0p0 Page 495
Page 495
At the lithography machine, the workers took out the small box containing the bottom plate (mask). The bottom plate shipped from Beijing was carefully installed in place, stuck, and its installation accuracy and levelness were repeatedly measured with instruments.
"For the first time, just [-] pieces. I want to see the change in the yield rate of this machine." Tang Hua said.
When the chip production line runs, there is high-frequency noise, but there is almost no vibration.
Vibration and low-frequency noise (also indicating that the machine is vibrating) are a big trouble for chip production, and the machine itself will definitely try to avoid vibration.In fact, chip production lines are also required to be built in places with few earthquakes as much as possible.
In the quiet but sharp noise, the 100 mm diameter monocrystalline silicon wafer was coated with photoresist, entered the production line and began to circulate.
"Old Wang, don't forget that for the new optical lens system, we must find a way to increase the diameter of the beam to 150 mm," Tang Hua said. "In Fushun, the 150 mm diameter monocrystalline silicon column is actually qualified."
Wang Dayan: "We are working on the design of 150 mm and making the manual of the process standard. ... From 100 mm to 150 mm, it is another tough battle."
……
Wafers of monocrystalline silicon flow between assembled lines, one machine to another.
In the afternoon, the first silicon wafer came off the production line, and then the second, third... almost went out with a clatter.All 200 silicon wafers were spit out, and the production line stopped.
That's it for test production.If this production line is running at full speed, one line can meet the computer needs of countries all over the world (actually China and Eastern Soviet Union) in 1954.
"Originally, I wanted to use the bottom board of the X5404 chip for testing," Tang Hua said. "But testing the yield rate of the X5404 is not that simple. Then we have to transport a group of skilled workers and a packaging workshop to Changchun."
Since this is not cost-effective, it is better to make a simple thing that can also test the yield rate of each silicon wafer.
Tang Hua picked up a silicon wafer with a tool, on which dozens of identical patterns were printed.Clamp this piece of silicon wafer to an instrument, turn on the instrument, and a joint will automatically extend out, attach to the silicon wafer, and align with the small pattern on the silicon wafer.
This is a separate compartment in the workshop, as dim as a darkroom where film is developed.Tang Hua took out a standard bright flashlight, turned it on, and let the light shine on the silicon wafer.
A small light bulb used for testing on the instrument lit up.
Tang Hua pressed the button of the flashlight twice, and the light intensity of the flashlight decreased by two levels.
The small light bulb on the instrument also dimmed a lot.
Tang Hua turned off the flashlight, and the small light bulb of the instrument was completely dark.
"Oh Huo?" Huang Kun said, "Einstein's photoelectric effect?"
Tang Hua: "It's much simpler than that, Photovoltaics, the photovoltaic effect."
Da Vinci's solar flashlight, no, the photovoltaic effect, was discovered in 1839.
In 1883, Charles Fritz covered the selenium sheet with an extremely thin layer of gold to make the first photovoltaic cell - selenium and gold foil formed a semiconductor metal junction, but the photoelectric conversion efficiency was only 1%.
If you want to spread, you can't use such precious things as selenium and gold leaf.Doping a certain amount of specific impurities in silicon can also have a photovoltaic effect, and the conversion efficiency is much higher than that of gold foil.
……
It is used to test the yield rate of lithography machines, which can be regarded as an alternative application of photovoltaic cells...
As long as you measure how much current a silicon wafer generates under a known light intensity, and how far it is from the theoretical value, you will immediately know the yield rate of the silicon wafer processed by the lithography machine/chip production line.
And if you engrave a batch of X5404, you have to test one by one how many of the dozens of processors on a silicon chip are working and how many are waste products.
Chapter 91, Laser Plane Interferometry
Using photovoltaic semiconductors to test the life of the bottom plate (mask) of a non-contact mask lithography machine is indeed very fast.
A total of 200 pieces of monocrystalline silicon were engraved in this day's trial production.The life of the direct contact mask lithography machines in Beijing and Shanghai is 200 times.
