Collaborating National Laboratories
Identity, develop, verify and validate machine learning and artificial intelligence tools and algorithms that may be used for proliferation data collection and decision making to analyze risks of the nuclear fuel cycle and to identify key sources of information.
Develop new, improve existing, and combine methods for characterization of materials in order to develop enabling technologies for nuclear material identification specific to fuel cycle stages and processes. These efforts are referred to in this proposal as "science of signatures." Specifically, investigators will advance and integrate radiochemical and mass and optical spectroscopy methods in nuclear forensics.
Conduct fundamental, materials science investigations of fissionable fuel forms, utilizing depleted uranium bearing compounds, and their non-radiological analogues, referred to here as surrogates. These materials will be processed using a variety of techniques and systematically hydrated and/or contaminated to represent materials of interest at various stages of the nuclear fuel cycle.