What should the fourth-generation lithography machine look like? What parameters should it have? These things have already been installed in Zhao Ye's mind at this time.
First of all, a lithography machine, as its name implies, must have light.
Light is a laser, and the lasers that can be used in industrial production only have a few wavelength bands. Currently, Zhao Ye still uses the same high-pressure mercury lamps that he used twenty years ago.
However, this kind of lamp has been updated several times, and the laser wavelength provided has now reached 436nm. The wavelength of the light source directly affects the chip manufacturing process.
To simply understand, this wavelength is like the width of a chisel. If you want to use a one centimeter wide chisel to carve a half centimeter wide line. This is almost impossible.
Almost everything is used here, but of course there are special circumstances. For example, Zhao Ye and the others have developed immersion technology. One millimeter of pure water is injected into the surface of the wafer, and the water's refraction of light is used to change the wavelength of the laser.
This technology is still top secret in the domestic lithography industry. All people who come into contact with and know this technology, to put it bluntly, are living under the supervision of the confidentiality department, but they don’t know or pretend not to know.
.
I say this because Zhao Ye wants to improve the manufacturing process, and the first step to solve is the problem of the light source.
High-pressure mercury lamps can still improve a few more times. This technology has not yet reached its limit.
Zhao Ye would not easily spit out chewing gum that did not taste clean. So the first task was to overcome the limits of high-pressure mercury lamps and bring the 436nm and 365nm wavelength lasers to the level of industrial production.
With the photolithography light source with a wavelength of 436nm, Zhao Ye can upgrade Lever's current process technology to between 800-1200nm, bringing Lever's high-end chip manufacturing process to the nanometer level.
His second wave is to advance the manufacturing process to 350-500nm. Correspondingly, a shorter wavelength, that is, a 365nm light source will be used.
It just so happens that 436nm and 365nm are the two spectral lines with the highest energy and the shortest wavelength in high-pressure mercury lamps. In this way, not only can industrial production be carried out more efficiently, but the performance of the chip can also be greatly improved.
Zhao Ye plans to use these two waves of high-end chips to directly impact the current chip market. As long as the timing of the two waves is properly grasped, Zhao Ye is confident that with just this, the photolithography machine manufacturers on the market will at least
More than half of it had to fall down.
The task of perfecting and developing mature high-pressure mercury lamps fell to Professor Dengquan, who has been focusing on lasers in recent years. Zhao Ye was not "stingy" this time, and gave everyone all the solutions that should be given. There is no reservation now.
Necessary, his latest plan is to abandon the mercury lamp route and develop a new light source system.
The new light source system group, without any surprise, fell on Xinran's head. This little girl, who is only 24 years old this year, is currently preparing to get married. For this reason, Zhao Ye personally found Professor Ci and asked him to do his grandson's work.
In the next one to two years, he is not ready to give Xin Ran a chance to leave the laboratory.
"Xinran, and everyone, next, your new project is to develop an excimer laser. Using this excimer laser, we can obtain high-power deep ultraviolet light that can be used for photolithography. At present, your task is
Make this kind of laser. After making it, we will select the wavelength we can use." After saying that, Zhao Ye began to distribute information to his students.
Looking at the information on the file, even Xinran felt his scalp numb. This kind of technical requirements is too high. At present, it seems that it is the same as alien technology.
Xinran was just shocked for a moment, then quickly calmed down and listened carefully to Zhao Ye's explanation of the technical difficulties and detailed ideas.
During Zhao Ye's explanation, Xinran and other researchers were shocked to discover that Zhao Ye, the instructor, had set the process accuracy below 350nm. This means that the wavelength of the light source they want to develop cannot exceed 300nm. This is also true.
, the goals given by Zhao Ye are the requirements of two bands, 248nm and 193nm.
This requirement has reached the technical level of his previous life in the late 1990s and the beginning of the new century. It is conceivable that in the current early 1980s, when the theoretical basis in all aspects is incomplete, the project team has to face
What a difficulty.
However, this did not discourage these researchers, because they did not know exactly what this meant. All they knew was that their mentors were back on the front line, and they could happily climb the peak of technology again.
These people have more confidence in Zhao Ye than Zhao Ye has in System Dad.
The light source is only the first step, and of course the most important step. Without this foundation, there is no need to think about other things.
After assigning tasks to the light source group, Zhao Ye's next target is the lens group.
