bookbashuk #419 - First-generation lithography machine breakthrough
#419 - First-generation lithography machine breakthrough
The automotive industry was about to take off. The following events couldn't be rushed; they needed time to slowly ferment and further refine the entire product. Ren Zhong stopped concerning himself with these details.
Ren Zhong shifted his focus to the development of communications.
For Dongda, breakthroughs in old-fashioned large-scale crossbar automatic telephone exchanges had long been achieved. Relying on its own production of this type of exchange, the first city network in the country had been established. However, due to the inherent technical weaknesses and patent restrictions of this crossbar exchange, it was destined to be unable to go abroad.
Therefore, in Ren Zhong's plan, there wasn't much development of this technology. He only built a simple communication network for government agencies and factories at the prefecture level before stopping the research and development of this old-fashioned exchange.
According to Ren Zhong's task for the Communications Research Center, the main focus was on researching and developing the next-generation program-controlled exchange from the Bright Sword world, taking a new technological route.
This technological route was also the direction of communication technology development that had been verified in the main world.
Although developing program-controlled exchanges was still somewhat challenging, under Ren Zhong's impetus, Dongda's accumulation of basic IT technology had already exceeded the level of the main world in the 1960s. With the technical advantages in transistors and integrated circuits, it was fully capable of developing program-controlled exchanges based on the current technological foundation.
Regarding the future development of communication technology, Ren Zhong didn't plan to skip stages of development. He didn't understand these aspects, so he completely copied the development steps of the communication industry in the main world.
The first step was to recreate the Bell No. 1 ESS system in the Bright Sword world.
This was the first commercial program-controlled exchange system in the history of the main world.
The technical data for this system had long been declassified. Ren Zhong spent some money to obtain detailed circuit diagrams and the main program for program control, and then brought these things into the Bright Sword world.
Although the technical accumulation of the communication research center team in the Bright Sword world was somewhat inferior to the engineers at Bell Labs in the main world.
However, what Ren Zhong brought into the Bright Sword world was not only the design of this system, but also basic knowledge of communication principles and outlines of principles from the main world, giving the engineers engaged in research and development at the Communication Research Center the opportunity to become self-taught.
Following the path of the main world's communication industry development would naturally not lead to mistakes. However, considering that the accumulation of the entire industry's development was not only about copying designs, but more importantly, about having opportunities for theoretical learning and improvement at the level of principles, to ensure the creation of a qualified talent team that truly understood the principles, knew the core technical points, and could also automatically develop derivative products in the future, the things that needed to be done now were not few.
Starting from learning the principles, then carefully studying and transforming the design data provided by Ren Zhong, and thoroughly understanding all the principles and technical points embedded in the technical data!
Because there were so many things that needed to be solved, the progress was relatively slow. The Communication Research Center had actually solved the design and production process of crossbar exchanges in the previous five-year plan, and had accumulated a lot of basic knowledge of telephone exchanges.
In this way, after they turned to the research and development of program-controlled exchanges, they absorbed a lot of knowledge of basic electronic components obtained from the design and production of crossbar exchanges. Now, to turn these functions into integrated circuits to replace them, for the Electronic Research Center, which had already broken through many design and production problems of integrated circuits, such new research and development needs were naturally not too big of a problem.
Now, the gradually specialized Electronic Research Center, Communication Research Center, and Computer Research Center, these three research institutions with strong correlations, have their own independent research topics, and are also constantly merging to jointly promote the most core research in the future IT industry.
With Ren Zhong's cheat-like support, Dongda took the first step in the development of transistor technology, creating the first transistor computer in the Bright Sword world, and moving faster and faster on the road of transistor development, making the development of transistor technology at least five years ahead of the Bright Sword world.
But this was just the first step in Dongda's leading IT technology.
Next, in terms of integrated circuits, Dongda also took the lead in breaking through the completion of integrated circuit design and production processes, further expanding the leading advantage of the IT industry, and realizing the mass production and large-scale application of integrated circuits several years earlier than many competitors in the Bright Sword world.
And applied for many patents for many proprietary function integrated circuits.
The functions of early integrated circuits were relatively simple, and the production process was also relatively simple. Usually, a complete set of planar process technologies such as grinding, polishing, oxidation, diffusion, photolithography, and epitaxial growth were used to simultaneously manufacture components such as transistors, diodes, resistors, and capacitors on a small piece of silicon single crystal wafer, and a certain isolation technology was used to electrically isolate the components from each other. Then, an aluminum layer was evaporated on the surface of the silicon wafer and etched into interconnection patterns using photolithography technology, so that the components were interconnected into a complete circuit as needed to make a semiconductor monolithic integrated circuit.
Dongda spent nearly 5 years to complete the development process of this process. By 55 years, it began to develop CMOS circuits composed of NMOS and PMOS symmetrical complementary devices, and developed the CMOS technology process that is familiar to the main world.
But when the technology developed to this stage, Ren Zhong introduced more advanced next-generation integrated circuit production technology to the Bright Sword world, and introduced the milestone 8080 chip!
By the 8080 stage, these new integrated circuits were already very close to the computer chips that Ren Zhong was familiar with.
In order to reduce the time for the entire technology iteration, Ren Zhong also found relevant professional technicians in the main world to sort out the complete set of production processes and equipment for the 8080.
The information on the multiple key steps and equipment involved in the production of 8080 chips, including wafer manufacturing, mask production, semiconductor manufacturing, packaging and testing, was made clear, and the goods were purchased to the Bright Sword world for secondary implementation, including re-developing these equipment.
