Chapter 281 Glue drop
Minqin County, a high-tech industrial park on the southwest side.
There are still many people staying to work during the Spring Festival.
Among them, the 521 Laboratory, 522 Laboratory and 523 Laboratory, composed of Hailufeng Company, Minqin Group, Changjiang Storage and several domestic semiconductor equipment companies, not only did not have a holiday, but also increased a lot of scientific researchers during the Chinese New Year.
The three joint laboratories currently have 493 researchers and 244 administrative and logistics personnel.
521 within the area of the laboratory.
With the aid of cryo-electron microscopy and other atomic scanning devices, several researchers finally combined dense silver-plated silicon nanotubes with some solid silicon carbide tubes.
These fine silver-plated silicon nanotubes are divided into special patterns by solid silicon carbide nanotubes. These patterns are part of the single-layer chip structure and the part that does not require lithography.
In essence, the photolithography process is to destroy part of the photoresist that needs to be removed through extreme ultraviolet light, and then remove the damaged photoresist through cleaning, and finally use etching, deposition and other processes to utilize these grooves.
Of course, in practice, these processes will be very complicated and require repeated lithography, etching, deposition, scrubbing dozens of times, or even hundreds or thousands of times.
At this time, the purpose of these researchers was to solve the problem that the country has been difficult to catch up with the West's top lithography machine technology through silver-plated silicon nanotubes.
"How?"
You can see the development equipment of atomic arrangement structure, and a very large structural picture is being displayed. The researcher in charge of the inspection raised his head: "Team leader, this piece of offset stamp has been checked, and no wrong structure has been found."
Team leader Ye Jin immediately ordered, "Then send it to the cutting!"
"Okay." Another researcher took over the cutting work.
In the wire cutting studio, the dropping mark with a length of 20 cm is cut little by little, and the thickness is divided into a very precise 1 cm.
These 1 cm thickness dropping printing templates are the core components of the dropping process.
Ye Jin and several other group leaders discussed the next arrangements.
"Old Liu, the template is now ready. Let's start testing. Are your silicon wafers ready?"
Lao Liu is a middle-aged Mediterranean man. He smiled and replied, "I have been ready for a long time, just waiting for your template."
"Then let's start the test as planned before! I hope everything goes well." Ye Jin was also very nervous at this time.
"Don't be nervous, the boat will be straight when it reaches the bridge." Old Liu said as he picked up the folder and looked at some information on it: "Is the glue on the 522 ready?"
The person who spoke was a researcher from a domestic photoresist company: "That's natural, don't underestimate us. Since there is no need to consider photolithography, we only need charged glue materials. This requirement is several orders of magnitude lower than photoresist."
Of course, in actual research, this glue material is not that easy, but it is at least much less difficult than photoresist. After some research, 522 Laboratory came up with several more suitable glue material.
When Lao Liu looked at the information, he suddenly felt a passionate feeling in his heart: "The etching machine of Zhongwei Semiconductor has been debugged, and now everything is ready, let's start!"
"Agree."
"Okay, it's a mule or a horse, you have to pull it out and try it."
"I feel that the success rate should be very high."
"You boy is confident. If you succeed, everyone will be famous in history."
After a while.
They came to the 523 laboratory responsible for the test.
As the offset printing template is inserted into the glue drip machine, the air in the glue drip chamber has been evacuated and replaced with a high-purity inert gas that will not react with the glue material and will not undergo ionization reaction.
At the same time, the dust removal equipment is also continuously operating, keeping the rubber dripping chamber highly clean and avoiding the dust being attracted by ionization, thereby affecting the performance of the rubber dripping machine.
The 8-inch silicon wafers washed with pure water are dried and fed into the glue drier by the machine.
The template of the glue dripping machine is also round, and there is also a just-fitting slot, which can be aligned with the silicon wafer to ensure that the silicon wafer and the template are completely fixed, and avoid subtle movements during the glue dripping mold printing process.
A researcher's expression was very nervous, and his palms were sweating. He trembled his fingers and gently pressed on the start switch.
drop!
