​Dr. Jeffery Aguiar applies machine learning to further the capability to predict and analyze combinatorial material science systems for nuclear applications. He is shortening the development cycle involved in discovering and qualifying new high-entropy alloy systems for molten-salt and fast-reactor applications. This early demonstration delivers on the mission to advance the economic competitiveness of nuclear energy and technology worldwide.

Dr. Aguiar's research enables machine learning to predict how a combination of materials will perform in an alloyed state.

​While neutron radiography using film is already practiced to examine nuclear fuels, digital imaging systems struggle to examine fuels because of the radiation that irradiated fuels emit. INL researcher Dr. Aaron Craft, in collaboration with Dr. Anton Tremsin from the University of California at Berkeley, successfully tested an advanced microchannel plate neutron-imaging system for its ability to examine irradiated nuclear fuel. Digital neutron radiographs of irradiated nuclear fuel were successfully acquired using a real-time digital imaging system for the very first time. These promising results could lead to routine use of modern digital neutron-imaging methods for examination of nuclear materials. 

The world’s first fully digital radiographs of highly irradiated nuclear fuel. The image above shows the top (left) and bottom (middle) caps of an irradiated fuel pin, as well as the top portion of the fueled region (right) that exhibited the highest measured dose rate of 550 R/hr.