At this time, the lenses used by Zhao Ye and others were products from an optical instrument research institute in Sichuan Province. On the photolithography machine they used, there were dozens of these lens groups, large and small, and the combined weight definitely exceeded a thousand kilograms.
Even so, Zhao Ye still put forward higher requirements. The lens itself also consumes energy. If it can not only improve the transmittance of the lens, but also ensure the flatness of the lens grinding, and reduce the need for specific wavelengths.
Laser loss is the biggest difficulty facing researchers and technicians.
For lens groups, not only researchers with good knowledge are needed, but also skilled workers with good craftsmanship.
For example, grinding lenses is basically done by hand at present. According to Zhao Ye's current requirements, for a lens with a diameter of 30 centimeters, the error range of the flatness is no more than three microns.
For this requirement, any machine is basically useless. We can only rely on the hands of workers to feel and polish bit by bit. In this field, there is basically no machine that can replace human hands. And, as it is now, in forty years
Years ago, it's still the same.
In terms of refinement, machines have never been able to surpass humans.
For this reason, Zhao Ye specifically applied to his superiors to merge this research institute into the Lihua Group. The staff of this research institute will not be transferred to the capital, which will only destroy the research institute.
What he has to do is to ensure the funding of this research institute in the coming days and take good care of the lives of these craftsmen living deep in the mountains.
What they need is quietness and tranquility. Prosperity can only make their hearts impetuous, which in turn affects their hands.
This may be cruel, but any profession has sacrifices. All Zhao Ye can do is make this sacrifice more worthwhile.
Regarding photoresist, Zhao Ye just told them the research goals of the light source team and asked them to develop corresponding photoresists respectively. Regarding this, Zhao Ye is actually the most confident. The purpose of arranging a special project team is to cultivate
Team. He can't do it by himself every time!
After all these things were arranged, Zhao Ye had only one other difficulty left, which was automation control.
The current photolithography machines have basically broken away from the scope of manual operation. No matter in terms of efficiency or accuracy, manual operation can no longer adapt to new requirements.
This is the most difficult part for Zhao Ye.
At present, Jingyi Machinery, which once rejected itself, is no longer in use. In the past ten years, they have hardly made any progress. The things they produce are still the same old, and no technological progress can be seen.
Shenji is better, but it is just better. Because there are many military tasks, some technological progress has been maintained to some extent. But in terms of automation control, there has been no upgrade for many years. Now they just want to pick up this area, more or less
The page is a bit difficult, because not only technology has experienced a generational decline, but talent has also experienced a generational decline.
Currently, if Zhao Ye wants to engage in technology research and development in this area, he has only two options. One is the Automation Research Institute in Zhennan, and the other is the Xi Er Machinery established a few years ago.
At present, the advantage of automation lies in its technology research and development capabilities, and the advantage of Xi'er Machinery lies in its team of technicians and front-line experience. Zhao Ye is an adult, so he has to make every choice.
These two units have a good relationship with Zhao Ye, so as soon as Zhao Ye's telegram for help was sent out, both units mobilized their best forces to provide support.
In fact, Zhao Ye already knew what the overall lithography machine should look like. He even sorted out the technical difficulties and solutions through the system.
In his virtual laboratory, these are not difficult tasks. The real difficulty is how to turn these things into physical objects and into real production equipment.
In a brand new test workshop, teams sent by the two companies are gathered here. The first thing they have to do is to build a reliable platform for their designs.
In the past, or in Zhao Ye's previous life, design and production could almost be placed in different places.
For example, for a piece of equipment, a designer can sit in an office building, conceive and calculate to his heart's content on a computer or drawing. Then he can realize the capabilities required by the design bit by bit and solve various difficulties in practical applications.
But in Zhao Ye's case, this doesn't work.
The reason is very simple. You can draw the drawings like this, but the factory may not be able to produce it. Now what Zhao Ye and the others are doing has almost surpassed the current mainstream production and processing capabilities in the world. In China, even a decent precision machine tool
No, no matter how good your design is, it may not be possible to realize it.
So what should we do? First, in theory, complete a basic design idea to prove that this method can achieve the functional requirements. Then based on this design idea, determine the required equipment. Buy what can be bought, and build what cannot. This is
It’s not over yet. Before these things, everyone needs to build an experimental platform to prove whether the theories of each sub-project can be formed and whether they are correct.
This is just about machining. A series of problems such as measurement, control systems, various sensors, etc. have to be solved by people like them alone.