After powering on, the glue drier template sucks the glue liquid in the storage on it, and quickly pulls the charged glue liquid from the top to bottom through electromagnetic acceleration.
In an instant, drops of tiny nano-scale rubber drops dripped from under the template on the surface of the silicon wafer. These rubber drops quickly formed drops, and these drops merged with each other to form patterns.
Then the glue dripper was powered off, and only a drop of glue dripped off. The remaining glue was wrapped in pipes and could not be dripped off at all.
The glue drier leaves above the silicon wafer, and then the photocuring device emits specific ultraviolet rays to the surface of the silicon wafer to lightly cure the glue drip pattern.
Ye Jin said impatiently: "Hurry, see if there is any problem with the pattern."
Old Liu shook his head: "Dr. Ye, don't be anxious! Make a few more pieces and send them to the test together."
"Okay!"
In less than half an hour, dozens of drop silicon wafers were produced continuously in the glue-drop chamber.
The research team responsible for testing the glue-drop pattern also sent batches of glue-drop silicon wafers to the laboratory for testing.
Everyone couldn't help but fall silent, they were all waiting for the result.
Time passed minute by minute.
About two hours later.
"Huhu..." A researcher ran over with a report.
Everyone stared at him in unison.
"Huhu...good news, there are no defects in the glue pattern. You can start testing of etching, deposition and other processes."
Suddenly the air became quiet.
Then there was a deafening whistle.
"Oh oh oh..."
"We succeeded!"
"Long live!"
"Hahaha, I just said the success rate is very high."
"Great, we did it!"
The excitement was almost deafening, and everyone was unable to calm down. They were expressing their inner excitement to their colleagues and colleagues around them.
Papa...After more than ten minutes, the older Lao Liu clapped his palms: "We have taken the first step towards success. Now is not the time to relax. Next, everyone will work harder and strive to complete the innovation of the entire set of craftsmanship."
"Don't worry! Old Liu, I'm full of strength now, more energetic than drinking a hundred cans of Red Bull." Hongguang, a researcher who was also in his forties, said with a face full of face.
Old Liu hurriedly reminded: "No, no, although work is very important, I don't want everyone to become ill due to overwork. After all, everyone doesn't want to lose the opportunity to go on stage to receive the award, right?"
"I know, do you combine work and rest? But it shouldn't be a big problem to work for ten hours a day. When I was at TSMC, I often stayed all night."
Ye Jin was also excited at this time: "Old Liu is right, see you at the celebration meeting!"
"See you at the celebration meeting!"
After being excited, everyone began to calm down.
Soon they used the laboratory equipment to start repeating glue drops, etching, deposition, grinding, cleaning, and ion implantation.
The first batch of memory chips produced in laboratories were produced half a month later.
The reason why it is so fast is mainly because the overall speed has been greatly improved after the glue drop process replaced the lithography process.
The chips they produced through the laboratory have their design models from Yangtze Memory, so they can only produce memory chips.
Not only the peripheral logic units of the memory, but also the internal memory units also adopt the same process, namely the 10-nanometer process.
At this time, several bosses from the Yangtze River Storage had arrived in Minqin County. Chen Nanxian, Chairman of Changjiang Storage, Yang Shining, Director Liu Xiukun and others looked at the memory chip in front of them, and their eyes were almost emitting.
Even after struggling in the semiconductor industry for half a lifetime, Yang Shining couldn't help but exclaim: "Is the storage density of 100G per square meter of millimeter? It's so terrifying..."
Chen Nanxian was also shocked: "Yes! It is equivalent to increasing the unit storage density by about 5 times, and the energy consumption has decreased by 36%, the reading speed has increased by 33%, and the cost has decreased by about 41%. This is simply an epoch-making product."
"Mr. Chen, I think we should enter the semiconductor foundry industry." Liu Xiukun, who was standing by, suggested.
"Semiconductor Foundry?" Chen Nanxian nodded, "Indeed, we cannot miss this opportunity. After the glue-drop process replaces the lithography process, the production cost of the entire chip has dropped by about 30% to 40%, and this is equivalent to directly mastering the 10-nanometer process. It is a pity that we did not enter the semiconductor foundry industry at this time."
