And there’s your issue. Digitimes is very tabloidy.
Chinese semiconductor industry
- Thread starter Hendrik_2000
- Start date
- Status
It seems hard to change our friends mind here.
Btw, people are lucky I am out today or I would have started deleting some posts already
That's hard to say. Depends how they implement on their threats. But banning ga effect on defense industry and new energy industry is massive. Again, let's see if they are willing to actually go forth with it fully.
At this point, you have already put in motion desires in western countries to source alternative to Chinese suppliers. As such, I see no reason to hold back on full implementation.
Not really. You can select cymer or gigaphoton. That's just standard choice any customers can make.
If asml onboards Chinese laser partner, then that's possible too.
Until asml says they expect noticeably lower sales to china (which would happen if they can't sell duvi to smic or ymtc), I don't see anything here as real obstacles.
Now the question is what happens if the sanction extends to all duvi product either by dutch themselves or Japan with banning of gigaphoton lasers.
My pt is that china isn't far away from their planned 14nm solution. So how much they get effected really depends on much they stock up and such. They are clearly stocking up in anticipation of this announcement so I expect q2 and q3 deliveries to look pretty good for asml.
Beyond that, it's up to smic to figure things out how to set process with domestic tools and maybe a couple of asml scanners.
As for lack of servicing, that will obviously be problematic, but they have been dealing with it wrt to American tools for a while.
if ASML refuses, then it must realize that its IP may no longer be enforceable, and the non-enforceability of its IP may not be limited to replacement of wear parts on their instruments.
This was already happening anyway
---
Europe is in a tougher situation when it comes to developing its own germanium output. Only zinc smelters using a specific process can recover it, but very few outside of China use that method and none in Europe do.
Trafigura-owned Nyrstar is among those assessing building a $150mn germanium and gallium recovery and processing facility at its zinc smelter in Tennessee, which could cover 80 per cent of US demand but would take at least two years to construct. Umicore, a Belgian advanced materials group that recycles germanium, said it is developing technologies based on “thin film” germanium to reduce use of the material.
Source
archive.ph/CI1nX#selection-2305.0-2309.412
---
So you might as well use Germanium and Gallium as a bargaining card now for the next 2-3 years.
By that time, you would expect Chinese DUVi equipment to be in production and equipping new Chinese fabs.
D
Deleted member 24525
Guest
SMIC has already proven its ability to continue 14/7nm production at workable yields despite lack of service for AMAT/LAM products. I wouldn't be too worried about the maintenance issue.
Spare parts are more problematic but they still have options. I remember reading from an industry analyst that China had something like 47 stockpiled arf machines as of late 2021. If they've kept the stockpile at that level then they could potentially cannibalize parts from spare machines if necessary. It's also much easier to localized individual components than the whole machine though that would take more time obviously.
More broadly, as @tphuang says, by this time next year China will very likely have indigenous 28nm capability and a year after that 14nm. SMIC only needs to keep things running for two more years and then the DUV problem will go from a technology issue to a capacity one, which China is excellent at quickly resolving. This alarmism just isn't warranted.
Spare parts are more problematic but they still have options. I remember reading from an industry analyst that China had something like 47 stockpiled arf machines as of late 2021. If they've kept the stockpile at that level then they could potentially cannibalize parts from spare machines if necessary. It's also much easier to localized individual components than the whole machine though that would take more time obviously.
More broadly, as @tphuang says, by this time next year China will very likely have indigenous 28nm capability and a year after that 14nm. SMIC only needs to keep things running for two more years and then the DUV problem will go from a technology issue to a capacity one, which China is excellent at quickly resolving. This alarmism just isn't warranted.
i am not sure if that is a good idea. US and western countries can then ignore the rare earth processing patents.
Until ASML is compelled to completely stop doing ABSOLUTELY ANY business with China. That will very likely soon happen, regardless of the consequences. The Americans political elites don't care about any of China's retaliatory capabilities on trade of anything - gallium, germanium, rare earths, etc - and the Europeans, Japanese, and other allies will either meekly or enthusiastically abide by them... SMEE and other Chinese IC chips and semiconductor related equipment manufacturers better hurry the entire indigenous development of all products related to high DUVi machines...
they can try, but the physical equipment is in China, its a complex process involving many tricky unit operations to separate elements of similar chemical reactivity, and many of the secrets in rare earth processing aren't patents but are trade secrets.
, they haven't because they can't, not because they're merciful. If it could harm China more than it would harm themselves they would've done it already.
- Status