An assistant professor in the Department of Chemistry, part of the Virginia Tech College of Science, wants to develop new computational modeling programs to help scientists study so-called “strongly correlated” molecules that have thus far been difficult to pin down.

Nick Mayhall, an assistant professor of chemistry in the Virginia Tech College of Science, will use his five-year National Science Foundation CAREER Award of more than $575,000 to better understand strongly-correlated molecules by building computer-based methods that can create new algorithms that can then be run on supercomputers.

“Computational models of molecules provide much more detail than lab bench experiments alone can achieve,” Mayhall said. “This means that if a computer model is accurate enough, we can simply run a computational simulation to make predictions about what will happen in nature, reducing cost of research, and providing deeper or at least a complementary insight compared to the laboratory experiment.”

The proverbial hard nut to crack: these strongly correlated molecules have remained difficult to model on a computer because they require simulations which begin with multiple states simultaneously, which precludes the use of most conventional computational algorithms, Mayhall said. Molecules which Mayhall hopes to better describe using models developed in this project are found in molecular magnets, organic photovoltaic materials, and everyday photosynthesis – or the process plants use sunlight to “eat.”

“Because of their small mass and size, molecules are governed by quantum mechanics, and all aspects of chemistry can thus be predicted by ‘simply’ solving the central equation of quantum mechanics,” Mayhall added. “Unfortunately, this equation is incredibly difficult -- or impossible -- to solve exactly on today’s computational resources, and it becomes necessary to make approximations to the solutions.”

Mayhall added that the central goal of the NSF-funded project is to exploit certain aspects of a molecule’s structure or approximate the point of separation to enable more accurate approximations for quantum chemistry simulations. 

“I decided to pursue this particular project because of how different it is from existing approaches,” he said. “The underlying physical concepts in this project connect old foundational ideas in our field, many-body expansions, to more current ideas such as tensor decompositions.”

He said his team of graduate students already has published one paper in the Journal of Chemical Theory and Computation, demonstrating that the main ideas in the project “hold promise.” Future steps include focusing on the development of the computer software and related theory so that both can be applied to industry problems across many scientific fields such as chemistry, biochemistry, and materials.

Mayhall earned his bachelor’s degree in chemistry from the University of Southern Indiana in 2006 and a Ph.D. in computational chemistry from Indiana University in 2011. He completed post-doctoral research work at the University of California Berkeley from 2011 to 2015, when he joined Virginia Tech.

The CAREER grant is the National Science Foundation’s most prestigious award, given to creative junior faculty considered likely to become academic leaders of the futureMayhall is one of three College of Science faculty to receive a CAREER Award in 2018, with the other recipients being Guoliang “Greg” Liu, also in chemistry, and Leah Johnson, an assistant professor in the Department of Statistics.

Related story

Chemistry researcher plans to layer nanofibers over glass to ease energy consumption in buildings, vehicles