Chapter 182 How to Make a Single Crystal (3k)

Duan Gong took the phone again and said to Gao Zhendong with a smile: "Director Gao, I won't say anything else. I'm bothering you again. Don't worry, I promise that whenever there is an order for your graphite components, I will personally keep a close eye on it."

This is a good thing, as it can ensure the source quality of single crystal furnace graphite components to the greatest extent.

The resources for confidential calls are still very valuable. After finishing the conversation and wishing each other a happy new year, they hung up the phone.

The Carbon Research Institute did not ask about other issues in the production of graphite materials, and Gao Zhendong could not actively talk about them. Firstly, perhaps their choice was better. Secondly, there was a confidentiality regulation. Even as a technical consultant, he could not ask questions other than the information that the other party had actively provided.

Look at the calendar, New Year's Eve is only two days away.

Gao Zhendong came out of the confidential room and did not return to the office, but went to the single crystal furnace test laboratory.

Yu Yuncheng is in charge of the specific affairs here.

With this professionally trained person around, Gao Zhendong felt relieved. He was ahead in ideas and principles, but the specific details were not necessarily better than Yu Yuncheng's.

Yu Yuncheng is more suitable to be responsible for the specific implementation process than him.

Seeing Gao Zhendong coming over, Yu Yuncheng also came over. He just happened to have some questions that he wanted to ask: "Director Gao, are you here?"

Gao Zhendong smiled and said, "Well, Engineer Yu, how is the situation now?"

Yu Yuncheng was also in a good mood: "It's going relatively smoothly. I'm installing and adjusting one by one."

Having said that, he took the opportunity to raise the questions in his mind.

"Director Gao, from a theoretical perspective, the zone melting method produces higher material purity than the Czochralski method. Why don't we just use the zone melting method?"

He was reluctant to ask this question at first, as it involved issues of principle and route. Seeing that Gao Zhendong's single crystal furnace design was so complete, he felt like he was overturning everything when he asked this question at first.

Gao Zhendong smiled and said: "There are many ways to make silicon single crystals, not only the Czochralski method and the zone melting method that we are currently using, such as the pedestal method, sheet single crystal growth, vapor phase growth, ingot casting method, liquid phase epitaxy, etc."

Upon hearing Gao Zhendong's words, Yu Yuncheng, who is currently considered a professional in silicon single crystal production, opened his mouth wide. He had never heard of quite a few of what Director Gao said. Just this one sentence was enough for him to ponder for many years.

"Are there so many ways?" He looked a little silly.

Gao Zhendong nodded: "That's right, but these methods all have problems of one kind or another. Generally speaking, in my opinion, the most suitable semiconductor materials used in electronic technology for quite some time are the Czochralski method that we are currently working on and the zone melting method you mentioned."

In fact, even decades later, the production methods of silicon single crystals for semiconductors are still mainly based on these two methods, among which the direct pulling method accounts for about 75% of the output.

Hearing Gao Zhendong affirm his idea, Yu Yuncheng felt secretly delighted.

Gao Zhendong then told him why he didn't use the zone melting method at the beginning: "But from these two methods, the Czochralski method has the widest application range, is the easiest to make the crystal diameter larger, and has simpler technology."

Yu Yuncheng had some knowledge of the zone melting method, but he did not understand its principles and specific implementation methods. It was probably at the level of being aware of it but never having seen it.

"Director Gao, what exactly is the zone melting method?"

Gao Zhendong simply introduced the principle of the zone melting method to Yu Yuncheng.

Simply put, the Czochralski method is to melt silicon in a pot and then pull the crystal upwards.

The zone melting method is to hang up the silicon rod, melt the lower end, and then pull the crystal downwards.

Yu Yuncheng was a little surprised when he heard it: "Ah, so that's it, it seems to be more difficult to control."

As expected of someone who is good at this, Gao Zhendong understood it immediately. He smiled and said, "Yes, the molten silicon liquid in this method is mainly kept from falling by surface tension, so it is more difficult to make the diameter larger than the direct pulling method. At the same time, the controllability of the entire crystal pulling process is poor, which can easily lead to crystal structure defects."

Yu Yuncheng understood: "Well, it seems that choosing a technical route cannot rely on a single indicator, but still requires comprehensive consideration."

Gao Zhendong smiled and said, "You are right. For now, the defects of the direct pulling method can be compensated or weakened through design methods, such as adding a quartz inner crucible. But its advantages are real. We are not doing theoretical research, which can only focus on the high points. We are doing applied research, so we still need to look around."

Yu Yuncheng understood more clearly: "Ah, I was wondering why we had to put another quartz crucible inside the graphite crucible. It turns out that I didn't think it through." Gao Zhendong said: "For our current needs, the advantages of the zone melting method are actually not very useful, but the disadvantages are real. The Czochralski method has high production efficiency, low difficulty in automated production technology, easier control of doping concentration, and easier preparation of large-diameter single crystals. These are what we urgently need now."

