 (entries 4, 10–12). At the above ratio, the suitable catalyst dosage for the reaction was 3 μmol RuCl3 and 37 μmol Ru0 (entries 4, 13, 14). In short, the appropriate pressure was 5 MPa CO2 and 5 MPa H2 (at room temperature) and the suitable catalyst dosages were 3 μmol RuCl3 and 37 μmol Ru0. Figure 3A demonstrates the impact of LiCl dosage on the reaction. Without LiCl, no C5+ product was produced. The catalytic activity and C5+ selectivity were enhanced evidently with the increase of the LiCl dosage, and 0.75 mmol was the optimal. The impact of LiI dosage was similar to that of LiCl, and 0.4 mmol LiI was suitable for the reaction (Figure 3B). The halogen anions are well known ligands in transition metal catalysis. The inhibition of the reaction by too much halides could be attributed to occupation of active sites by excess Cl− and/or I−. Figure 3C illustrates the effect of reaction temperature. At 120°C, the CO selectivity was very high (> 80%), and only a small amount of C5+ n-paraffins (about 5%) was produced. The reaction rate and paraffin selectivity rose dramatically with increasing temperature. At 180°C, the C5+ selectivity reached 71.1 C-mol % and the TOF of the reaction was as high as 9.5 h−1. Moreover, no CO could be detected in the product. Beyond 180°C, the increase of catalytic performance became minor. Thus, the appropriate reaction temperature was 180°C. The time course of reaction is shown in Figure 3D. CO appeared rapidly at the beginning of the reaction and formed at a stable speed, then the paraffins were generated by FTS reaction at a higher and steady rate. With passing time, more and more CO was consumed, and the selectivity of C5+ paraffins rose continuously. At 12 h, the CO generated in the reactor was used up, and the C5+ selectivity reached 71.1 C-mol %. The C5+ selectivity did not change obviously, and the increase of the product became slower after 12 h. The Ru0 catalyst after the target reaction was washed with acetone five times and dried in a vacuum oven at 50°C for 12 h, then the catalyst was used directly for the next run. The results of the recycling test revealed that the catalytic performance did not change considerably after five cycles (Figure S10). The transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), XRD, XPS, and XAFS characterizations indicated that the particle size, crystal structure, and surface properties of the Ru0 catalyst did not alter obviously during the reaction (Figures S1, S2, S4–S6). The homogeneous RuCl3 could also stably catalyze the RWGS reaction. After the reaction, the gaseous sample was released, and the amount of CO was determined by GC. The water and small amount of methanol generated were removed from the reaction solution under vacuum at 70°C, then the reactant CO2 and H2 were recharged into the reactor to start the next reaction. It was found that production of CO had no obvious decrease after five cycles (Figure S11). The stability of the homogeneous catalyst was further verified by determining the contents of the catalytic components (cationic Ru, Li+, Cl−, and I−) before and after the reaction (Table S3). In the presence of lithium halides, the homogeneous RuCl3 catalyst could efficiently promote the RWGS reaction to generate CO. The Ru0 catalyst could not catalyze the production of liquid fuel in CO2 hydrogenation, although it is known as an excellent FTS catalyst.28Xiao C.X. Cai Z.P. Wang T. Kou Y. Yan N. Aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanocluster catalyst.Angew. Chem. Int. Ed. Engl. 2008; 47: 746-749Crossref PubMed Scopus (168) Google Scholar,29Li W.-Z. Liu J.-X. Gu J. Zhou W. Yao S.-Y. Si R. Guo Y. Su H.-Y. Yan C.-H. Li W.-X. et al.Chemical insights into the design and development of face-centered cubic ruthenium catalysts for Fischer-Tropsch synthesis.J. Am. Chem. Soc. 2017; 139: 2267-2276Crossref PubMed Scopus (99) Google Scholar Inspired by this, we tested the CO hydrogenation using Ru0 catalyst, and the results showed that remarkable C5+ paraffins were generated (Table 1, entry 26). These indicated that, in this work, the liquid fuel was generated via RWGS reaction catalyzed by RuCl3 and subsequent FTS reaction accelerated by Ru0. This reaction pathway was also proved by the time course of the reaction (Figure 3D). The scheme of the cascade reactions is shown in Figure 4. The time course study also revealed that trace C5+ paraffins began to form at 1 h, suggesting that the induction period of the catalytic system for the target reaction was 1 h. In comparison, trace CO appeared at the very start of the reaction and it increased with time. The electrospray ionization mass spectrometry (ESI-MS) study of the reaction solution further confirmed that RuCl3 was immediately converted to Ru carboyl halide species at the beginning of the reaction and they kept stable during the reaction (Figure S12). Nearly all the reported catalysts of RWGS reaction and FTS reaction were heterogeneous and operated in flow reactors.25Su X. Yang X. Zhao B. Huang Y. Designing of highly selective and high-temperature endurable RWGS heterogeneous catalysts: recent advances and the future directions.J. Energy Chem. 2017; 26: 854-867Crossref Scopus (87) Google Scholar,26Dry M.E. The Fischer-Tropsch process: 1950–2000.Catal. Today. 