Computational screening of small-molecule organic semiconductors

A computational screening workflow for small-molecule organic semiconductors which starts from a defined search space of molecules and ends with a set of proposed molecules was made. The MolBuilder program was developed to use an evolutionary algorithm to optimise the molecular structures of a population of molecules constrained to a search space defined by MolBuilder using a set of molecular fragments. We successfully applied the MolBuilder program to a search space of nitrogen substituted polycyclic aromatic hydrocarbons, and indenofluorenedione derivatives to obtain high-performance n-type organic semiconductors by using a fitness function that optimises for low reorganisation energies and specific electron affinities. In both cases, the computational screening workflow was made to take the best molecules proposed by the evolutionary algorithm through a crystal-structure prediction and electron mobility calculation stage for further evaluations. Based on the results of the computational screening workflows: for the search space of nitrogen substituted polycyclic aromatic hydrocarbons, suggested molecules were found to be competitive to a set of azapentacenes designed from computational considerations; and for the search space of indenofluorenedione derivatives, suggested molecules were found to be superior to a set of previously synthesised indenofluorenedione molecules used in organic electronic applications