Understanding Photoexcitation in Artificial Photosynthetic Systems: A GPU-accelerated First-principles Approach

Aaron Sisto, Stanford University

Photochemical conversion of solar energy involving light-harvesting molecules has been identified as a promising pathway to practical alternative energy technologies. Multi-chromophoric phenylacetylene dendrimers exhibit the prerequisite chemical properties for solar energy conversion due to strong light absorption in the visible spectrum and efficient excitation energy transfer. Furthermore, dendrimers can be synthesized with a high degree of order and a wide range of chromophoric units. However, the design of higher efficiency artificial photosynthetic systems and energy conversion devices necessitates a detailed understanding of molecular photoexcitation and energy transfer processes. Here, we use time-dependent density functional theory (TDDFT) to describe the electronic structure and excitation processes in dendrimeric systems. The implementation of this methodology on GPUs has facilitated a fully ab initio description of excitation processes in large-scale molecules. As a result, absorption spectra of a wide range of dendrimers are computed and agreement with recent experiments is discussed. Additionally, the correlation between molecular conformation at finite temperature and photoabsorption and electron relaxation is determined. These results also are examined for varying dendrimer structure and number of chromophoric units to gain insight into the electron transfer mechanisms from the periphery to the core molecule.

Abstract Author(s): Aaron Sisto, Alexander Witt, Todd Martinez