China's homegrown DUV tool jolts chip industry, but commercial yield test still looms
Seoul, August 6
Recent reports that China has developed its own immersion deep-ultraviolet lithography system have jolted the global semiconductor industry, suggesting that Beijing may be closing one of the most formidable technology gaps maintained by US export controls, according to a news report by The Korea Herald.
The news report mentioned that Korean experts cautioned that building the machine is only the first step. China could still need years to prove that the equipment can operate reliably, process wafers at competitive speeds and deliver commercially viable chip yields.
Still, Beijing's steady progress toward semiconductor self-sufficiency should not be underestimated, Korea's Industry Minister Kim Jung-kwan said Thursday. He called the reported advance "shocking and faster than expected."
"China is advancing step by step in semiconductors despite the limited profitability of many Chinese chipmakers backed by massive state support," the news report quoted Kim.
Blocked from buying advanced lithography systems, Chinese chipmakers relied largely on older dry DUV equipment from ASML, used mainly to produce legacy chips on 20- to 28-nanometer processes. Chinese companies reportedly pushed older equipment to produce 7-nanometer-class chips through multipatterning, though their yields and production economics remained unclear.
A domestically developed immersion DUV system could help China move toward smaller process nodes. But determining whether the equipment is suitable for stable mass production could take at least two to three years, according to Shim Dae-yong, an electrical engineering professor at Dong-A University and former SK hynix vice president specializing in high-bandwidth memory.
"Lithography is widely considered the most difficult of the roughly 800 steps involved in chipmaking, which also include deposition, etching and cleaning," the news report quoted Shim.
"Having the machine does not mean the work is finished," Shim added. "Even SK hynix needed two to three years to tune its processes and achieve meaningful yields after adopting ASML's ArF immersion DUV equipment."
China's system may eventually be capable of producing advanced chips, but initial yields could remain at 10 to 20 per cent, Shim noted, while profitable mass production generally demands levels closer to 90 per cent.
Kim Hyung-jun, an emeritus professor at Seoul National University and head of the state-led Next Generation Intelligence Semiconductor Foundation, offered a similar assessment regarding the early single-stage architecture of the tool.
"What matters is throughput -- how quickly the equipment can process wafers," the news report quoted Kim. "Given that it is a single-stage tool, it will take considerable time before China can produce meaningful results."
However, state support could offset these early mechanical limitations.
"They may lack the yields, and producing one advanced chip with older tools can become extremely expensive," Kim said. "But if the government requires electronics manufacturers to use domestic chips and creates a funding cycle around the industry, companies can continue building scale."
China's reported lithography advance poses little immediate threat to Korean chipmakers like Samsung Electronics and SK hynix, which are already pushing into next-generation memory architectures. Yet experts noted Beijing's willingness to absorb low profits could steadily narrow the technology gap.
— ANI
Reader Comments
Interesting piece, but honestly, a 10-20% yield isn't going to scare TSMC or Samsung anytime soon. The Korean experts make a good point - commercial viability is a different beast entirely. Let's see if China can actually scale this up profitably.
The state support angle is key here. Unlike private companies, China can absorb losses for years. They proved this with solar panels and EVs. Western analysts keep saying "it won't work profitably" - but when has that stopped them? Strategic self-sufficiency isn't just about profits.
This is a wake-up call for India too. We're still importing most of our chips while China is developing their own lithography tools. We need to seriously rethink our semiconductor strategy and not just rely on building fabs with foreign help.
The Korean professor's comparison is spot on - even SK hynix took 2-3 years to tune ASML equipment. China starting from scratch will face even bigger challenges. But honestly, 10-20% yields don't matter when you have unlimited government funding. #ChipWar
Guys, don't underestimate the power of persistence! China has shown they can catch up in anything they put their mind to. But I agree with the article - there's a long way from prototype to mass production. The real test will be in the next 5 years.
We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.