Chinese semiconductor industry

Status
Not open for further replies.

hvpc

Junior Member
Registered Member
Ahhhh, I see.

I remember 10years ago I heard some asml guys talk about something that could be similar to this...was something like the size of a soccer field that could power more than 10 EUV systems. There was concern about environmental impact of something. But I guess they didn't go with this method. I wonder if it's the same as the SSMB technology that you are talking about. Have you heard about this before?

Let me do some research on SSMB to see if I can find anything useful for everyone.

It was proposed by Stanford in 2015, a breakthrough was archived in 2020 by a team from Tsinghua and a German institute. If you read previous comments we have discussed the idea here.

Please, Log in or Register to view URLs content!

View attachment 85354

View attachment 85355
Ok. I went back and read all the posting from the beginning of this year.

SSMB was indeed investigated by asml, but deemed not economically/environmentally/industrially viable. It was a novel idea that was floated around. My supplier told me it's just not feasible given the timeline of their product release timing. So, my inference is that this is not something that's easy to implement in reality. Sounds good on paper, but hard to industrialize.

Being in the semi industry for almost 30years, my view is not as optimistic as most of you on this forum. I've seen the industry transition from stepper to scanners, from iLine to KrF, to ArF, ArF immersion, and EUV. There is a lot of difficulties going from a concept in a lab to an actual product. Most people are still talking about power source "break through", or even the twin stage...but let me tell you, there's a lot more to it than that. Could SMEE assemble a system to image a wafer and take SEM images? Absolutely. I think delivery of those system today is believable. But to be able to build a system that could withstand the complexity of semiconductor wafer production within the next five years....I have my doubt.

I've evaluated Nikon's system and ASML's system....the hardware are about the same. I would even say the Nikon system's hardware is more robust. But this is not what set ASML apart from Nikon.

Also, I was told SMEE has no system in production, excluding backend applications. Not even their iLine system. So I don't understand how they could jump from having no system in frontend IC manufacturing to having the capability to ship a production worthy 28nm system? Maybe I'm just too conservative, but could someone help me understand why there's so much optimism here for Chinese domestic DUV and even EUV system by 2025? Everyone I talked to in the industry are not so optimistic...even those in China.
 

tokenanalyst

Brigadier
Registered Member
Ok. I went back and read all the posting from the beginning of this year.

SSMB was indeed investigated by asml, but deemed not economically/environmentally/industrially viable. It was a novel idea that was floated around. My supplier told me it's just not feasible given the timeline of their product release timing. So, my inference is that this is not something that's easy to implement in reality. Sounds good on paper, but hard to industrialize.

Being in the semi industry for almost 30years, my view is not as optimistic as most of you on this forum. I've seen the industry transition from stepper to scanners, from iLine to KrF, to ArF, ArF immersion, and EUV. There is a lot of difficulties going from a concept in a lab to an actual product. Most people are still talking about power source "break through", or even the twin stage...but let me tell you, there's a lot more to it than that. Could SMEE assemble a system to image a wafer and take SEM images? Absolutely. I think delivery of those system today is believable. But to be able to build a system that could withstand the complexity of semiconductor wafer production within the next five years....I have my doubt.

I've evaluated Nikon's system and ASML's system....the hardware are about the same. I would even say the Nikon system's hardware is more robust. But this is not what set ASML apart from Nikon.

Also, I was told SMEE has no system in production, excluding backend applications. Not even their iLine system. So I don't understand how they could jump from having no system in frontend IC manufacturing to having the capability to ship a production worthy 28nm system? Maybe I'm just too conservative, but could someone help me understand why there's so much optimism here for Chinese domestic DUV and even EUV system by 2025? Everyone I talked to in the industry are not so optimistic...even those in China.
I think their had been selling their i-line and their KrF for while, i even had seen some photos of production sites in China were one of the Scanners match their frontend offerings, the problem is because they are not a public company and the owners are the Shanghai electric group, SMEE is pretty obscure about sales of their products. Nobody knows really how much SMEE sell.
There have been announcement and rumors over the immersion scanner since early 2020, doing my research, i think they are developing an immersion system from a while back and the development had REALLY accelerated after the sanctions, some companies have been set for this project to mass produce subsystems and parts not only for the immersion scanner but also for the other systems they have. I Think SMEE planning to do an upgrade to all their systems, the SSB520 was an upgrade from the SSB500 with better resolution and accuracy. Others may follow.
The EUV issue just my conviction that they are putting an herculean effort to develop an EUV scanner, there has been a big output of immersion lithography patents in 2020-2021 by the lithography companies in China but not papers by research institutes, now they are just publishing EUV papers like the ones i posted before, so i just guessing that they could have a prototype by 2025. But even that could put a lot of pressure to ASML to ship their low NA EUV to China, who have fought hard to secure the Chinese market.
I think Export controls are usually adjusted depending potential of China of developing something, if not the best Scanner that China would have been allow to import would be a KrF scanner.

1647473951308.png

I am not 100% sure about anything in life but that is just how i view the situation. So don't pay to much attention to us we are just a group of enthusiasts.
 
