China advances domestic immersion DUV lithography, but faces hurdles before challenging ASML

China’s efforts to develop domestic immersion deep-ultraviolet lithography systems have reached a new stage, yet significant technical and logistical challenges remain before these tools can rival the industry leader, ASML, in commercial high-volume production.

China’s effort to build its own chipmaking tools has moved into a more exacting phase. On 27 July, The Information reported that a Shanghai state-backed group had begun making immersion deep-ultraviolet lithography systems. By 2 September, however, UBS, as cited by TrendForce and Bloomberg, said China might still need two to five years to reach high-volume production and was unlikely to produce a domestic extreme-ultraviolet rival within the next decade. The immediate question, then, is no longer whether China can assemble an immersion DUV machine, but whether it can qualify one for sustained industrial use.

That matters because immersion DUV sits in the most commercially important tier of chipmaking equipment still realistically within China’s reach. These machines use water between the lens and wafer to sharpen 193-nanometre light and print finer features than dry DUV systems. They cannot match EUV at the leading edge, but they can make a broad range of mainstream semiconductors and can be pushed further through multiple patterning, in which the same layer is exposed several times. Reuters and The Information said a successful domestic tool would give Chinese fabs an alternative if Washington and allied governments tighten restrictions not only on sales, but also on the servicing of foreign lithography systems in China.

The reported customer list shows why the development has drawn so much attention. The Information said the first deliveries are due in 2026 to Semiconductor Manufacturing International Corp, Hua Hong Semiconductor and memory maker ChangXin Memory Technologies, with output of about five machines this year and roughly 20 in 2027. Reuters later identified the manufacturer as Shanghai Aishengna Electronic Technology Group, a little-known state-backed company that has drawn together teams from Shanghai Micro Electronics Equipment and Yuliangsheng. Tom’s Hardware reported that SMIC has been testing a Yuliangsheng immersion tool since September 2025, indicating that the present milestone follows a longer qualification programme rather than a single abrupt breakthrough.

Even so, the reporting around “mass production” has been hedged by unusually strong scepticism. Paul Triolo told the South China Morning Post that the real challenge is keeping machines running “24 hours, seven days a week, for an entire year”. Reuters said Aishengna’s tool still needs further testing and remains well behind ASML’s established systems. Tom’s Hardware added that while most components are said to be domestic, some critical parts still come from Japan, and delays at Chinese suppliers have already constrained this year’s output. In other words, a tool delivered to a customer is not yet the same thing as a tool trusted on a high-volume production line.

The scale comparison with ASML helps explain why analysts do not see an immediate commercial upset. China is reported to be targeting five immersion systems in 2026; ASML, by contrast, expects to ship about 130 immersion tools this year, roughly in line with 2025, according to comments by finance chief Roger Dassen cited by Tom’s Hardware after the company’s July results. ASML also plans to raise immersion capacity by 30 per cent in 2027. TrendForce, citing UBS, said an immersion DUV machine sells for close to US$90 million, while a top-end EUV system costs more than US$200 million. China is therefore still competing first on strategic availability, not on production scale or technical parity.

Markets still treated the July report as significant. The South China Morning Post said ASML shares fell more than 8 per cent in US trading before closing 5.8 per cent lower, while a Reuters syndication carried by Investing.com said the stock was on track for its worst day since 8 June 2026. The reaction reflected strategic exposure as much as near-term financial damage. TrendForce said China accounted for 14 per cent of ASML’s net system sales by shipment location in the second quarter of 2026, down from 19 per cent in the first. A local substitute that is merely good enough for some customers would still matter in one of ASML’s most politically fraught markets.

For Beijing, the project fits a much wider industrial campaign. Reuters described the Aishengna effort as part of President Xi Jinping’s semiconductor self-sufficiency drive, and TrendForce said wafer-fab equipment spending in China is forecast to rise from about US$44.3 billion in 2026 to US$53 billion in 2027 and US$60.7 billion in 2028. The same assessment pointed to a rise in Chinese patent activity in areas such as light sources and laser subsystems, which UBS regards as one sign of growing technical maturity. Yet the bank still judged China’s lithography capabilities to be roughly where ASML stood around 2004, suggesting real progress without anything close to catch-up.

China is also pursuing EUV, but that track remains far longer. Reuters reported in late 2025 that a domestic EUV prototype had been completed in early 2025 and was undergoing testing. Even so, the latest UBS view is that a Chinese EUV system comparable with ASML’s best products is unlikely within a decade. The practical conclusion is that China may become much harder to shut out of mature and mid-range chip production well before it can challenge the top of the market. That could eventually lower costs in some mainstream electronics components, but it also points towards a more fragmented and duplicated global semiconductor equipment stack rather than a swift convergence with the industry’s technological frontier.