CAREER: First-Principles Thermodynamic Models for Defect Phase Diagrams

Nontechnical summary This CAREER award supports theoretical and computational research and education on the role of defects in materials and facilitating their use to control properties in materials design. Solid materials are often made of atoms arranged in regular, repeating patterns known as crystal structures. These atomic arrangements play a central role in determining how materials behave, including how they conduct electricity, withstand heat, emit light, store energy, or respond to their environment. Although imperfections in these structures are often thought of as flaws, atomic-scale imperfections can be beneficial when they are understood and controlled. For example, missing atoms, extra atoms, or atoms in the “wrong” location can be used to tune the properties of a material. This concept is known as defect engineering: the deliberate control of atomic-scale imperfections to create or improve useful material properties. Defect engineering is central to many technologies, including microprocessors, memory devices, light-emitting diodes, solar cells, sensors, and energy materials. However, predicting which defects form, how many are present, and how they respond to temperature, chemical environment, composition, and other conditions remains a major challenge. This CAREER project develops a new framework for predicting and controlling defect populations in solid materials. By advancing the ability to anticipate how atomic-scale defects form and evolve under different conditions, the project helps to accelerate the design of materials with targeted properties and functions. The resulting advances support broader progress in defect engineering across a wide range of materials, including semiconductors, insulators, and complex oxides. This capability can open new pathways for designing materials used in electronics, energy conversion, sensing, and other emerging technologies. The project also includes an integrated education and outreach effort designed to make defect science more accessible to learners at multiple stages. A public educational website brings together learning modules, data sets, visual materials, and teaching resources that explain how atomic-scale physics and chemistry influence the properties of real materials. These resources support students from middle and high school through undergraduate levels through graduate students and postdoctoral researchers. By connecting fundamental concepts to practical technologies, the project increases participation in materials science and helps train the next generation of scientists and engineers in predictive materials design. Technical Summary This CAREER award supports theoretical and computational research and education on the role of defects in materials and facilitating their use to control properties in materials design. Understanding and predicting defect thermodynamics in materials is essential for determining their functional properties and underlying phenomena. However, first-principles thermodynamic models for defect phases are lacking due to the disconnect between fundamental defect chemistry/physics and macroscopic thermodynamic descriptions. Ab initio and atomistic models accurately determine defect energetics at the fundamental electronic and atomic levels but remain disconnected from macroscopic thermodynamic models for defect phases, which are mainly empirical and adjust their parameters to fit experimental and computational phase boundary and thermochemical data. This project aims to bridge this gap by developing a physics-based, first-principles thermodynamic framework for defect phases. This framework, which is to be demonstrated on target materials with various defect types and chemistries, overcomes current model limitations through parameterization derived from fundamental atomic and electronic interactions, without reliance on empirical defect data. It extends the CALPHAD (CALculation of PHAse Diagram) methodology to defect phases and nonstoichiometric regimes by accounting for defect-induced structural variations and long-range interactions. Included in this project is the development of a new supercell structure to enable efficient first-principles modeling of defect-induced structural disorder, as well as bulk amorphous and liquid phases. The education plan involves developing learning resources integrated with the research activities, including tutorials that connect defect-level physics and chemistry to materials thermodynamic stability, automated workflows for literature-based defect data mining, and an artificial intelligence educator focused on materials defects. Open-access repositories and a custom educational website make developed resources broadly available. 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.