Cang Yaozu still knows a lot about the development of lithography technology, whether it is immersive lithography or extreme ultraviolet lithography EUV, but there is no one in Cang Yaozu's hands. It seems that you have to fool Lao Yang. Give Lao Yang eye medicine.

When it comes to lithography technology, I can't fail to mention Intel, the advertisement of Intel Pentium inside is very awesome, give the computer a galloping core, hehe, no, you have to mess with Intel.

Cang Yaozu directly found Lao Yang's office, not Yang Chongshan, he can only be regarded as Xiao Yang, Lao Yang is Yang Sizheng, the principal of Qing University, and then directly teased, no, it was chatting: "Lao Yang, in less than 20 years, the domestic semiconductor import value will definitely exceed the scale of 500 billion Chinese yuan, and the global semiconductor output value will at least exceed 3 trillion Chinese yuan. ”

"Aren't you a sinner?"

"Of course I'm not, I'm not in the semiconductor business. ”

"Well, you're really not in the semiconductor business. But the lithography machine is not easy to do. People have a first-mover advantage, but now it is a market economy, and everyone puts their interests first. We have developed a generation of lithography machines, and people have sold lithography machines that are stronger and cheaper than us into China, they are a virtuous circle, and we are a vicious circle. R&D has fallen into a source of no water, and I can't afford to invest it. Enterprises also hold the idea that it is better to buy than to build, and it is better to rent than to buy, and they do not want to invest in research and development at all. ”

"Lao Yang, do you know the boiled frog in warm water? We Huaxia is this frog, when the research and development of the lithography machine is completely cut off, people will not sell it to you, not only the machine will not be sold to you, the chip will not be sold to you, you should know, the cold war, the trade war, the local war, the probability of these occurrences is 100%, when the time comes, will those companies that are out of cooking poke you Qingda and the backbone of the Chinese Academy of Sciences and scold your mother. ”

"By that time, I'll be retired, maybe I'll be in the ground, and I won't be able to scold my head. ”

"Hehe, even if you can't hear it, it will hang you on the pillar of shame in history. It is indeed the market instinct of enterprises to seek advantages and avoid disadvantages, but can the country have the same thinking as enterprises? Can universities have the same thinking as enterprises? If this development continues, our semiconductor industry in China will become a cow that only knows how to plough the land, but the nose ring is in the hands of others.

"Don't tell me these big truths, I know more than you, you kid just tell me what to do, I know you're going to do something again, come, come, my old Yang is waiting, waiting for you to take the money and stone me to death." ”

"Beautiful to you, why should I take the money, can I afford it? The 500 billion, trillions of dollars market will not be given to me, I can't do it, just the so-called government in his position, I am neither the leader of the country, nor the president of Qing University, let me take the money? I can't take it. I'm going to give you ideas today, so that you can live with a clear conscience for decades to come. ”

"Okay, okay, you say, as long as your idea is useful, I will go to ask grandpa to tell grandma, go and kneel, and I will go and get the money. ”

"No need, you have so many alumni, you see how much money Yan Post has raised, your size, it is impossible not to raise some funds, start it first, and then talk about it, there is a state and a bank in the back." There are only two ways to break this vicious circle, one is that the state is desperate to throw money at all, and it doesn't care how it loses, and can it lose 500 billion yuan a year? In this way, we should concentrate our forces to do great things, mobilize all the forces that can be mobilized, do not play cards according to the routine, develop in overdrive, and catch up with them in one fell swoop. The second way is to wait, when the leader encounters a bottleneck, they have a hurdle that is difficult to overcome, and you grit your teeth and throw money to chase after it, you will definitely be able to catch up. ”

"You said it lightly, the country is now a hole everywhere, and it has to develop everywhere, and the semiconductor industry cannot be invested by investment. Your second option is even more unreliable. ”

"The first way is to have strength, the greatest force. The second method is really reliable, and they are about to hit a roadblock. It is widely believed that 193nm lithography cannot be extended to the 65nm technology node, and 157nm will become the mainstream technology. However, 157nm lithography has encountered great challenges from lithography machine lenses. This is due to the fact that most materials strongly absorb 157nm light waves, so this is their problem. ”

"Tell me carefully. ”

"Now several large foreign manufacturers are using DUV, that is, deep ultraviolet light, through the krypton fluoride KrF excimer laser to produce a wavelength of 248 nm light, this lithography machine is now the mainstream, their laboratory has now basically developed the argon fluoride ArF excimer laser to produce the next generation of light waves with a wavelength of 193 nm, 193nm lithography machine is no problem. But using the F2 excimer laser to produce light with a wavelength of 157nm is a big pit, and they can't get by, and the technology can't solve it. The bottleneck has appeared, and the opportunity for us to catch up has come. ”

"And what's their solution?"

"Research and development of EUV lithography machine, that is, extreme ultraviolet lithography technology. Extreme ultraviolet lithography uses soft X-rays with a wavelength of 13.5 nm to lithography, which is equivalent to using a smaller, sharper 'knife'. Extreme ultraviolet lithography is a necessary process for more advanced processes such as 5nm. However, because EUV cannot penetrate the lens, it can only be focused with special mirrors, so the energy loss of EUV will be large, and it will require a longer exposure time, which will seriously affect the production speed, which is why it is difficult for EUV lithography machines to be mass-produced. Therefore, I am convinced that EUV lithography technology will not be able to complete the prototype in 15 years, and there is no hope of mass production in 20 years. ”

"Are you sure?"

"I'm pretty sure yes, yes. The information I learned is that the Dutch ASML company will cooperate with several companies in Citigroup to solve the problem of EUV, and they will not take us to play, of course, they do not play on behalf of Japanese manufacturers. Nikon and Canon will die 157nm lithography machine, this is a dead end, and when they decline, we can join them to build factories in China, absorb their technology, and then develop EUV lithography machine together, 7nm below the process, DUV lithography machine will not work. ”

"Wait, you say DUV lithography can't reach below 7nm? Do you mean that DUV lithography can break through 65nm, or even 28nm, 14nm and even 7nm?"

"Yes, yes, even 14nm, 7nm is fine, but it's more troublesome. ”

"How? How did you do that?"

"You don't care how you research. You can't say what I'm going to say next, including on the Internet, you see that I came to your office to talk to you, and you don't even need ICQ, you know the seriousness of the problem. ”

"Is it necessary to be so cautious?"

"Of course it's necessary. Previously, the upgrading of the chip manufacturing process was based on a three-year cycle, but Intel took the lead in shortening this cycle to two years. Intel achieved mass production of 0.35 micron process in 1995, 0.25 micron product in 1997, 0.18 micron process in 1999, and 0.13 micron product mass production in 2001. Intel is faithfully practicing Moore's Law. ”

"Every two years, this is going to kill AMD. ”

"That's good, so that we can win over AMD and fight against Intel. However, this may only be a fantasy, there must be a tacit understanding between Intel, AMD and Microsoft, they will join forces to kill everyone else, and then they will play slowly themselves, but Intel and AMD will not be played to death, otherwise they will be suspected of monopoly. ”