UC Berkeley

Both [Co^II(qpy)(H₂O)₂]²⁺ and [Fe^II(qpy)(H₂O)₂]²⁺ (with qpy = 2,2':6',2″:6'',2‴-quaterpyridine) are efficient homogeneous electrocatalysts and photoelectrocatalysts for the reduction of CO₂ to CO. The Co catalyst is more efficient in the electrochemical reduction, while the Fe catalyst is an excellent photoelectrocatalyst ( ACS Catal. 2018, 8, 3411-3417). This work uses density functional theory to shed light on the contrasting catalytic pathways. While both catalysts experience primarily ligand-based reductions, the second reduction in the Co catalyst is delocalized onto the metal via a metal-ligand bonding interaction, causing a spin transition and a distorted ligand framework. This orbital interaction explains the experimentally observed mild reduction potential and slow kinetics of the second reduction. The decreased hardness and doubly occupied d_z²-orbital facilitate a σ-bond with the CO₂-π* in an η¹-κC binding mode. CO₂ binding is only possible after two reductions resulting in an EEC mechanism (E = electron transfer, C = chemical reaction), and the second protonation is rate-limiting. In contrast, the Fe catalyst maintains a Lewis acidic metal center throughout the reduction process because the metal orbitals do not strongly mix with the qpy-π* orbitals. This allows binding of the activated CO₂ in an η²-binding mode. This interaction stabilizes the activated CO₂ via a π-type interaction of a Fe-t_2g orbital and the CO₂-π* and a dative bond of the oxygen lone pair. This facilitates CO₂ binding to a singly reduced catalyst resulting in an ECE mechanism. The barrier for CO₂ addition and the second protonation are higher than those for the Co catalyst and rate-limiting.