Chapter 301 298. Isn’t this a must-lose game?

Zhang Rujing said, "We are already working on the next generation of chip production processes, but this must take a time process, and it is probably not that fast."

"The lithography machine we are using now is the world's most advanced 193-nanometer wavelength lithography machine, the same as TSMC's."

Speaking of this, Zhang Rujing thought about it, "This is equivalent to everyone being carving people, the tools they use are similar, and the quality and precision of the products they make are based on the level of craftsmanship.

If we want to produce the next generation of 45nm chips, judging from the current equipment, what we can do is to improve our technology, at least next year we can break through 60nm."

Zhang Rujing took a sip of wine and his face turned slightly red.

"Since SMIC has several ways to improve the precision of chips, one is to obtain chips with higher precision through multiple exposures."

"Hey?" Cao Yang suddenly raised his head, feeling a little shocked.

Because he had never heard of this technique in his previous life.

Is there any such technology?

"This method can indeed produce chips of 45 nanometers or even lower nanometers, but the disadvantages are obvious," said Zhang Rujing. "Multiple exposures have a geometric exponential increase in technology requirements.

It is equivalent to using the knife to cut in the same place when doing micro-carving, and each knife must be extremely accurate.

This is a very difficult thing. At the same time, due to multiple exposures, the yield rate will definitely drop sharply, resulting in an increase in production costs. I estimate that even if we can surpass TSMC and produce 30 and 20 nanometer chips, if such a chip is sold to Bomi, the price may be as high as more than 1,000 or even 2,000 yuan, which is probably difficult for you to bear.

It is difficult to mass production, which is the biggest problem encountered by multiple exposures at present, but we are also exploring. Perhaps our multiple exposure technology will be able to make breakthroughs in a few years, and mass production may be carried out at that time."

Um...

Cao Yang held his chin and fell into thinking.

In fact, he is calculating a problem-

Now SMIC gives Bomi chips 90nm, 380 yuan per piece, and 45nm per piece, or even 30nm, we still have some profits. We will increase the price a little more, and use the dimensionality reduction to crush Apple to establish an advantage and then promote chip manufacturing in reverse?

Is this road feasible?

"There is another kind," Zhang Rujing was a little excited today, and continued with a smile, "the immersion technology fills the lower surface of the last lens of the photolithography projection objective lens and the photoresist on the silicon wafer with a high refractive index liquid.

This is the photolithography method that TSMC is now using.

It is also interesting to think about it. Four years ago, when TSMC Linmoma proposed the infiltration technology, the industry even sneered at it, believing that this is an impossible technology method.

However, two years ago, their immersion technology was successfully studied. Now, because we do not have specific production processes and techniques, we can only rely on ourselves to explore it."

Cao Yang still has some impression of this.

The most important reason why TSMC has opened up a gap with other major manufacturers is that the immersive lithography technology invented by Lin Benma.

Before this technology emerged, Nikon, the world's leader in the field of micro-film, was Higashima. When Lin Benma proposed the feasibility of immersion technology, it was equivalent to slapping Nikon and IBM in the face.

Because they are still studying how to use X-rays to perform lithography, because X-rays are shorter than deep ultraviolet rays, but Lin Benma told them that it is actually not that complicated. Based on the current deep purple DUV lithography machine, add a little water to soak it, and change the refractive angle can create a chip with higher precision.

This undoubtedly tells them that the path you are taking is wrong.

How can IBM and Nikon stand this?

They have invested dozens or tens of billions of dollars in research and development funds, and have been suddenly questioned.

The emergence of any disruptive new technology will always be subject to huge resistance from traditional forces. Lin Benjian's immersive lithography has been banned by almost all giants such as Nikon, Canon, IBM, etc. Nikon even put pressure on TSMC to hide Lin Benjian.

The fall of giants is always the same. Kodak first produced digital cameras, but Kodak made a decision because of his fear of digital cameras threatening his film business - he must hide it well and not let others know.

Nikon's IQ is exhausted before the huge existing economic benefits.

Although Nikon still occupies the leading position now, Cao Yang knows that in the near future, when TSMC's 45nm appears, Asma lithography machine will pull Nikon off its market share and cause a giant to die.

Now there are actually such signs.

Although SMIC knows that Lin Benma has used immersion technology to help TSMC achieve major breakthroughs below 60 nanometers, SMIC does not know the specific practices and process technologies.

So now it is also the stage of crossing the river by feeling the stones.

"Immersion technology is not complicated in principle, but it is still the same problem and takes time." Zhang Rujing took a sip of wine with a little sadness and said uncomfortable, "We are trying to use immersion technology, and it should not be a problem to increase it to the 60-nanometer level next year."

Using immersion technology to make chips does not require multiple exposures. The yield rate is guaranteed, and the production cost should be similar to the current 90-nanometer chips.

So even if Bomi does not have a 45-nanometer chip next year, it can still buy 60-nanometers from SMIC.

However, Cao Yang will definitely not be satisfied with this. Bomi's chip design needs to continue to dig deeper and keep leading over Apple. Then, TSMC's 45-nanometer chip must be introduced first.

Unless SMIC can overtake.

If we want to achieve technological breakthroughs and surpass TSMC, we have to make breakthroughs from the lithography machine itself. However, so far, there is no company that can independently produce lithography machines. To be precise, there is no independent production of lithography machines for 193 nanometers or even shorter bands.

