7.9
IntellectualThe Intellectual
EUV lithography machine, may be the most cutting-edge technical achievement that human science and technology can reach so far | Source: ASML.com
Editor's note
At the regular press conference of the Ministry of Foreign Affairs held on July 6, a Bloomberg reporter once again mentioned the issue of the United States requiring Dutch company ASML not to export the most advanced lithography machine to China. At the recent 2022 SPIE Advanced Lithography Conference, Asmay introduced that its new EUV lithography machine is under development, and the next 2-nanometer chip is no problem. TSMC also announced its 2nm manufacturing technology in mid-June this year, and plans to start production in 2025. Chinese chip manufacturing companies have no chance to obtain these most advanced chip manufacturing technologies. By reviewing history, this article hopes that readers can understand from it that the research and development process of lithography machines can be said to be "frozen three feet, not a day's cold."
Written by Chen Qi Editor | Di Lihui
● ● ● ●
chip becomes "Bao Cai Price"
Figure 1 PE 100 Team | Source: chiphistory.org
There were two types of lithography machines at that time, contact type and progressive type.
Contact lithography machine covers the photomask plate on silicon wafer coated with photoresist, open the light source, and "click" to complete the exposure. Obviously, there is a problem with this method. photoresist can easily contaminate mask board, and as the number of exposure increases, the mask is very easy to damage, so the failure rate is very high, the chip yield is extremely low, and the cost is expensive.
. The progressive lithography machine has the photomask plate that does not come into direct contact with the silicon wafer. A measurement tool is added to the lithography machine to make the two as close as possible. But the problem that follows is that light has a diffraction effect, and the edges become blurred during projection, resulting in a decrease in accuracy and a large projection error.
Because the yield is too low, a 4-inch silicon wafer cannot produce a few chips at all, so the chips back then were extremely expensive and even the wealthy US military could not afford it. To reduce costs, technological innovation is needed.
Micralign 100 is not complicated. It is a reflective projection system. It uses two coaxial spherical mirrors to project the figure on the mask plate onto the silicon wafer after three reflections. This symmetric spherical mirror can eliminate most of the aberrations generated and achieve ideal resolution. The birth of
Micralign greatly improved the yield of lithography technology, from about 10% of contact lithography technology to 70%. The yield rate has increased significantly, and the chip price has dropped sharply. A year ago, the Motorola processor 6800 sold for US$295, and a year later, the 6502 processor 6502 sold for only US$25. Cheap chips have prompted microcomputer to emerge like mushrooms after a rain.Although the price of
is three times that of progressive lithography machines, the actual yield is improved, and the ultimate production of more chips can be produced, allowing all customers who purchase Micralign 100 to make a fortune. Orders from Intel , Deyi and other companies are flying like snowflakes.
entered the field of semiconductor equipment in less than three years, and Perkins Elmer became the largest semiconductor equipment company at that time.
However, the good times did not last long. Perkins Elmer's 1:1 pure reflection lithography machine has become increasingly obvious. The reflector cannot completely eliminate spherical aberration, and the image resolution is too low to further meet the reduction process. The stepper lithography machine (stepper) based on lens group has begun to emerge, and lithography technology has entered the era of scaling projection.
GCA's highlight moment
Stepper lithography machine uses a scaling projection method to reduce the pattern of the mask plate to the original 1/4 to 1/5 and then project it onto the surface of the silicon wafer, greatly improving the upper limit of exposure intensity and resolution, allowing the lithography accuracy to enter the micron level. For scaling projection, there is no need for the mask plate and the line width of the transistor, and the requirements for the mask plate are also reduced, which simplifies the mask plate production requirements and reduces costs.
In the early 1970s, GCA replaced the lens supplier with Nikon , but in order to obtain the telecentric lens needed, GCA abandoned Nikon and turned to cooperate with German Zeiss . The reason is that they think that the lens accuracy in Japan is not good, and the lithographic aberration is too affected by the focal length.
In addition to the objective lens system, GCA also needs something new - automated silicon wafer workpiece table (stage) . The exposure area of the stepper lithography machine is no longer the entire silicon wafer, but a small area of it. Therefore, the lithography must be completely broken into zero. Every time it is exposed, the silicon wafer must be moved to the next area and exposed again, and advance step by step until the entire silicon wafer completes the exposure work.
