Photoreduction of carbon dioxide enhanced by Cu atoms doped in a Pd cluster supported on TiO2: mechanism, selectivity, and catalytic descriptor†

Solar-powered transformation of CO₂ into hydrocarbon fuels stands out as a promising strategy to address the global energy crisis and to inhibit the increasing accumulation of atmospheric CO₂. Recent experimental study found that the isolation of Cu atoms in a Pd lattice supported on TiO₂ can remarkably enhance the photoreduction from CO₂ to CH₄ (selectivity, 96%). However, the photocatalytic reaction mechanism at the interface remains elusive and the physical origin for the high selectivity is unknown. Herein, we carry out a comprehensive investigation on the CO₂ photoreduction at the Pd₁₁Cu₂@TiO₂ interface (totally 11 models) using combined DFT calculations and microkinetic simulations. On the basis of the results, we find that the Cu doping induces substantial electron redistribution, which eventually enhances the adsorption and activation of CO₂. In addition, the Cu doping reduces the energy barrier of the highest-barrier step compared to Pd₁₃@TiO₂. Importantly, we find that the Cu–Pd sites perform better for CO₂ photoreduction than the Cu–Cu sites. Moreover, specific spatial arrangements of Cu atoms within Pd₁₁Cu₂ can, to different extents, modulate the CO₂ reduction. Through detailed investigation on the CO₂ reduction paths, the hydrogenation of *C*O to H*C*O is identified as the most critical elementary reaction and the negative value of the integrated crystal orbital Hamilton population of *C*O (i.e., −ICOHP_C–O) is proposed as a descriptor to predict the CO₂ reduction activity. Microkinetic simulations clearly show that the side products, such as CO, H₂, HCOOH, and CH₃OH, cannot compete with CH₄. In-depth analysis reveals that the high selectivity originates from the complicated competition among the relevant elementary reactions having different energy barriers, desorption energies and so on. Methodologically, the present work demonstrates that microkinetic simulations are indispensable in certain cases where potential energy profiles from static electronic structure calculations are inadequate to obtain correct reaction mechanisms. The obtained computational findings not only benefit the microscopic understanding of CO₂ photoreduction by metal-cluster supported catalysts, but also help rationally and precisely design bimetallic-cluster supported catalysts with superior catalytic activity and selectivity.