The bottom plate is in direct contact with the silicon wafer/photoresist, no matter how gently it is handled, abrasion and contamination are inevitable.Starting from the 40th piece, the bottom plate pattern was damaged, and the pattern exposed by the photoresist was flawed, and the yield rate of the chips produced gradually decreased.The 200th piece is a critical point. If you continue to use this base plate, the yield rate is so low that it is not worthwhile. This base plate should be scrapped and replaced with a new one.
But the price of making a base plate is not cheap...
For this production line in Changchun, the yield rate of the 200th monocrystalline silicon wafer has not changed significantly from that of the first wafer.The yield rate of each monocrystalline silicon is made into a table. The 1th piece has a small high point, and the 20th piece also has a small high point, but the entire function curve is very flat.
This is in line with previous theoretical calculations.Of course, the life of the bottom plate of the non-contact lithography machine is not unlimited. It will slowly age and become brittle under long-term strong laser baking, but it is much lighter than physical wear after all.The theoretically estimated base plate life is more than 2500 pieces, and only 200 pieces have been engraved. Of course, there will be basically no decline in the yield rate.
Wang Dayan and Huang Kun praised the photovoltaic cells produced.
"Let me see, there are only these three solar cells on the circuit, and the three small ones can light up the light bulb. If it is a palm-sized solar cell, it may be able to light up the light bulb in the house. "Wang Dayan said.
"Would it be possible to run a computer with a solar cell the size of an umbrella?"
"If you don't include the display, you probably don't need the size of an umbrella."
Huang Kun was thinking about another thing: "Let's go under the sun and try the power later. The light intensity of the glare flashlight is actually far worse than that of the sun."
At this time, Wang Dayan was still researching silicon wafers of photovoltaic semiconductors: "Silicon wafers can be made into processors, volatile memory, memory, and solar cells. It is really amazing."
Wang Dayanyuan specializes in optoelectronic instruments in Changchun, and is not particularly familiar with the latest types of chips that Team X is tinkering with.
Tang Hua: "Semiconductor technology will have more magical applications in the future."
Just wait, CCD is already planning.
……
Changchun Optical Instrument Factory.
Changguang has split into two factories, but the headquarters still has a lot of business. In the past two years, it has continuously recruited technical secondary schools and college students, and the talent gradient is very complete.
"Now that the Kunming Optical Instrument Factory has been rebuilt and put into production, Kunming will focus on the production of ordinary military telescopes in the future. After all, the technical content is a little bit worse."
Tang Hua met the factory director, chief engineer and deputy chief engineer in Changguang.
"However, Su Liu and Su Ba still keep a small workshop to maintain low-volume production, old and new, and make some telescopes, as a way to give new workers a long-term skill."
"Changguang's new mission is to manufacture optical lenses for cameras and movie cameras."
The factory manager is used to Changguang always receiving all kinds of unimaginable new tasks.The chief engineer was even eager to try, and asked:
"Mr. Tang, well, besides making optical lens sets, Changguang can also open a camera branch factory to produce cameras, and then there are movie cameras... Isn't Changchun Film Studio right here?"
Tang Hua: "Now Beijing, Shanghai, and Tianjin are all making cameras, but the camera factories in these three cities all use imported optical lenses. Tianjin and Beijing use Soviet lenses, and the Shanghai Camera Factory even uses original American lens groups. In the future Changguang will be the supplier of lenses and lens groups for these camera factories, Changchun will not make the whole camera, but Changguang will make individual lenses in the future.”
"Understood, Minister Tang."
Tang Hua: "The core component of a camera is the lens. If you make this thing and do it well, you will become the most irreplaceable leader in the national camera and video camera industry."
……
Chang Guang, the seventh optical lens workshop.
The seventh workshop is where the optical system of the lithography machine is produced.
It took half a year to manufacture the experimental optical system last year. Afterwards, Changguang became familiar with the process and the workers were also proficient. In addition, two groups of people worked hand in hand, and now it can produce an optical system in an average of 75 days.