In this way, the positioning accuracy of the workpiece table affects the alignment accuracy, the speed of movement determines the silicon wafer exposure per hour of the lithography machine, and the time of stable operation without faults determines the overall efficiency of the lithography machine. The accuracy and efficiency of the lithography machine depend entirely on the technical level of the workpiece table.
DSW 4800 is the first step-by-step lithography machine equipped with an automated silicon wafer workpiece table.
Figure 2 GCA's stepping lithography machine DSW 4800 | Source: chiphistory.org
This lithography machine uses a 436-nanometer G-line light source and a 10:1 shrink mask plate. Although the lithography speed is not fast and the price of $500,000 is far higher than the price of $98,000 for Micralign, it has overwhelming advantages in stability, resolution, numerical aperture, and intercalation accuracy, which has won the favor of customers of major manufacturers including IBM, Xiantong , and Deyi as soon as it was born.
Although Perkins Elmer also launched the Micralign 500 model this year, which can expose 100 silicon wafers in one hour, giving full play to the advantage of speed, the industry is evolving according to the "script" of Moore's Law . The size of transistors continues to shrink, and the accuracy of lithography is becoming more and more important, and Micralign can no longer keep up.
Immediately afterwards, Japanese two-man Nikon and Canon also began to make efforts. But Perkins Elmer turned a blind eye to all this and continued to indulge in past successes. He neither listened to customers' opinions seriously nor invested in the research and development of the next generation of lithography machines, and was self-satisfied. After the market lost, Perkins Elmer finally woke up, but it was too late. After that, he took the initiative to develop more advanced EUVs, but failed.Afterwards, Perkins Elmer's semiconductor lithography machine division was sold to SVG (Silicon Valley Group) . SVG was acquired by Asmay for US$1.6 billion in 2001. The lithography boss 40 years ago was committed to the current boss. Asmay's most advanced EUV lithography machine has returned to reflector , and technology is like this "spiral" rising.
Although Perkins Elmer failed in subsequent competition, its contribution is undeniable. It is precisely because the Micralign 100 lithography machine has greatly improved the yield of chip manufacturing that chip prices have dropped and more electronic products have entered ordinary people's homes.
The first lithography machine battle, GCA has the last laugh. During this period, several American manufacturers chased each other, reflecting the strong scientific and technological strength and innovation ability of the United States in the field of lithography during this period.
3
Japan counterattack
Figure 3In February 1980, NSR-1010G was launched | Source: nikon.com
Just as the two American manufacturers were in a hot game, in 1982, a lithography machine from Japan appeared in the factory of IBM and Deyi. Whether it is an optical system or a silicon wafer workpiece table, it looks almost the same as GCA's stepper lithography machine. This is the NSR-1010G launched by Nikon that year.
Why can this Japanese manufacturer, a lens accessories, break the GCA's technological monopoly and develop its own stepping lithography machine in just a few years? The story starts in 1976.
In the four-year plan, Japan has chosen several key breakthrough technical routes, and lithography technology and equipment are one of them. As old optical manufacturers, Nikon and Canon have not joined the project on the spot, but have also begun their respective lithography machine research and development tasks under the cooperation framework of the industry-wide organization.
Canon previously mainly made camera lens , and it still lacked in the precision measurement part, so the imitation is the Micralign alignment device with a lower threshold.
. Nikon's predecessor was Japan Optical Co., Ltd., founded in 1917. It has a profound technical foundation in lens manufacturing and precision measurement technology. It can not only make high-resolution camera lenses, but also make astronomical telescope , and even provide military-grade optical rangefinders for battleship . GCA has used Nikon lenses for a while, allowing Nikon to understand the latest technologies in semiconductor optics.
Although its own technology has been well accumulated, it is not easy for Nikon to achieve it from scratch. Fortunately, there is full support from friendly forces. Japan Electric secretly handed over the GCA lithography machine he bought to Nikon for disassembly and analysis, but it was unable to install it after disassembly. Because this lithography machine is extremely precious, Japan Electric could only ask GCA for repair with shame, but the GCA engineer found that the machine had been dismantled, and the scene was once very awkward.
No matter how embarrassed it is, Nikon still learned a lot of valuable knowledge from dismantling the GCA lithography machine and launched its first step lithography machine in 1980.
The first generation of Nikon lithography machine had many problems, but Japan Electric and Toshiba were still very supportive. After buying it, the technicians of both sides worked together to promptly feedback many problems in the actual work process, helping Nikon quickly update and iterate the technology, so that the level of Nikon lithography machine was rapidly improved.