2002; 71: 227-241Crossref Scopus (1635) Google Scholar The excellent performance of this work lies in the elegant synergy of the catalytic components and the above two reactions in a batch reactor. Synergy of the catalytic components is the basis for the success of this work. The RuCl3-catalyzed RWGS reaction could be promoted by LiCl and LiI, respectively (Table 1, entries 27, 28). The Cl− was more effective than I− in accelerating the reaction rate. Compared with LiCl, LiI was more effective in increasing the CO selectivity and inhibiting methanol byproduct. When LiCl and LiI were adopted together, both the catalytic activity and CO selectivity were markedly improved (Table 1, entry 5). Obviously, synergistic effect existed between LiCl and LiI in promoting the RuCl3-catalyzed RWGS reaction. The LiCl and/or LiI were also indispensable for the stability of the RuCl3 catalyst. Without the lithium halide, the RuCl3 was reduced to Ru0 in situ, and methane, instead of CO, was produced (Table 1, entry 4). As for the Ru0-catalyzed FTS reaction, the C5+ selectivity could be evidently enhanced by adding LiCl (Table 1, entries 26, 29). The Cl− on the Ru0 catalyst surface may lead to a higher fraction of COad species adsorbed on Ru atoms with low coordination numbers, which could accelerate the FTS reaction.37González-Carballo J.M. Pérez-Alonso F.J. García-García F.J. Ojeda M. Fierro J.L.G. Rojas S. In-situ study of the promotional effect of chlorine on the Fischer-Tropsch synthesis with Ru/ Al2O3.J. Catal. 2015; 332: 177-186Crossref Scopus (16) Google Scholar Although LiI has little effect on the Ru0-catalyzed FTS reaction (Table 1, entries 26, 30), it could effectively promote the C5+ selectivity in CO2 hydrogenation in the presence of RuCl3/Ru0/LiCl (Table 1, entries 1, 7). The promoting effect of LiI on this reaction is much better than that of LiBr, indicating that I− played a key role in enhancing C5+ selectivity of the CO2 hydrogenation (Table 1, entries 1, 21). The effect of I− may be ascribed to its larger ion size and stronger nucleophilicity.38Maitlis P.M. Haynes A. James B.R. Catellani M. Chiusoli G.P. Iodide effects in transition metal catalyzed reactions.Dalton Trans. 2004; 2004: 3409-3419Crossref Scopus (71) Google Scholar The CO cleavage is a basic step in FTS reaction and can be strongly promoted by alkali metals.27Maitlis P.M. Zanotti V. The role of electrophilic species in the Fischer-Tropsch reaction.Chem. Commun. 2009; 13: 1619-1634Crossref Scopus (116) Google Scholar It was also reported that the alkali metal additives could improve the C5+ paraffin selectivity and suppress the formation of olefins and oxygenates in Ru0-catalyzed FTS reaction.39Wang C. Zhao H. Wang H. Liu L. Xiao C. Ma D. The effects of ionic additives on the aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanoparticle catalyst.Catal. Today. 2012; 183: 143-153Crossref Scopus (30) Google Scholar As discussed above, Li+ proved to be the best cationic species of the halides, which could be ascribed to its small size and/or strong Lewis acidity.5Qian Q. Zhang J. Cui M. Han B. Synthesis of acetic acid via methanol hydrocarboxylation with CO2 and H2.Nat. Commun. 2016; 7: 11481Crossref PubMed Scopus (91) Google Scholar The NMP as a solvent may remarkably affect the catalytic performance. It possesses stronger electron donating ability, which may accelerate the homogeneous Ru-catalyzed RWGS reaction.40Tsuchiya K. Huang J.-D. Tominaga K.-I. Reverse water-gas shift reaction catalyzed by mononuclear Ru complexes.ACS Catal. 2013; 3: 2865-2868Crossref Scopus (52) Google Scholar Only n-paraffins were formed when the Ru0 accelerated FTS reaction was conducted in NMP. Whereas, considerable olefin and/or oxygenates (i.e., aldehyde and alcohol) are usually observed when the reaction was conducted in other solvents.28Xiao C.X. Cai Z.P. Wang T. Kou Y. Yan N. Aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanocluster catalyst.Angew. Chem. Int. Ed. Engl. 2008; 47: 746-749Crossref PubMed Scopus (168) Google Scholar,29Li W.-Z. Liu J.-X. Gu J. Zhou W. Yao S.-Y. Si R. Guo Y. Su H.-Y. Yan C.-H. Li W.