Organic semiconductors (OSs) have attracted world-wide attention in the past few decades, owing to their unique properties, which lead to innovative advancements in technologies like solar cells, luminescent solar concentrators and organic light emitting diodes. Photoluminescence quantum yields (PLQY) of OSs have been proven to be strongly correlated with the performance of these energy devices.
However, the development of new materials with high PLQY has been inefficient and demanding, because of the complicated synthesis of these materials and experimental results that fall short of expectations. The recent development of Machine learning (ML) may significantly optimize this R&D process. ML has been proven successful in predicting properties of systems without knowledge of the underlying physical principles. Currently I have been building a scraper to gather data regarding chemical structures and their PLQYs, which I will use to develop a deep learning algorithm predicting structure-property relationships of OSs in the future.
Advisor: Christine Luscombe — Materials Science & Engineering