Career: neighborhood driving forces for fatigue crack propagation at the microstructure scale

The National Science Foundation Division of Materials Research awarded the University of Wisconsin System's Office of Sponsored Programs (doing business as University of Wisconsin–Milwaukee) $361,058 on August 15, 2026, under the Mathematical and Physical Sciences program (CFDA 47.049).

The award funds a CAREER project investigating neighborhood driving forces for fatigue crack propagation at the microstructural scale. The research combines high-energy X-ray methods to observe metal deformation and failure, computational modeling of material response to repeated stress at microscopic scales, and machine learning tools to identify complex patterns in material behavior. The work aims to advance understanding of crack initiation and propagation in structural metals used in transportation, manufacturing, energy, and health sectors.

The project includes workforce development and educational components: hands-on training for graduate and undergraduate researchers in experimental techniques, modeling, and data analysis; modernization of university courses; and outreach programs delivering "crushing device" demonstrations to middle and high school classrooms in Milwaukee and neighboring communities.

Performance period runs through July 31, 2031, with work conducted in Milwaukee, Wisconsin.

Abstract

Non-technical summary modern technologies rely on materials that must withstand repeated use without failing. Yet even today, scientists and engineers cannot fully predict when and where tiny cracks will form and grow inside structural materials. This NSF career project is uncovering how early damage begins at the microscopic level and is creating powerful new tools leading to safer, longer lasting, and more efficient metals. This research project is combining three advanced capabilities: high-energy x-RAY methods that let US ?see? Inside metals as they deform and fail, computer models that simulate how materials respond to repeated stress at the microscopic scale, and machine learning tools that can identify complex patterns within materials that go beyond what humans can logically predict. Altogether, these approaches are generating a clearer picture of how cracks initiate and spread, which is essential knowledge for building more reliable metals used in the transportation, manufacturing, energy, and health sectors. In addition to advancing scientific understanding, this project directly invests in the next generation of scientists and engineers through hands-on training in state-of-the-art experimental techniques, modeling, and data analysis. This project is also modernizing university courses and supporting outreach programs that bring engaging ?crushing device? Demonstrations to middle and high school classrooms across milwaukee and neighboring communities. These efforts are inspiring and engaging students in stem while seeding scientific curiosity in our future workforce. By creating new knowledge, training future innovators, and supporting educational outreach, this project is advancing scientific progress with respect to understanding how materials fail while contributing to the nation?s long-term prosperity and well-being. Technical summary despite decades of progress in fatigue modeling, existing approaches still struggle to accurately predict the nucleation sites, rates, and mechanistic origins of microscale damage that grow and lead to material failure. This NSF career project is advancing the fundamental understanding of early stage, microstructure sensitive fatigue damage by integrating high fidelity synchrotron diffraction measurements, substructure enhanced crystal plasticity finite element modeling, and machine learning frameworks. Synchrotron based high energy x RAY diffraction and complementary high resolution microscopy are being used to quantify intragranular lattice strains, dislocation substructures, grain boundary interactions, and local stress?strain evolution both preceding crack formation and during crack propagation. These data are informing and validating enhanced crystal plasticity models that incorporate microscale substructure formation, grain boundary interactions, and evolving anisotropic hardening laws. A key intellectual contribution is the development of reduced?order modeling strategies that retain microstructure sensitivity while significantly lowering computational cost. Machine learning frameworks are being used to predict fatigue relevant fields and crack initiation metrics directly from microstructural topology and loading conditions. Uncertainty quantification is being employed to assess the influence of measurement fidelity, model assumptions, and parameter selection on fatigue prediction certainty. Model transferability is also being demonstrated across multiple FCC alloy systems. The resulting framework is providing a computationally efficient pathway for microstructure-aware fatigue prediction and materials design. Beyond the technical contributions, this project provides substantial educational and societal impact. This research project is modernizing university courses in fatigue with interdisciplinary training spanning advanced characterization, computational mechanics, and data science. Moreover, this work is supporting outreach programs that bring engaging ?crushing device? Demonstrations to middle and high school classrooms across milwaukee and neighboring communities. Collectively, these efforts are strengthening the materials workforce pipeline by engaging students at various levels in engineering and materials science. 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.