China Develops Domestic EUV Tool, ASML Monopoly in Trouble

[Daniel Nenni]

China's domestic extreme ultraviolet (EUV) lithography development is far from a distant dream. The newest system, now undergoing testing at Huawei's Dongguan facility, leverages laser-induced discharge plasma (LDP) technology, representing a potentially disruptive approach to EUV light generation. The system is scheduled for trial production in Q3 2025, with mass manufacturing targeted for 2026, potentially positioning China to break ASML's technical monopoly in advanced lithography. The LDP approach employed in the Chinese system generates 13.5 nm EUV radiation by vaporizing tin between electrodes and converting it to plasma via high-voltage discharge, where electron-ion collisions produce the required wavelength. This methodology offers several technical advantages over ASML's laser-produced plasma (LPP) technique, including simplified architecture, reduced footprint, improved energy efficiency, and potentially lower production costs.

The LPP method relies on high-energy lasers and complex FPGA-based real-time control electronics to achieve the same result. While ASML has refined its LPP-based systems over decades, the inherent efficiency advantages of the LDP approach could accelerate China's catch-up timeline in this critical semiconductor manufacturing technology. When the US imposed sanctions on EUV shipments to Chinese companies, the Chinese semiconductor development was basically limited as standard deep ultraviolet (DUV) wave lithography systems utilize 248 nm (KrF) and 193 nm (ArF) wavelengths for semiconductor patterning, with 193 nm immersion technology representing the most advanced pre-EUV production technique. These longer wavelengths contrast with EUV's 13.5 nm radiation, requiring multiple patterning techniques to achieve advanced nodes.



However, this Huawei system must still answer questions about resolution capabilities, throughput stability, and integration with existing semiconductor manufacturing flows. However, commercializing an alternative EUV lithography tool will challenge ASML's position. ASML's latest High-NA EUV tool costs around 380 million US Dollars. No matter the cost for Chinese R&D centers, the Huawei EUV machine will deliver the much-needed upgrade path for the older DUV scanners, which previously limited domestic chip production. Despite China's development of solid IP, its manufacturing progress was limited, but it could experience a "DeepSeek" moment very soon. Leading fabs like SMIC are working with Huawei to integrate the EUV scanners into existing workflows. A solid semiconductor manufacturing workflow takes years to build, so we have to see what the final result will be.

China Develops Domestic EUV Tool, ASML Monopoly in Trouble

China's domestic extreme ultraviolet (EUV) lithography development is far from a distant dream. The newest system, now undergoing testing at Huawei's Dongguan facility, leverages laser-induced discharge plasma (LDP) technology, representing a potentially disruptive approach to EUV light...

www.techpowerup.com

 

[nghanayem]

Considering native Chinese DUV systems have a lower field size than industry standard, and worse throughout and overlay than a Nikon. I smell a wiff over of optimism in the air. I wonder if relying on complex multi paterning schemes with more sophisticated imported tools is better than simpler sub par EUV tools. After all EUV with ASML throughput and overlay isn't always better than quad paterning from a TCO perspective (see DRAM bitcells until the late 1x nm generation where the layouts are simple and you don't need to be in block/cut hell), and is only sometimes better than DUV double paterning from a TCO or TVO basis (interconnects and gate paterning on 5nm class processes being easy examples where double DUV is often used instead direct print EUV). Why would I assume a worse performing EUV system would be used for anything beyond where it is absolutely needed (like nano sheet paterning). And we are also assuming that this tool is 0.33 NA. After all the first ASML prototype/demonstration "production" units had an unusable NA of 0.25. Not to say this tool has an NA that bad. But without field size, NA, overlay, and throughput saying you have a tool doesn't mean anything. As an extreme example nobody was making wafers with those EUV "tools" from the 90s even though they technically produced 13.5nm light.

 

[Jert]

ASML must be shivering uncontrollably as it helplessly witnessing the end of its monopoly...

