The radios inside a future phone, smartwatch, or skin patch may no longer have to be rigid to be fast. A team has built a flexible transistor, the tiny switch at the heart of all electronics, that runs at the blistering frequencies 6G will need, while bending like a sheet of plastic. Until now, those two qualities, flexible and fast, were treated as a trade-off. This result says they do not have to be.
What they actually built
Researchers at Peking University in Beijing, working with collaborators at Stanford, made radio-frequency transistors on a bendable plastic film instead of a stiff silicon or quartz wafer. The switches are built from carbon nanotubes, which are essentially sheets of carbon rolled into hollow tubes thousands of times thinner than a hair. Lined up in dense, neat arrays, these tubes carry electrical signals extremely well.
The headline number is speed. A transistor’s “cut-off frequency” is the point at which it stops being useful as an amplifier, so higher is better. These devices reached a current-gain cut-off of 152 GHz and a power-gain cut-off of 102 GHz. Both clear the 100 GHz mark that engineers see as the entry ticket for 6G, the next generation of wireless after 5G. They did it while sipping power, under 200 milliwatts per millimetre of device, and the team also used them to build a working flexible amplifier. The results were published in Nature Electronics on 12 May 2026.
Why heat was the hidden wall
Here is the part most coverage skips, and it is the real story. Flexible radios were not slow because nanotubes are weak. They were slow because plastic cannot get rid of heat.
A fast transistor is a small space with a lot of energy moving through it, and that energy turns into heat. A rigid wafer like quartz or silicon pulls that heat away quickly, like a metal pan spreading warmth off a stove. The soft plastic films that make electronics bendable do the opposite. They trap heat like a foam cup. A transistor that overheats slows down, drifts, and eventually fails.
The team’s fix was not a new material but a redesign of how heat escapes, which they call electrothermal co-design. They reshaped the contacts, the gate, and the device geometry so the heat has a path out, then tuned the electrical layout around that. The proof is in their own control test. Identical nanotube transistors built without the heat-aware design reached only about 44 GHz, and more than half of them broke permanently the first time they were measured. With the redesign, the same devices ran above 100 GHz and kept working. The lesson is blunt: the ceiling on flexible high-speed electronics has been thermal, not electronic.
What this changes
The practical payoff is where fast radios can now live. 6G is expected to use very high frequencies that carry enormous amounts of data but travel only short distances, which means the world will need far more small antennas and radios, placed closer to people and objects. If those radios can bend, they can sit on a wrist, curve around a vehicle panel, line a wall, or laminate onto a medical patch, rather than living only inside flat rigid boxes.
It also matters for who made it. This is not a simulation or a lone lab curiosity. It is a fabricated, measured, repeatable device from a leading carbon-electronics group, with a respected Stanford thermal team involved. That combination makes it a credible engineering milestone rather than a press-release promise.
The honest limits
This is a proof of concept, not a product. The hard part of carbon nanotube electronics has always been manufacturing: growing perfectly aligned, pure nanotube arrays uniformly across large areas, at low cost, again and again. The paper does not solve that, and any stray metallic nanotubes degrade performance. There is also a long road from a single high-speed transistor to a complete flexible radio system that survives real-world bending, sweat, and time.
So the right read is measured. The barrier everyone assumed was the material turned out to be heat, and heat is an engineering problem with engineering answers. That is a more hopeful place to be stuck than a fundamental physical limit. Flexible 6G is still years away, but this removes one of the reasons people thought it might never arrive.
Sources
Xia, F., Xia, T., Su, H., Gan, L., Hu, Q., Wang, W., Huang, R., Bai, T., Chen, Y., Ma, C., Long, G., Wang, S. X., Pop, E., Peng, L.-M., & Hu, Y. (2026). Flexible radio-frequency carbon nanotube transistors operating at frequencies above 100 GHz. Nature Electronics. Published 12 May 2026. DOI: 10.1038/s41928-026-01632-1. https://www.nature.com/articles/s41928-026-01632-1
Ray Jackson holds a BSc in Electrical Engineering from the University of Manitoba and a PhD in Physics from Carleton University. His reporting interests include Current and Future Technologies, Engineering and Artificial Intelligence.