ENG-AI: Harnessing Deformation in Nanocomposites via Artificial Intelligence-Assisted and Processing-Driven Far-from-Equilibrium Amorphous-Crystalline Interfaces

This award aims to revolutionize the design and manufacturing of advanced materials using artificial intelligence (AI), by improving the mechanical performance of nanocomposites (advanced materials made by combining different substances at extremely small scales). Research enabled by this award focuses on understanding and controlling a specific type of internal structure in these materials – called amorphous-crystalline interfaces – that can significantly enhance strength, durability, and reliability. If successful, the research findings could impact a wide range of critical applications in the areas of energy, defense, transportation, and others. By applying AI and advanced manufacturing techniques, the award seeks to uncover how processing methods can be used to tailor the structure and behavior of these interfaces for optimal performance. This award aims to develop a physics-based framework for the tunability of metastable amorphous-crystalline interfaces (ACIs) in nanocomposites through physical vapor deposition (PVD) processing. Research tasks focus on investigating how PVD parameters – such as chemical composition, deposition rate, temperature, and incident velocity – affect the local structural and chemical environments at ACIs, which in turn control deformation mechanisms like plasticity and shear banding. Self-propelling energy landscape sampling algorithms are employed to explore atomic-scale rearrangements without prior assumptions, combined with transition state theory to quantify the kinetics of transitions among various metastable micro-states. Machine learning models and Bayesian optimization will guide intelligent data acquisition and accelerate exploration of complex phase spaces. These computational approaches will be integrated with precision magnetron sputtering experiments, high-resolution electron microscopy, and nanomechanical testing to validate predictions. The resulting predictive, testable processing-structure-property loop could enable the design of high-performance, ACI-rich nanocomposites for advanced manufacturing applications. 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.