Enhanced electrocatalytic CO2 reduction via field-induced reagent concentration

Abstract

Electrochemical reduction of carbon dioxide (CO) to carbon monoxide (CO) is the first step in the synthesis of more complex carbon-based fuels and feedstocks using renewable electricity. Unfortunately, the reaction suffers from slow kinetics owing to the low local concentration of CO surrounding typical CO reduction reaction catalysts. Alkali metal cations are known to overcome this limitation through non-covalent interactions with adsorbed reagent species, but the effect is restricted by the solubility of relevant salts. Large applied electrode potentials can also enhance CO adsorption, but this comes at the cost of increased hydrogen (H) evolution. Here we report that nanostructured electrodes produce, at low applied overpotentials, local high electric fields that concentrate electrolyte cations, which in turn leads to a high local concentration of CO close to the active CO reduction reaction surface. Simulations reveal tenfold higher electric fields associated with metallic nanometre-sized tips compared to quasi-planar electrode regions, and measurements using gold nanoneedles confirm a field-induced reagent concentration that enables the CO reduction reaction to proceed with a geometric current density for CO of 22 milliamperes per square centimetre at -0.35 volts (overpotential of 0.24 volts). This performance surpasses by an order of magnitude the performance of the best gold nanorods, nanoparticles and oxide-derived noble metal catalysts. Similarly designed palladium nanoneedle electrocatalysts produce formate with a Faradaic efficiency of more than 90 per cent and an unprecedented geometric current density for formate of 10 milliamperes per square centimetre at -0.2 volts, demonstrating the wider applicability of the field-induced reagent concentration concept.

Catalyst electro-redeposition controls morphology and oxidation state for selective carbon dioxide reduction

“…It has been shown previously that sharp tips can improve bubble nucleation, concentrate stabilizing cations and exhibit high local fields, all of which increase current densities [27][28][29][30] . This high current density also promotes high local pH, limiting the protonation of bound CO that leads to methane formation 54 .…”

“…Furthermore, high-curvature structures, such as nanoneedles, promote nucleation of smaller gas bubbles 27 , and benefit from field-induced reagent concentration [28][29][30][31][32] , where high local negative electric fields concentrate positively charged cations to help stabilize CO 2 reduction intermediates 33 , enhancing CO 2 RR. However, combining high-curvature morphology with Cu + promotion to enable selective chemical conversion has yet to be explored.…”

“…To model ERD Cu, we constructed a mixed Cu:Cu 2 O slab with 25% Cu + species, closely matching the optimal amount of Cu + determined from sXAS studies. The ERD Cu(111) and ERD Cu(211) slabs serve as model systems for flat and high-curvature ERD Cu surfaces, respectively 28 . It was found that the Gibbs free energy of formation of the OCCOH* intermediate was lowest (1.13 eV) on the ERD Cu(211) system, 0.76 eV lower than the Cu(211) system, suggesting the presence of Cu + is favourable for ethylene production (Fig.…”

The Virtuous CO2 Circle or the Three Cs: Capture, Cache, and Convert

“…In fact, metal [86] was used to produce a nanoporous silver (np-Ag) catalyst able to reduce CO 2 electrochemically to CO in an efficient and selective way [87]. Nanostructuration of other metals such as gold was also studied both theoretically and practically, and it was proved that the increase of concentration of alkali cations due to manipulation of electric field allows an increase in electrocatalytic CO 2 reduction [88]. Another route is the production of formate/formic acid.…”

Use machine learning to find energy materials

“…It has already been used to predict the results of quantum simulations to identify potential molecules and materials for flow batteries, organic light-emitting diodes 6 , organic photovoltaic cells and carbon dioxide conversion catalysts 7 . The algorithms can predict results in a few minutes, compared with the hundreds of hours it takes to run the simulations 8 .…”