Excited-State Properties

Accurate prediction of excited-state properties is important for the discovery of organic semiconductors for applications such as light-emitting and energy-conversion technologies. Commonly used TD-DFT methods allow large numbers of molecules to be screened efficiently, but can become less reliable for molecules with more complex electronic structures. In this project, TD-DFT calculations were used to identify molecules most likely to be poorly represented by conventional methods, which were then recalculated using higher-level multireference CASSCF and NEVPT2 approaches. This resulted in a consistently generated database of more than 1500 organic semiconductors containing low-lying singlet and triplet excitation energies and oscillator strengths. Comparison between the different levels of theory showed that some excited-state properties are considerably more sensitive to the computational method than others, particularly singlet excitation energies, oscillator strengths and energy differences between excited states. These differences were also found to depend on molecular structure, with particular chemical features showing stronger method dependence. Importantly, the choice of computational method affected the identification of molecules with technologically interesting behaviour, including thermally activated delayed fluorescence, inverted singlet-triplet gaps and anti-Kasha behaviour. In several cases, the higher-level calculations identified promising molecules and chemical structures that were not recovered using the original TD-DFT screening. The resulting database is now also being used to develop machine-learning approaches that aim to retain the efficiency required for high-throughput screening while improving the accuracy of excited-state predictions across large chemical spaces. Overall, this work aims to make accurate excited-state screening more practical across large chemical spaces and support the discovery of organic semiconductor materials with unusual and technologically useful electronic properties.

For more information about the project contact Dr Tahereh Nematiaram (tahereh.nematiaram@strath.ac.uk) Chancellor’s Fellow in Materials and Computational Chemistry at the Department of Pure and Applied Chemistry at the University of Strathclyde.

For a list of the research areas in which ARCHIE-WeSt users are active please click here.