The October 2022 issue of IEEE Spectrum is here!

Close bar
Cross-section rendering

In this artist’s depiction of a new molybdenum disulfide transistor, the bright point [center] represents the transistor gate electrode.

Fan Wu

For decades, silicon transistors become smaller and smaller, but they are fast approaching the point at which they can no longer shrink the lengths of their gates—that is, how far current must travel in these devices. Now, by using atomically thin materials, scientists in China have created a transistor with a record-breaking gate length of just roughly one-third of a nanometer wide, only as thick as a single layer of carbon atoms, shedding light on how much smaller—if at all—transistors can possibly get.

In all transistors, current flows from the source to the drain, and that flow is controlled by the gate, which switches on and off in response to an applied voltage. The length of the gate is a key marker of a transistor’s size.

Since the first integrated circuit was built in the 1950s, silicon transistors have shrunk following Moore’s law, helping pack more of these devices onto microchips to boost their computing power. However, transistors are now approaching their theoretical limits when it comes to the sizes of their gates. Below roughly 5 nanometers, silicon can no longer control the flow of electrons from sources to drains because of a quantum-mechanical effect known as tunneling.

“This could be the last node for Moore’s law.”
—Tian-Ling Ren, Tsinghua University

Recently, scientists began exploring two-dimensional materials for next-generation electronics, including graphene, which consists of single layers of carbon atoms, and molybdenum disulfide, which is made of a sheet of molybdenum atoms sandwiched between two layers of sulfur atoms. For example, in 2016, scientists created a transistor with gates each just 1 nm long using carbon nanotubes and molybdenum disulfide.

Now scientists in China have created a transistor using graphene and molybdenum disulfide with a gate length of just 0.34 nm by exploiting the vertical aspect of the device. “We have realized the world’s smallest gate-length transistor,” says study senior author Tian-Ling Ren, an electrical engineer at Tsinghua University in Beijing.

To envision the new device, imagine two steps in a stairwell. On top of the higher step is the source, and on top of the lower step is the drain, both of which are made of titanium-palladium metal contacts. The surface of the stairwell, which acts as the electrical channel connecting the source with the drain, is made of a single layer of molybdenum disulfide. Underneath that surface is a thin layer of electrically insulating hafnium dioxide.

Inside the higher step is a sandwich of several layers. The bottom layer is a sheet of graphene, which consists of a single layer of carbon atoms. On top of it is a block of aluminum covered in aluminum oxide, which keeps the graphene and the molybdenum disulfide largely separated, except for a thin gap on the vertical side of the higher step. Both the higher and lower step sit on a silicon dioxide layer on a 5-centimeter silicon wafer.

“It will be hard to beat this record for quite some time.”
—Huamin Li, State University of New York at Buffalo

When the gate is set to its on state—so electricity can essentially flow down the steps from the source past the graphene to the drain—the gate is effectively only 0.34 nanometers wide, the same width as the graphene layer.

“In the future, it will be almost impossible for people to make a gate length smaller than 0.34 nm,” Ren notes. “This could be the last node for Moore’s law.”

In 2021, a different group reported a vertical transistor made using molybdenum disulfide with a gate length of 0.65 nm in its on state. This new work pushes the scaling limit for gates further to “just the thickness of a single layer of carbon atoms,” says Huamin Li, a nanoelectronics scientist at the State University of New York at Buffalo, who did not take part in this study. “It will be hard to beat this record for quite some time.”

In transistors, gate on and off states typically differ in length when electric fields are applied, but this effect is usually not significant at larger scales. In this new device, when a voltage is applied to the gate to switch it to its off state, this makes the gate effectively 4.54 nanometers long, a difference that can prove an advantage.

“A longer channel with higher resistance at the off-state would help prevent leakage current,” Li says. “In contrast, a shorter channel length with lower resistance at the on state would boost the on-state current density.”

In the future, the researchers plan to create larger-scale circuits with their new transistors. “The next goal is making a 1-bit CPU,” Ren says. A likely challenge is fabricating higher-quality, larger-area molybdenum disulfide, as well as the material’s current high cost, he notes.

All in all, “this prototypical work stands as a new attempt to explore the vertical architecture of transistors, following the evolution of FinFET technology,” Li says. “Hopefully it will inspire more creative ideas to fully explore the potential of 2D materials and extend Moore’s law for high-performance energy-efficient nanoelectronics.”

The scientists detailed their findings in the 10 March issue of the journal Nature.

The Conversation (1)
Caleb Adewole15 Mar, 2022

Nice stuffs.

The First Million-Transistor Chip: the Engineers’ Story

Intel’s i860 RISC chip was a graphics powerhouse

21 min read
Twenty people crowd into a cubicle, the man in the center seated holding a silicon wafer full of chips

Intel's million-transistor chip development team

In San Francisco on Feb. 27, 1989, Intel Corp., Santa Clara, Calif., startled the world of high technology by presenting the first ever 1-million-transistor microprocessor, which was also the company’s first such chip to use a reduced instruction set.

The number of transistors alone marks a huge leap upward: Intel’s previous microprocessor, the 80386, has only 275,000 of them. But this long-deferred move into the booming market in reduced-instruction-set computing (RISC) was more of a shock, in part because it broke with Intel’s tradition of compatibility with earlier processors—and not least because after three well-guarded years in development the chip came as a complete surprise. Now designated the i860, it entered development in 1986 about the same time as the 80486, the yet-to-be-introduced successor to Intel’s highly regarded 80286 and 80386. The two chips have about the same area and use the same 1-micrometer CMOS technology then under development at the company’s systems production and manufacturing plant in Hillsboro, Ore. But with the i860, then code-named the N10, the company planned a revolution.

Keep Reading ↓Show less