Yang Shining reminded: "Mr. Chen, what do Academician Jiang say?"
Upon hearing this, Chen Nanxian knew that this matter could not be avoided by Jiang Miao and Hai Lufeng Company, so he picked up the phone and called.
Doudu…
"Hey, Mr. Chen, what's the matter?"
"Academician Jiang, the project we cooperated with has been initially successful. I would like to ask what plans are Hailufeng Company next?"
Jiang Miao, opposite the phone, naturally had received relevant news, but he did not have the idea of entering the semiconductor industry, so he replied:
"Does Mr. Chen want to enter other semiconductor industries?"
"That's right, I wonder if Academician Jiang is interested in investing?"
"Yes, Hailufeng Company will invest 20 billion! We will do financial investment, and the stock ratio will be calculated based on the investment ratio of each party."
"No problem." Chen Nanxian knew very well that if Hailufeng Company was willing, it would be fine if it was willing to provide hundreds of billions of working capital. However, it seemed that Jiang Miao did not plan to control this semiconductor foundry company, it was really just a simple financial investment.
After all, for the semiconductor industry, 20 billion yuan can only build a small wafer factory.
Although the outside world has not received any news about the technological breakthrough this time, the official must have known it.
Therefore, the majority of investment accounts for the majority of the National Semiconductor Industry Fund and State-owned Assets Investment Company. In addition, Changjiang Storage, as the operating management party, will definitely obtain a part of its shares.
He estimated that Hailufeng Company's 20 billion yuan can only account for about 10% of its shares.
Of course, this does not mean that Hailufeng Company is losing money, because 51% of Minqin Precision Instruments Company's shares belong to Hailufeng Company.
Of the remaining 49% of the shares of this precision instrument company, Changjiang Semiconductor only received 7% of the shares, Liangzhou State-owned Assets Investment Company 10%, China Micro Semiconductor 8%, and several other cooperative companies that contributed their talents and efforts received 24% of the shares.
In the future, the domestic semiconductor industry chain will replace the lithography process, and the ecological niche of this process will be led by Minqin Precision Instrument Company.
After chatting for a while, Chen Nanxian hung up the phone.
"Academician Jiang said that as equipment suppliers, they can do it, and then they plan to invest 20 billion yuan in our semiconductor foundry company. As a simple financial investment, they do not intend to enter semiconductor production."
Yang Shining secretly breathed a sigh of relief. Fortunately, Hailufeng Company had no idea, otherwise it would be troublesome. After all, Hailufeng Company's status in China is too special
Especially for their state-owned enterprises like Changjiang Storage, they definitely don’t want to offend Jiang Miao behind Hailufeng Company.
Liu Xiukun narrowed his eyes and said, "Mr. Chen, I don't think we should announce technological breakthroughs immediately."
Chen Nanxian immediately understood his thoughts: "Do you want to raise some of the equipment before launching a fierce attack?"
"Well, the outside world doesn't know that we have succeeded now. I don't think it is necessary to announce it immediately. After all, it is announced now, and there is no benefit at all. Instead, it will make Western forces react. They will definitely be anxious and increase the ban on purchasing other equipment."
Yang Shining nodded: "I agree with Xiu Kun's opinion. Now that we have got the match point, we should calm down, rather than revealing the trump card in advance."
After thinking for a while, Chen Nanxian felt that he could not act irritatedly: "Okay, I will communicate with Hailufeng Company later, and they should cooperate with us."
"There is one more thing. I think we should contact HiSilicon now." Liu Xiukun pointed out.
"Haisi?" Chen Nanxian nodded immediately: "Yes, contact them secretly, but don't tell them the key information. Please send the core management to come over and discuss the cooperation between the two sides."
The reason why HiSilicon was invited is that Changjiang Memory does not have the design capabilities of other logic chips.
The emergence of the glue-drop process instantly raised the chip processing process of Yangtze Memory to the 10-nanometer level.
