NeuroAI NEUROAINEUROAI.SITE
China AI, reported clearly
1,300+ verified stories on models, robotics, BCI and the business of AI.
Browse stories
ESC

How a Chinese university team cut contact resistance in 2D transistors by 94 percent

A HUST-led team used oxygen doping to slash contact resistance in monolayer WSe2 p-type transistors, reaching high drive current in a 16nm device.

2026-11-16 · 1021 words · NeuroAI
How a Chinese university team cut contact resistance in 2D transistors by 94 percent

Key takeaways

  • A Huazhong University of Science and Technology–led team reported oxygen doping in monolayer tungsten diselenide (WSe2) p-type transistors.
  • The approach reduced contact resistance by 94 percent, according to the report.
  • A device with a 16nm channel reached a drive current of 1,635 microamps per micron (μA/μm).
  • The finding was reported on 2026-11-13 and points to progress in two-dimensional semiconductor physics.

What happened

On 2026-11-13, news emerged that a research team led by Huazhong University of Science and Technology, referred to as HUST, had applied oxygen doping to monolayer WSe2 p-type transistors. The result, as reported, was a 94 percent reduction in contact resistance. In the same work, a transistor with a 16-nanometre channel length achieved a drive current of 1,635 μA/μm. The source for this account is SCMP Tech on 2026-11-13. That is the factual foundation available; no further numbers, institutional partners, funding bodies, or publication venues beyond this report are asserted here.

The significance begins with the material itself. WSe2 belongs to a class of materials known as transition metal dichalcogenides, or TMDs. When reduced to a single layer—a monolayer—it behaves as a two-dimensional semiconductor. Unlike silicon, which is a three-dimensional bulk material that the industry has engineered for decades, a monolayer crystal is only one atom thick in the vertical direction. That thinness is what makes it attractive for future logic devices: as transistors shrink, keeping control over the channel becomes harder in bulk silicon, but an atomically thin body offers electrostatic control that does not degrade in the same way.

Yet the thinness also creates a practical problem. A transistor is not only a channel; it also has contacts, the metal interfaces where current enters and leaves the semiconductor. In two-dimensional materials, the contact region is often where performance is lost. If the interface between metal and monolayer WSe2 is poor, resistance rises, and the device cannot deliver the current that the channel itself would otherwise allow. Contact resistance is therefore not a minor detail but a central bottleneck for turning promising laboratory materials into viable circuit elements.

Why it matters

The reported 94 percent reduction addresses that bottleneck directly. A 94 percent cut means the resistance at the contact is only a small fraction of what it was before the doping step. In a p-type transistor, the carriers are holes rather than electrons, and WSe2 has been studied as a candidate for p-type behaviour because many two-dimensional materials are easier to make n-type. Getting good p-type contacts in monolayer form is part of the broader challenge of building complementary logic, which needs both p-type and n-type devices working together, much as silicon CMOS does today.

The drive-current figure gives the result a concrete anchor. A 16nm-channel device achieving 1,635 μA/μm means that, per micron of channel width, the device can push that much current when switched on. Channel length is a proxy for how far a technology has been scaled; 16nm is not the smallest node in production silicon, but for a research device on a two-dimensional monolayer, reaching this length while still reporting a high current density is notable. It suggests the oxygen-doped contacts are not merely helping a long-channel laboratory curiosity but are compatible with a scaled geometry.

Oxygen doping is the mechanism named in the report. Doping, in semiconductor language, means introducing atoms or molecules into a crystal to change its electrical properties. Oxygen, when introduced into the contact region or the material system around it, can alter the band alignment or the chemical interaction at the metal interface. The precise physical pathway is not detailed in the available summary, so this analysis does not claim a specific microscopic model. What can be said plainly is that adding oxygen changed the contact behaviour enough to produce the measured drop in resistance and the reported current.

What remains unknown

Several things are not established by the source report. The full composition of the team beyond HUST's leadership role is not given. The exact device stack, the metal used for contacts, the doping method's duration or concentration, and the conditions under which the drive-current figure was measured are not specified beyond what is quoted. Whether the result has been replicated by independent groups, or published in a peer-reviewed journal, is not stated in the news item available here. The long-term stability of oxygen-doped contacts under operating conditions, and whether the method can be integrated with existing fabrication tools, are open questions that the report does not answer.

For readers outside China, the context is that Chinese universities are active in two-dimensional semiconductor research, and HUST is one of the institutions producing work in this field. A result on contact resistance does not by itself mean a commercial technology is near. It means a specific obstacle has been measured and reduced in a specific experimental setting. The path from a single device metric to a manufacturing process is long, and nothing in the source report claims otherwise.

The wider picture for post-silicon logic

The angle here is a breakthrough in 2D semiconductor physics pointing toward post-silicon logic pathways. Silicon will not disappear quickly; the installed base of design, fabrication, and packaging knowledge is enormous. But the industry is interested in materials that can extend scaling or reduce power when silicon reaches its limits. Two-dimensional semiconductors offer one such route because their thinness helps with short-channel control and because they can be stacked or combined with other materials in ways bulk crystals cannot.

Contact engineering is the gatekeeper for that route. A perfect channel with bad contacts delivers nothing. By reporting a 94 percent drop in contact resistance and a high drive current in a 16nm-channel monolayer WSe2 p-type transistor, the HUST-led team has supplied evidence that the gatekeeper can be pushed back. The finding is a step in device physics, not a product announcement. Its value is in showing what the numbers can be when the interface problem is attacked directly, giving other researchers a target to match or exceed in the work that follows.


Sources: SCMP Tech,2026-11-13 "HUST-led team cuts contact resistance 94% in monolayer WSe2 p-type transistors with oxygen doping" — original report

Related coverage

Cite this article

NeuroAI (脑机网), "How a Chinese university team cut contact resistance in 2D transistors by 94 percent," 16 November 2026, https://neuroai.site/a/na-auto-2026-11-16-hust-led-team-cuts-conta

Machine-readable: BibTeX · RIS · JSON · Markdown

More in “Chips & Compute” → · Back to home · Markdown version

China AI, reported clearly
1,300+ verified stories on models, robotics, BCI and the business of AI.
Browse stories