Simulation of photoinduced processes in photoelectrochemical cells for solar fuel production Jan Paul Menzel 1 , Wan Jae Dong 2,3 , Zetian Mi 2 , Victor S. Batista 1 1 Department of Chemistry, Yale University, 225 Prospect St, New Haven, CT 06520, USA, 2 Department of Electrical Engineering and Computer Science, University of Michigan, 1301 Beal Avenue, Ann Arbor, MI 48109, USA, 3 current adress: Dep. of Integrative Energy Engineering, Korea University, Seoul 02841, Rep. of Korea Solar-to-fuel devices such as Photoelectrochemical Cells (PECs) have shown great promise in the conversion of solar energy into chemical fuels, and have the potential to play a key role in the transformation to a carbon- neutral, clean-energy society. For example, PECs based on GaN nanowires decorated with catalytic clusters have been successfully employed in a wide range of photoelectrochemically driven production of chemical fuels, such as hydrogen evolution, [1] urea production [2] and nitrate reduction [3,4] . However, the efficiency of such devices is inherently connected to the properties of all components involved. Computational investigations can provide a framework for investigating, designing, and optimizing promising candidates and therefore facilitate the search for novel photoelectrochemical and catalytic materials. The atomic resolution of ab initio simulations allows for zooming in on the processes at the molecular level, elucidating electronic structure, dynamics, catalytic mechanisms and can provide design principles for more efficient material assemblies. Modelling and investigating key processes in these complex systems that involve surfaces, solid-liquid interfaces, molecular components and catalysts however remains challenging due to the different timescales and extended character of the system. Here, we apply a wide range of computational methods, ranging from DFT(+U), real-time excited state dynamics [5,6] and semi-empirical methods to investigate photoinduced processes at the material interface. Our computational mechanistic studies reveal the underlying molecular reasons for observed efficiencies of photoelectrochemical conversion at the catalytic surface, elucidating key intermediates, while providing insights into the role of surface structure and competing side reactions. We find that (de-) stabilization of key binding modes on the catalytic surface determines product selectivity. We elucidate the process of photoelectrochemical defect formation and its influence on the chemical conversion. Through a combination of semi-empirical molecular dynamics for the nuclei and quantum dynamics based on an Extended Hueckel Hamiltonian, we follow the interfacial charge transfer in photoelectrodes in real time, providing insights into the ultrafast photoinduced charge separation and how nuclear motion and molecular design influence the injection rate. This multiscale modeling of photoelectrochemical processes in catalytic materials provides valuable insights into molecular-level processes, offering generalizable design principles for more efficient solar energy conversion systems. References
1. J. P. Menzel, W. J. Dong et al. , ACS Catalysis 2024 , 14 , 13314-13323; 2. W. J. Dong, ǂ J. P. Menzel ǂ et al ., ACS Catalysis 2024 , 14 , 2588-2596; 3. W. J. Dong, ǂ J. P. Menzel ǂ et al ., accepted at Nature Communications ; 4. W. J. Dong, ǂ J. P. Menzel ǂ et al ., Small 2025 , 21 , 2412089; 5. J. Jayworth, ǂ C. Decavoli ǂ et al . Appl. Mat. Interfaces 2024 , 16 ,14841–14851; 6. J. P. Menzel et al ., J. Phys. Chem. C 2020 , 124 , 27965–27976;
P61
© The Author(s), 2025
Made with FlippingBook Learn more on our blog