Metal–Nitrogen–Carbon Catalysts by Dynamic Template Removal for Highly Efficient and Selective Electroreduction of CO2

Abstract

The renewable electroreduction of CO2 to CO is a key component of future clean energy scenarios. These scenarios allow for the recycling of carbon emissions into value-added chemicals which achieves the joint goal of reducing greenhouse gase(s) while producing valuable chemical product(s). A catalyst which has a high activity and selectivity for the electroreduction of CO2 to CO is highly desired for these applications. Nonprecious metal catalysts (non-PGM) and specifically metal–nitrogen–carbon (M–N–C) catalysts are prime cathode candidates as they are selective for CO and H2 formation with only trace amounts of other products such as CH4. The traditional method of production of atomically dispersed M–N–C proceeds either through a sacrificial poymer approach or through hard-templating the catalyst with silica. The other is through the direct pyrolysis of nonabundant metal macrocycles such as MOF-based precursors via a soft-templating approach. These syntheses have substantial industrial limitations as they require harsh acid or basic solvents for postpyrolytic removal of the support or they require rare chemical precursors. The method herein uses mechanochemical mixing of a fluorine-containing polymer with common pyrolytic precursors for the in situ removal of the template during the first pyrolysis. Further ball-milling and post-treatment in ammonia atmosphere yield a highly selective catalyst for CO2 reduction. The role of the metal center in these M–N–C catalysts in promoting CO2 reduction is explored (M = Fe, Ni, Co, Mn) vs the performance of metal-free N–C. A mechanistic pathway for CO2 reduction on the various M–N–C catalysts is suggested. The champion catalyst in terms of overall selectivity/activity (Ni–N–C) boasts a 98.9 ± 0.2% faradaic efficiency for CO formation (FEco) at −1.1 V vs RHE and an unmatched selectivity for CO2 reduction (FEco > 85%) even at low overpotential (E = −0.3 V vs RHE) compared to traditional Ni–N–C. The catalysts synthesized here present a promising class of electrocatalysts which may be explored for a range of electrocatalytic applications.