Exploring Synthesis–Structure Relationships in Epitaxially–Grown Semiconductors with Quantum and Classical Learning Algorithms

Substantial interest exists in the application of machine learning technologies to rapidly identify optimal materials designs and synthesis conditions. In this work, data describing hundreds of plasma–assisted molecular beam epitaxy growth experiments each of ZnO as well as various nitride semiconductors have been organized into separate, composition–specific data sets. For each growth record, the complete set of experiment design parameters are associated with binary measures of both crystallinity and surface morphology as determined by in–situ reflection high–energy electron diffraction. A continuous Brag–Williams measure of lattice disorder (S²) is included as an additional figure of merit for investigation. Supervised learning algorithms, including those written to incorporate quantum computation, are trained on the data sets to study which growth parameters are most statistically important for influencing each structural property. The probabilities of obtaining monocrystalline and atomically–flat thin film crystals are forecasted for each composition across processing spaces of the two most important synthesis parameters. When predicting the presence of grain boundary defects in epitaxially–grown GaN thin film crystals, the quantum machine learning algorithms demonstrate improved accuracy over the conventional models. The machine learning predictions of the growth conditions agree with published values for obtaining single crystalline and flat semiconductor thin films.