UNIVERSITY PARK, Pa. — China recently limited the export of gallium nitride, a type of semiconductor used to manufacture a variety of consumer power electronics, such as cellphones and computers, as well as medical devices, cars, wind turbines, solar farms, LED lightbulbs and more.  

China produces nearly 98% of the world’s supply of semiconductor gallium, according to the U.S. Geological Survey. In response to the country’s move to limit exports on the compound, the U.S. Department of Energy (DOE) put out a call for proposals to U.S. scientists to engineer a solution to replace gallium nitride-based semiconductors and awarded four teams $1 million for a one-year research sprint.  

Patrick Lenahan, distinguished professor of engineering science and mechanics at Penn State, will investigate the possibility of replacing gallium nitride-based devices with boron nitride. Co-principal investigators from Ohio State University, the University of Iowa and QuantCAD — a quantum technology startup with locations in Iowa City, Iowa, and Chicago — will join Lenahan in the investigation.  

“We believe boron is a potentially good alternative to gallium in some technologically critical gallium nitride-based applications, but we don’t know exactly how this may work out,” Lenahan said. “Our team, which involves theorists and experimentalists, will work to determine how boron nitride works, physically, in systems of real interest, and what could be the potential drawbacks in using it in place of gallium nitride.” 

Consisting of 50% gallium and 50% nitrogen, gallium nitride has the advantage of being a wide bandgap semiconductor, Lenahan said, meaning it can withstand higher electric fields and sustain higher voltages and temperatures than lower bandgap semiconductors, like silicon. Boron nitride has comparable, but far less understood, physical characteristics to gallium nitride, and may have the potential to perform similarly in some critical applications.

“Boron does not have the same importation restrictions as gallium, so it could be a long-term replacement in multiple applications,” Lenahan said.  

Lenahan and the students in his lab, most of them pursuing master’s or doctoral degrees in engineering science, will use techniques — such as electrically detected magnetic resonance, near zero field magnetoresistance and conventional electron paramagnetic resonance in conjunction with a variety of electrical measurements — to assess semiconductor devices manufactured with boron nitride. All three techniques provide information about the ways electrons interact with imperfections in boron nitrides by testing a property of electrons called spin. Such understanding will be key in evaluating boron nitride’s effectiveness as a semiconductor, Lenahan said. 

“This project is very much a team effort, involving device fabrication, a variety of measurements and theoretical analysis,” Lenahan said.  

Ezekiel Johnston-Halperin, professor of physics and co-director of the Center for Quantum Information Science and Engineering at Ohio State, will fabricate devices built up from two-dimensional, or one atom-thick, layers of boron nitride and explore electronic transport through a process called electron tunneling.   

“In electron tunneling, an electron can ‘tunnel’ past an extremely thin insulating barrier, a process that gets exponentially weaker as the barrier gets thicker,” Johnston-Halperin said. “My lab has piloted a process for building very thin structures that take advantage of this quantum property of electron tunneling to sensitively probe the properties of atomic impurities in boron nitride. By picking a thickness for boron nitride just beyond the range of this tunneling process, the current through the device becomes extremely sensitive to the presence of crystalline defects.”  

Lenahan and Johnston-Halperin, along with Sina Soleimanikahnoj from QuantCAD and Durga Paudyal, a researcher from the University of Iowa, will work together to obtain a unique “fingerprint” of impurities in boron nitride in order to optimize the design of the compound’s chemical structures.  

“As a potential alternative material to gallium nitride for power electronics, boron nitride hits a lot of the right spots, such as its large electronic bandgap and impeccable thermal conductivity,” Soleimanikahnoj said. “To determine if this potential can be put into practice, QuantCAD will perform detailed microscopic modeling of boron nitride-based power electronic devices to benchmark them against their gallium nitride counterparts. We will then integrate the microscopic modeling into a macroscopic electrical device simulation to design optimal device structures.”  

With the tight timeline on this one-year research sprint, the collaborators will work in tandem to establish whether a realistic path forward exists for boron nitride and to broadly determine how scientists and technologists could most effectively proceed on this path.