Military Industry of Great Powers: Build the Strongest Flanker

Entering the photolithography machine workshop of Tsinghua University for the first time, Zhao Guoqing was a little excited and went straight to the topic, a few key technologies.

"Professor He, what kind of light source does your lithography machine use?"

"You really know how to use photolithography machines!"

Zhao Guoqing smiled: "It's introduced in the book!"

He Zhenhua said: "The high-pressure mercury lamp source used is a 436-nanometer blue mercury g-line. This type of high-pressure mercury lamp source has now reached the international advanced level!"

Well, at the international advanced level, the world has begun to study purple laser power sources with shorter wavelengths ten years ago.

Zhao Guoqing thought of the KMBBF crystal. In his previous life, he discovered that the KMBBF crystal showed superconducting phenomena at minus 213 degrees Celsius.

Just when he used X-rays to penetrate the KMBBF crystal and wanted to detect the molecular structure of the KMBBF crystal, he discovered that the KMBBF crystal that entered the superconducting state could actually reflect X-rays.

This is an incredible discovery.

Because X-rays have strong penetrating power, currently no lens can allow them to shift direction. However, if the KMBBF crystal can reflect X-rays, it can reduce the image of the mask through a concave mirror, which will subvert the light source of the lithography machine. of change.

It's a pity that after discovering this matter, I gave up soon and did not continue to study it.

Now, though, he has plenty of time to continue his research.

Professor He Zhenhua continued his introduction, and from what he heard in his tone, he was far ahead.

"The internationally advanced non-contact lens exposure method is used, and the imaging quality is high. Compared with the contact exposure produced by H Da three years ago, the service life of the mask can be significantly improved, and the lens exposure can be compared with the mask. With a 5:1 ratio of photolithography patterns, more complex chips can be manufactured."

Zhao Guoqing nodded. Although he was not an expert on lithography machines, he could be said to be well-informed.

He has been in chip manufacturing plants of Intel and IBM. From the 1990s to 2023, he is still aware of the evolution of foreign lithography machines.

"With a 436-nanometer blue mercury g-line light source and a non-contact mask exposure process, it is theoretically possible to manufacture chips with a one-micron process!"

If multiple exposures are used, it is even possible to manufacture chips with a 500-nanometer process, or a water film infiltration process. Of course, the technical difficulty is higher, and it is possible to manufacture 250-nanometer chips. There is no way to say this.

As for the use of KBBF crystals invented in the 1990s to upgrade extreme ultraviolet light and duplex tables to improve commercial manufacturing efficiency, they are still out of the question.

Can you start the layout now?

If you haven’t said it yet, you will definitely do it!

Professor He Zhenhua did not answer Zhao Guoqing's question, but Zhao Guoqing also discovered some problems.

"The calibration is done manually!"

The photolithography process is not a one-step process and sometimes needs to be repeated multiple times. Therefore, if there is a slight change in the position of photolithography again, it often means that the manufacturing fails, so calibration is required.

The calibration method is to leave a mark on the wafer and use this mark to calibrate.

He Zhenhua shook his head.

"The most advanced Modu motor factory in China does not have suitable servo motors. The domestic light pressure sensors are not sensitive enough and can only be calibrated manually. We use a three-axis manual calibration device, and the error can be accurate to one hundred nanometers."

Zhao Guoqing shook his head. This photolithography machine had no commercial prospects at all. The XYZ three-axis manual calibration, although I didn't watch them operate it, would definitely take a long time.

"Photoresist, etching hydrogen fluoride..."

"Imported from Japan?"

"Where's the lens for exposure!"

"Imported Japanese Nikon!"

Well, Zhao Guoqing probably understood it.

“Is it because of the aluminum wires connecting the transistors that it is impossible to manufacture chips with a one-micron process?”

He Zhenhua looked at Zhao Guoqing in surprise. He did not disclose this matter to Zhao Guoqing at all.

Gao Changming also sighed. Xiao Zhao refreshed his knowledge again and again.

Just look at He Zhenhua's expression and you will understand that when they first met, he had a condescending attitude. Today, he has treated Zhao Guoqing as an equal, and now he seems to be looking up to him a little bit.

"How did you know?"

Zhao Guoqing said nonchalantly: "There are no problems with the light source, lens, calibration, colloid, and etching materials. It can only be a problem with the aluminum interconnection of the transistor. Everyone knows that the silicon substrate and metal aluminum will diffuse, and the smaller the process. The spread gets worse until you can no longer work!”

"If my guess is correct, your three-micron or even five-micron chips may not be able to achieve the designed frequency!"

"..."

Everyone knows, why don't I know? Gao Changming is speechless.

"How did you know?"

Zhao Guoqing said: "Professor He, you can look through the 1974 "Microelectronics Industry" journal, there is an introduction on it!"

"..."

Once you know the crux, it will be easier to solve it. Zhao Guo smiled and said to He Zhenhua: "Professor He, I have a solution, do you believe it?"

"Do you have any idea?"

"Just add a layer of isolating metal!"

A layer of metal separates the aluminum from the silicon to prevent diffusion from each other.

He Zhenhua shook his head and said: "We also thought about using a metal isolation layer, but we haven't found a suitable material yet!"

Sometimes it's just such nonsense. Many times, just because of a small thing, technological progress has been blocked for many years. Only after it is made public do we realize that it is bullshit.

"Professor He, try using tungsten. The metal atom still you used for light-emitting semiconductors last time can be used. On the silicon substrate on the wafer, first deposit a layer of tungsten film, and then add an aluminum wire layer."

"Is it really possible?"

He Zhenhua looked straight at Zhao Guoqing. If it was really possible, it would be of great significance to directly progress from a five-micron process to a one-micron process.

Gao Changming couldn't sit still either.

"On National Day, the Ministry of Science and Technology has just issued a document, which states that we need to consolidate five microns, tackle three microns, and move towards a one-micron process. With your words that are neither salty nor bland, the document issued by the Ministry of Science and Technology has turned into a joke."

"Do you have this document? I thought we could produce chips with a one-micron process?"

"..."

He Zhenhua shook his head.

"It's not a hassle. We have all the equipment. Let's give it a try!"

…………

Zhao Guoqing took a long breath. There was only so much he could do to help. However, these lenses and photolithography materials were all imported from Japan, which was quite worrying.

There is also process automation. If it is not solved, the photolithography machine project will never be commercialized, which means that there will never be automatic hematopoiesis capability.

Zhao Guoqing said to Gao Changming: "Director Gao, if possible, Institute 14 still needs to purchase several Nikon photolithography machines."

Today's ASML is still a junior, and the most advanced ones are Japanese lithography machines. It is still necessary to get a few of them for reverse engineering and accumulate some technical foundation!