He isn’t on the lists of leaders, brilliant technicians and ASML heroes, but Cees Krijgsman gave the young company a push at decisive moments in the 1980s. Behind the scenes, he put the pre-ASML stepper team in contact with its first customer. Later, he himself signed the contracts for the first litho machines from Veldhoven.
In the summer of 1984, Gjalt Smit and George de Kruiff both had a problem. ASML director Smit was waiting for the green light from his shareholders Philips and ASM International for 50 million dollars in financing. Commissioner De Kruiff had the task of convincing senior Philips management to cough up half that amount.
That wasn’t obvious as De Kruiff didn’t have a convincing story. At Elcoma, where Philips had housed its chip activities, they were never satisfied with the home-made steppers. The other joint venture partner, ASM International, had a bad reputation amongst the management of the chip factory in Nijmegen. Smit had gone there for a coffee in the summer and had gotten the door slammed in his face. The largest European site for semiconductor production had told him that they preferred to buy American litho equipment. The Perkin-Elmer machines were running just fine there. They weren’t interested in ASML’s steppers.
Smit not only needed money, but he was also desperate for a good sales pitch. Nobody wanted his hydraulically powered steppers. Why should they say yes to a new machine that was still only on the drawing board? If the chip business of shareholder Philips didn’t even buy ASML, Smit would have a difficult story to tell potential customers.
Coincidence or not, exactly in that year, a formidable opportunity presented itself. Philips and the German company Siemens decided in early 1984 to collaborate on memory chips. The two companies set out to catch up with the Japanese. They invested a total of 1.5 billion guilders (700 million dollars) in the project that was to deliver the first megabit DRAMs in 1989. Hence the name: Mega-project or Megabit-project.
The herculean task required lithographic machines that could image details smaller than a micron. These didn’t yet exist, but ASML was one of the companies that had dived into them. Smit and De Kruiff understood that they had to talk to the man in charge of the Mega project. This was Cees Krijgsman, a man well versed in the world of chips. As the director of Elcoma, Krijgsman regularly visited competitors in the US and he was good friends with Chuck Harwood, the boss of Signetics, which had been acquired by Philips in 1975. Harwood never failed to describe in great detail to Krijgsman what he saw when he looked out the window in Sunnyvale, California. Compared to the Elcoma executive, Smit and De Kruiff were just getting started in the chip market.
Philips elite
Cees Krijgsman belonged to the Philips elite. He was one of the men who worked around the clock for the multinational, but who also met informally on the golf course in Valkenswaard, at tennis clubs and in fancy restaurants such as De Karpendonkse Hoeve.
Back then, Philips was an unwieldy bureaucratic apparatus, with factions fighting each other fiercely. Many managers were primarily concerned with their own careers and company politics. But Krijgsman was part of a select group with an entrepreneurial spirit, of a group of Philips men that wanted to conquer the world. He was determined to give the Japanese a run for their money. The Mega project, therefore, fit him like a glove. For Krijgsman, it was an absolute necessity. “To be competitive, we simply have to be strong in semiconductors,” was his holy conviction. During the interviews for the book “Natlab”, former director of Philips Research Kees Bulthuis said about this, “Krijgsman wanted Philips to be at least in the global top-3 of semiconductors.”
Krijgsman was well acquainted with ASML’s stepper activities. A few years earlier, he’d put the stepper activity at Philips S&I in contact with IBM in Burlington, Vermont, where the computer company was running the world’s most advanced wafer fab. This had resulted in the very first delivery of a Philips stepper to an external customer in 1982. However, IBM had never placed any follow-up orders.
In 1984, it was clear that Mega was the key to financial resources and ASML’s market entry. With 700 million dollars to spend, Krijgsman had a virtually blank check. If De Kruiff could get him excited about ASML’s steppers, that would send a positive signal to other Philips executives. And so, soon after Smit presented his business plan to the supervisory board, De Kruiff suggested they visit Krijgsman, who was in charge of Elcoma’s chip activities.
At the premises of the Philips Research Laboratories (Natlab), the development of the megabit memory chip was in full swing. A test fab was being erected where Philips researchers had access to the most advanced equipment to put Europe back on the global semiconductor map. Natlab was even developing a couple of steppers for the project, as a backup.
The schedule was tight, and it applied to the project’s machinery suppliers, too. On 1 April 1986, Philips wanted to receive the first systems for use in producing megabit memories from several companies at its test site. It would then decide which system to buy three months later. “Then we’ll decide which machines we’ll use for mass production,” Krijgsman told Smit and De Kruiff when they met at the Elcoma offices at De Hurk in Eindhoven.
Smit unveiled his plans: since its internal deadline for the PAS 2500 was 1 January 1986, ASML could meet the April deadline. By now, he’d given his pitch multiple times, and it ran smoothly. Krijgsman was impressed, but Elcoma’s director was also frank about his requirements. He wanted to receive the first PAS 2500 on 1 April 1986 and not a day later. “Otherwise, we’ll go to Nikon,” he warned.
If ASML met the requirements and Elcoma approved the machines, Krijgsman would buy many more PAS 2500 steppers for the Mega project. Smit was astonished. He knew Krijgsman hadn’t been forged in the Philips culture – he’d come to the project from Honeywell – but still, he was surprised by Krijgsman’s constructive attitude. But the requirements were aggressive. ASML had just eighteen months left to develop and produce a machine. From that moment on, a razor-sharp deadline was in effect.
René Raaijmakers is the author of “ASML’s architects,” which is available in Dutch, English and Chinese. He also wrote the book “Natlab,” together with Paul van Gerven, about the history of Philips Research. In this series, Raaijmakers describes notable, striking and never-publicized events from ASML’s illustrious history. In addition to well-known milestones and stumbling blocks, these are often stories that didn’t make it into his books, but that are noteworthy, inspiring, interesting or just plain fun.