Skip to content Skip to navigation

Using Math to Search for a 'Needle in a Haystack' to Make Better Solar Cells

Accelerated searches, made possible by machine learning techniques, are of growing interest in materials discovery. A suitable case involves the solution processing of components that ultimately form thin films of solar cell materials known as Hybrid Organic-Inorganic Perovskites (HOIPs). The number of molecular species that combine in solution to form these films constitutes an overwhelmingly large “compositional" space (at times, exceeding 500,000 possible combinations). Selecting a HOIP with desirable characteristics involves choosing different cations, halides, and solvent blends from a diverse palette of options. An unguided search by experimental investigations or molecular simulations is prohibitively expensive. In this work, we propose a novel Bayesian optimization method that overcomes challenges where data is scarce, and in which the search space is given by binary variables indicating whether a constituent is present or not. We demonstrate that the proposed approach identifies HOIPs with the targeted maximum intermolecular binding energy between HOIP salt and solvent at considerably lower cost than previous state-of-the-art Bayesian optimization methodology and at a fraction of the time (less than 10 per cent) needed to complete an exhaustive search. We find an optimal composition within 15 iterations (plus or minus 10) in a HOIP compositional space containing 72 combinations, and within 29 iterations (plus or minus 11) when considering mixed halides (240 combinations). Exhaustive quantum mechanical simulations of all possible combinations were used to validate the optimal prediction from a Bayesian optimization approach. These proofs of concept demonstrate the undeniable promise of the novel Bayesian optimization methodology, presented here, to the field of materials discovery.
H. Herbol et al., Nature Comp. Materials, under review (2018)

CCMR researchers have used mathematical methods, typically used in business forecasting, to suggest which combination of components will make the best solar cell materials in a “perovskite” arrangement. These materials are made in solution, essentially in a beaker, at room temperature. This makes them far more energy-conservative than traditional silicon solar cells. But researchers are spoiled for choice in terms of components that could be put into the “soup;” too many to make in the lab. Instead, theorists have developed software that predicts the ingredients that lead to the best performing material. They typically find the optimal solution after testing only 10-15% of all the options. Researchers are now tackling even more complex solutions where the optimal solution cannot be known in advance.