Branco Weiss Fellow Since
2019
Research Category
Computational Physical Organic Chemistry, Computer Science
Research Location
Schulich Faculty of Chemistry, Technion, Haifa, Israel
Background
Human quality of life is defined, to a large extent, by the type of compounds and materials that can be made to fulfill various needs: drugs, composites, fuels, adhesives, packing and insulation materials, energy harvesting compounds, etc. To optimize these functions and make them more suitable for sustainability, better compounds are required. But owing to the limited effectiveness of the strategies currently employed in molecular discovery, the development of new compounds is an arduous and expensive task. Inverse design — the process of constructing a target molecule to meet the desired property or function — has the potential to radically change the way new molecules are discovered. To locate a molecule with specified properties, it’s necessary to be able to uniquely map structural features to chemical behaviors. The main challenge therefore lies in the fact that our understanding of the very complex relationship between structure and function is limited.
Details of Research
To address these challenges, Dr. Gershoni-Poranne will combine chemistry and computer science techniques to investigate the relationship between the structure of polycyclic aromatic hydrocarbons and their electronic behavior. Polycyclic aromatic hydrocarbons are pervasive in chemistry and materials science, and are especially important in the field of organic electronics. Dr. Gershoni-Poranne will use high-throughput quantum chemical calculations to construct a database which will supply the data needed for the application of machine learning algorithms and the training of deep generative models. The overarching goal is to design optimal candidates for various organic electronic-based uses — including photovoltaics, field-effect transistors, and light-emitting diodes — as a demonstration of the utility of deep learning within the realm of chemistry, and for inverse design in particular. This will enable resource-efficient molecular discovery, paving the way to more effective and environmentally responsible molecules and materials in the future.