Just as Nikon was making great progress, GCA's production capacity problem has not improved for a long time, because its lens supplier Zeiss was at a low point at the time, with frequent quality problems in the lens and delayed delivery. After one rises and the other falls, other manufacturers gradually lose patience with GCA.
The three American heroes, which still occupied most of the country in the early 1980s, were already shaky by the end of the 1980s and were on the verge of collapse. The two Japanese lithography machines Nikon and Canon rose strongly, replacing American manufacturers, and occupying more than 70% of the market share of .
4
Why did the United States fail?
Why is the once-popular American manufacturers overtaken by Japanese manufacturers in less than ten years? We may be able to summarize some lessons.
First of all, the two Japanese companies have extremely deep foundation in optical equipment and precision machinery. This is the pre-emptive skill for developing lithography machines and the prerequisite for Japan to quickly catch up with the United States.
Second, Japan has concentrated its efforts to accomplish major tasks and provided extremely generous policies and generous funds. The VLSI project pays, hires people, and sets up joint laboratories to achieve joint cooperation between upstream and downstream manufacturers and make common progress. American chip manufacturers are unwilling to share information with GCA for fear of leaking technical details, which has led to GCA keeping it in the dark and not knowing how the real situation of the customer has reached.
Third, Japanese manufacturers vertically integrate degrees higher. Whether it is lenses, platforms, or automation technology, Nikon and Canon are all their own technologies, solving demands from the source, communication of R&D faster, technology iteration is more accurate, and production costs are cheaper. GCA relies entirely on Zeiss' lenses, and once Zeiss has problems with quality control and communication, it will cause disastrous consequences. Indeed, Zeiss was in trouble for a while and had a lot of problems. There were also problems in the early cooperation between Asmay and Zeiss, and almost dragged Asmay into the water. Fortunately, Asmay completely transformed Zeiss.
4 is the conservative and arrogance of American manufacturers. Perkins Elmer is self-conceited and GCA is arrogant. When customers report that Nikon and Canon's equipment performance is better, the management just blames the sales team, rather than reflecting on why their equipment is not as good as competitors.
5. Japanese manufacturers have better business philosophy and service support. GCA's service team in Asia are all sent Americans and cannot integrate into local customers at all. Nikon attaches great importance to customer needs. When the first equipment was delivered to American customers in 1982, it began to hire local engineers to establish Silicon Valley service centers.
6: The early stage of the development of the semiconductor industry in the late 1970s, with various lithography technology routes emerging one after another, and lithography machines are still in the rash era. Japan can surpass its American competitors with efficient and low-cost strategies.
7: Under the economic background at that time, the United States was in an economic recession caused by the oil crisis . GDP has fallen, the unemployment rate has risen, Federal has to implement monetary tightening policies, while Japanese economy is in a prosperous cycle, especially the semiconductor industry.
Japanese manufacturers in the 1980s not only rubbed American manufacturers on the ground in the field of lithography machines, but also conquered other chip markets such as memory, causing American chip companies to retreat steadily. In those years, whether it is morality, fairy child, or AMD, it was difficult for the clay bodhisattva to cross the river to protect itself. Intel was even forced to lay off more than 2,000 employees because it withdraws from the memory market.
But the good times didn't last long. In 1985, the United States forced Japan to sign the " Square Agreement ". Immediately afterwards, the US-Japan Semiconductor Agreement was issued in September 1986, which forced a punitive tariff on Japanese chips, and the United States also supported companies in Taiwan and South Korea.
Although Japanese companies can still rely on previous technical accumulation to continue to maintain their market share, once the two agreements were released, the Japanese chip industry has been in a prosperous state and was forced to retreat under the sword of the United States' policy. They can still eat meat in the industry's prosperity cycle. Once an economic downturn occurs, Japanese manufacturers will suffer countless casualties.
Under the attack of Taiwan's TSMC foundry and South Korea's memory, Japanese companies have been losing streak in memory and advanced digital chip manufacturing.
In 2012, the only Japanese memory company, Miaoer Bida, was acquired by Micron for a bargain price of 2.5 billion US dollars. In 2017, Toshiba Storage was also eventually acquired Holdings by US Capital and changed its name to Kioxia due to the financial problems of Toshiba Group . The once glorious Japanese storage disappeared. Japan has maintained its technological advantages in lithography machines for nearly 20 years until the inconspicuous European small company, which was full of wings, launched three major technical battles, namely "immersion system", "TWINSCAN system" and "EUV light source", and finally defeated Japanese manufacturers and ascended the throne.
