Well from my findings RSLaser patent activity has been the highest among of all the companies working in the DUVi project. Looks like the are gearing to mass produce not only ArF lasers but also KrF lasers.
Chinese semiconductor industry
- Thread starter Hendrik_2000
- Start date
- Status
There will be haters that will make outlandish remarks about China's capability to build DUVL & EUVL scanners. These are folks that use superlatives in denouncing China's capabilities or something outlandish like the latter half of the prediction @lube referenced (....even with max espionage efforts"). Those types of "critiques" that are racist and nationalistic in nature are easy to spot, we should just outright ignore.
The "more objective" critics would never say 'never' but will attach boundary conditions to their opinions. Conditions such as "workable", "by 2030".
Now, I think the disparity between the western critics and the general public is the difference in our perspectivesand the different measuring stick we use to assess the situation. For example, what we think "workable" means.
If "workable prototype" simply means something that uses EUV light source to image some simple test pattern on a wafer, then I believe we already have that in some university research lab.
If "workable prototype" means a system that could be used to build functioning 5nm-node transistors (but not in HVM condition or efficiency), then this would be considered a herculean effort for anyone, including the American & the Chinese. Personally, I don't take offense to this type of critique because it's truly not as easy as everyone in this forum seemed to think.
Each of our expectations probably fall between the two extreme examples above so we all have varying degree of prediction. My observation that there is an obvious difference between the general public's optimism and industry professional's more pessimistic view. Ask enough people that work with scanners, and you will very very unlikely find someone that would speak with the type of conviction I see in this forum that we will have a working EUV prototype by 2025 or 2030. Is 2030 a possibility, yes. Would I bet my kid's life on it, hell NO (maybe I'd bet the Mrs.'s after taking out a life insurance policy for her, but not the kid........LMAO).
Last edited:
Good news. SMIC new domestic equipment fab is in machine moving in phase.
Yes, they could develop a working prototype that can produce "5nm" transistors in a few years, i dont see it as a big deal, but having a machine with the productivity that can mass produce chips is another deal at least with LPP (if the Tsinghua SSMB project is sucessful that will be another conversation), going from 10-20 WPH to 120 WPH of current ASML offerings is going to be the hardest part.
broadsword
Brigadier
That's kinda sobering downer. I was feeling @ZeEa5KPul projection of three years was already slow as my own wildly optimistic guess was two years for a production model.
The biggest worry I have isn't the semiconductor production volume. That will come by itself and even if not, it will create huge pressure on foreign suppliers to sell cheap and drive their margins down.
I think the most important to get right:
1. the production equipment (not just lithography but everything) - this is going somewhat well. lithography is a bit lagging but etch, CVD, is good. I think Naura PVD and ALD is not leading edge yet but they're getting there. This I'm pretty confident in.
2. the scientific instruments - there's some good progress in CD-SEM but this is a highly specialized market. Otherwise the instrumentation market is dominated by US, EU and Japan. This is not good because this actively limits the progress rate. The key isn't in just access to commodity instruments, its the capability to order a custom instrument.
Suppliers on the chemistry side (GC, LC, ICP, FTIR, Raman, gas monitors, pressure gauges) exist but they are small relative to major western/Japanese companies like Agilent/Thermo/MKS or even European/Japanese companies (Shimadzu, Hitachi, Zeiss, Horiba, Bruker). You really need at least a mid sized company here so people can trust your supply continuity and service.
Same with analytical SEM. No major analytical TEM companies. and , they're relatively small (benchtop XRD and XRF only, QC grade only, no TXRF for wafer, not analytical quality).
Good start, a long way to go. It must be noted this is a high tech, high barrier to entry, relatively low margin business (if at market rates; when monopolized, it has a mediocre margin). Maybe a bigger state backed player could do the trick.
3. the components. We've seen alot of progress here on this thread so this is something to watch for. There's good (ultraclean quartz and ceramic parts up to 300 mm wafer showerheads, good vacuum chamber design, etc) but some of the supporting capabilities like ultraclean fittings, valves, pumps, build to order vacuum chambers, large scale exotic materials machining (titanium, pure silicon) etc. needs work.
The semiconductor supply chain, like aerospace, automotive, petrochemical, pharmaceutical, etc. is heavily tied with basic machinery, optics, electrical and chemical manufacturing stuff that was started 100-150 years ago. Every country strong in the overall semiconductor supply chain (US, Japan, EU) was a developed country at the time. China is currently doing the impossible.
I think the most important to get right:
1. the production equipment (not just lithography but everything) - this is going somewhat well. lithography is a bit lagging but etch, CVD, is good. I think Naura PVD and ALD is not leading edge yet but they're getting there. This I'm pretty confident in.
2. the scientific instruments - there's some good progress in CD-SEM but this is a highly specialized market. Otherwise the instrumentation market is dominated by US, EU and Japan. This is not good because this actively limits the progress rate. The key isn't in just access to commodity instruments, its the capability to order a custom instrument.
Suppliers on the chemistry side (GC, LC, ICP, FTIR, Raman, gas monitors, pressure gauges) exist but they are small relative to major western/Japanese companies like Agilent/Thermo/MKS or even European/Japanese companies (Shimadzu, Hitachi, Zeiss, Horiba, Bruker). You really need at least a mid sized company here so people can trust your supply continuity and service.
Same with analytical SEM. No major analytical TEM companies. and , they're relatively small (benchtop XRD and XRF only, QC grade only, no TXRF for wafer, not analytical quality).
Good start, a long way to go. It must be noted this is a high tech, high barrier to entry, relatively low margin business (if at market rates; when monopolized, it has a mediocre margin). Maybe a bigger state backed player could do the trick.
3. the components. We've seen alot of progress here on this thread so this is something to watch for. There's good (ultraclean quartz and ceramic parts up to 300 mm wafer showerheads, good vacuum chamber design, etc) but some of the supporting capabilities like ultraclean fittings, valves, pumps, build to order vacuum chambers, large scale exotic materials machining (titanium, pure silicon) etc. needs work.
The semiconductor supply chain, like aerospace, automotive, petrochemical, pharmaceutical, etc. is heavily tied with basic machinery, optics, electrical and chemical manufacturing stuff that was started 100-150 years ago. Every country strong in the overall semiconductor supply chain (US, Japan, EU) was a developed country at the time. China is currently doing the impossible.
Last edited:
By the way, it's not my projection. It's not even @WTAN's projection. It's the Trump administration's projection. When they were formulating their technology blockade in 2018 they asked how long it would take China to develop its own EUV and the answer was eight years, i.e. 2026. And this is from people who underestimate China for a living.
It's a sobering downer if you believe this person knows anything about what he's talking about. Even if he does, it doesn't matter. Even if it's the consensus of everyone who's even looked at an EUV scanner, it doesn't matter. You know why? Because China. As I said, it surprises me that anyone could be dumb enough to still bet against China but here we are.
China is unleashing the power of 1.4 billion people, for me it's a numbers game. Given Asian cultures embracing the scholarly type redirects a lot of students into STEM things that enhance the material world.
Personally I will just keep Made in China 2025 as a target for some form of EUV that is productive enough and just profitable.
China has about 17 years left of 15-18 million potential university graduates entering the labor force annually. After that it starts to drop precipitously. In the 2040s the 1980s generation (the last generation before the decline in China's birth rate set in) starts to retire in earnest. If China hasn't leaped ahead technologically by then, it's in trouble. It could honestly extend this a little by winning more international talent and scientists to come to China, but it would have to improve living standards and overcome covid challenges for these people to immigrate.
- Status