A step closer to the electrochemical production of liquid fuels.

The efficient and large-scale synthesis of liquid fuels from renewable sources is one of grand challenges of modern chemistry. One important low-toxicity liquid fuel that can be produced sustainably is ethanol, in particular cellulosic ethanol made from second-generation biomass sources that do not compete with food production. Large-scale plants for producing cellulosic ethanol already exist, demonstrating the viability of this technology. The main advantage of the production of cellulosic ethanol is that we can make extensive use of (photo-)catalytic technology that nature has already developed: the photosynthesis of sugar monomers and cellulose polymers by plants through carbon dioxide fixation, and the subsequent enzymatic “cracking” of cellulose to ethanol. However, from a purely chemical point of view, the production of cellulosic ethanol is inelegant and wasteful. The detour from carbon dioxide via cellulose to ethanol is highly unfavorable energetically; even by the most optimistic estimates, the efficiency of the conversion of photon energy to chemical energy is not more than 1%. Besides the chemical inefficacy, tremendous agricultural investments will be needed if we wish to produce such biofuels on a large scale. A promising long-term alternative for the large-scale production of liquid fuel is to replace nature s catalytic technology by tailored man-made catalytic technology, that is, to convert CO2 to ethanol (or another liquid fuel) through a limited number of sensible intermediates using efficient, durable, and cost-effective synthetic catalysts, which preferably operate at ambient temperature. In contrast to the production of cellulosic ethanol, this new technology does not yet exist, but researchers worldwide are making progress in understanding the intricacies of two key reactions in the process—CO2 reduction and water oxidation. In a recent letter to Nature, the research group lead by Matthew Kanan at Stanford University reports on an exciting new development in a possible final step of such a future technology: the selective conversion of carbon monoxide to ethanol and other oxygenates on a copper electrode. Copper has been known to be a good catalyst for the electrochemical conversion of CO2 and CO to methane and ethylene since the seminal work of Hori in the 1980s. Kanan et al. have now shown that nanocrystalline oxide-derived copper (OD-Cu) electrodes, prepared from the reduction of thick Cu2O layers, produce mainly ethanol, as well as acetate and n-propanol, at low overpotentials ( 0.25 to 0.5 V versus RHE (reversible hydrogen electrode)) with an unprecedented Faraday efficiency of up to 57%. No C1 products were observed, indicating rapid C C coupling at low overpotential. By comparing their OD-Cu to electrodes composed of commercial Cu nanoparticles, Kanan et al. conclude that the exceptional CO reduction activity of OD-Cu is related to the constrained environment of the grain boundaries formed during its synthesis. In addition, OD-Cu has a substantially lower hydrogen evolution activity than Cu nanoparticles, which also contributes to the high selectivity of the OD-Cu catalyst towards hydrocarbons. Another copper surface that mediates the highly selective formation of C2 products through CO reduction is the Cu(100) single-crystal electrode, on which the selective formation of ethylene can be observed at 0.3 V versus RHE. In good agreement with the results of Kanan et al., this selective C2 formation at low overpotential is also observed only in alkaline media, whereas at higher overpotentials, C2 selectivity decreases and methane is formed on Cu(100) as well. The low-potential C C coupling in alkaline media has been explained by the formation of surface-bound CO dimers through a rate-limiting proton-decoupled electron transfer, a reaction step which, according to recent density functional theory (DFT) calculations, strongly prefers (100) surface sites. The subsequent hydrogenation of this dimer leads to the formation of a CH2CHO(ads) intermediate, a kind of oxymetallacycle that binds to the copper surface through both C and O (see Figure 1). The DFT calculations suggest that this intermediate can be either reduced to ethylene, as observed on Cu(100) single-crystal electrodes, or to ethanol, as observed by Kanan et al. The factors determining whether ethylene or ethanol is formed are not yet understood, but presumably the pH and the local surface [*] Dr. K. J. P. Schouten, Dr. F. Calle-Vallejo, Prof. Dr. M. T. M. Koper Leiden University, Leiden Institute of Chemistry P.O. Box 9502, 2300 RA Leiden (The Netherlands) E-mail: m.koper@chem.leidenuniv.nl