Why a next-gen semiconductor doesn’t fall to pieces

A new class of semiconductors that can store information in electric fields could enable computers that run on less power, sensors with quantum precision, and the conversion of signals between electrical, optical and acoustic forms—but how they maintained two opposite electric polarizations in the same material was a mystery.

Now, a team led by engineers at the University of Michigan has discovered the reason why the materials, called wurtzite ferroelectric nitrides, don’t tear themselves apart.

“The wurtzite ferroelectric nitrides were recently discovered and have a broad range of applications in memory electronics, RF (radio frequency) electronics, acousto-electronics, microelectromechanical systems and quantum photonics, to name just a few. But the underlying mechanism of ferroelectric switching and charge compensation has remained elusive,” said Zetian Mi, the Pallab K. Bhattacharya Collegiate Professor of Engineering and co-corresponding author of the study in Nature.

“How is the material stabilized? It was largely unknown.”

Electrical polarization is a bit like magnetism, but while a bar magnet has a north and south end, an electrically polarized material has a positive and negative end. The new semiconductors may start out polarized in one direction. Exposure to an electric field can switch the polarization of the material—the positive end becomes negative and vice versa—and once the electric field is off, the reversed polarization remains.

But often, it’s not the whole material that switches polarization. Instead it’s divided into domains of the original polarization and the reversed polarization. Where these domains meet, and especially where two positive ends come together, researchers didn’t understand why the repulsion didn’t create a physical break in the material.

“In principle, the polarization discontinuity is not stable,” said Danhao Wang, U-M postdoctoral researcher in electrical and computer engineering and co-corresponding author of the study. “Those interfaces have a unique atomic arrangement that has never been observed before. And even more exciting, we observed that this structure may be suitable for conductive channels in future transistors.”

With experimental studies led by Mi’s team and theory calculations led by the group of Emmanouil Kioupakis, U-M professor of materials science and engineering, the team found that there is an atomic-scale break in the material—but that break creates the glue that holds it together.

At the horizontal joint, where the two positive ends meet, the crystal structure is fractured, creating a bunch of dangling bonds. Those bonds contain negatively charged electrons that perfectly balance the excess positive charge at the edge of each domain within the semiconductor.

“It’s a simple and elegant result—an abrupt polarization change would typically create harmful defects, but in this case, the resulting broken bonds provide precisely the charge needed to stabilize the material,” said Kioupakis, also the Karl F. and Patricia J. Betz Family Faculty Scholar and a co-corresponding author of the study.

“What’s remarkable is that this charge cancellation isn’t just a lucky accident—it’s a direct consequence of the geometry of tetrahedra,” he said. “This makes it a universal stabilizing mechanism in all tetrahedral ferroelectrics—a class of materials that’s rapidly gaining attention for its potential in next-generation microelectronic devices.”

The team discovered this with electron microscopy that revealed the atomic structure of the particular semiconductor they used, scandium gallium nitride. Where the domains met, the usual hexagonal crystal structure was buckled over several atomic layers, creating the broken bonds. The microscopy showed that the layers were closer together than normal, but density functional theory calculations were needed to reveal the dangling bond structure.

In addition to holding the material together, the electrons in the dangling bonds create an adjustable superhighway for electricity along the joint, with about 100 times more charge-carriers than in a normal gallium nitride transistor. That highway can be turned off and on, moved within the material, and made more or less conductive by reversing, moving, strengthening or weakening the electrical field that sets the polarization.

The team immediately noticed its potential as a field effect transistor that could support high currents, good for high power and high frequency electronics. This is what they plan to build next.