5
The giant born in the bungalow
Figure 4 At the beginning of its establishment, ASMA | Source: youtube.com
In the early 11980s, , which had poor financial conditions, Philips , finally decided to shut down its lithography machine project, so he went to the United States to discuss with Perkins Elmer, GCA, Cobilt, IBM and other companies, but no one was willing to cooperate.
On April 1, 1984, Philips finally joined hands with chip machine manufacturer Advanced Semiconductor Materials International (ASMI) to establish a new company, Asmay, right in a rain leaking factory shed near Philips' office in Eindhoven, Netherlands. With some technical accumulation that began to be developed in the 1970s, Asmay released the first lithography machine, PAS 2000, in the same year.
When the new company was established, although it was with the halo of Philips Natlab, at that time, no one cared about Asmay's lithography machine. The only one I have bought is Elcoma, which is Philips Semiconductor and Materials Division. It is considered to be my own care, but there are too many problems due to the use of oil pressure, so I can only stay idle most of the time.
, Asmay advanced into the Asian market. In 1987, TSMC was born and is a joint venture between Taiwan Institute of Technology and Philips.
But in the face of fierce competition, Asmay has few customers. What's worse is that shareholder ASMI was unable to maintain high investments and had a slim return. He decided to withdraw, and his old boss Philips also announced a large-scale cost-cutting plan.As life hangs on the wire, Asmer executives contacted Philips board member Henk Bodt, who convinced his colleagues to lend a helping hand.
soon, Asmay launched PAS 5500. With its industry-leading productivity and resolution, the PAS5500 is profitable. In 1995, Asmay was listed on Amsterdam and New York Stock Exchange , becoming a completely independent listed company.
In February 11995, the first PAS 5500 arrived at the Samsung factory, when Samsung was already the largest memory manufacturer. Samsung has developed a 0.25 micron process for the production of 16Mb memory particles using the PAS 5500. In the following years, the Koreans fully trusted the Dutch stepper lithography machines.
PAS 5500 was very excellent. At that time, except for the Japanese and American markets, ASMA was unstoppable. However, due to customer inertia, Intel and IBM still chose Nikon's lithography machines more. Does ASMA still have a chance to realize its ambitions?
6
Challenge Nikon
Time came to the late 1990s. With the continued evolution of Moore's Law, the process began to start from 130nm to 90nm, and the size of wafer has also been upgraded from 8 inches to 12 inches. At the same time, the wavelength of the lithography machine has also entered from 248nm to 193nm, but unexpectedly, the industry has been stuck in the 193nm wavelength for nearly 20 years.
Until July 2002, at the 157nm microfilm technology seminar held in Brussels, Belgium, TSMC's Lin Benjian gave a special speech on the "Principle of Infiltration". He said, "It's amazing, I found a light wave with a wavelength of 134nm."
Figure 5 Dr. Lin Benjian, who invented the immersion scheme | Source: Lin Benjian
The so-called "infiltration principle" is to add a layer of water above the wafer photoresist. The refractive index of the water is 1.44, so 193nm/1.44≈134nm. Therefore, without changing the wavelength of the lithography machine, the wavelength of 193nm can be equivalent to the wavelength of 134nm!
Nikon has been working on light sources, focusing on how to reduce wavelengths. At this time, Nikon announced that its 157nm product and EPL product prototype were completed, but the actual situation after the test was not satisfactory.
In December 2004, the Japan Semiconductor Exhibition SEMICON Japan was opened. Asmay officially launched the prototype of the immersion lithography machine and proved that the immersion lithography machine solution is feasible.
In 2006, Asmay's XT 1400i entered Intel and successfully passed the verification of the 40nm process. A year later, Intel placed a large order, and other manufacturers followed suit and purchased Asmay's more mature products. Nikon, who was still the leader in the lithography field in 2000, was overtaken by Asmay in 2009, with less than 30% of the market share remaining, while Asmay's market share was nearly 70%.
Nikon's dry 157nm defeated Asmay's 193nm plus immersion scheme. You may ask, why didn’t Nikon change the track and cut to the route of immersion lithography?
The reason is that at that time, the last lens of ASMA 193nm was flat and could seamlessly connect to the immersion system, while Nikon was a curved lens, and the entire objective lens system had to be redesigned, which would take at least 2 years.
