October 17, 2012 - In another move to spur EUV lithography development, ASML Holding HV (NADSAQ: ASML) has agreed to acquire lithography light source provider Cymer Inc. (NASDAQ: CYMI) for €1.95B (US $2.5B) in cash and stock, calling the deal "the natural evolution" of their year-long cooperation in developing EUV technology.
Combining Cymer’s EUV light sources with ASML’s tool-building expertise will more tightly integrate development work, reducing risk, simplifying the supply chain, and accelerating the introduction of EUV litho technology, according to the two companies, who call this "a natural next step" from their R&D collaboration. Cymer also has an established business for dry deep-ultraviolet (DUV) systems that it sells to other scanner systems providers; ASML says this unit will "remain a significant and growing engine of sales and profit and will be well positioned to support and balance customer needs for EUV and immersion multiple patterning." ASML pledges to keep Cymer’s commercial operations as an independent division based in the US.
"We believe that this transaction will improve our capabilities to bring new technologies to our customers" while also benefiting shareholders, said ASML president/CEO Eric Meurice. "We expect the merger to make EUV technology development significantly more efficient and simplify the supply chain and integration flow of the EUV modules." He also emphasized benefits in Cymer’s growing business for immersion lithography systems and established customer base for dry deep-ultraviolet (DUV) systems.
"We are very encouraged that ASML’s resources will enable the combined company to continue to develop and successfully commercialize EUV on an accelerated time frame," echoed Cymer chairman/CEO Bob Akins.
Details of the deal are $20.00 cash for each Cymer share, plus a fixed ratio of 1.1502 of ASML ordinary shares; that price translates to nearly $80/share for CYMI, a 61% premium over the past 30-day average and 52% over the past 90 days. (As has become commonplace with M&A activity, law firms are already angling for better terms.) Most of that premium has been erased; CYMI stock is up roughly 50% at midday trading. (ASML shares are down about 6%.) The deal, approved by both companies’ boards but awaiting approval from Cymer’s shareholders and US and international regulators, is expected to close in 1H13, and be accretive to ASML’s earnings in the second year after closing.
What it means: Accelerating EUV by consolidating leaders
Many believe EUV lithography is the next key process step in semiconductor manufacturing approaching and surpassing the 2Xnm nodes, as conventional optical lithography reaches its limits and must rely on additional and costly steps (e.g. multipatterning) to work. Thanks to ASML and top chipmakers who have pledged multiple billions of dollars in part to push EUV development up to production capabilities, EUV appears to be a matter of if and not when.
Unfortunately, EUV’s promise has been known for years, and its progression toward volume-manufacturability has repeatedly lagged behind expectations. Source power has been the main culprit, still many factors below what is required to achieve cost-effective throughputs for most chipmaking applications. ASML and Cymer say they have proven "a sustained 30W source exposure power potential" which would translate into 18 WPH in ASML’s NXE:3300B system. Targets for 2014 and volume production remain 105W and 69 WPH, according to the companies. Around 50-70 wafers/hour could be acceptable for some applications, while others need well in excess of 100 WPH to make economic sense, likely starting with DRAM manufacturers and next probably foundries.
In its statements, ASML said it will continue to deliver and service DUV and EUV sources for all customers "on an arm’s length basis," and that its litho systems "will continue to interface with light sources from all manufacturers." Indeed, Cymer and its laser-produced plasma (LPP) technology are not the only EUV source developer in the game. Ushio/Xtreme reported at the recent EUV Symposium some notable performance improvements with its laser-assisted plasma (LDP) source, which is installed on an ASML EUV tool at imec — but it still is not demonstrating sustained high power, duty cycles, or continuous mode operation. Gigaphoton, meanwhile, is making progress with its own EUV source, but remains well behind the other two.
One has to wonder, then, with ASML literally buying into Cymer — the two far-and-away leaders in their respective EUV technology development — what this says about the readiness and future expectations for these other EUV source technologies.
What the analysts are saying
Given EUV’s plodding progress toward volume production readiness — it was supposed to be ready several nodes ago, and might still end up several nodes away from full adoption — ASML’s continued efforts to consolidate EUV technology development efforts is not only smart, it might be keenly necessary to getting the technology off the ground. "Cymer’s light source is critical to EUV success and given recent slippage of key metrics, we think it makes sense for the technology to move in-house at ASML," notes Barclays analyst CJ Muse. "With ASML now having both R&D and equity investments from TSM/SEC/INTC, and CYMI pulling back from its stated EUV targets, it makes sense that there is more urgency for ASML to get EUV right."
Muse notes that regulatory issues might not be so much a problem as in typical M&A, since ASML has built an arguable monopoly position in EUV systems and Cymer was progressing toward that in the light source area, so "customers will push this deal to get done with regulators." He also doesn’t see any other bidders swooping in for a play, which would both delay and raise the price.
Deutsche Bank’s Vishal Shah points out the timing of the M&A is perhaps significant given the overall caution and concern about "cyclical headwinds" (i.e. persistent market sluggishness and lack of visibility). This might be why ASML structured the Cymer deal as a stock/cash transaction and not an all-cash deal, he says.
And although any M&A action causes the market to look around at the next potential target(s), Shah concludes that this is very specific to litho and EUV, and shouldn’t be interpreted more broadly into other semiconductor capital equipment trends — though, as always, "some smaller semicap companies could be viewed as next M&A targets," he supposes.