Machine Learning Based Methodology to Predict Point Defect Energies in Multi-Principal Element Alloys

Multi-principal element alloys (MPEAs) are a new class of alloys that consist of many principal elements randomly distributed on a crystal lattice. The random presence of many elements lends large variations in the point defect formation and migration energies even within a given alloy composition. Compounded by the fact that there could be exponentially large number of MPEA compositions, there is a major computational challenge to capture complete point-defect energy phase-space in MPEAs. In this work, we present a machine learning based framework in which the point defect energies in MPEAs are predicted from a database of their constituent binary alloys. We demonstrate predictions of vacancy migration and formation energies in face centered cubic ternary, quaternary and quinary alloys in Ni-Fe-Cr-Co-Cu system. A key benefit of building this framework based on the database of binary alloys is that it enables defect-energy predictions in alloy compositions that may be unearthed in future. Furthermore, the methodology enables identifying the impact of a given alloying element on the defect energies thereby enabling design of alloys with tailored defect properties.

A straightforward one-pot approach to two new defect energy levels in ZnS

ZnS with double defect energy levels shows an amazing visible light photoactivity.

Extracting bulk defect parameters in silicon wafers using machine learning models

The performance of high-efficiency silicon solar cells is limited by the presence of bulk defects. Identification of these defects has the potential to improve cell performance and reliability. The impact of bulk defects on minority carrier lifetime is commonly measured using temperature- and injection-dependent lifetime spectroscopy and the defect parameters, such as its energy level and capture cross-section ratio, are usually extracted by fitting the Shockley-Read-Hall equation. We propose an alternative extraction approach by using machine learning trained on more than a million simulated lifetime curves, achieving coefficient of determinations between the true and predicted values of the defect parameters above 99%. In particular, random forest regressors, show that defect energy levels can be predicted with a high precision of ±0.02 eV, 87% of the time. The traditional approach of fitting to the Shockley-Read-Hall equation usually yields two sets of defect parameters, one in each half bandgap. The machine learning model is trained to predict the half bandgap location of the energy level, and successfully overcome the traditional approach’s limitation. The proposed approach is validated using experimental measurements, where the machine learning predicts defect energy level and capture cross-section ratio within the uncertainty range of the traditional fitting method. The successful application of machine learning in the context of bulk defect parameter extraction paves the way to more complex data-driven physical models which have the potential to overcome the limitation of traditional approaches and can be applied to other materials such as perovskite and thin film.

Electron-Beam-Induced Current and Cathodoluminescence Study of Dislocations in SrTiO3

Electron-beam-induced current (EBIC) and cathodoluminescence (CL) have been applied to investigate the electrical and optical behaviors of dislocations in SrTiO3. The electrical recombination activity and defect energy levels of dislocations have been deduced from the temperature-dependent EBIC measurement. Dislocations contributed to resistive switching were clarified by bias-dependent EBIC. The distribution of oxygen vacancies around dislocations has been obtained by CL mapping. The correlation between switching, dislocation and oxygen vacancies was discussed.