Asst. Prof. Stephen T. Lam, Collaborators to Use Particle Accelerator, Machine Learning and Laser Spectroscopy to Gain New Insights

Stephen Lam at his desk
Chemical Engineering Asst. Prof. Stephen T. Lam’s research interests include nuclear materials, advanced reactor technologies, materials chemistry, simulation, multi-scale modeling, machine learning and energy storage and conversion.

By Edwin L. Aguirre

Over the past decade, nuclear power has been making a comeback as an important source of low-cost, carbon-free energy. This renaissance is fueled by the world’s ever-growing energy needs and by efforts to cut down on greenhouse gas emissions. 
At the same time, the safety and reliability of nuclear reactors have been enhanced substantially, along with the development of advanced reactor technologies. Among them are molten salt reactor systems, which use mainly chloride-based or fluoride-based liquid salt mixtures that can operate at high temperatures. The molten salts are used as the primary coolant and/or fuel for the reactors.
Now, Chemical Engineering Asst. Prof. Stephen T. Lam and researchers from MIT and Stony Brook University are working to improve our understanding of molten salt chemistry and the safety and efficiency of these types of reactors. The team recently won a three-year, $1 million U.S. Department of Energy (DOE) grant to fund their research.
“Our work will reduce the technical risk associated with molten salt reactors,” Lam says. “Furthermore, it will enable fuel recycling technologies so we can maximize our utilization of nuclear fuel.”
As molten salt reactors operate, nuclear reactions create new chemical entities from fission, the splitting of heavy atoms such as uranium or plutonium into lighter elements that release tremendous amounts of energy in the process.
Molten salt reactor experiment Image by ORNL
A top view of the Molten Salt Reactor Experiment that was developed at the Oak Ridge National Laboratory in Tennessee.
“A key technical challenge in developing molten salt technologies is understanding how chemical properties evolve with temperature and composition, and how to separate nuclear fuel, called actinides, from fission byproducts in order to recycle the unspent fuel,” says Lam.
The project, which is led by researchers at MIT, with Lam as the principal investigator for UML, is supported by the Oak Ridge, Los Alamos and Brookhaven national laboratories. The grant is part of the investment by the DOE of more than $59 million to support 25 American colleges and universities in advancing research and development in nuclear science and technology in the country. 
Lam and his co-researchers will combine state-of-the-art methods and capabilities – from particle accelerator experiments at Brookhaven to machine learning-enhanced simulations and laser-induced breakdown spectroscopy in the lab – to investigate the structure and chemical properties of actinides under reactor operating conditions.
“We hope to obtain critical data and gain new insights into the chemical redox behavior of actinides that will enable their separation and recovery and allow us to predict their phase behavior,” Lam notes. “This knowledge is essential for the safe and economical operation of molten salt reactors and to augment their fuel cycles.”
According to Lam, this is the first time this type of research is being conducted in the country using a unique combination of experimental, computational and machine learning methods.
“It is challenging to predict and obtain data on how low concentration of impurities from fission products would impact the chemistry of molten salt due to limitations in lab measurements and simulation methods,” he explains. “Furthermore, in practice, it is quite challenging to work with actinide materials in the first place.”
Although the team’s research focuses more on the fundamental aspect, in the long term, it could also help with the commercialization of molten salt reactors, which could potentially be significantly safer, more economical and more reliable than conventional reactor systems, says Lam.
“These reactors could help Massachusetts and the rest of the country achieve their goal of cutting down carbon emissions,” he says.
Lam plans to hire a Ph.D. student as well as undergraduate research assistants to work on the project.
“This will provide UML students opportunities to engage in collaborative research with other universities and national laboratories in the areas of materials simulation, characterization and machine learning,” he says. “The job will also include travels to the collaborators’ sites as well as attending national conferences.”