Revisiting Electrochemical CO2 Reduction on Copper Via Embedded Correlated Wavefunction Theory

Electrochemical CO₂ reduction to useful chemicals could contribute in a circular economy to reducing greenhouse gas emissions. Copper remains the only metal catalyst for CO₂ reduction to produce hydrocarbons but it requires a high overpotential for reasonable rates and faradaic efficiency. In addition, copper exhibits poor selectivity to useful products. Quantum mechanics simulations can aid in understanding the reaction mechanisms of CO₂ reduction on copper, thus providing insights for rational design of more efficient electrocatalysts. All theoretical efforts in this area to date have been within the framework of approximate density functional theory (DFT), which suffers from electron self-interaction error. DFT predicts that the key intermediate, CO, adsorbs in a hollow site while atop site adsorption has been observed experimentally on copper, suggesting that DFT-predicted reaction mechanisms and energetics could be problematic. Instead, embedded correlated wavefunction (ECW) theory, which provides a local correction to exchange-correlation error inherent in DFT, successfully reproduces the observed CO atop site adsorption. We are the first to apply ECW theory to study electrochemical CO₂ reduction on copper. Within ECW, the extended surface is described by an embedding potential derived from density functional embedding theory, while the catalytically active center is treated with CW theory in the presence of the derived embedding potential. We will present ECW-predicted reaction mechanisms of CO₂ reduction on copper, including the critical CO reduction and C-C bond formation steps. We will compare rate-limiting steps predicted by DFT and ECW theory, as well as structural and mechanistic behavior predicted by these two methods.