Grain boundaries (GBs) are ubiquitous defects that can have profound effects on the mechanical and electronic properties of most polycrystalline materials. Although there have been significant improvements in experimental imaging and image analysis, it is still exceedingly difficult to unambiguously interpret the atomic structures within a GB region without well-established theoretical models for the interfaces. This challenge calls for the development of computational modeling approaches that can describe the GB region with high fidelity across the full five-parameter GB crystallographic space. This project will develop a multiscale approach to study GB structure, and apply the developed methodology to study the GB region and its properties in the methylammonium lead triiodie perovskites (MAPbI3) that are of high interest for photovoltaics and optical applications. Here the project aims to develop and conduct a massively parallel, multiscale computational approach based on density functional theory and reactive force fields (RFFs) to determine the structure and properties of GBS.

The research team will develop and use a genetic code to determine GB structure, investigate the properties of GB regions for defect segregation and their barriers at T=0K and room temperature, and develop a RFF for multicrystalline MAPbI3.