The rise of nanometers

Just when the special plane of Li Fuzhen, Park Shengli and his party arrived at Pengcheng Bao'an Airport.

Negotiations between Korea Petroleum Corporation and the Energy Alliance on the Yellow Sea natural gas pipeline have also reached a deadlock.

The Energy Alliance is well aware of this. The current situation is not that the Petroleum Corporation does not want to negotiate, but that the "father" behind them does not want to see the construction of the Yellow Sea Natural Gas Pipeline.

In order to appease the National Petroleum Corporation, North America has given a batch of relatively low-priced shale gas and shale oil special quotas to the Petroleum Corporation to alleviate the recent high crude oil and natural gas prices in Korea.

Similarly, oil and gas companies in Japan have also received a batch of crude oil quotas of about 75 yuan per barrel.

However, this price is still higher than the psychological price of the two places. After all, the crude oil price obtained by the Energy Alliance is only about half of their price. This price makes Korea Petroleum Corporation and Japanese oil and gas companies envious and jealous.

But there is nothing that can be done in North America. After all, the production cost of shale oil and shale gas is as high as 30 to 45 yuan per barrel. You can't lose money to supply crude oil, right?

As soon as he got off the plane, Li Fuzhen received information about the situation from the Petroleum Corporation, and his face suddenly showed a hint of helplessness.

Park Shengli, who specializes in international trade, shook his head and smiled bitterly: "Vice president, now the hope is even slimmer."

"Just take it one step at a time!" Li Fuzhen could only resign himself to his fate.

In this kind of confrontation between great powers, Samsung is just a pawn. If the Suiren system had not risen before, Samsung, which has mastered chip manufacturing technology, could still have its way.

Now that Suiren's chip technology has developed, Samsung, which has no use value, has lost its original importance to those two giants.

In fact, before Lee Fuzhen and Park Seungri came over, Samsung invested huge amounts of R\u0026D funds, hoping to upgrade the chip manufacturing process to 16 nanometers within a year to recover lost foundry orders.

It’s just that Samsung doesn’t know that at this time, a new generation of textile technology is already being developed at the Long Totem Semiconductor Base in Fair Town, Shanmei.

After returning from the Northern Tour, Huang Xiuyuan, Lu Xuedong, Zhang Jingjian and others have been studying new textile techniques.

Inside a semiconductor manufacturing laboratory.

Everyone has now improved the process accuracy of chip textile machines to 12 nanometers, but mass-produced chip textile machines can only reach 16 nanometers.

At 12 nanometers, we have actually reached the limit of photolithography. Why is 12 nanometers the limit of photolithography?

Huang Xiuyuan, who has memories of the future, naturally understands the tricks involved.

In the future, the so-called 10nm/7nm/5nm of Samsung and TSMC actually have a process of secretly changing the concept, and Intel is too lazy to pay attention to their "new process" because it has no foundry business.

In fact, in terms of performance comparison of CPUs produced by Samsung, TSMC, and Intel, Intel is even better.

Why can’t Samsung and TSMC’s 5nm technology compete with Intel’s 10nm technology?

This is the different process standards between the two parties, which leads to the seeming gap between Intel and Samsung and TSMC. In fact, it is Samsung and TSMC that mess up the standards.

The limit process of photolithography is around 12 nanometers. After all, photolithography relies on deep ultraviolet light and etching, which is determined by the physical properties of ultraviolet light.

The limit of the textile method depends on the cross-sectional diameter of the nanowire. Theoretically, the cross-sectional diameter of a single-atom nanowire is about 0.5 nanometers.

However, Huang Xiuyuan understands that the 0.5-nanometer chip process is basically impossible to achieve at this stage. The quantum tunneling effect alone is enough to scrap the chip.

Why do humans pursue smaller chip technology?

The answer is energy consumption.

More refined transistors can effectively reduce chip energy consumption.

If you simply pursue the improvement of computing power, you can actually use a supercomputing blade server construction method to continuously increase the number of chips.

In commercial applications, although computing power is a big indicator, the real core factor is the energy consumption per unit of computing power, which is what we often call the energy efficiency ratio.

Suppose there are two chips A and B, both with floating point computing power of 1 billion times per second. Chip A uses a 28-nanometer process, while chip B uses a 22-nanometer process. The unit energy consumption of the two is, usually A is higher than B. .

It may not be obvious to individual users, but those major Internet companies have to pay hundreds or even billions of electricity bills a year, so they have to consider energy consumption.

This is one of the reasons why, after the birth of Lilong 1, major Internet companies had to purchase it, including energy consumption considerations.

Huang Xiuyuan looked at the chip spinning machine in front of him. Although the 12-nanometer process can produce chips, the speed has dropped sharply. On average, it can only process two to three thousand transistors per second.

Compared with the amazingly fast Zhenyi-1 and Zhenyi-2, the speed of Zhenyi-4 is obviously too slow.

"Xiuyuan, I plan to introduce the three-dimensional process of the Second Semiconductor Laboratory into the 12-nanometer process. What do you think?" Lu Xuedong suggested.

Putting down the ballpoint pen in his hand, Huang Xiuyuan thought for a while: "The three-dimensional technology of the Second Semiconductor Laboratory must not be mature, right?"

"Yes, heat dissipation is a big problem." Lu Xuedong spread his hands helplessly.

The textile method has natural advantages on three-dimensional chips, but three-dimensional multi-layer chips cannot be stacked casually.

Originally, the chip generates serious heat during operation. Once multi-layer stacking is used, the chip will generate even more heat during operation.

This kind of working heat has little effect on the upper and lower layers of the surface and the edges. However, heat will continue to accumulate in the chip layers in the middle, resulting in a rapid decrease in the service life of the chip, or even directly burning the chip.

The current 16nm process can stack up to 5 layers of chips, and excessive heat dissipation is a problem.

The operating voltage is even lower and the transistors are smaller in the 12nm process. The limit should be 8 to 9 layers.

The main reason why it is necessary to develop three-dimensional chips instead of increasing the chip area is that three-dimensional chips have advantages that flat chips do not have, that is, they occupy a small space and can also arrange more pins.

Huang Xiuyuan knows the future trend. Three-dimensional chips are a big trend. To solve the heat dissipation problem of chips, we can only start with materials.

"Xuedong, the current heat dissipation materials are not good enough. New heat dissipation materials must be developed so that the inside of the chip can fully dissipate heat."

Lu Xuedong nodded: "The idea is feasible, but the requirements for this material are very high. Even our thermoelectric refrigeration materials can't do it."

Zhang Jingjian, who came back from the second laboratory, also joined the topic. Everyone discussed it for a long time, but still did not come up with a suitable plan.

Although Huang Xiuyuan knew that some materials could do it, he did not want to launch it too quickly. He could only temporarily shelve the three-dimensional structure plan and let the Second Semiconductor Laboratory continue to conduct in-depth research.

Near lunch time, Xiao Yingnan reported to him about the arrival of the Samsung delegation.

"Are these sticks thinking about eating peaches?"

"That……"

Huang Xiuyuan waved his hand: "Just let Baijie deal with it. Samsung is not qualified to negotiate with us. They are just puppets of Wall Street. There is no need to waste time."

"I'll tell Mr. Lin right now." Xiao Yingnan nodded.

Li Fuzhen and his party, who had just arrived in Shanmei, did not know their fate. They had already been clearly arranged and were still trying their best to negotiate.

Thank you for your support (ω｀)