Photosynthesis is a promising method for H₂O₂ production, but its application in pure water is limited by slow oxidation kinetics and rapid photocarrier recombination of photocatalysts. Herein, a novel defective carbon nitride photocatalyst (D-C_3−xN₄) containing the C vacancies and the frustrated Lewis pairs (B and N of cyano group) is designed for H₂O₂ photosynthesis, and the role of C vacancies on the electron transfer mechanism during photocatalysis is systematically investigated. The D-C_3−xN₄ exhibits a H₂O₂ production rate of 140.1 μmol·g⁻¹·h⁻¹ in pure water, which is 87.6 times that of C₃N₄. Such superior performance for H₂O₂ photosynthesis is found to arise from the C vacancies and frustrated Lewis pairs (FLPs). The C vacancies have strong electron-trapping ability, which greatly enhances the separation of photocarriers. The C vacancies can also effectively reduce O₂ to ^*OOH via a proton-coupled process, which significantly accelerates the O₂ reduction kinetics. Meanwhile, the FLPs show an outstanding catalytic activity for H₂O oxidation. This study not only provides a new structure for highly active photocatalysts, but also deepens the understanding of the electron transfer mechanism of photocatalysts with trapped sites.

Plasmon-generated hot electrons show great potential for driving chemical reactions. The utilization efficiency of hot electrons is highly dependent on the interaction of the electronic states at the interfaces between plasmonic nanoparticles and other materials/molecules. Strong interaction can produce new hybridized electron states, which permit direct hot-electron transfer, a more efficient transfer mechanism. However, Au usually has very weak interaction with most molecules because of its inertness, which makes direct hot-electron transfer impossible. Herein, the improvement of the hot-electron transfer efficiency from Au to N₂ is demonstrated by introducing a Ru bridging layer. Both the N₂ fixation rate and Faradic efficiency (FE) are enhanced by the excitation of plasmons. The enhancement of the N₂ fixation rate is found to arise from plasmon-generated hot electrons. Theoretical calculations show that the strong interaction of the Ru electronic states with the N₂ molecular orbitals produces new hybridized electronic states, and the Ru d electrons also strongly couple with the Au sp electrons. Such a bridging role of Ru makes direct hot-electron transfer from Au to N₂ possible, improving the FE of nitrogen fixation. Our findings demonstrate a new approach to increasing the utilization efficiency of plasmonic hot electrons for chemical reactions and will be helpful to the design of plasmonic catalysts in the future.