Organic light-emitting diodes (OLED) are and have been the future of display technology for a minute. Looking back, display technology has moved from cathode-ray tube displays (CRTs) to liquid crystal displays (LCDs). Whereas CRT displays were clunky and had quite high power- consumption, LCDs were thinner, lighter and consumed less energy. This technological shift has made it possible to create smaller and more portable screens, aiding in the development of personal electronics. Currently, however, LCDs place at the top of the display hierarchy is being challenged by OLED displays, providing higher pixel density and overall higher performance. OLED displays consist of thin layers of organic semiconductors, and are instrumental in the development of folding displays; small displays for virtual reality and augmented reality applications; as well as development of displays that are energy-efficient.

In the creation of OLED displays, the organic semiconducting material is vaporized and adhered to a thin film through vapor deposition techniques. One way of aiding in the creation of organic electroluminescent (OEL) materials and OLEDs is through in silico analysis of sublimation temperatures through machine learning. This master’s thesis inhabits that space, aiming to create a deeper understanding of the OEL materials through sublimation temperature prediction using ensemble learning (light gradient-boosting machine) and deep learning (convolutional neural network) methods.

Through analysis of experimental OEL data, it is found that the sublimation temperatures of OLED materials can be predicted with machine learning regression using molecular descriptors, with an R2 score of ~0.86, Mean Absolute Error of ~13°C, Mean Absolute Percentage Error of ~3.1%, and Normalized Mean Absolute Error of ~0.56.