 

[chenyj]

I checked the original article and found the posted picture of an EUV interferometer (in a key national optical lab) was from TikTok where you can find tons of party's propaganda materials. EUV light source is way way from a robust high-volume production worthy EUV scanner, not to mention the LDP source is still on the early stage of research and development. I would just ignore the word of "Huawei" since some media just used it to catch your eyeballs. In general, it will take at least 10-15 years for China to come up with a production worthy EUV scanner. Regardless of whatever DeepSeek or Manus can do, it remains to be one urgent question how Chinese semiconductor industry can follow the continuing scaling pace of high performance computing when they only have access to DUVL.

 

[Fred Chen]

Their 65nm scanner has 8nm overlay, not consistent with being competitive with ASML.

 

[Paul2]

Considering native Chinese DUV systems have a lower field size than industry standard, and worse throughout and overlay than a Nikon. I smell a wiff over of optimism in the air. I wonder if relying on complex multi paterning schemes with more sophisticated imported tools is better than simpler sub par EUV tools. After all EUV with ASML throughput and overlay isn't always better than quad paterning from a TCO perspective (see DRAM bitcells until the late 1x nm generation where the layouts are simple and you don't need to be in block/cut hell), and is only sometimes better than DUV double paterning from a TCO or TVO basis (interconnects and gate paterning on 5nm class processes being easy examples where double DUV is often used instead direct print EUV). Why would I assume a worse performing EUV system would be used for anything beyond where it is absolutely needed (like nano sheet paterning). And we are also assuming that this tool is 0.33 NA. After all the first ASML prototype/demonstration "production" units had an unusable NA of 0.25. Not to say this tool has an NA that bad. But without field size, NA, overlay, and throughput saying you have a tool doesn't mean anything. As an extreme example nobody was making wafers with those EUV "tools" from the 90s even though they technically produced 13.5nm light.

Leading edge is largely inconsequential for mainland fabs.

Most what they were doing for the last decade is trying to swallow the legacy market, which is 80%+ of commercial IC volume.

And they were failing miserably even at 180nm service.

Now, Taiwanese fabs let their guard down by chasing the leading edge, and losing market, and mainland pounced at that.

 

[samwu1998]

If a country’s semiconductor industry still relies on imported dry pumps, what’s there to talk about with EUV?

 

[Paul2]

If a country’s semiconductor industry still relies on imported dry pumps, what’s there to talk about with EUV?

Optics is more important than substance. Think why does China even shows all that in its English language (state) media.

Because they want Americans to see that.

Why they want Americans to see that?

If it was really important, we wouldn't have known that at all. Most of mainland's government 5 year plan is secret. What they release in public is a severely redacted version.

 

[Fred Chen]

China's domestic extreme ultraviolet (EUV) lithography development is far from a distant dream. The newest system, now undergoing testing at Huawei's Dongguan facility, leverages laser-induced discharge plasma (LDP) technology, representing a potentially disruptive approach to EUV light generation. The system is scheduled for trial production in Q3 2025, with mass manufacturing targeted for 2026, potentially positioning China to break ASML's technical monopoly in advanced lithography. The LDP approach employed in the Chinese system generates 13.5 nm EUV radiation by vaporizing tin between electrodes and converting it to plasma via high-voltage discharge, where electron-ion collisions produce the required wavelength. This methodology offers several technical advantages over ASML's laser-produced plasma (LPP) technique, including simplified architecture, reduced footprint, improved energy efficiency, and potentially lower production costs.

The LPP method relies on high-energy lasers and complex FPGA-based real-time control electronics to achieve the same result. While ASML has refined its LPP-based systems over decades, the inherent efficiency advantages of the LDP approach could accelerate China's catch-up timeline in this critical semiconductor manufacturing technology. When the US imposed sanctions on EUV shipments to Chinese companies, the Chinese semiconductor development was basically limited as standard deep ultraviolet (DUV) wave lithography systems utilize 248 nm (KrF) and 193 nm (ArF) wavelengths for semiconductor patterning, with 193 nm immersion technology representing the most advanced pre-EUV production technique. These longer wavelengths contrast with EUV's 13.5 nm radiation, requiring multiple patterning techniques to achieve advanced nodes.



