How NREL’s Brooks Tellekamp Explores the Next Generation of Materials for More Efficient LEDs and Power Electronics
Brooks Tellekamp found his way to science through his high school's computing infrastructure. He helped the school's information technology expert, a retired Honeywell engineer who had worked on NASA contracts, to address computing infrastructure needs.
Those early experiences set Tellekamp on a clear path—he targeted engineering school as a result.
"I found myself in a cooperative work program at a zinc oxide semiconductor bulk growth startup," said Tellekamp, a materials science researcher at the National Renewable Energy Laboratory (NREL). "An employee at the company recommended that I take classes with Alan Doolittle, who would go on to suggest I pursue a graduate degree in his laboratory. Along the way, I never considered research as a possibility, but getting the opportunity to work in the Doolittle lab certainly changed that."
That unlikely path brought Tellekamp to NREL, and now he has been recognized by the Journal of Physics D: Applied Physics as an emerging leader in a special issue of the journal.
Synthesizing Key Materials
Tellekamp clearly recalls an early breakthrough in his career: In graduate school at the Georgia Institute of Technology, after years trying to synthesize a material whose synthesis secrets had been lost, he successfully synthesized lithium niobite (LiNbO2). The material is a memristor—a device used to mimic synapses for neuromorphic computing. This approach to computing mimics the human brain’s functioning.
Tellekamp then arrived at NREL as a postdoc to work on research in an area that was new to him: ternary nitrides. Tellekamp has carried his past work on neuromorphic computing into his work at NREL, reviewing chalcogenide materials for this application and demonstrating the memristive properties of another oxide material, NdNiO3. From there, he has focused on next-generation oxide and nitride semiconductors for power electronics applications.
In another breakthrough, Tellekamp and team were able to synthesize a gallium oxide (Ga2O3)-indium (III) oxide alloy, another semiconducting material with power electronics applications.
"We were having trouble getting the indium to go into the Ga2O3 crystal," Tellekamp said. "We typically use a method of polishing away some of the Ga2O3 crystal with gallium at high temperatures to clean the surface before growth, and after another unsuccessful growth (we can monitor in real time with electron diffraction), I asked Stephen Schaefer, a postdoc working on the project, to try to etch away the film using gallium and just start over again. This led to the development of a cyclical growth/etch method that we used to rapidly span the growth space, which increased our throughput by about six times and our substrate utilization by about 46 times. That was a fun ‘aha’ moment."
Tellekamp's work on ternary nitride semiconductors has brought him and colleagues good attention. Ternary nitrides have the potential to provide more energy-efficient light-emitting diodes (LEDs). His postdoctoral work focused on ternary nitrides—synthesizing them at high quality, growing them on different substrates to understand their optical properties, and demonstrating high-quality interfaces between ZnGeN2 and Gallium Nitride (GaN). All of this has led to the demonstration of growth of GaN and ZnGeN2 superlattices through molecular beam epitaxy—an important step toward more efficient LEDs. This work was featured as part of the emerging leader recognition. ZnGeN2 poses an alternative to GaN, a compound semiconductor in LEDs that are very efficient at making blue light but not so efficient for green and amber light needed for warm white indoor lighting. ZnGeN2 has a similar structure to GaN—making it highly compatible—and it may potentially enable more energy-efficient green and amber LEDs without the need for rare earth elements currently used to translate blue light into white light.
Semiconductors: The Next Generation
The next generation of ultrawide-bandgap semiconductors currently occupies Tellekamp's time at NREL. He is working on semiconductors with more applications across power electronics and extreme environment electronics. Ultrawide-bandgap materials have traditionally been insulators—sometimes not great ones—but Tellekamp noted that the ability to control conductivity in these compounds has turned out to be important for an electrified economy. "Electrification only makes sense if you can manage all the power requirements in an efficient and cost-effective way, and these ultrawide-bandgap materials enable this efficiency," he said.
"One way we are pursuing this right now is through the discovery and design of new ultrawide-bandgap materials, which some of my collaborators theoretically predicted and I am now trying to synthesize. Another way is through the development of manufacturing methods for vertical (rather than lateral) devices using a well-studied material, (Al,Ga)N (AlGaN), which is an alloy of gallium nitride (all of your LEDs are made of this) and aluminum nitride. I had a project funded internally a few years ago to figure out how to do this, and we came up with a way to deposit tantalum carbide thin films in a cost-effective way but make them high enough quality to grow AlGaN on top of them."
Tellekamp and fellow researchers are taking this research into a recently funded Energy Frontier Research Center program led by NREL’s Nancy Haegel: A Center for Power Electronics Materials and Manufacturing Exploration (APEX). Tellekamp will colead the work on this topic.
Tellekamp aspires to continue making progress on vertical AlGaN electronic devices.
"I hope we can get to a point where the devices can be lifted off from our designed substrates, enabling both better thermal management and substrate reuse," Tellekamp said. "There are so many experiments along the way to understanding how interfaces of dissimilar materials like to go together, and then you have to learn how to manage new types of defects, but I hope that within the next five years we will have demonstrated this as a viable technology to improve the energy efficiency of power-demanding applications.
"On the oxide (rather than nitride) side of things, I think we will see gallium oxide take some pretty drastic leaps in terms of cost, performance, and adoption. Already, the pace has been exhausting—the first gallium oxide transistor was only developed 12 years ago—yet I think in the next five years we will see gallium oxide devices hitting the market while research-level devices start to regularly exceed 10-kV breakdown voltages. I hope that our basic science investigations of Ga2O3 and the interfaces formed with other materials play a part in that advance."
Learn more about Brooks Tellekamp's work and about NREL's materials science research.