The atomic motion in chemical systems that arises due to temperature and solvent interaction has a large effect on the system’s observed electronic properties. Ensemble-averaged properties of the system can be calculated by performing conformational sampling. However, calculating ensemble-averaged properties in this way requires performing a huge number of electronic structure calculations, making conformational sampling on large chemical systems computationally infeasible. We have developed the Effective Stochastic Potential (ESP) method to overcome these challenges associated with conformational sampling. The ESP method performs conformational sampling efficiently by treating fluctuations in the system as a stochastic perturbation which is constructed from ab-initio calculations of thermally-accessible states. In this work, we have combined the sampling capabilities of the ESP method with linear-response time-dependent Density Function Theory (LR-TDDFT) in order to accurately and efficiently calculate ensemble-averaged excitation energies of a variety of molecules at finite temperature. In the ESP-LR-TDDFT method, the single-particles states and energies are obtained from a stochastic Kohn-Sham Fock operator, and the A and B matrix elements are constructed using the stochastic KS-orbitals. The ESP-LR-TDDFT method was used to obtain the distribution of excitation energies for a set of organic dyes and semiconductor nanoparticles in aqueous environment at T=300K using a sample size of Nsample=104-105. We demonstrate the importance of sampling over sufficiently many molecular configurations by comparing the ensemble-averaged quantities at non-zero temperatures to those obtained from performing calculations on minimum energy structures at 0K.

Development of stochastic linear-response TDDFT method for obtaining distribution of excited state properties

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.