For more than half a century, silicon has been foundational to electronics, ushering in the era of personal computers, the Internet, and electric cars. However, as technologies have advanced, silicon chips are reaching their limits and inventing more powerful electronics will likely require new approaches—and new materials.
Wide-bandgap materials, a class of semiconductors that may allow power converters to become smaller, faster, and more efficient than their silicon-based counterparts, have emerged as one such viable option.
When assistant professor Juan Rivas-Davila and his research team member Grayson Zulauf, PhD ’20, discovered a mutual interest in exploring the scalability of wide-bandgap semiconductors, they decided to apply for a research seed grant from the TomKat Center for Sustainable Energy.
Together they set out to invent a “power electronics cell”—a bidirectional converter module that can be fabricated entirely on a printed circuit board and connected in a variety of configurations.
Most electronic devices—from cell phones to electric cars—operate on batteries that charge using direct current (DC). Because they plug into outlets that are typically wired in alternating current (AC) to send electricity across power lines, nearly all electronic devices use converters to change AC to DC and deliver the correct amount of power to the device. But these converters are purpose-built; their designs and components are specific to one use, and technology improvements don’t transfer easily.
Rivas-Davila and Zulauf’s concept is to invent a converter cell that could be used across applications. Each converter is only about the size of a wallet and able to handle hundreds of watts, and can link together like building blocks, providing AC to DC or DC to DC conversion at the levels needed to electrify vehicles—tens of kilowatts or more.
With the TomKat Center funding in place, they purchased an off-the-shelf gallium nitride power semiconductor device, and set to work. The semiconductor had been marketed as a wide-bandgap material ideal for high switching frequencies, and so they anticipated most of their time on the project would be spent developing the passive components around it.
“That’s not how it turned out,” says Rivas-Davila.
After repeated tests of the device, Zulauf continued to find unexplained losses in the power device. As the researchers increased their switching frequency, the device finally failed completely after overheating. Curious, they purchased and tested competing brands of gallium nitride semiconductors, and found that the results were consistent across all devices. Given that these losses had been previously unreported and unmeasured by manufacturers, they realized that in order to advance the new technology, they first needed to establish a baseline of their actual performance.
By now, Rivas-Davila and Zulauf are midway through the field’s first comparison of silicon, silicon carbide, and gallium nitride semiconductors operating at high- and very-high-frequencies.
“We had to take one step back in order to take two steps forward,” says Zulauf.
Bandgap is an electronic property of a substance that is closely related to how easily electrons pass through it. Materials like copper and other conductors have little to no bandgap and transmit electricity easily. Materials like rubber, plastic, and other insulators have large bandgaps and stop electrons in their tracks. And materials that fall in the middle are the ever-important semiconductors, which can behave like a conductor and then be quickly switched to behave like an insulator.
Silicon has a bandgap of 1 to 1.5 electron volts (eV), whereas wide-bandgap semiconductor materials have 2 to 4 eV—and this higher energy gap gives devices made from these materials the ability to switch larger voltages much faster. Further, these newer materials are more robust, can operate at higher temperatures.
“In many automotive and aerospace applications, for example, there are lots of power converters that use bulky components,” explains Rivas-Davila. “Wide bandgap materials could create lighter and more compact power converters that can improve autonomy and fuel efficiency.”
Rivas-Davila and Zulauf’s ultimate goal is to build unequivocally better power electronics at previously unattainable switching frequencies.
“We are getting close to the point of building these systems,” says Rivas-Davila, and their progress has attracted the attention of Airbus, a corporate partner, as well as other manufacturers of drones and helicopters that are in search of compact, highly efficient power components. The technology could open the door to electrifying flight and transforming other industries that are dependent on fossil fuels.
“What’s been great about the seed grant is the chance to explore these concepts that may or may not work,” says Zulauf. “The funding is unique in supporting the discovery process that is inherent to these early, exploratory stages of research.”