Lab scientist wins DOE Howes Scholar award for work in computational science

Lawrence Livermore National Laboratory (LLNL) postdoc Kyle Bushick has garnered the 2024 Frederick A. Howes Scholar award for technical excellence, leadership and character in his innovative use of atomistic simulations to understand quantum processes and as a champion of inclusivity and community building.

“Learning I was selected as a Howes Scholar was an enormous honor. The DOE Computational Science Graduate Fellowship (CSGF) community has played an instrumental role in my scientific and personal journey throughout – and now beyond – my graduate studies, through the relationships with peers and alumni across scientific disciplines and exposure to the national lab system,” said Bushick, who works in LLNL’s Quantum Simulations Group.  “To receive such a recognition from this community, not only for my technical contributions but more importantly for my efforts to improve the experiences of graduate students at Michigan and share the wonder of looking at the world through a materials science lens with the broader public, is incredibly meaningful.”

The award was established in remembrance of the late Frederick Anthony Howes who managed the Department of Energy's Applied Mathematical Sciences Program during the 1990s. Howes was highly respected and admired for his energy, dedication and personal integrity. To honor Howes’ memory and his commitment to the DOE CGSF program, one or two alumni are chosen each calendar year as a Howes Scholar. This award is bestowed upon a recent graduate of the DOE CSGF program who has shown outstanding leadership, character and technical achievement in the field of computational science.

“This recognition reinforces my drive to apply an open and collaborative spirit to the parallel pursuits of scientific research and working to improve the communities around me,” he said.

At the Lab, Bushick works on a Laboratory Directed Research and Development Strategic Initiative project focusing on the construction, training, assessment and application of interatomic potentials. His efforts are centered around building the computational infrastructure to unify and automate the different facets of this work, helping researchers employ multiple independent methods and codes using a shared interface.

“I’m also interested in developing methods for quantifying the potential energy surfaces of machine-learned interatomic potentials to be able to understand their applicability to different physical systems and conditions,” he said. “Through this work, I hope to accelerate and improve the quality of interatomic potential deployment to answer fundamental questions about material properties.”

In the future, Bushick plans to continue building and developing computational tools to both accelerate and scale-up simulations while enabling scientists to answer new scientific questions.

“I’m particularly interested in understanding the properties of materials relevant for energy applications, from storage and generation to more efficient use in functional applications,” he said. “By creating and refining tools to computationally study these materials at the atomic level, I can help provide mechanistic insights into their behavior, which in turn allow us to engineer materials with superior performance.”

Bushick graduated with his Ph.D. in materials science and engineering and scientific computing from the University of Michigan.

He will detail his research as a featured presenter during the fellowship’s annual program review on July 15 in Washington, D.C.