Jensen Huang has been repeating it over and over again for many years: Moore’s law is dead. The NVIDIA co-founder and CEO isn’t the only leading technologist championing this vision, though he is one of those who do so most vehemently. However, there are also those who affirm the opposite and assure that innovation in semiconductors is managing to overcome the difficulties to delay the prediction made by Gordon Moore in 1965. This is Intel’s official position.
What Moore realized a little over five and a half decades ago was that the number of transistors in integrated circuits would double every year, while their relative cost would drop drastically. Ten years later he amended his observation by increasing the time frame required for such integration technology development to take place to twenty-four months, not one year. Since then his prognosis has come true quite accurately.
Beyond what these two opposing visions defend, there is something undeniable: every day we are a little closer to exhausting the capacities that have been placed in our hands for decades. silicon technology. And the companies that are involved in the semiconductor industry know this better than anyone. Fortunately, all is not lost. The future is hopeful because very promising innovations are on the table that seek to restore health to Moore’s law.
2D semiconductors come to the rescue
Research on materials whose two-dimensional (2D) versions take on the characteristic properties of semiconductors goes back a long way. Some of the elements that have attracted the attention of many groups that develop their research in this area are arsenene, antimonene or phosphorene, but there is a problem: producing them is very difficult. Fortunately, exfoliation techniques are constantly improving, hopefully making it possible to get these and other materials beyond the walls of research labs.
Sang-Hoon Bae and his research colleagues have developed a set of techniques that enables 2D materials to be produced on an industrial scale.
Precisely a group of researchers led by Sang-Hoon Bae, who is a professor of mechanical engineering at Washington University in St. Louis (United States), has published a very interesting article in Nature in which he claims to have solved three of the biggest challenges facing poses large-scale production of 2D materials candidates to break through in the semiconductor industry. Broadly speaking, these challenges are the precise control of the fine-tuning layer by layer of the 2D material, the need to preserve an absolute uniformity of each layer, and, finally, the meticulous execution of the number of layers necessary in the production of the wafers.
Sang-Hoon Bae and his research colleagues have come up with a set of techniques that, according to their paper, makes it possible to produce 2D materials on an industrial scale. This is very important. In fact, if it is confirmed that his proposal can really be used in the large-scale production of this type of semiconductor materials, the chip industry could receive the boost it has been seeking for many years. Interestingly, Intel is one of the companies that is investing the most resources in this area of research.
The company currently run by Pat Gelsinger has been engaged in 2D transistor tuning, as well as in refining the packaging of its integrated circuits and in the design of flexible systems of interconnection of the chiplets with a purpose: to use these technologies as soon as possible in the manufacture of its next microprocessors. And it looks like it’s on the right track. This is, at least, what their current strategy tells us, which is extraordinarily ambitious.
Intel plans to have microprocessors with a trillion transistors by 2030
At the end of last October, Gelsinger assured during the interview he gave to The Wall Street Journal that Intel plans to have the best transistors and the most advanced integration technology in the world by 2025. But this is not all. He also envisions having chips with a trillion transistors by 2030. And achieving this requires, among other things, developing the technology of two-dimensional semiconductor materials that are called upon to fill in the cracks that are already beginning to fracture silicon technology. Until then, much remains to be done, but the outlook is hopeful.
Cover Image: Intel
More information: Nature
