A Russian research facility is creating a lithography scanner that may manufacture devices using 7nm-class fabrication techniques. The machine is currently being developed, with a target completion date of 2028. When complete, it ought to be more effective than the Twinscan NXT:2000i tool created by ASML, whose development took more than a decade.

Taiwan acted quickly to halt supply of cutting-edge chips to the country on February 24, when Russia started its brutal conflict against Ukraine. The United States, the United Kingdom, and the European Union then imposed sanctions that effectively forbid almost all contract chipmakers with cutting-edge fabs from cooperating with Russian businesses. Additionally, businesses like Arm are unable to licence their technologies to chip designers in Russia. As a response, the Russian government launched a national initiative to create the nation’s own 28nm-class fabrication technology by 2030, reverse engineer the greatest number of international chips, and train local talent to work on home chips.

A production node in the 28nm class, however, poses a challenge by 2030. The most advanced chip factory in Russia is capable of fabricating semiconductors using a 65nm process. Due to sanctions, American and European manufacturers of fab tools are unable to supply their products to Russia, thus the nation must design and construct its own local wafer fabrication machinery if it wishes to use the 28nm node. In essence, it will take around eight years to complete what has taken organisations like ASML and Applied Materials decades to create and iterate.

According to their intentions, which were posted on the Nizhy Novgorod Strategy Development website, the Russian Institute of Applied Physics of the Russian Academy of Sciences apparently expects to surpass all expectations and build a 7nm-capable lithography scanner by 2028. (via Cnews).

A modern lithography scanner that can process wafers using a 7nm-class process technology is a highly complicated device that includes, to name a few essential components, a high-performance light source, sophisticated optics, and precise metrology. IAP, however, is certain that it can create such a tool quickly given its position as one of Russia’s top universities for applied physics.

When compared to scanners made by firms like ASML or Nikon, the tool will be somewhat different. For instance, IAP intends to employ a >600W light source (total power, not intermediate focus power) with an exposure wavelength of 11.3nm (EUV wavelength is 13.5nm), requiring much more advanced optics than are now available. The tool will be more portable and simpler to construct because the light source is relatively low power. However, it also indicates that compared to contemporary deep ultraviolet (DUV) instruments, its production of the scanner will be significantly lower. IAP says that might not be a problem.

IAP could be a little too optimistic about timing. Chipmakers utilise immersion lithography for anything smaller than 32 nm (which is essentially a booster to DUV tools). To make 65nm logic circuits and 70nm half-pitch DRAMs, ASML unveiled their first immersion lithography equipment, the Twinscan XT:1250i, in late 2003 with a goal of shipping one in Q3 2004. The business announced its 32nm-capable Twinscan NXT:1950i in late 2008, with client deliveries beginning in 2009. It took them around five years and a new generation of tools.

The market leader then took about nine years to release its Twinscan NXT:2000i DUV tool in 2018 with support for 7nm and 5nm. Although TSMC’s first-generation N7 manufacturing process used multi-patterning and less sophisticated tools, the timing of ASML’s debuts shows how challenging the move from 65nm to 7nm is. ASML needed 14 years to transition from 65nm to 7nm. IAP now plans to create a 7nm-capable machine for volume production from scratch in around six years despite having no prior experience in chip fabrication and no connections to chipmakers. Although the proposal doesn’t seem possible, it appears that IAP is enthusiastic.

By 2024, IAP hopes to have a fully operational alpha scanner. This one will only need to function and appeal to potential investors; it won’t need to deliver maximum productivity or resolution. By 2026, IAP hopes to have a beta version of the scanner with increased output and resolution. Although this machine should be capable of large production, its productivity is not anticipated to be at its highest level. It is predicted that the litho scanner’s ultimate version will appear in 2028. It should receive a high-performance light source, which will improve metrology and all of its capabilities. How many of these machines IAP and/or its manufacturing partners will be able to create by 2028 is unknown.

The use of fab equipment is not just restricted to lithography scanners, it should be highlighted. Other machine types that are used for etching, depositing, resist removal, metrology, and inspection tasks and are not manufactured in Russia exist. Additionally, there is also less sophisticated equipment like ultrapure air and water generators that are also not made in Russia. Even if IAP RAS succeeds in creating a lithography tool, Russia will still require a few hundred tools to create a cutting-edge fabrication facility. Additionally, fabs require ultrapure raw materials made in nations that won’t export to Russia.

Also Read:

source