Excellence in research: backreaction of electron excitation/DE-excitation on perovskite lattices, light-induced defect creation/annihilation and charge trap states

The National Science Foundation Office of Integrative Activities awarded Jackson State University $500,000 on September 1, 2026, under the Integrative Activities program (CFDA 47.083) for computational research into metal halide perovskites and machine learning applications in optoelectronic materials design.

The project supports research investigating how light illumination affects the performance of metal halide perovskites in solar cells and optoelectronic devices. The work integrates light-induced trajectory surface hopping (LITSH) and time-dependent density functional theory methods to develop the L-T-A model, which will advance understanding of degradation mechanisms in these materials when exposed to light. Research outcomes aim to inform the design of next-generation perovskite materials with improved properties for energy generation and medical imaging applications.

The award strengthens research and teaching capabilities at Jackson State University, a historically black college and university in Jackson, Mississippi, and enhances connectivity among research efforts through artificial intelligence integration. The project includes hands-on student training in AI development and computational methods, with research modules designed to serve as training platforms for future use at other HBCUs. Student participation supports workforce development in artificial intelligence within Mississippi, positioning graduates for careers as AI engineers or STEM educators in middle and high schools.

The period of performance runs from September 1, 2026, through August 31, 2029, with performance at Jackson, Mississippi.

Abstract

Nontechnical summary this excellence in research award supports computational research into the properties of a class of optoelectronic materials known as metal hydride perovskites. IT also benefits students by enhancing research and teaching capabilities at jackson state university and improves connectivity among research efforts at JSU utilizing artificial intelligence (AI). The PI aims to apply machine learning combined with computational methods to investigate how illumination by light affects the performance of the optoelectronic materials in applications including solar cells and optoelectronic devices. The project exploits AI to help inform the design of next generation perovskite materials with better properties and performance to improve the quality of human life from energy generation to medical imaging. The research provides unique opportunities for students to participate in collaborative and interdisciplinary research, leading to a broader range of career development options. IT also provides hands-on student training in AI development. Student involvement is through research modules developed in this project which may serve as platforms for training more students in the future at other hbcus. The project helps develop the stem workforce in a state that is sorely underrepresented in the AI area, as these students will go on to teach stem courses in middle and high schools in mississippi or particularly become AI engineers. Technical summary this excellence in research award supports computational research to advance understanding of the mechanisms that degrade the optoelectronic properties of metal halide perovskites (mhps) when illuminated by light. A new model, the l-t-a model, is being created by integrating the light induced trajectory surface hopping (litsh) and time-dependent density functional theory plus GPW/GTH pseudopotentials (tddft/GPW/GTH) together with a deep artificial neural network (ANN). The resulting model can do a long-time nonadiabatic dynamics simulation for large systems. With the l-t-a model, the PI: 1) investigate the light-induced defect annihilation/creation process at low/high photon fluence in the MHP thin film; 2) shows that the excited electron produced by illumination turns an oxygen molecule into a superoxide which reduces the charge recombination time; the superoxide is then transformed into a peroxide that enhances optoelectronic properties but causes MHP degradation; 3) elucidates the impact of the light on water and MHP thin films so as to determine if low level humidity increases, but high level humidity decreases, the power conversion efficiency (PCE); and 4) characterizes how dopants in mhps under light soaking induces shallow trap states and stabilizes the MHP structure. Starting with the l-t-a model the PI converts a complex computational project to one that is down-sized into multiple modules, each of which can be used as a standalone research topic for students. The results of the project yield insights into MHP thin films and inform the synthesis of new perovskites that may lead to technological advances such as solar cells with high PCE and photodetectors with clear digital images. This project advances application of machine learning to enable nonadiabatic molecular dynamics simulation and stimulates new ideas regarding stem education of students through training on the application of AI to advance stem research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.