-X. et al.Chemical insights into the design and development of face-centered cubic ruthenium catalysts for Fischer-Tropsch synthesis.J. Am. Chem. Soc. 2017; 139: 2267-2276Crossref PubMed Scopus (99) Google Scholar Synergy of the RWGS and FTS reactions in the batch reactor is another key for the peculiar catalytic performance. The RuCl3 could catalyze CO hydrogenation to produce methanol (Table 1, entry 32). The time course of the RWGS reaction also revealed that CO was first produced, then methanol was formed slowly (Figure 5). However, the control tests showed that methanol did not participate in the subsequent FTS reaction, and it inhibited the reaction (Table 1, entries 31, 33). Therefore, if methanol was formed in CO2 hydrogenation, it should have been retained in the product, and the reaction rate should have been much lower. But little methanol was detected in the target reaction, and the catalytic activity was even higher than that in Ru0-catalyzed FTS reaction (Table 1, entries 1, 31). It can be known from Figures 3D and 5 that the paraffins could form at a markedly higher rate than methanol at 180°C. This suggested that CO generated in situ was converted to paraffins before methanol was formed. Moreover, the rate of FTS reaction was nearly the same at different CO contents, which explains that CO can be consumed completely in the reaction (Figure 3D). The major contribution of this work is high-efficiency production of liquid fuel at mild condition by coupling homogeneous and heterogeneous catalysis, and the selectivity of C5+ paraffins was the highest and the reaction temperature was the lowest up to date, which can be attributed to synergy of the catalytic components and synergy of the cascade reactions. More detailed results and mechanistic discussions of the homogeneous Ru-catalyzed RWGS reaction40Tsuchiya K. Huang J.-D. Tominaga K.-I. Reverse water-gas shift reaction catalyzed by mononuclear Ru complexes.ACS Catal. 2013; 3: 2865-2868Crossref Scopus (52) Google Scholar, 41Li W. Guo S. Guo L. Theoretical investigation of reverse water gas shift reaction catalyzed by ruthenium halogen carbonyl complexes.Catal. Surv. Asia. 2017; 21: 185-197Crossref Scopus (6) Google Scholar, 42Tominaga K. Sasaki Y. Hagihara K. Watanabe T. Saito M. Reverse water-gas shift reaction catalyzed by ruthenium cluster anions.Chem. Lett. 1994; 23: 1391-1394Crossref Google Scholar, 43Tominaga K. Sasaki Y. Watanabe T. Saito M. Ethylene oxide-mediated reverse water-gas shift reaction catalyzed by ruthenium complexes.Energy. 1997; 22: 169-176Crossref Scopus (7) Google Scholar and heterogeneous Ru0-accelerated FTS reaction17Xu Y. Li J. Li W. Li W. Zhang X. Zhao Y. Xie J. Wang X. Liu X. Li Y. et al.Direct conversion of CO and H2O into liquid fuels under mild conditions.Nat. Commun. 2019; 10: 1389Crossref PubMed Scopus (15) Google Scholar,27Maitlis P.M. Zanotti V. The role of electrophilic species in the Fischer-Tropsch reaction.Chem. Commun. 2009; 13: 1619-1634Crossref Scopus (116) Google Scholar, 28Xiao C.X. Cai Z.P. Wang T. Kou Y. Yan N. Aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanocluster catalyst.Angew. Chem. Int. Ed. Engl. 2008; 47: 746-749Crossref PubMed Scopus (168) Google Scholar, 29Li W.-Z. Liu J.-X. Gu J. Zhou W. Yao S.-Y. Si R. Guo Y. Su H.-Y. Yan C.-H. Li W.-X. et al.Chemical insights into the design and development of face-centered cubic ruthenium catalysts for Fischer-Tropsch synthesis.J. Am. Chem. Soc. 2017; 139: 2267-2276Crossref PubMed Scopus (99) Google Scholar, 30Long H.C. Turner M.L. Fornasiero P. Kašpar J. Graziani M. Maitlis P.M. Vinylic initiation of the Fischer-Tropsch reaction over ruthenium on silica catalysts.J. Catal. 1997; 167: 172-179Crossref Scopus (48) Google Scholar,37González-Carballo J.M. Pérez-Alonso F.J. García-García F.J. Ojeda M. Fierro J.L.G. Rojas S. In-situ study of the promotional effect of chlorine on the Fischer-Tropsch synthesis with Ru/ Al2O3.J. Catal. 2015; 332: 177-186Crossref Scopus (16) Google Scholar,39Wang C. Zhao H. Wang H. Liu L. Xiao C. Ma D. The effects of ionic additives on the aqueous-phase Fischer-Tropsch synthesis with a ruthenium nanoparticle catalyst.Catal. Today. 2012; 183: 143-153Crossref Scopus (30) Google Scholar have also been discussed in literature. In summary, liquid fuel (C5+ n-paraffins) was very efficiently synthesized via CO2 hydrogenation by coupling homogeneously catalyzed RWGS reaction and heterogeneously catalyzed FTS reaction. The catalytic system consisting of RuCl3-Ru0-LiCl-LiI and NMP exhibited outstanding performance. The TOF of the reaction was as high as 9.5 h−1, and the selectivity of C5+ n-paraffins in total product could reach 71.1 C-mol % at 180°C. Interestingly, the products were all n-paraffins, and no CO was detected in the final product. The synergy of catalytic components and/or the consecutive reactions was crucial for the outstanding performance of the catalytic system. This work opens a new avenue for high-efficient production of liquid fuel via CO2 hydrogenation at milder condition. We believe that the protocol to combine homogeneous and heterogeneous catalysts can also be used to design some other efficient catalytic systems.