Traveling through the sword to engage in military industry.

Chapter 314 Semiconductor Technology from Scratch

For a century-old navy, this can't be rushed. After visiting the Fletcher-class destroyer, Ren Zhong solved a small conflict between the shipyard and the fleet and revealed his plan a little bit.

Professor Liu and Li Changxing are both key figures in future development. If we make them clear about their ideas, many things will be easier to handle, and there will be no bumps in future cooperation.

Now Ren Zhong can accept experimental failure, but basically does not accept unnecessary human interference.

In fact, the main combat power of future warships will come from command and combat systems and weapon systems.

There is one technology that both of them need in common, which is the evolution of radar technology.

Whether it is search radar or fire control radar, there is an urgent need for further technological evolution to miniaturize the radar and improve its accuracy.

The current aiming error of up to 1000 meters is basically useless for naval operations.

So after Ren Zhong sorted out the Doppler radar concepts and related technologies, he handed them over to the radar team for research.

Because there are too many disciplines involved, from materials science to electronics, communications and even computer operations, it was only after the breakthrough of transistor computers that we barely had the preliminary capabilities of spectral analysis and clutter filtering, and completed the first generation of miniaturized Doppler radar.

The first application of this system was actually in artillery, where it was the first to complete the research on artillery-aiming radar!

Of course, this system is not small. It needs three vehicles to tow it. The power generator and the computer command system of the gun-aiming radar are on one vehicle, and the transmitting unit and the receiving unit are on another vehicle. This is naturally a compromise due to insufficient materials, technology and workmanship. However, with this system, the reaction capability for detecting artillery trajectory can now reduce the position estimation error of the artillery position to about 1 meters within 100 minute.

Considering the current artillery counterattack capabilities, this is no different from cheating.

A large number of new transistorized Doppler radars are beginning to show their initial power.

"It's too difficult. To obtain modern microwave transistors, they must have fine geometric dimensions of micrometers or submicrometers. There are not many good ways to achieve this processing technology even if you buy it with money.

This manufacturing process requires the development of micro-processing technologies such as thin-layer epitaxial technology, shallow junction diffusion or ion implantation technology, projection exposure, extreme ultraviolet exposure, X-ray exposure, and electron beam exposure. There are too many prerequisite technologies that need to be developed.

And in this regard, microwave transistor processing and integrated circuit processing have begun to have a lot of overlap.

It seems that it is not feasible to achieve it all at once, and we need to sort out the development context of various supporting technologies and start from scratch.”

Ren Zhong looked at the collected information and had to say that it was too difficult for a liberal arts student to solve these problems.

So Ren Zhong did not plan to do it himself, but instead resorted to spending money and began to buy technical information.

The first thing to be solved is the ion implanter. This device has a long history. In the main world, it can be traced back to the 50s. Semiconductor ion implantation equipment is one of the important equipment in semiconductor manufacturing. It can implant ions into semiconductor materials to change their conductive properties. This process needs to be carried out in a vacuum environment to avoid the influence of impurities on semiconductor materials. It can be said that semiconductors and subsequent chip manufacturing are inseparable from such equipment.

Specifically, the gaseous doping compound raw materials are introduced into the reaction chamber, and the electric field and magnetic field are added to interact to form plasma; after the ion beam is extracted from the reaction chamber, it is pulled forward by the electric field and accelerated, and is accelerated again after passing through the magnetic field to increase the range of the ion beam; the required ion source is screened through a mass analyzer; the ion source uses a precise ion scanning system to ensure that the doped ions can be evenly injected into the entire silicon wafer.

The whole process is very precise, but fortunately, the basic model of this equipment now has relatively popular technology and can be purchased for a small amount of money.

In terms of the core material of transistors, the first generation of transistors used germanium to manufacture junction transistors, but this first generation of transistors had a fatal flaw: serious problems would occur when the temperature exceeded 80°C. This was its inherent defect, and there was no other way except to cool the surface, so the power amplification of germanium transistors was very limited.

But relatively speaking, germanium transistors are relatively easy to realize. Germanium has a lower melting point, which means that its crystals are easier to grow, and it is relatively easy to produce transistors. So Ren Zhong still chose this material and quickly realized the first generation of transistor computers. Of course, there are more fans on the back for heat dissipation.

In the future, the research direction of semiconductor technology will be to conquer silicon transistors!

This is the truly mainstream technology that can be used into the 21st century.

Silicon transistors are npn structures and need to be manufactured through a growth junction process. Silicon has a larger bandgap, which enables it to operate at higher temperatures (see Table 1). Secondly, it has a significant synergistic effect with its oxide, silicon dioxide (SiO2).

A high dielectric strength, electrically insulating SiO2 layer can be inexpensively formed by simply heating silicon in an oxygen-containing atmosphere. This SiO2 layer is very stable chemically and mechanically, effectively passivates the surface states of silicon, forms an effective diffusion barrier for commonly used dopants, and can be easily etched or deposited on silicon.

Because of such excellent performance, silicon transistors have become the darling of the semiconductor era, dominating the semiconductor industry for most of the era. The production of semiconductor silicon wafers alone has become a market worth tens of billions of dollars, supporting a trillion-level semiconductor market.

However, it is extremely difficult to produce semiconductor-grade silicon. The purity of silicon needs to be purified to an extremely high level, usually more than 9 nines (i.e. 9%), before it can be used to manufacture semiconductor transistors. For the time being, this is a hellish level of difficulty.

In the main world, there are currently only two processing methods that can achieve commercial production.

The CZ method is the main method for preparing semiconductor-grade silicon. Its principle is to heat high-purity silicon material to a molten state, and then form an interface between the crystal rod and silicon by rotating the crystal rod and controlling the temperature. The crystal rod is slowly pulled down, and at the same time, a thin silicon rod is pulled out between the crystal rod and the silicon. Its internal structure and lattice are exactly the same as the crystal rod. Because the area between the crystal rod and the silicon is extremely clean, high-purity semiconductor-grade silicon materials can be prepared by this method.

The FZ method is another commonly used method for preparing semiconductor-grade silicon. Its principle is to add a strong magnetic field around the silicon crystal, melt the silicon material through inductive heating, and then form a certain area of ​​melting in the silicon material by controlling electromagnetic induction and movement direction. A wide strip-shaped dissolution layer is formed around the molten area, and the dissolution layer gradually separates from the solid silicon crystal layer above to form a silicon rod. This method can prepare high-purity semiconductor-grade silicon materials.

Both processes are energy-intensive and require very complex equipment. Getting this set of equipment ready is a systematic project.
Not only is the production method cumbersome, but in order to ensure the quality of semiconductor-grade silicon, the testing methods are also quite demanding.

At present, common detection methods for silicon crystals include thermal absorption, mass spectrometry, atomic fluorescence, etc. Among them, thermal absorption is one of the most commonly used methods. It can determine the impurity content by testing the amount of gas released when the silicon wafer is heated. Mass spectrometry and atomic fluorescence can directly detect the impurity content in silicon wafers with high sensitivity and accuracy. However, it is obvious that these detection methods need to be supported by a set of sophisticated instruments and equipment.
One problem after another made Ren Zhong's journey in developing semiconductors extremely difficult and painful.

Of course, he knew that this was inevitable. If he wanted to walk the path that others had walked in twenty or thirty years in three to five years, he would have to put in more effort. Even if it was just copying the homework from the main world, it would not be an easy task.

First update, please give me a monthly ticket



(End of this chapter)