Elusive diamond transistors could spark multifaceted electronics performance gains

// Electrical & Computer Engineering

Tags: 2019, Faculty, News, research

Photo of diamond bipolar transistor chip

Made from diamond, and measuring smaller than a penny coin, newly developed devices can transmit signals with higher dynamic ranges than any other existing technologies. Image credit: Sang June Cho

Share this story:

Diamonds, it turns out, can be an electrical engineer’s best friend.

And with a little polish, a multi-university team of researchers plans to use diamond to create high-power communication devices that can transmit signals with higher dynamic ranges than any other existing technologies.

Photo of Zhenqiang “Jack” Ma
Zhenqiang “Jack” Ma. Photo credit: Renee Meiller.

It’s a project they hope will lead to more than 4-fold improvements in state-of-the-art bipolar transistors—a technology for enhancing the nation’s military capability that could also improve our cell phones, home appliances and the power grid.

Led by Zhenqiang (Jack) Ma, a professor of electrical and computer engineering at the University of Wisconsin-Madison, the team is working with support from a $4.85 million grant from the U.S. Defense Advanced Research Projects Agency.

“Our team is unique,” says Ma. “We are the only group using diamond.”

On paper, bipolar transistors made with diamond could achieve much greater linearity and power than the silicon devices commonly used in today’s electronics.

But for more than four decades, researchers have not actually been able to harness diamond’s promising properties—especially in bipolar transistors.

There’s a really good reason diamond doesn’t hold the same luster for other research groups. The precious stone has one major shortcoming: It’s rather stuck in its ways.

Bipolar transistors need two types of charge carriers to function. And to achieve those two types of charge carriers, they must contain junctions between two different types of materials, doped p-type and n-type.

Engineers add substances called “dopants” to create the p-type and n-type versions of materials, and some, like silicon, are happy to play along, readily forming both varieties.

Unfortunately, diamond is discerning. And despite more than 40 years’ of trying, researchers have failed incorporate an effective n-type dopant.

Additionally, mixing p-type diamond with the n-type of a different semiconductor material, such as silicon or germanium, has long been impossible because the atoms at semiconductor junctions must align perfectly. Even the slightest mismatch between the crystalline structures of the p-type material and the n-type material will destroy its semiconducting properties.

In other words, only materials whose atoms arrange themselves into similarly spaced patterns can be combined in heteroepitaxy—and that limits which materials will work for many types of transistors, photon-sensing devices or light-emitting devices.

The difference between the atomic structures of diamond and silicon is vast.

“There is a huge mismatch. One is basically in the south pole the other is basically in the north pole,” says Ma.

Ma devised an ingenious strategy to overcome that mismatch. He inserted an ultrathin intermediate layer, measuring mere atoms across, to bridge the gap between the two different semiconducting materials—without interrupting the flow of current carried by the two types of charge carriers.

That intermediary—an ultrathin oxide—would, under normal circumstances, block current between the p-type and n-type layers. But Ma leveraged a phenomenon of quantum mechanics called quantum tunneling to allow electricity to flow through the narrow oxide layer.

Voila! Using their expertise in materials fabrication, Ma and colleagues combined diamond with silicon. And while they were at it, they also combined diamond with gallium arsenide, which is currently used in the most state-of-the-art transistors.

Moving forward, the team will develop diamond-based transistors, with the goal of achieving 4-fold improvement on radio-frequency power density over current best semiconductor devices. They are patenting this unique technology through the Wisconsin Alumni Research Foundation.

However, the researchers won’t focus on diamonds forever: Their ultrathin oxide strategy makes it possible to match up almost any pair of p-type and n-type semiconductors—and in total, the team has created more than 10 novel combinations of materials using the approach.

In the commercial world, that could pave the way to ultra-high-performance electronic devices.

Each combination opens up new electronics or optoelectronics application fields and drastic performance improvement possibilities for LEDs, high-power communications, and many, many other devices.

“Every time you combine new materials you find something new,” says Ma. “This is the highest achievement of my life so far.”

Zhenqiang (Jack) Ma is the Lynn H. Matthias Professor in Engineering & Vilas Distinguished Achievement Professor in the Department of Electrical and Computer Engineering at the University of Wisconsin-Madison. Collaborators include John Albrecht, John Papapolymerou and Timothy Grotjohn in the Department of Electrical and Computer Engineering at Michigan State University; Jung-Hun Seo in the Department of Material Design and Innovation at the State University of New York, Buffalo; and Michael Becker at Fraunhofer USA. The research is funded by the Defense Advanced Research Projects Agency under the DREaM Program.

Author: Sam Million-Weaver