Now Playing: MUSiC, the First Multi-user Silicon Carbide Fabrication Facility in the U.S.

University of Arkansas invites you to play in its new SiC sandbox

December 1, 2025

Sebastian Blanco

The University of Arkansas Power Group opened MUSiC, the first multi-user silicon carbide (SiC) fabrication facility in the U.S., at its Fayetteville campus in mid-November. (Sebastian Blanco)

With plenty of references to national security and international competition, the University of Arkansas held the grand opening of MUSiC, the first multi-user silicon carbide (SiC) fabrication facility in the U.S. in mid-November. Helmed by Dr. H. Alan Mantooth, the founding director of the UA Power Group (UAPG), the new MUSiC facility will expand the UAPG’s capabilities thanks in part to $20 million in funding from the National Science Foundation’s Mid-scale Research Infrastructure-1 Program (Mid-scale RI-1), with additional support from the ARL’s Army Research Office and X-FAB. Local congressman Steve Womack (R) made it clear that, even in an era when the federal government is pulling back funding dollars for research efforts across the country, he still supports the work at MUSiC.

“This facility is vital,” Womack said during MUSiC’s opening ceremony. “Vital to our nation, vital to our national security, vital to our economic development, and we’re in a fast-paced cavalry battle right now with near peers around the globe for research on these issues. It is essential we win that race.”

The pace may be increasing, but the UAPG has been working on developing better power electronics for decades. The group was started in the wake of the 2003 Northeast power outage as an interdisciplinary institute that involves electrical, mechanical, and computer engineering. It has since collaborated with over 75 companies, including Toyota, Ford, and Caterpillar, on projects like microgrids, power electronics, and advanced cooling systems. The UAPG has experience packaging pre-commercial prototypes with thermal management for the automotive industry, among others. A decade or so ago, for example, NCREPT (National Center for Reliable Power Transmission), part of UAPG, built a 6.6-kW, air-cooled charger for a Toyota Prius PHEV.

Devices produced with the help of the UAPG on display in the new MUSiC building. (Sebastian Blanco)

“[Toyota’s] current unit, at the time, was 1.6 kW, the size of a pizza box and water cooled,” he said. “So there's DNA in the Toyota Prius from Arkansas. Of course, Toyota redesigned and commercialized it, but that’s what got them started.”

The UAPG’s next chapter, MUSiC, will focus on creating a SiC fabrication facility that operates as an open national sandbox, a place where “we can unleash the creativity of all these investigators around the country that have no place to build their stuff,” Mantooth said. “You can’t go to Cree, [because] Cree is building product. Can’t go to Microchip, they’re building product. You can go to GE a little bit, but it has to be under a program that GE is interested in. So if I’m partnered with Siemens, I can’t really go to GE because they’re competitors.”

Companies of many sizes have expressed an interest in working at MUSiC when the lab is up and running, Mantooth said. UAPG has already been contacted by a number of small companies that have federal Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs, but they’re not the only ones.

“They have responded to a call from the government, and they want to come in and use the facility and work with us to prove out the idea, to see if they have something,” Mantooth said, pointing to how Monolith Semiconductor did something similar with UAPG around a half-decade ago, but then had to build their products at X-FAB Texas. “What they did was they proved it out, and that company grew; it was acquired [by Littelfuse, Inc.]. This is the model that you want to see, so MUSiC is part of that innovation engine. Maybe we’re just the fuel in the engine, but it’s part of that innovation engine for those small businesses.”

Mantooth continued, posing, “What about large businesses? Well, believe it or not, large businesses come to us because they don’t have such a facility. You know, Toyota, Ford Motor Company, different OEMs, they’re not going to build silicon carbide devices, but they want to understand how it can impact their vehicles. So they want to do research and work with their supply chain to bring it online and have a differentiation in the market. Even large companies want to build stuff here and prototype things. We think both will play a key role.”

Housing MUSiC at a university as a nonprofit makes the space more open and, since UAPG “spent no small nickel on cyber infrastructure and compartmentalization of protecting IP,” Mantooth said, it will be able to provide academics and industry a place to experiment with new concepts without worrying about security and with shorter timelines.

Mantooth said an undisclosed company has approached UAPG to build models of every piece of equipment at MUSiC to create digital twins to be able to stand up processes faster and with greater control than is currently possible.

“So you think, what does it take to stand up a new CMOSS process, maybe year-and-a-half for all the qualifications and stuff,” he said. “It’s a lot of engineering work, but with digital twins, simulation and other things, we have great insight and great data, and we have machine learning models that we’re using in this setting. As a result of that, we can gain insight and make process stand up happen in six months, maybe three months. That means if we need to move a new technology out, we can move a new technology out. That means we’re building bridges to high-volume manufacturing much faster.”

The digital twins project could start as early as the spring of 2026, pending federal funding, Mantooth said. The fab’s first material trial for a power device run should take place in January 2026, followed by a CMOSS run in February that will be “just a lot of test structures, no circuits,” to characterize MUSiC’s lithography capability, among other things.

“My hope is that by mid-year 2026 we’re able to say, ‘Okay, let’s run some circuits and CMOSS, and let’s run a different power device,’” Mantooth said. “We ran the 1.7 kV – that’ll be our first run – [then] maybe we’ll be ready to trial a thicker Epi, 3.3 kV or higher. We will already be doing experiments with our Epi growth to try to grow thicker Epi for higher-voltage devices. Our target is 10 and 15 kV devices, that means 100 microns, 150 microns of Epi.”