This level of chip manufacturing process is already at the international advanced level. As for the 7-nanometer and 5-nanometer process, it is not really 7-nanometer or 5-nanometer, but is called equivalent 7-nanometer and equivalent 5-nanometer.
In other words, if Yangtze River Storage continues to study, they can also achieve the so-called equivalent 7 nanometers and equivalent 5 nanometers through similar technical means.
The problem facing Changjiang Storage now is to find the right customer.
If you don’t have the right customers and have huge chip production capacity, you will still lose money.
At this stage, many domestic companies are unreliable in mobile phones and computers. The reason is very simple. These companies need to worry about the international market and Ameika's attitude.
After all, these companies use software systems, chips and technologies of Western companies, so it is difficult to avoid the influence of Western capital.
The only company that doesn't need to consider this issue is Pollen Company, after all, they are already on the blacklist of Ameika.
Other companies have no determination to go all out.
Therefore, Yangtze Memory has chip production capacity and also needs a stable customer; while Pollen Company has advanced chip design capabilities and chip foundry companies that also require advanced processes.
The niches of the two can be highly complementary.
If there are no customers, then Yangtze Memory can only learn from Intel and do its own chip design and chip processing.
As for the chips needed by state-owned enterprises, most of the market actually do not need to use 10-nanometer chips.
For example, chips needed by the power grid and chips needed by automobiles, and many products at the level of more than 20 nanometers to more than 40 nanometers are basically enough.
10-nanometer chips are generally used on computers, mobile phones, and servers.
This is actually the difference between industrial-grade chips and consumer-grade chips.
Consumer-grade chips require performance, power consumption, and high added value, so they can naturally afford 10-nanometer chips.
However, industrial-grade chips often do not require such good chips. For example, automotive chips. If you feel that the computing power is not enough, you can directly superimpose a few more chips. Anyway, the car space is large enough and there is also a car radiator.
After all, advanced process chips are also very costly.
However, Yangtze Memory can also consider using low-priced promotion methods to replace low-level chips with advanced process chips. Anyway, after switching to the glue-drop process, the production cost of the entire chip will drop by 30% to 40%. It can be completely used to directly use the advantages of high performance and low cost to directly eat up this mid- and low-end market.
Anyway, China imports a large number of mid- and low-end industrial chips every year.
However, it is indeed very troublesome to seize the market of mid- and low-end chips. The reason is that from beginning to end, Changjiang Storage is not a professional logic chip design company. Without strong logic chip design capabilities, it means that they have to start from scratch.
This is much more difficult than doing OEM.
Moreover, many professional chips can not be designed by recruiting hundreds of engineers, such as those automotive chips. Why do European chips still have chip companies? It is because they also have good professional chip design capabilities.
For example, STMicroelectronics, Infineon, etc., these are the best professional chips.
If it is not necessary, Yangtze Memory plans to replace TSMC's ecological niche instead of becoming a full-system semiconductor company similar to Intel.
At the same time, in terms of mid- and low-end chips, domestic SMIC and other companies have good strength. Changjiang Storage does not need to focus its energy on mid- and low-end chips, but should focus on advancing to advanced processes.
After all, from a process perspective, in the process of developing equivalent 7nm and equivalent 5nm chips, the dropping process is easier to implement than the lithography process.
Because light diffraction requires a lot of precision lenses, and the control accuracy requirements are very high.
The glue-drop process is different. The difficulty of this thing lies in how to make a defect-free glue-drop-print template. As long as the template is made, the subsequent process will be very simple.
At the same time, the glue material of the glue dripping process is not as harsh as photoresist, and some special designs can even be achieved through different glue materials.
Chen Nanxian and others also had to sigh that materials are the cornerstone of modern technology.
Instead of leaving Minqin, the three of them left one person to coordinate the work, while the other two rushed back to the headquarters of Yangtze River Storage.
While secretly dispatching troops and deploying generals, they are deploying the semiconductor foundry industry here.
On the other hand, we have negotiated cooperation with many semiconductor processing equipment suppliers, requiring domestic equipment suppliers to speed up technical research and increase money to allow foreign equipment suppliers to ship goods as soon as possible.
(This chapter ends)