For single crystals used in semiconductor devices, doping of the substrate is one of the commonly used processes. In this regard, CZ melting is much more convenient.

After that, he added: "Wait until the vertical pull is successful, then do zone melting. After all, zone melting has its advantages."

Yu Yuncheng was convinced by Gao Zhendong's class. He felt that he had made the right decision by coming here. Not only did he learn the technology, but he also learned the ideas of scientific research.

Therefore, he had no shame in asking Gao Zhendong to learn from him. If he didn't take advantage of Director Gao's free time to learn more, Director Gao's visit would not be in vain.

"Director Gao, in my opinion, as long as the crucible heating temperature is stable, the thermal field is stable. The crucible heating temperature can be controlled by power supply power, and the pulling rotation and height of the crystal can be controlled by the seed crystal. Why do we need to add complex crucible lifting and rotation control on the crucible side? Doesn't this greatly increase the complexity of the system?"

When Gao Zhendong heard this, he realized that he had really thought deeply about it. He had given relatively in-depth thoughts on the functions of each subsystem he mentioned and the role it played in the system. However, he was still mainly relying on experience to think about the problem and lacked certain theoretical support.

For some people, it is not easy to clearly distinguish the functions and effects of subsystems.

Simply put, function refers to what this subsystem can do.

The effect is the function of this subsystem and what benefits or impacts it can bring to the entire system.

The former is the cause and the latter is the effect.

Sometimes some people easily fall into the trap of function worship. The more functions the better, and the more complex the better. As for what role the functions can play, whether they are beneficial to the system, or whether it is worth paying such a high price for such a small benefit, they have never thought about it or have not thought it through.

Gao Zhendong’s mentor in his previous life once gave an example. In a certain project, in order to transmit less than 500k bytes of data every few minutes between two locations 1 meters apart, the design unit decided to pull a single-mode optical fiber between the two locations.

Then they recounted how much they had considered for this optical fiber, such as ground wire composite optical cable lightning protection, optical cable armor to prevent physical damage, deep digging and deep burial to prevent pressure
Gao Zhendong's supervisor was reviewing the project, so he asked a question: "Don't you know how to use VHF data transmission radio? It has guaranteed reliability at low rates, is easy to install, requires no maintenance, and can guarantee a very good data arrival rate through protocol control. It is also cheap, and you can get a pair of very good ones for just a few thousand yuan."

The other party was speechless, but perhaps it was not the design unit that was stupid, but Gao Zhendong’s mentor.

So when Gao Zhendong heard Yu Yuncheng ask this question, he was very happy and explained it to him in detail.

"I added this system not for the crystal pulling rotation control, but for the thermal field control."

"The graphite crucible rotates partly to ensure uniform heating. After all, no matter how good the control system is, it cannot guarantee uniform temperature everywhere in the heating furnace. One problem we need to pay attention to is that during the crystal pulling process, the crystal formation is greatly affected by the temperature gradient of the heat field."

"Temperature gradient?"

"Yes, it is the change in temperature in a certain direction. There are radial temperature gradient and longitudinal temperature gradient. You can think of the former as the temperature gradient on the circular cross section of the crystal, and the latter as the temperature gradient in the height direction."

Now, Yu Yuncheng understood a little.

Gao Zhendong took a pen and paper and drew a temperature gradient curve for Yu Yuncheng during the crystal pulling process, and explained in detail the source and impact of the temperature gradient.

Yu Yuncheng and the research team members who were watching and learning around him felt that they had opened the door to a new world, and that many of their previous relevant experiences could be explained in it.

Finally, Gao Zhendong concluded: "In principle, the radial temperature gradient should be as uniform as possible, which is part of the reason why I just mentioned that the crucible should be rotated."

"The lifting and lowering of the crucible is similar. You have just seen the impact of the radial temperature gradient. It should be higher so that the crystallization latent heat of single crystal silicon growth can be dissipated in time and the interface temperature can be kept stable. But it cannot be too high, otherwise it may cause polycrystalline or a large number of structural defects."

Seeing that everyone understood this principle, Gao Zhendong explained the reason for adding a crucible lifting system: "One of the simplest ways to increase the temperature gradient is to increase the height of the crucible. In other words, the simplest way to adjust the vertical temperature gradient is to lift the crucible. Of course."

Before Gao Zhendong finished speaking, Yu Yuncheng had already learned to answer: "Also, can it cooperate with other control systems to adjust the furnace temperature, improve the crystallization conditions and environment, and adjust the crystallization state?"

Gao Zhendong smiled and said, "Yes!"

After saying this, the core control conditions of single crystal crystallization have actually been explained clearly. Those present here may not make a single crystal furnace in the future, but they may still have ideas on repairing a single crystal furnace.

As for people like Yu Yuncheng, who are already great, he benefited a lot.

(End of this chapter)