This was not a simple matter. Even on the basis of having previously researched a large number of integrated circuit production equipment, the equipment needed for 8080 production research was still very difficult.
The first to bear the brunt was the manufacture of high-purity wafers.
This process includes crystal pulling, wafer slicing, wafer grinding, etching, silicon wafer polishing, cleaning, and wafer epitaxy, involving high-purity materials and strict temperature control to ensure the quality and purity of the wafer.
The mask production is to use photolithography technology to expose the photosensitive material to ultraviolet light, and then form the required circuit pattern on the surface of the chip through chemical etching or deposition. This step is the key to ensuring the accuracy and reliability of the chip design. In the chip manufacturing process, this level is the most core so-called photolithography link.
The manufacturing process involves multiple process steps such as deposition, corrosion, and cleaning. The technical difficulty is not small, especially ion implantation. This step is a key step to change the conductivity of the silicon wafer and form electronic components such as transistors. This process requires precise control of the concentration and temperature of various chemical substances to ensure the performance and stability of the chip.
Theoretically, solving this problem isn't difficult. Intuitively, it's as simple as lifting the photomask a little to prevent it from contacting the photoresist.
Early engineers thought so too. So, they added a horizontally and vertically movable platform to the contact lithography machine, as well as a microscope to measure the distance between the photomask and the silicon wafer and the overlay, so that the two would be as close as possible during photolithography, but without direct contact. This is the proximity lithography machine. It prevented the photoresist from contaminating the photomask, but it brought a new problem: the accuracy of the lithography machine decreased due to the diffraction effect of light.
Macroscopically, we believe that light travels in a straight line, but microscopically, it doesn't. Light has wave properties and will diffract, or bend, when passing through small holes or encountering tiny obstacles, deviating from its original straight-line propagation and shining on places it shouldn't. The larger the light source wavelength compared to the narrow slit, the more severe the diffraction phenomenon. It's like you're holding a knife and want to cut a 100-nanometer wide slit in the silicon wafer, but you find that the blade is 400 nanometers wide and can only be used to cut melons.
The newest novels are first published on 69shu ba!
In proximity lithography, the lithographic accuracy is not only limited by the wavelength but also depends on the distance between the photomask and the silicon wafer. The greater the distance, the greater the error between the projection on the silicon wafer and the graphic on the mask. Lifting the photomask reduces the lithographic accuracy; putting the photomask down increases the lithographic cost. This period of wavering between contact and proximity lithography is the shadow mask era in semiconductor history. These two early lithography machines are collectively called Mask Aligners. The 1:1 photomasks they use are just light shields. The lithography machine only needs to shine the light shadow on the silicon wafer, with a simple structure and no need for any complex optical system.
Contact Lithography Machine
Although this contact lithography machine could also be used to produce chips, and the early SRAM memory chips and the first commercial CPU in the early days were produced using this equipment, no one could afford it at the time because the production yield of contact lithography was too low and the damage to the photomask was too great, resulting in excessively high chip prices, only suitable for cost-insensitive scenarios such as scientific research and military industry.
Obviously, if lithography technology were only like this, the popularization of chips would be empty talk, so scientists in the IT industry did not stop.
They hoped to make a lithography machine that was both high in accuracy and didn't require pressing the photomask onto the photoresist.
This is the true starting point of modern lithography technology. At this step, the lithography machine officially transitioned from contact/proximity to the projection era, realizing the construction of the first modern lithography machine.
The new change wasn't complicated. The main world's new generation of lithography machines, Micralign, adopted a reflective projection method, using two coaxial spherical reflecting mirrors to project the graphics on the photomask onto the silicon wafer after three reflections. This symmetrical optical path design could eliminate most of the aberrations generated by the spherical mirrors, allowing the lithographic graphics to achieve the ideal resolution.
Thus, the birth of the latest model projection lithography machine, Micralign, increased the yield of chip production from about 10% for contact lithography to 70% overnight, a sevenfold increase. This leap in lithography technology led to a plunge in chip prices. In the history of the main world, Motorola had a 6800 microprocessor that was produced using a contact lithography machine, with a unit price of $295 per chip.
The following year, eight engineers ran away from Motorola to MOS Technology. They took the design circuit diagram of the 6800, simply modified it, and directly used the newly born projection lithography machine to make a chip with the same architecture as the OS6502. The chips produced not only had stronger performance but also greatly reduced the price, selling for only $25!
The price was reduced by more than 90% at once, a price cut that could be described as a bone fracture!
The power of technological progress is evident.
Ren Zhong took this process out for the R\u0026D engineers of the Electronic Technology Research Center to go through from beginning to end. Different from the main world, all of this had a feasible path, so it greatly shortened the time for the entire process to mature, and the first generation of lithography machines was brought out faster.
As for packaging and testing, this is the last step in chip manufacturing and is considered the step with the lowest technical difficulty. The main task is to connect the chip to external pins and encapsulate it in a protective casing for connection and installation with other electronic devices.
Chip testing is relatively complex, so it has become a separate independent stage. This stage mainly involves comprehensive testing of the chip's functions, performance, and reliability to ensure that it meets the design requirements.
In the early days, it was intertwined with chip production, but later it became an independent step, making the professional division of labor in chip manufacturing more detailed and efficient.
However, in the early stage of chip development in the Bright Sword world, it is naturally impossible to separate it out. Instead, after researching and improving the process equipment for packaging and testing, a dedicated packaging workshop and testing workshop will be established to do the last link of chip manufacturing.
After several years of development, the process designs and equipment for these steps have finally been equipped.
The first real CPU produced was the 8080 that Ren Zhong 'borrowed' from the main world! (End of chapter)