However, this Huawei system must still answer questions about resolution capabilities, throughput stability, and integration with existing semiconductor manufacturing flows. However, commercializing an alternative EUV lithography tool will challenge ASML's position. ASML's latest High-NA EUV tool costs around 380 million US Dollars. No matter the cost for Chinese R&D centers, the Huawei EUV machine will deliver the much-needed upgrade path for the older DUV scanners, which previously limited domestic chip production. Despite China's development of solid IP, its manufacturing progress was limited, but it could experience a "DeepSeek" moment very soon. Leading fabs like SMIC are working with Huawei to integrate the EUV scanners into existing workflows. A solid semiconductor manufacturing workflow takes years to build, so we have to see what the final result will be.

China Develops Domestic EUV Tool, ASML Monopoly in Trouble

China's domestic extreme ultraviolet (EUV) lithography development is far from a distant dream. The newest system, now undergoing testing at Huawei's Dongguan facility, leverages laser-induced discharge plasma (LDP) technology, representing a potentially disruptive approach to EUV light...

www.techpowerup.com

I didn't see any threat to ASML.

1. ASML doesn't sell EUV to China. So this won't change.
2. China doesn't ship any competitive ArF systems, buys ASML systems instead. Their 65nm dry ArF system doesn't have sufficient overlay. Immersion system also never panned out despite multiple false alarms. These are much easier to develop than an EUV system.
3. LDP was evaluated by ASML against LPP and did not win out. A key limitation is that LDP requires two big electrodes for the discharge; these block angles for EUV light collection (LPP can collect from back and front if you see above drawing).

 

[samwu1998]

I didn't see any threat to ASML.

1. ASML doesn't sell EUV to China. So this won't change.
2. China doesn't ship any competitive ArF systems, buys ASML systems instead. Their 65nm dry ArF system doesn't have sufficient overlay. Immersion system also never panned out despite multiple false alarms. These are much easier to develop than an EUV system.
3. LDP was evaluated by ASML against LPP and did not win out. A key limitation is that LDP requires two big electrodes for the discharge; these block angles for EUV light collection (LPP can collect from back and front if you see above drawing).

SMEE sold total $48M of litho and mask making tools in 2023. That is equivalent to ~2 units of NXT1998i @ 50% discount. I guess almost all their tools go to packaging houses.

 

[Paul2]

China doesn't ship any competitive ArF systems, buys ASML systems instead. Their 65nm dry ArF system doesn't have sufficient overlay. Immersion system also never panned out despite multiple false alarms. These are much easier to develop than an EUV system.

China's commercial interests are not cutting edge chips, but exploiting the huge legacy chip shortage.

Even 180nm steppers are all sold out at second hand equipment dealers.

 

[benb]

China is pretty strong in electrochemistry (PV, battery) that's probably why the LDP with the electrode is attractive to them.
Agree this announcement does seem 2008ish. 10 more years of hard work ahead. I wouldn't bet against them getting there eventually though.

 

[kevin01]

China is pretty strong in electrochemistry (PV, battery) that's probably why the LDP with the electrode is attractive to them.
Agree this announcement does seem 2008ish. 10 more years of hard work ahead. I wouldn't bet against them getting there eventually though.

I saw on X another theory about this so-called news: it was being used to pump stock prices of certain companies in Chinese A shares market.

 

[KevinK]

China should go for it - will probably cost them a few hundred billion dollars plus 10 years to get something useful for manufacturing.

 

[lefty]

Interesting tweet here:

https://twitter.com/x/status/1899042593264320976