In lithography, ASML is a monopoly. As a lead user of ASML’s extreme ultraviolet (EUV) lithography machines, TSMC has become a monopoly in high-end (5nm as of 2021) silicon processing. Consequentially, the ASML TSMC nexus has been at the root of fueling the global semiconductor monopoly. However, unlike many other cases, this nexus has been due to the offering of higher quality transistors at a decreasing cost. It did not happen due to big ideas, radical steps, or revolutionary innovation. It did not occur due to collusion or political decision either. This is the outcome of a consistent journey systematically exploiting existing technology cores. The innovation value proposition out of benefiting from science with precision engineering has created the blue ocean, making the competition irrelevant to both ASML and TSMC. Due to its systematic nature in creating revolutionary success, a series of connections linking these two giants deserve investigation.
The semiconductor chip shortage slowing down more than 200 sectors has surged interest to know this $600 billion industry. Unlike many other industries having monopolistic competition, the semiconductor industry’s value chain is highly segmented and distributed all across the globe. Furthermore, no single country or firm poses the material supply or has vital patents. The underlying reason has been the development of specialization out of the exploitation of science through precision engineering in a systematic manner. Consequentially, there has been a monopoly in all significant value chain layers. However, notable ones are ASML’s monopoly in photolithography and TSMC’s unbeatable edge on processing the smallest transistors. It happens that TSMC needs the latest machine of ASML to keep sharpening the edge. On the other hand, without TSMCs early adoption and close cooperation, ASML cannot proceed to improve its machine further. Hence, there has been ASML TSMC nexus.
ASML TSMC monopoly status:
In the photolithography market, ASML had an 80 percent maker share in 2021. However, in 2005, all the three competing firms, ASML, Nikon, and Canon, had a more or less equal market share. But in the EUV market segment requiring a 13.5 nm light source, ASML has 100 percent market share, a true monopoly. On the other hand, TSMC’s market share is almost 54 percent in the Foundry market segment, followed by Samsung, UMC, and GlobalFoundries. But in the 5nm silicon processing, TSMC has 100% market share, as Samsung’s 5nm service is questionable. Hence, both ASML and TSMC have a 100 percent share in their respective high-end markets. But none of these two monopolies started with big ideas. Instead, they kept growing simple ideas in making them grow out of systematic exploitation of science and precision engineering.
Underlying science, formation, and shareholding of ASML:
Photolithography is not a new technology by exploiting which ASML has become a global monopoly and the 32nd most expensive firm in 2022. The concept is about the change of chemical composition of material due to exposure to light. In the 1830s, Niépce developed the portable camera obscura using it. As the formation of a transistor requires area-specific treatment of the substrate, in the 1950s, Bell Labs started the experimentation of using photolithography for having images of design on photoresist covered silicon wafer.
To make the transistor increasingly smaller, in the 1960s, a race started to improve key performance indicators (KPIs) of the photolithography technique. The most important of these KPIs are (i) making the wavelength of the light source smaller and (ii) improving lenses and optics for increasing numerical aperture for gathering and projecting light. In this race led by American and Japanese firms, ASML is a late entrant.
ASML stands for Advanced Semiconductor Materials Lithography. Its parents are Philips and ASM international. Among them, Philips has a long history of producing light sources. It also has strength in optics due to its innovation in microscopy and scientific instrumentation. On the other hand, Arthur del Prado founded ASM as ‘Advanced Semiconductor Materials’ in Bilthoven, the Netherlands, in 1964. This company had business in trading and making semiconductor material processing equipment such as chemical vapor deposition equipment. However, to focus on lithography, ASM international and Philips got together and gave birth to ASM Lithography (ASML) in 1984—a joint venture. Inauspiciously, it began the journey in a leaky shed next to a Philips office in Eindhoven, the Netherlands. Although both the parents sold most of ASML shares, three significant customers, such as TSMC, Samsung, and Intel, are among ASML’s strategic shareholders.
ASML’s growth from the exploitation of science through precision engineering:
Due to having roots in ASMI and Philips, in the same year of its formation in 1984, ASML launched its first product—PAS 2000 stepper. However, in the market dominated by Nikon, Canon, and American trios, the product became obsolete within just one year. Due to continued focus on sharpening resolution by improving light source and optics, ASML reached duopoly status by the end of 1990s.
However, for turning the competition firmly in its favor, ASML had to decide to accelerate progress. The focus was on the light source. Target was set for reaching from deep ultraviolet (DUV) region that occupies wavelengths between 280 nm and ~200 nm to extreme UV (EUV) region ranging from 124 nm to 10 nm wavelength. This decision demanded finding new ways to produce, gather and project light of such a small wavelength. Hence, it required cooperation with lead users.
