Chinese semiconductor industry

Wahid145 said:

Wow, this looks huge! Does it mean that we don't need Tin droplet to generate the laser (which is incredibly complex).

Click to expand...

The abstract is light on the details without getting at the full paper, but my guess is not in a time frame before China succeeds in productionization of LPP or synchron SSMB light sources. Getting a EUV light source for lithography is a lot more than generating some EUV photons - it requires generating the light at a high enough power level, consistently, reliability and at a scale that can be used in a commercial production environment.

As with most things in science, it's usually the ability to move from lab to real world that makes or breaks cutting edge scientific discoveries(which is what this is)

tokenanalyst said:

Extreme Ultraviolet Laser by Single Photon Process​

Daobiao Hong
Chinese Academy of Science (CAS) - Shanghai Institute of Microsystem and Information Technology
Bingke Xiang
Fudan University
Tong Wu
Fudan University
Zhonghao Liu
Chinese Academy of Sciences (CAS) - Shanghai Institute of Microsystem and Information Technology
Zhensheng Tao
Fudan University
Yihua Wang
Fudan University
Shan Qiao
Chinese Academy of Sciences (CAS) - Shanghai Institute of Microsystem and Information Technology

Abstract​

Generating laser with short wavelength is a bottleneck problem in laser technology. The current applicable table top extreme ultraviolet (EUV) lasers are all generated by multi-photon process, with low efficiency and the record short wavelength of 113.8 nm do not meet some applications. By utilizing metastable helium atoms excited by microwave and irradiated by a resonant laser, here we report the development of a practical 58.4 nm laser by single-photon-excitation related anti-Stokes Raman scattering (ASRS). The conversion efficiency is much higher than that of high harmonic generation (HHG). The same divergence of 1.4 mrad as that of excitation laser indicates its stimulating character. Our results show an applicable path towards up-conversion by single-photon process to generate table top EUV lasers.

Please, Log in or Register to view URLs content!

Click to expand...

Any Raman scattering has a tiny cross section, even lower for anti Stokes Raman scattering. This is unlikely to be a high power source.

tokenanalyst said:

Looks like they are localizing as much as they can. Apart from lithography, metrology is, I think, probably the next most difficult, so the fact that they are buying, testing and verifying local made tools is encouraging.

Click to expand...

Is there a company that does validation for these tools? Or do the buyers have to verify the tools themselves with no 3rd party references?

latenlazy said:

For scanners you still would. The laser in the paper is 50-60 nm. Current EUV scanners are at 13.5 nm.

Click to expand...

Is an step to a more simple EUV light source and this could be useful for EUV inspection and metrology purposes.

HighGround said:

Is there a company that does validation for these tools? Or do the buyers have to verify the tools themselves with no 3rd party references?

Click to expand...

To my knowledge fabs themselves do the verification of the tools and materials that they buy.

HighGround said:

Is there a company that does validation for these tools? Or do the buyers have to verify the tools themselves with no 3rd party references?

Click to expand...

Typically, any tools whether digital, analog or mixed are built to a general set of requirements or specs intended for a range of uses within an industry or across industries. These would be verified by the OEM to perform within a certain range of capabilities or specifications. An end user would evaluate those capabilities and see if it meets their needs. The OEM would provide general instructions on how to operate the tool within those capabilities but it would be the responsibility of the end user to configure the tool and steps in operating it in conjunction with other tools to achieve the outcomes it desires. The exception to this would be tools specifically developed for a certain customer in mind, they would have been designed to fit with the customer workflows and desired outputs to begin with. Otherwise it is up to the user to squeeze the necessary performance out of it. For example, a 3D printer may be rated for certain materials, volume, etc but depending on the skill of the user it can be configured to print a shoe or aerospace parts.

paiemon said:

Typically, any tools whether digital, analog or mixed are built to a general set of requirements or specs intended for a range of uses within an industry or across industries. These would be verified by the OEM to perform within a certain range of capabilities or specifications. An end user would evaluate those capabilities and see if it meets their needs. The OEM would provide general instructions on how to operate the tool within those capabilities but it would be the responsibility of the end user to configure the tool and steps in operating it in conjunction with other tools to achieve the outcomes it desires. The exception to this would be tools specifically developed for a certain customer in mind, they would have been designed to fit with the customer workflows and desired outputs to begin with. Otherwise it is up to the user to squeeze the necessary performance out of it. For example, a 3D printer may be rated for certain materials, volume, etc but depending on the skill of the user it can be configured to print a shoe or aerospace parts.