Even Nikon could have come up with a similar immersion lithography machine at that time, it might not be able to turn the tables. New equipment always takes several years to work hard by multiple manufacturers. Others mass-produce it earlier than you, so they will have more time to improve problems and improve yields than you, which makes the latter more difficult to surpass, one step behind, one step behind.
Nikon lost 157nm and lost its first chance.
In 2020, Asmay shipped the first dry NXT system in history. This is the first lithography system that can handle more than 300 wafers per hour - thanks to the latest TWINSCAN platform technology on the system.
TWINSCAN, a dual-scan workpiece table, is one of the biggest secrets for Asmay to maintain competitiveness.
As early as the early 1990s, when the PAS 5500 was shipped, Professor Bert van der Pasch, an expert in precision instruments and interferometer , was also conducting research on interferometer systems and wafer transfer modules. He then joined ASMA to become an expert in the position measurement system of lithography scanners.
Bert leads the team to continue to innovate and keeps the PAS 5500 in the leading position, providing the industry-leading productivity and resolution at the time. With the arrival of the new era, Asmay realizes that a revolutionary innovation is needed to help customers like TSMC, which have a strong demand for high production capacity, achieve the next stage of leap.
Looking back, the solution is actually very simple. Before the pattern is exposed to the wafer, the wafer must be accurately measured. Both measurement and exposure take time. In order to reduce the time required for each process, why not start measuring and alignment of the latter wafer while exposing it? In this way, the TWINSCAN system was born.
TWINSCAN is the first lithography system with a dual wafer working platform. The wafers are alternately loaded onto the platform. When one wafer is exposed, the other wafer is loaded onto the No. 2 platform for alignment and measurement. Then the two platforms exchange positions. The wafers on the No. 2 platform are exposed, and the wafers on the No. 1 platform are unloaded. The new wafer is then loaded for alignment and measurement work.
This parallel scheme of measurement alignment and exposure is carried out simultaneously greatly improves the production capacity of the lithography machine within an hour. In 2001, the first lithography machine using this revolutionary technology was shipped - the TWINSCAN AT:750T lithography machine. After
After the TWINSCAN system spans various platform models of ASMA, expanding the technical scope, allowing all chip layers to be exposed on the new platform.
Asmai relies on the TWINCAN system and the immersion system to form a combination of two swords, which completely defeated Nikon. To this day, Nikon, which is subject to patents and technology, is still struggling to find a better solution that can match TWINSCAN. They once chose to use plug-in components, but the effect was not satisfactory and still could not overcome this threshold.
7
moves into EUV
The next step in lithography technology is EUV. Since the development began in the 1980s, EUV has finally matured today. What would happen if the world had no EUV? Maybe it will always be stuck in the 7nm process.
EUV lithography machine may be the most cutting-edge technological achievement that human science and technology can achieve so far.
Now ASMAI's NXE series EUV lithography machine will be sold for more than 100 million US dollars, and the high-NA version of the EXE 5000 series has been sold for 3000 to 400 million US dollars each.
In the history of EUV research and development, the birth of the EUV LLC alliance in the 1990s was very important.
This organization was led by Intel and the U.S. Department of Energy, and brought together Motorola and AMD, which were still in its prime at the time, as well as the three prestigious American national laboratories: Lawrence Livermore Laboratory, Lawrence Berkeley Laboratory and Sandia National Laboratory, investing $200 million to theoretically verify the possible technical problems of EUV.
Intel also invited Asmay and Nikon to join EUV LLC because the US lithography equipment company had basically died at that time. But this move was obstructed by the US government because they were reluctant to let foreign companies share the most cutting-edge American technology and believed that there should be no "outsiders".
At this time, Asmay showed amazing technological prospects and must squeeze into EUV LLC, although the goal of this organization is to prove the feasibility of EUV technology, not mass production. Asmay lobbies strongly and offers conditions that are difficult to refuse - Asmay invests in building factories and R&D centers in the United States, and ensures that 55% of the raw materials are purchased from the United States.
The final result was that Nikon was excluded, and the more open Asmay was allowed to join after making a lot of promises to contribute to the United States. Another exceptional non-US company was the Infineon , which was then the memory overlord, and it was allowed to join EUV LLC with Micron.
From 1997 to 2003, EUV LLC scientists published hundreds of papers in 6 years, verifying the feasibility of EUV lithography machines. After the theoretical verification is completed, the EUV LLC alliance announced its dissolution. Next, all the practical problems were thrown to the industry.
