MolDesign · Molecule design for next generation solar cells using machine learning approaches trained on large scale screening databases

Research in organic electronics has already generated important applications, such organic light emitting diodes (OLEDs) in mobile displays that are already indispensable in our everyday life. Other applications, such as large scale displays and lighting, lightweight and flexible organic photovoltaics, carbon-based electronic paper, organic sensors and RFID tags, are under intense investigation.The multifunctional character of these applications poses enormous challenges for the development of novel materials, which are hard to meet with the present day trial-and-error strategies. The MolDesign project thus aims at computational material design by combining accurate but involved materials simulation methods with inexpensive novel machine learning methods to enable large scale guided materials screening. These methods will be used to improve small-molecule organic semiconductors, which are used as absorber materials in vapor-deposited organic solar cells. While some materials of this class are almost at the photovoltaics market, there is much room for improvement regarding properties such as charge carrier mobility as well as the integration of organic material into completely new applications, such as hole transport materials in highly promising perovskite solar cells.