Terahertz waves are hard to tame. Ranging between microwaves and infrared light, these electromagnetic frequencies promise ultrafast wireless links but are very difficult to create and manipulate efficiently. Now, new research sheds light on a promising candidate to harness these waves: mercury telluride (HgTe), a material that converts two incoming frequencies into terahertz outputs with record room-temperature efficiency.

Tatiana Uaman Svetikova, a Ph.D. candidate at the quantum technologies department of the Helmholtz Center Dresden-Rossendorf in Dresden, Germany, says the newly published research marks a non-cryogenically-cooled first. She says the team was aiming to prove that mercury telluride worked “intrinsically, not just in simulation or under special lab conditions.”

A number of exotic materials (for example, spintronics and nonlinear crystals) have been studied to try to close the terahertz gap, harnessing terahertz frequencies for use in practical technologies.

German researchers developed chip-ready tech to convert low-frequency signals into terahertz frequencies, suitable for high-frequency data links.Oliver Killig/HZDR

In their experiment, Uaman Svetikova and her team shot two laser beams onto a 70-nanometer-thick mercury telluride film, and the film changed the incoming beams into terahertz waves.

Can Terahertz Links Replace Data Cables?

“It’s easy to do mathematical simulations to work with terahertz, but it’s extremely difficult to get experimental results. Just the fact that this group did it is a feat in itself,” says Arjun Singh, director of the Wireless and Intelligent Next Generation Systems (WINGS) research center at the State University of New York Polytechnic Institute. “This study actually got to true terahertz, and there are very few experimental studies that actually get there,” adds Singh, who did not take part in the study.

Singh says the mercury telluride device’s development represents a step toward on-chip terahertz sources—a prerequisite for miniaturizing today’s bulky tabletop systems into components suitable for consumer or data-center use. “We still don’t have a ‘terahertz laser’ or a Wi-Fi router equivalent,” he says.

He adds that short-range terahertz links could eventually serve high-capacity “wireless-wire” connections between servers or devices. “Imagine eliminating hundreds of cables in a data hall,” Singh says. “It’s not just faster—it saves weight, space, and power.”

Even so, Singh cautions that terahertz will not form the backbone of 6G networks. “Most early 6G deployments will still rely on low- and mid-band frequencies already in use,” he says. Instead, terahertz waves are more likely to appear in very dense environments—stadiums, city hubs, or AI data centers—where extreme data rates and low latency justify the added complexity. “Think of it less as the foundation of 6G and more as a specialty layer for extreme-data situations,” Singh adds.

The research team tested their ultrathin film mercury telluride device at room temperature, making the technology more practical for real-world applications.Oliver Killig/HZDR

“This is fundamental science,” says Uaman Svetikova. “But every time we improve efficiency or understand these materials better, we move one step closer to practical terahertz technology.”

How Can Efficiency of Terahertz Devices Be Improved?

Georgy Astakhov, head of the quantum technologies department at Helmholtz-Zentrum, says that while the device’s efficiency is only 2 percent, that’s in part because of the thickness of the material. “We expect that if we have a thicker material, or maybe a multilayered film…we can get it close to 100 percent. That’s our hope,” he says.

As a comparison, he adds, in some tests with materials like silicon, researchers got similar results but with much thicker films: “If we have high-quality mercury telluride, specifically of this thickness, we will get much better results,” he says.

But then the issue, Uaman Svetikova and Astakhov say, is the availability of such material. Mercury telluride is quite expensive to produce and is used mostly in military detectors, Astakhov says.

So the availability of the material for nonmilitary use is quite limited, even for ultrathin films like the ones in this experiment. “We can optimize the parameters and find cheaper materials, or less expensive ones, that can be done thicker or on the wafer scale—and then we can hope to increase the efficiency,” says Astakhov.

The researchers described their work in a recent issue of the journal Nature Communications Physics.