Last edited:

tokenanalyst

Brigadier
Registered Member
Could SMEE assemble a system to image a wafer and take SEM images?
IDK if SMEE can but others sub-systems providers could, from my POV looks like SMEE is just acting as a integrator. Like i say before we here is this entire forum not 100% know about anything, here we just scout the surface of an ocean from the beach and that apply to everything. So do not pay to much attention to us we are just enthusiasts.
But to be able to build a system that could withstand the complexity of semiconductor wafer production within the next five years....I have my doubt.
The fear of U.S. sanctions is fueling the drive of China semiconductor industry. Only time will tell.
1647475600572.png

Please, Log in or Register to view URLs content!
 

latenlazy

Brigadier
Ok. I went back and read all the posting from the beginning of this year.

SSMB was indeed investigated by asml, but deemed not economically/environmentally/industrially viable. It was a novel idea that was floated around. My supplier told me it's just not feasible given the timeline of their product release timing. So, my inference is that this is not something that's easy to implement in reality. Sounds good on paper, but hard to industrialize.

Being in the semi industry for almost 30years, my view is not as optimistic as most of you on this forum. I've seen the industry transition from stepper to scanners, from iLine to KrF, to ArF, ArF immersion, and EUV. There is a lot of difficulties going from a concept in a lab to an actual product. Most people are still talking about power source "break through", or even the twin stage...but let me tell you, there's a lot more to it than that. Could SMEE assemble a system to image a wafer and take SEM images? Absolutely. I think delivery of those system today is believable. But to be able to build a system that could withstand the complexity of semiconductor wafer production within the next five years....I have my doubt.

I've evaluated Nikon's system and ASML's system....the hardware are about the same. I would even say the Nikon system's hardware is more robust. But this is not what set ASML apart from Nikon.

Also, I was told SMEE has no system in production, excluding backend applications. Not even their iLine system. So I don't understand how they could jump from having no system in frontend IC manufacturing to having the capability to ship a production worthy 28nm system? Maybe I'm just too conservative, but could someone help me understand why there's so much optimism here for Chinese domestic DUV and even EUV system by 2025? Everyone I talked to in the industry are not so optimistic...even those in China.
Tbh I think the SSMB concept is still mostly vaporware and will take time to materialize. But the reason why I think 2025 is a doable, albeit difficult, timeline for China to build their own EUV scanner is because they’ve already built a concept EUV scanner at the Changchun Institute of Optics all the way back in 2017, and seems to already have some prototype options for an industrial grade light source, so most of the work they have left to do just to have something usable is component and system integration. In terms of performance I would expect that such an EUV scanner is unlikely to have the same performance and reliability as ASML’s but if you’re SMIC or another Chinese fab it’s not like you’re swimming in alternatives, and some of the business pains from adopting a less reliable or performant scanner can be mitigated somewhat by government support.
 

ansy1968

Brigadier
Registered Member
can anyone help by telling me what city Huawei's 28nm fab is located? are they using all domestic manufacturing systems including SMEE's litho scanner?
@hvpc Shanghai in collaboration with ICRD, From engadget.com from the excerpt below:

Huawei might have a way to avoid some of the worst consequences of
Please, Log in or Register to view URLs content!
, provided it’s willing to be patient. Financial Times
Please, Log in or Register to view URLs content!
claim Huawei is planning a dedicated chip factory in Shanghai that would make parts for its core telecom infrastructure business. It would be run by a partner, the city-backed Shanghai IC R&D Center, and would be considered experimental until it’s ready to make chips Huawei can use.

The plant would start by making chips based on a very old 45-nanometer process before moving to 28nm chips by late 2021. That would be advanced enough to make chips for smart TVs and Internet of Things devices, the tipsters said. It would reach 20nm by late 2022, when it could make “most” of its 5G cellular hardware.

Please, Log in or Register to view URLs content!




Nov 1, 2020 — Financial Times sources claim Huawei is planning a dedicated chip factory in Shanghai that would make parts for its core telecom infrastructure ...
 

tokenanalyst

Brigadier
Registered Member
Tbh I think the SSMB concept is still mostly vaporware and will take time to materialize. But the reason why I think 2025 is a doable, albeit difficult, timeline for China to build their own EUV scanner is because they’ve already built a concept EUV scanner at the Changchun Institute of Optics all the way back in 2017, and seems to already have some prototype options for an industrial grade light source, so most of the work they have left to do just to have something usable is component and system integration. In terms of performance I would expect that such an EUV scanner is unlikely to have the same performance and reliability as ASML’s but if you’re SMIC or another Chinese fab it’s not like you’re swimming in alternatives, and some of the business pains from adopting a less reliable or performant scanner can be mitigated somewhat by government support.
We don't know man, they could be hitting gold or could be just vaporware. Maybe the researchers on Tsinghua could be just blind by their ambition. But in times of necessity there is only just one way to find out and money is non an issue by the urgency that China has.
 
Status
Not open for further replies.
Top