It is now 2006. In the field of lithography machines, Nikon from Dongdao is the leader, and the future lithography machine king ASMA Amsl has not yet risen. The first EUV lithography machine will not be produced until 2012.

In terms of time, maybe we still have a chance.

At least that's what Cao Yang thought.

...

"The infiltration method is to achieve the purpose of shorter light waves by changing the refraction of light, right?" Cao Yang asked.

"Yes." Zhang Rujing smiled, "This is a junior high school physics problem. The wavelength of refracted light is equal to the wavelength of incident light divided by the refractive index. The larger the refractive index, the shorter the final refractive wavelength.

Now the immersive lithography technology is to separate short wavelengths by increasing the refractive index and refractive angle of the incident medium, thereby improving the fineness of the chip."

Thinking of this, Cao Yang smiled slightly, "Then why not use light with shorter wavelengths directly?"

Zhang Rujing shook his head, "No, because IBM has studied it before. X-ray waves have shorter wavelengths for deep ultraviolet light now, but in this case, the refractive effect of water medium on X-rays is very poor. When the light reaches a certain wavelength, water is not completely transparent to them, and occlusion and noise will occur. This is something that cannot be solved."

"No, I mean, wet lithography is not used, but dry lithography, but shortens the light wavelength to double digits," said Cao Yang.

"ha?!"

Not only Zhang Rujing, but all the big guys here all looked like they saw a ghost.

Ni Guangnan and Liu Qiang are from chips. Although Huawei and Mr. Ren have not been engaged in technology research and development in this area, they have a little knowledge. Moreover, chips are equivalent to the heart and are indispensable for the entire high-tech field. They can deal with chips no matter how they have not done them, and they may have some understanding of them.

this……

You should know that it is now the limit to be able to produce 193 nanometers of deep ultraviolet light, and deep ultraviolet light can be increased to up to 157nm.

If the infiltration method is used and pure water is used as the immersion liquid, its refractive index is 1.44, so the equivalent wavelength drops from 193nm to 134nm (193nm/1.14=134nm), and the resolution is also improved, and the process nodes can move towards 45-22nm.

Now Cao Yang actually said that he used a double-digit number, that is, a wavelength light source below 100 nanometers, which is simply a fantasy.

So Zhang Rujing was stunned for a moment and laughed, "If there was really such a light source, we could produce single-digit nanometer-scale chips."

At that time, the improvement in chip processing performance brought about can be described as a leap.

"Yes, now China can't even create stable deep ultraviolet light. Where can I get such a short-wavelength light source?" Professor Ni also said.

Um……

Now not only does not exist in China, but also in foreign countries. It will not appear until ASML acquires Cymer, a top light source company in 2012, extreme ultraviolet stable light sources with a wavelength of 13.5nm in light waves.

This is a difficult problem for the whole world, and it also means opportunities.

But now is there any way to grasp it.

Cao Yang's impression is very vague. After all, he did not study these majors in his previous life. It was just that because of the sanctions of Huawei around 20 years ago, he learned a lot about lithography machines, and he only had a slight understanding. He only understood the literal meaning of some principles, and stayed at the level of comparative theory, so he did not understand the specific content at all.

but……

He thought he could wait a few days to ask at his alma mater. If Cao Yang remembers correctly, in fact, by 2021, our country's Tsinghua University will be able to independently develop stable extreme ultraviolet light.

It is less than 10 years behind AMSL.

"Hahaha, I just have this idea, I haven't thought about how to implement it in detail."

Cao Yang just smiled and did not continue this topic. It felt more feasible to bring together the big guys to jointly develop it.

...

After chatting for a while, everyone had high expectations for Bomi's smartphones and was very optimistic about Bomi and the future prospects of the smartphone industry.

It was almost 9 o'clock in the evening that everyone left the seat.

Cao Yang sent the big guys into the car one by one.

Finally, when seeing off Zhang Rujing, Cao Yang hesitated and said, "Mr. Zhang, I heard that the lawsuit between TSMC and SMIC recently seemed to be quite serious. Is this the case?"

Zhang Rujing was stunned for a moment, and a cloud of sorrow appeared on his face.

In Cao Yang's opinion, SMIC's biggest trouble is that it is in a lawsuit, and Zhang Rujing has been accused by TSMC of infringement of intellectual property rights.

Because in terms of equipment, SMIC currently uses the same DUV lithography machine as TSMC, the only difference is in the immersion technology and production process.

If Zhang Rujing did not leave SMIC due to TSMC's accusation in his previous life, and then SMIC entered an era of no leader and serious internal friction, but was devoted to scientific and technological research and development, then what would it be possible to catch up with TSMC?

SMIC is really a tragic future. Founder Zhang Rujing has been in a lawsuit and cannot devote himself to technological development.

Later, after Zhang Rujing left SMIC, SMIC encountered many tragedies, the founder left, and the old team fell apart. During the Tang Dynasty, he took advantage of the position of major shareholders and played Game of Thrones.

It can be said that if there is less trouble, SMIC should not be behind for so many years in terms of chip manufacturing.

"Yes..." Zhang Rujing smiled bitterly in frustration, "This is all an old grudge with TSMC. I guess they will stop until I leave SMIC."

"Never do not," Cao Yang said quickly, "SMIC needs you. Giving SMIC a stable development environment is more important than anything else."

"Yes..." Zhang Rujing sighed.
To be continued...