Formation and humble beginning of TSMC:
In 1987, Dr. Morris Chang took the lead to form a semiconductor fab–TSMC. The business was of delivering silicon wafer processing services as per the design of customers. Small chip-making companies without having fabs became TSMC’s customers. Unlike integrated device makers (IDMs) like Intel, AMD, or Motorola, fabless companies were demanding older generation processing technologies. Hence, in the beginning, TSMC was not the customer for state-of-the-art lithography equipment. For sure, patronizing the next-generation lithography was out of the question for TSMC.
However, in the 2010s, TSMC’s appetite for higher precision silicon processing technology started growing. This is due to the growing demand for fabless companies. Smartphone makers were asking them to supply processors with increasing chip density, higher frequency, and lower power consumption. Hence, TSMC started accelerating the processing technology for decreasing the feature dimension. But still 2010, in adopting lithography technology, TSMC was far behind Intel and Samsung. However, TSMC anticipated that situation would rapidly change due to the growing complexity of smartphone processors.
ASML TSMC nexus for managing risk and expediting learning for exploiting EUV:
The concept development of EUV started in the 1990s and early 2000s in NTT in Japan, the Bell Labs and Lawrence Livermore National Laboratory in the U.S., and the University of Twente in the Netherlands. But ASML’s EUV journey of innovating practical machines unfolded with the release of Alpha Demo Tools in 2006. TSMC sent that complex machine to imec in Belgium and Suny Polytechnique Institute in the USA so that researchers could test and share results for empowering ASML to fine-tune a new system that involves a plasma source and reflective optics operating in a vacuum.
Subsequently, in 2010, ASML shipped TWINSCAN NXE:3100 machine to a major customer in Asia 2010. Using 27 nm wavelength, this machine promised to produce a 7nm feature dimension. But soon, ASML learned that step from the R&D fabs of chipmakers, in evaluating and fine-tuning EUV machines, to the high-volume manufacturing fabs was a big one.
By 2012, ASML concluded that although the EUV scanners could print chip features for the next technology nodes, many issues the production was facing demanded one by one attention. Hence, transferring R&D outputs into the manufacturing process became highly challenging and required further R&D.
Therefore, it became clear that ASML’s next-generation photolithography development, using EUV, demanded prolonged high investment in R&D. It also needed close cooperation with lead users of fine-tunning design. Furthermore, there was also a risk in demand, as the number of customers was also very low. Hence, ASML embarked on forming a collaboration with lead users. Among others, TSMC, Intel, and Samsung responded to it by investing in ASML. In 2012, these three companies agreed to contribute EUR 1.38 billion to ASML’s research and development. Furthermore, ASML also received EUR 3.85 billion for issuing shares to the three participating customers under the Co-Investment Program.
ASML TSMC nexus has created virtues cycle:
On the one hand, ASML was desperate to move with EUV. But it could not do it alone without the active involvement of lead users in testing and helping in fine-tuning the next-generation photolithography machines. Coincidentally, TSMC’s desperation for getting lithography machines for moving to the next node of feature dimension also started accelerating. This is due to the accelerated rise of fabless companies’ demand in offering silicon processing services for decreasing chip dimensions. It intensified with Apple’s decision to engage TSMC foundry to process silicon for its series of chips for iPhone and iPad. Hence, a resonance between ASML and TSMC urged to quickly move to the next node leading to 5nm in 2020. In desperation, even Intel fell behind TSMC in two-generation of chip technology.
Hence, in addition to co-investment, ASML started vigorously getting R&D inputs from TSMC to accelerate the progress, making it the only company offering EUV photolithography machines. On the other hand, TSMC’s active role in fine-tuning and adopting following generation machines has made its global monopoly in 5nm and following 3nm chip processing technologies.
Further to testing future generations’ process technologies and helping ASML refine the design, TSMC also contributed to developing concepts. For example, TSMC made an active contribution in developing immersion lithography technology.
There have been powerful R&D partnerships between TSMC and ASML in two critical phases of developing next-generation EUV photolithography machines. And both of them were aggressive in this cooperation due to their complementary business interest. As the problem was so big, people of both sides actively helped each other—creating this nexus of creative destruction. In the end, the ASML TSMC nexus has resulted in monopolies in two major segments—Lithography and Foundry.
...welcome to join us. We are on a mission to develop an enlightened community by sharing the insights of wealth creation out of technology possibilities as reoccuring patters. If you like the article, you may encourage us by sharing it through social media to enlighten others.
- Semiconductor Value Chain–globally distributed ecosystem
- Semiconductor Monopoly Due to Winning Race of Ideas
- Intel Falling Due to PC and Mobile Waves
- ASML Lithography Monopoly from Sustaining Innovation
- Taiwan’s Semiconductor Monopoly – How did it arise?
- ASML TSMC Nexus Fuels Semiconductor Monopoly
- ASML Monopoly in Semiconductor — where is magic?