b'Passive strain measurementsMultiple length scale mechanical testing, computational modeling, for experiments inand machining learning algorithms correlate microscopic radiation environments features to mechanical failure modes within materials.U nderstanding the mechanical response of materials under radiation is essential to qualifying any material for use within a nuclear reactor. In a radiation environment, such as in a reactor core, the radiation damage degrades many instruments and thus it limits the instrumentation options. The goal of this project was to attempt to take advantage of anisotropic material responses to radiation such that a directional response could be used as a means PROJECT NUMBER:for instrumentation. Scanning electron microscopes used digital image correlation 21A1050-060FP for a micron scale evaluation to test zirconium single crystals. Transmission electron TOTAL APPROVED AMOUNT:microscope mechanical testing performed micropillar compression tests on the $1,288,000 over 3 years same zirconium crystals for nanoscale behavior. The mechanical responses from both microscale and nanoscale mechanical testing were within ranges reported in PRINCIPAL INVESTIGATOR:literature for bulk materials. Single crystal specimens were also irradiated at the Boone Beausoleil Massachusetts Institute of Technology Reactor to doses in the range of 131024CO-INVESTIGATORS: neutrons per cm. These samples were prepared for mechanical testing in both a Nedim Cinbiz, INL scanning electron microscope and a transmission electron microscope but were not Stephanie Pitts, INL completed due to problems with the mechanical stages in both instruments. Yachun Wang, INL For the single crystal materials to be useful as a probe in the proposed Djamel Kaoumi, North Carolinainstrumentation a robust plasticity model also needed to be developed. A crystal State University plasticity model was implemented within the MOOSE that was used to represent Ju Li, Massachusetts Institute of Technology stress responses of hexagonal closed packed materials such as zirconium. This model was used to replicate the low length scale mechanical testing of the scanning electron microscope and a transmission electron microscope tests. Additionally, stochastic tools within MOOSE were used to assess geometric sensitivities in mechanical tests. These results were shown to be highly beneficial in determining the efficacy of mechanical test results in conventional and low length scale experiments. Finally, mechanical test results were used to develop a machine learning algorithm to determine the constitutive relationship between mechanical stress and strain within materials. This improves the ability to make predictive assessments of materials based upon simple parameters. Although the final objective of developing a passive in-pile strain sensor was not achieved, this work made significant contributions to the understanding of irradiation effects on mechanical properties.34'