There is widespread interest in the hydrogenation of CO2 to energy-rich products such as formate. However, first-row transition metal catalysts for the hydrogenation of CO2 to formate remain rare. Copper complexes are widely used in the reduction of organic substrates but their use in the catalytic hydrogenation of CO2 has been limited. Here we demonstrate that the copper(I) complex LCu(MeCN)PF6 is an active catalyst for CO2 hydrogenation in the presence of a suitable base. Screening of bases and studies of catalytic reactions by in operando spectroscopy revealed important and unusual roles for the base in promoting H2 activation and turnover.The development of catalysts for efficient hydrogenation of CO2 is an active area of research, with the potential to reduce our dependence on fossil resources for the production of chemical fuels and feedstocks. Catalysts that can convert the electrical energy from intermittent energy sources, such as wind and solar, into chemical fuels could provide a valuable energy storage mechanism by producing fuels during periods of excess supply that can be used during periods of excess demand. 1 When produced from the reduction of carbon dioxide, carbon-based fuels such as formic acid (usually trapped as formate) 2 and methanol 3 are attractive targets for energy storage, as these fuels have higher volumetric energy densities and can be stored and transported more efficiently and safely than hydrogen. 4 With an energy input of renewably generated H2 or electricity, the overall cycle can in principle be carbon-neutral, as equal quantities of CO2 are first taken up and then emitted over the life cycle of the fuel. 5Numerous transition-metal catalysts for CO2 hydrogenation to formate have been developed over the previous few decades, with turnover numbers (TON, or mol formate per mol catalyst) easily exceeding 10 4 and turnover frequencies (TOF, or turnovers per unit time) in some cases greater than 10 6 h -1 . 6-8 The most active catalysts are typically complexes of either ruthenium or iridium. The greater abundance and lower cost of first-row transition metals would make them better suited to the large-scale production of fuels, if they could be made sufficiently active as catalysts. Several examples of first-row catalysts for CO2 hydrogenation are now known, and, in particular, complexes of Fe 9-15 and Co [16][17][18][19] have been shown to be quite active. There are reasons to believe that copper complexes should be effective in CO2 hydrogenation. For instance, copper dispersions on ZnO/Al2O3 are widely used in the industrial conversion of syngas to methanol; 20,21 mechanistic studies have revealed that this reaction occurs primarily through hydrogenation of the CO2, rather than CO, in these CO2/CO/H2/H2O mixtures. 22,23 Moreover, homogeneous copper phosphine and carbene complexes are highly efficient catalysts for the reduction of CO2 to CO, 24,25 hydroboration of CO2 to form boryl formates, 26 and hydrosilylation of CO2 to form silyl formates. 27-29 These reactions are driven by t...

The copper(I) triphosphine complex LCu(MeCN)PF6 (L = 1,1,1-tris(diphenylphosphinomethyl)ethane), which we recently demonstrated is an active catalyst precursor for hydrogenation of CO2 to formate, reacts with H2 in the presence of a base to form a cationic dicopper hydride, [(LCu)2H]PF6. [(LCu)2H](+) is also an active precursor for catalytic CO2 hydrogenation, with equivalent activity to that of LCu(MeCN)(+), and therefore may be a relevant catalytic intermediate. The thermodynamic hydricity of [(LCu)2H](+) was determined to be 41.0 kcal/mol by measuring the equilibrium constant for this reaction using three different bases. [(LCu)2H](+) and the previously reported dimer (LCuH)2 can be synthesized by the reaction of LCu(MeCN)(+) with 0.5 and 1 equiv of KB(O(i)Pr)3H, respectively. The solid-state structure of [(LCu)2H](+) shows threefold symmetry about a linear Cu-H-Cu axis and significant steric strain imposed by bringing two LCu(+) units together around the small hydride ligand. [(LCu)2H](+) reacts stoichiometrically with CO2 to generate the formate complex LCuO2CH and the solvento complex LCu(MeCN)(+). The rate of the stoichiometric reaction between [(LCu)2H](+) and CO2 is dramatically increased in the presence of bases that coordinate strongly to the copper center, e.g. DBU and TMG. In the absence of CO2, the addition of a large excess of DBU to [(LCu)2H](+) results in an equilibrium that forms LCu(DBU)(+) and also presumably the mononuclear hydride LCuH, which is not directly observed. Due to the significantly enhanced CO2 reactivity of [(LCu)2H](+) under these catalytically relevant conditions, LCuH is proposed to be the catalytically active metal hydride.

The thermodynamic hydricity of a metal hydride can vary considerably between solvents. This parameter can be used to determine the favourability of a hydride-transfer reaction, such as the reaction between a metal hydride and CO2 to produce formate. Because the hydricities of these species do not vary consistently between solvents, reactions that are thermodynamically unfavourable in one solvent can be favourable in others. The hydricity of a water-soluble, bis-phosphine nickel hydride complex was compared to the hydricity of formate in water and in acetonitrile. Formate is a better hydride donor than [HNi(dmpe)2](+) by 7 kcal mol(-1) in acetonitrile, and no hydride transfer from [HNi(dmpe)2](+) to CO2 occurs in this solvent. The hydricity of [HNi(dmpe)2](+) is greatly improved in water relative to acetonitrile, in that reduction of CO2 to formate by [HNi(dmpe)2](+) was found to be thermodynamically downhill by 8 kcal mol(-1). Catalysis for the hydrogenation of CO2 was pursued, but the regeneration of [HNi(dmpe)2] under catalytic conditions was unfavourable. However, the present results demonstrate that the solvent dependence of thermodynamic parameters such as hydricity and acidity can be exploited in order to produce systems with balanced or favourable overall thermodynamics. This approach should be advantageous for the design of future water-soluble catalysts.

The nature of the iron-iron bond in the mixed-valent diiron tris(diphenylforamidinate) complex Fe(2)(DPhF)(3), which was first reported by Cotton, Murillo et al. (Inorg. Chim. Acta 1994, 219, 7-10), has been examined using additional spectroscopic and theoretical methods. It is shown that the coupling between the two iron centers is strongly ferromagnetic, giving rise to an octet spin ground state. On the basis of Mössbauer spectroscopy, the two iron centers, formally mixed-valent Fe(II)Fe(I), are completely equivalent with an isomer shift δ = 0.65 mm s(-1) and quadrupole splitting ΔE(Q) = +0.32 mm s(-1). A large, positive zero-field splitting D(7/2) = 8.2 cm(-1) has been determined from magnetic susceptibility measurements. Multiconfigurational quantum studies of the complete molecule Fe(2)(DPhF)(3) found one dominant configuration (σ)(2)(π)(4)(π*)(2)(σ*)(1)(δ)(2)(δ*)(2), which accounts for 73% of the ground-state wave function. By considering all the configurations, an estimated metal-metal bond order of 1.15 has been calculated. Finally, Fe(2)(DPhF)(3) exhibits weak electronic absorptions in the visible and near-infrared regions, which are assigned as d-d transitions from the doubly occupied metal-metal π molecular orbital to half-occupied π*, δ, and δ* orbitals.