Click to expand...

So not third party that does independent verification. OEM (Original equipment manufacturer) is the manufacturer of said tools.

Seems like there would be a market for that sort of thing, especially in China, with all that money pouring in and new companies flooding the space.

HighGround said:

So not third party that does independent verification. OEM (Original equipment manufacturer) is the manufacturer of said tools.

Seems like there would be a market for that sort of thing, especially in China, with all that money pouring in and new companies flooding the space.

Click to expand...

A third party can perform independent verification, but it is typically redundant because if you don't believe an OEM's claims about their product, you wouldn't be buying from them. You may test it our or ask for demos, but it would be rare to send it to a third party to verify (imagine buying a new car and then sending to a mechanic to go over it with a fine tooth comb). I could see that being more applicable to used tools, because you would want reconfirmation that everything is up to spec after it is refurbished.

Regardless, even if a third party confirms a tool is what is claims to be, the end user still has to figure out how to optimize its use to achieve the desired outcome. Its like how giving your average driver an F1 racecar won't give you the F1 performance, because they don't have the skills to use it to its full capability. Similarly, from what I can tell CXMT suffers not just from obtaining tools, but squeezing the performance from its existing ones. After all, until recently both CXMT and YMTC were able to access the same tools as their peers but YMTC got alot more performance out of theirs and was able to catch up to its peers while CXMT did not.

HighGround said:

So not third party that does independent verification. OEM (Original equipment manufacturer) is the manufacturer of said tools.

Seems like there would be a market for that sort of thing, especially in China, with all that money pouring in and new companies flooding the space.

Click to expand...

there is a market for 3rd party verification of components but not of whole tools. For deposition/etch/cleaning that's for things like metallurgy/cleaning QC and supplier qualification. I imagine for lithography or metrology they do some optics stuff that I don't know about. If you don't trust the OEM's tool will do what they say it will do, you ask for a tool rental to do a demo run at your facility. If you still don't trust them after the demo, then you just don't buy from them.

Current semiconductor tooling in China, what is being use, what is in verification and for what process, what node or industry.

tokenanalyst said:

Current semiconductor tooling in China, what is being use, what is in verification and for what process, what node or industry.

View attachment 113616

Click to expand...

seems like they have a lot of tools for 14nm, 128+ Layers & 17/19nm DRAM that are in the process of getting validated. Maybe we will get some good news about what they can do with de-americanized line over the next few months.

Just looking t this, I get the sense that 19nm production line with ASML scanners may be possible later this year. That would remove CXMT's vulnerability to being put on entity list.

speaking off the devil

In the past five years, records show that Micron has lobbied the US government at least 170 times to raise questions about China’s competition. Some Chinese memory chip makers put on Washington’s blacklist are either direct or potential competitors of Micron. Such tactics have backfired.

Click to expand...

170 times??? Micron is the absolute worst
nobody comes close

Micron stands to lose its 28 per cent lock on the Chinese market share of DRAM memory chips – which are used in all types of electronics from smartphones to computers – behind Samsung’s 43 per cent.

The entrance of the Micron Technology automotive chip manufacturing plant in Manassas, Va. Photo: AP
The move against the semiconductor giant, which may well see it eventually ejected from the Chinese market, involves particular bad blood.

Click to expand...

Well, some of these Chinese participants need to get used to CXMT DRAMs. Similarly, how much longer are people like Apple going to keep selling products in China with Micron SSDs and DRAMs?