Computational Discovery of Organic LED Materials
Automation, software and machine learning are enabling a data-driven revolution in areas such as self-driving cars, logistics, manufacturing and finance. In this chapter, we describe how these tools are being combined for computer-driven discovery of thermally activated delayed fluorescence materials. We analyze the increasingly automated deployment of robust and accurate computer simulations to assess candidate molecules virtually and identify leads for experimental characterization. Recent advances in machine learning techniques to accelerate the screening process and to increase its accuracy are also described. The role of user-experience and custom experiment–theory interaction tools are described. Finally, we report how these computer-based efforts have resulted in novel high organic light-emitting diode materials.