Chinese semiconductor industry

Status
Not open for further replies.

FriedButter

Major
Registered Member
Please, Log in or Register to view URLs content!

Scientific Hypotheses and Huawei’s Business Vision for the Future​

Global Huawei Analyst Summit 2022​

To reduce power consumption and improve reliability of semiconductor components, we are also working with researchers to explore the thermal mechanism in semiconductor components to see if we can create conditions to accelerate the conversion of optical phonons to acoustic phonons, reducing the temperatures of the gate electrode and drain of semiconductors.
Chip design
In chip design, most of the constraints on components are local. In this area, Georgia Institute of Technology researcher Peng Yang and his colleagues created an advanced algorithm with a computational complexity of the 2.3316 power of n. For this achievement, Peng won the 2021 Best Paper Award from SODA, a leading symposium on computational theories. A few months ago, our mathematicians created a revised algorithm, further reducing the complexity to the 2.28 power of n, 0.0516 power of nless than Peng’s algorithm. What does this improvement signify? It means that if nrepresents one million, our algorithm can further reduce the computational complexity by about 45%.
Based on these four pairs of hypotheses and visions, we have identified two scientific questions and eight technological challenges that we can work on together, focusing on innovations that fall in the Pasteur’s quadrant.

  • How can we develop new processes that surpass complementary metal-oxide-semiconductor (CMOS) fabrication, cost less, and are more efficient?
1651025749109.jpeg

1651025817756.jpeg
 

bettydice

Junior Member
Registered Member
But Huawei currently is not in a good position to get into semiconductor manufacturing business. Besides the huge capital required, they also lost their huge smartphone business division. So any hope of becoming an IDM a la Samsung is unrealistic now.
Huawei's smartphone sales have shrunk but that doesn't mean Huawei lost any smartphone business division. What smartphone business division did Huawei lose? Then which division in Huawei is developing, designing, manufacturing and marketing all those new Huawei smartphones being launched?

The reason Huawei's smartphone sales have shrunk is because Huawei can't get access to semiconductor chips. The decline in smartphone sales is the RESULT of the problem of not having semiconductor manufacturing business. I don't understand how that conversely leads to the reason not to get into semiconductor manufacturing business. You don't make sense how "get into semiconductor manufacturing business" is not appropriate for a company whose biggest problem is lack of access to semiconductor chips. Samsung started semiconductor manufacturing business even before smartphone.
 

FairAndUnbiased

Brigadier
Registered Member
This is a major prospect I've been sitting on for a while too.

The question being -- by the time that SSA800 and the domestic 28nm supply chain and machinery are ready for commercial production and use by domestic Chinese companies including SMIC, would the production of 28nm node even be profitable? I suppose being a supplier for domestic companies and users who otherwise couldn't acquire 28nm nodes would still be very useful, and with multipatterning they could attain 14nm and give them some breathing space.

And in a national security and technological sanctioning context, continued 28nm production (or 14nm with multi patterning) is still essential.


But the question about whether SMIC and others are able to stay relatively profitable and expand and advance in context of a potential industry scale oversupply at the 28nm node, and get through to be able to reap the intended emergence of equipment allowing them to produce more advanced nodes, is something to consider.
There's no way around it, 28 nm has a ton of advantages in cost such as being essentially the best planar transistor before FinFET. FinFET requires new materials and processes that represent a large leap in cost.

However 28 nm also has a form factor advantage that makes it hard to get rid of. For instance many 200 mm fabs move up to 300 mm for productivity and thus transition from say 130 nm to 90 nm on more modern tools which also have an automation advantage. But 28 nm is already using modern 300 mm tools with full automation.

Also, 28 nm is already good enough or even overboard for many products like analog, microcontrollers, power, RF, optoelectronics, etc. So there's no real worry of 28 nm disappearing, it represents a "frontier" node (best planar transistor). Watch for the in-between nodes like 45 nm to decline instead.
 

FairAndUnbiased

Brigadier
Registered Member
At this point they are. They should still be in 2 years time. But I doubt they will get access to EUVL by then.
One thing that China probably should consider is to go into 450mm wafers. They are probably the only market that can justify that kind of volume. And 450mm wafer tools don't exist anywhere else. It is a huge fixed cost to get into it but requires little new actual technological development.
450 mm tools have significant challenges in etch and deposition tools due to uniformity issues. For example, to just maintain temperature uniformity over a 300 mm wafer at current process nodes you need 100 independently controlled temperature zones. Uniform heating impacts everything else such as plasma density, adsorption of inert species, etc.

Please, Log in or Register to view URLs content!

That's not all. For capacitive plasmas the etch rate is anisotropic. It peaks in the center. This was even a problem at 200 mm. The increase to 450 mm severely exacerbate the problem.

Please, Log in or Register to view URLs content!

The other problem is that lithography becomes the limiting step as exposure field doesn't increase, so each lithography step takes longer.

Please, Log in or Register to view URLs content!
 

Minm

Junior Member
Registered Member
is there a chinese company that makes 300m wafers at production scale?

SMIC, Huahong, Changxin, Cansemi, Nexchip, Tsinghua Unigroup, YMTC .. want more ? ;)

What is actually your intention with that question ? :rolleyes:
I think he's talking about the wafer production, not the chip manufacturing using wafers as an input. It's currently dominated by Japanese companies, but there's a Chinese competitor called zing semiconductor and the nominally Japanese ferrotec which is focused on production in China, has a Chinese company president and praises the made in China 2025 Initiative on its website

Please, Log in or Register to view URLs content!
 

FairAndUnbiased

Brigadier
Registered Member
You can increase the brightness of the light source to reduce the exposure time. Just like with any lithography machine.
but Chinese lithography producers are not leading edge, so I am not sure if they have the capability to increase brightness. Every manufacturer in general tries to have the brightest light source possible so what we see right now is the leading edge.
 
Status
Not open for further replies.
Top