The conversion of carbon dioxide (CO2) into fuels and chemicals using renewable energy is a potential pathway to mitigate increasing CO2 concentration in the atmosphere and acidification of the oceans (1). In a process that is essentially the reverse of combustion and is analogous to photosynthesis, CO2 can be electrochemically reduced to hydrocarbons by using renewable power sources such as wind and solar (2). This process would not compete with direct use of renewable energy as electricity, as the objective is to store excess energy for later use. On page 783 of this issue, Dinh et al. (3) show that ethylene can be generated selectively via electrochemical CO2 reduction at rates that could yield a technologically feasible process.
The thermodynamics of reducing CO2 are similar to those of splitting water into hydrogen and oxygen, which has been done commercially with an energetic efficiency as high as 80% (4). However, CO2 reduction is considerably more challenging because of the unreactive nature of the CO2 molecule and the demands of controlling multiple electron and proton transfer events (12, in the case of ethylene) on the surface of the electrocatalyst. Copper catalysts bind carbon monoxide (CO) and other reaction intermediates in such a way as to produce two-carbon products such as ethylene and ethanol (5). However, it has been difficult to steer the reaction toward any one product. Moreover, most experimental studies provide CO2 to the electrode from aqueous solution, where its finite solubility leads to an upper, diffusion-limited current density of a few tens of milliamperes per square centimeter—far below what would be commercially relevant.
A breakthrough in efficiency can be achieved through intensification of mass transfer within the process. Process intensification is a chemical engineering approach that can achieve manyfold increases in product throughput by eliminating mass and energy transport limitations and exploiting potential synergies, such as combining multiple functions (for example, reaction and separation) (6). Use of a gas diffusion electrode similar to those in fuel cells can greatly reduce the mass-transfer constraint for CO2 and has enabled current densities above 500 mA/cm2 for formation of one-carbon products such as CO (7). However, careful management of the gas phase CO2, liquid electrolyte, and solid electrocatalyst is required to maintain selectivity and minimize parasitic reactions such as water reduction.
Higher pH conditions can increase the yield of two-carbon products (8), but CO2 itself is acidic, setting an upper limit to the pH attainable in a conventional experiment. Dinh et al. show that very high hydroxide (OH−) concentrations can be maintained at the catalyst surface, provided that the electrochemical conversion is faster than the homogeneous reaction of CO2 with OH− to form bicarbonate. An optimal balance between these competing processes is attained through the use of very thin (∼25 nm) Cucatalyst layers deposited on the gas diffusion electrode (see the figure). The collocation of the electrocatalyst and high CO2 and OH− concentrations led to about 70% current efficiency to ethylene at current densities up to 750 mA/cm2. When the authors used thicker catalyst layers (for example, 100 nm), a region of lower OH− concentration formed, resulting in lower selectivity for ethylene.
To be economically viable, the process would need to operate continuously. Dinh et al. found that the highly basic conditions required to enhance ethylene yield led to deterioration of the carbon-based electrode material within an hour. As an alternative, they designed and implemented a gas diffusion electrode consisting of base-stable polytetrafluoroethylene (PTFE), with copper nanoparticles as the catalyst and carbon nanoparticles and graphite providing electrical contact. Use of this electrode led to a lower current density (∼300 mA/cm2), but it was stable for 100 hours.
A full electrochemical CO2-reduction system must also oxidize water to oxygen at the anode for sustained operation. Dinh et al. performed such a full-cell experiment, using a nickel iron oxide (NiFeOx) to catalyze the oxygen-evolution reaction at the anode. They measured a full-cell energy conversion efficiency, which captures all losses (overpotentials at the cathode and anode and electrical resistance of the electrolyte), of 34%. This value is lower than the 60 to 80% achieved for water splitting but is comparable to CO2-reduction cells, which make one-carbon products such as CO or formate and have lower cathode overpotentials (9).
Although the work of Dinh et al. is an important step toward chemical storage of renewable energy, challenges remain. Their reactor, and indeed nearly all CO2-reduction reactors in the literature, makes products which are either entrained in the CO2 stream or dissolved in the electrolyte, leaving product separation as an unsolved challenge (10).
There is a lively discussion in the literature regarding the prospective economics of electrochemical CO2 reduction (11). Although there is consensus that a carbon tax would be required to provide an incentive for CO2 conversion, opinions diverge on the economic viability of the conversion targets (such as CO and/or syngas, ethylene, and ethanol). Benchmark demonstrations such as that of Dinh et al. can be used to focus the discussion.
The products of electrochemical CO2 reduction are simpler than those of natural photosynthesis, yet they are the most ambitious targets of preparative electrosynthetic chemistry; most work over the past 100 years or more has focused on simpler transformations involving far fewer electron transfers (12). Demonstrations such as that of Dinh et al., combined with increasing understanding of the mechanism, could lead to a commercially viable electrochemical CO2-reduction process for mitigating rising atmospheric CO2 concentrations and promoting the use of renewable energy.