Forget for a moment the clamor around the decline of Moore’s Law. It’s had a brilliant run, something to be marveled at given it’s not a law at all. Squeezing out the last bit of performance that roughly corresponds to Moore’s Law may fall to nanosheet transistors expected to make their debut with 3 nm process technology.
A trio of prominent researchers – Peide Ye (Purdue), Thomas Ernst (CEA-Leti) and Mukesh V. Khare (IBM) – has written an interesting summary article in IEEE Spectrum (The Nanosheet Transistor Is the Next (and Maybe Last) Step in Moore’s Law) describing the strengths of nanosheet transistor on which all three have worked. While much of the material may be familiar, their article is a good (and fast) read tracing the evolution of FETs and recent work on nanosheet transistors. Today, of course, FinFET 7 nm technology is the leading edge in process technology for logic.
Unlike FinFET technology, in which the gate surrounds the channel on three sides to help control current leakage, in nanosheet technology the gate completely surrounds the channel providing even better control of current leakage. Use of the new shape and potentially new materials may extend Moore’s Law.
In the their article, Ye, Ernst, and Khare note the nanosheet approach, “goes by a variety of names—gate-all-around, multibridge channel, nanobeam—but in research circles we’ve been calling it the nanosheet. The name isn’t very important. What is important is that this design isn’t just the next transistor for logic chips; it might be the last. There will surely be variations on the theme, but from here on, it’s probably all about nanosheets.”
Here’s a brief excerpt:
“One of the astounding things about the nanosheet design is that it may extend Moore’s Law so far that it actually outlasts the use of silicon in the channel. To a large degree, what’s at issue here is heat.
“Transistor density is still increasing with every technology node. But the amount of heat an IC can reasonably remove—the power density—has been stuck at about 100 watts per square centimeter for a decade. Chipmakers have gone to great lengths to keep from surpassing this fundamental limit. To keep the heat down, clock rates don’t exceed 4 gigahertz. And the processor industry moved to multicore designs, correctly reasoning that several slower processor cores could do the same job as a single fast one while generating less heat. If we ever want to be able to ramp up clock speeds again, we’ll need more energy-efficient transistors than silicon by itself can deliver.
“One potential solution is to introduce new materials into the channel region, such as germanium or semiconductors composed of elements from columns III and V of the periodic table, such as gallium arsenide. Electrons can move more than 10 times as fast in some of these semiconductors, allowing transistors made from these materials to switch faster. More important, because the electrons move faster, you can operate the device at a lower voltage, which leads to higher energy efficiency and less heat generation.
The article is best read directly. To a large extent manufacturing nanosheet transistors involves processes and equipment that are already familiar and available. “All in all,” write the researchers, “stacking nanosheets appears to be the best way possible to construct future transistors. Chipmakers are already confident enough in the technology to put it on their road maps for the very near future. And with the integration of high-mobility semiconductor materials, nanosheet transistors could well carry us as far into the future as anyone can now foresee.”
Authors of IEEE Spectrum article:
Peide Ye is the Richard J. and Mary Jo Schwartz professor of electrical and computer engineering at Purdue University. Thomas Ernst is the scientific director at CEA-Leti in Grenoble, France. Mukesh V. Khare is vice president of semiconductor and AI hardware at IBM Research.
Link to IEEE Spectrum article: https://spectrum.ieee.org/semiconductors/devices/the-nanosheet-transistor-is-the-next-and-maybe-last-step-in-moores-law