In order to develop EUV systems, Asmay invested tens of billions of US dollars in R&D expenses in nearly 20 years. In fact, Asmay can continue to make money on DUV lithography machines, but the Dutch still maintains its original intention, refuses to lie flat, and chooses to launch an impact on the peak of technology again.
EUV light has the biggest problem that it cannot pass through any object, including air. Therefore, the machine has to be pumped into a high vacuum state. For this reason, Asmay and Zeiss spent a lot of money to jointly establish the world's largest vacuum cavity laboratory to simulate various problems encountered in a vacuum environment.
In addition, for EUV lithography machines, how to generate light sources and control the lens part of the optical path forward are also two major challenges.
The manufacturing process of these reflective lenses is quite complex. Different materials are stacked layer by layer by layer using coating technology, up to dozens of layers. The surface needs to be almost perfectly smooth and clean. Each nanolayer needs to have a precisely defined thickness. The surface error is within 0.01nm, which is equivalent to making a railway track from Beijing to Shanghai, with no more than 1 mm of ups and downs.
In 2006, an EUV prototype appeared in Asmay's laboratory. Four years later, in 2010, the first EUV engineering prototype of human beings born in Asmay's hands: NXE 3100.
Figure 6 EUV lithography | Source: asml.com
The first experimental EUV lithography machine designed 13 reflectors, which were then reduced to 9. Due to the loss of each reflection, the remaining energy after 9 reflections was only a pitiful 1.007%. 9 reflectors are already the limit, and there is no need to be less, and the resolution is not enough.
So the first verification machine, NXE 3100, can only expose 30 pieces per hour, which of course cannot meet customer requirements. To increase production capacity, you must increase the power of the light source.
At that time, ASMA's light source supplier was the American company Cymer.When developing EUV light sources, Cymer felt that this was an impossible task, which was quite perfunctory. At the beginning, the light source had only 30W of power, and it could not be exposed in a few wafers in an hour. After several years of work, it was still on the spot.
After Cymer's molten worker, ASMA acquired Cymer for US$2.6 billion in 2012, making Cymer a subsidiary of ASMA. After countless efforts, the light source power was finally increased from 30W to 250W, and the production capacity was increased to 125 pieces per hour, barely reaching the commercial standard.
EUV light source generation method is different from previous excimer lasers. In order to generate ultra-short wavelength light with a wavelength of 13.5nm, scientists thought of a solution: use a carbon dioxide laser with a wavelength of 9.2-10.8 microns to continuously bombard the metal tin droplets falling from the air. The diameter of this tin droplet is less than 13 microns. After the tin droplet is stimulated and vaporized, it can generate the required 13.5nm light.
The first pulse bombardment flattens the tin droplets into a cake shape, and the light receiving area becomes larger; the second bombardment of the cake-shaped tin droplets vaporizes them, and the two high-energy laser pulses can instantly heat the tin droplets to 50,000K, thereby adding the tin atoms to the high-energy state and returning to the ground state to release 13.5nm ultraviolet photons.
In order to ensure the sustainability and intensity of light, the frequency of this tin droplet being ejected from the jet is 50,000 drops per second. Because it is bombarded in two times, it is equivalent to accurately hitting tin droplets in one second, which is not bad at all.
After solving a series of problems, the EUV lithography machine was finally carefully moved into major wafer factories and began to work for the most core lithography process in the chip manufacturing process, so the current 7nm and 5nm chips were found. Even the future 3nm, 2nm, and 1nm will be produced by EUV lithography machines.
Asmay's success in EUV has completely cut off all Nikon's thoughts. In the lithography machine market, Asmay accounts for more than 90%, while Nikon survives in the corner. Canon has long stopped playing high-end lithography machines and has only made a living in lower-end fields.
However, although Nikon currently loses to Asmay in the lithography machine market competition, he is still a major player in the lithography machine industry. Nikon has many other products and is also very successful.
In the past 20 years, relying on the three major battles of "TWINSCAN system", "immersion system" and "EUV system", Asmay completely stepped on the former giant Nikon. The inconspicuous small company in the bungalow back then became the absolute overlord today.
(This article was reviewed by Wei Ya, a researcher at the Institute of Microelectronics, Chinese Academy of Sciences. I would like to thank you.)
Plate editing | ginger duck
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-