Israeli High Tech Continues Its Progress
(Part I)

Alexander E. Braun -- Semiconductor International, 9/1/2006

As a country, it is not the largest or most populous: It is slightly bigger than Massachusetts and has a population of 6.4 million, 30 million fewer than California. Yet, like little David of old, Israel has overcome enormous odds not only surviving, but also becoming a technology powerhouse in the process.

In a recent tour of Israel, courtesy of the Israel Export and International Cooperation Institute (IEICI, Tel Aviv), I had the opportunity to meet with the heads of various companies who are counted among our industry's leaders. They shared their views on the origin of the Jewish State's high-tech sector, as well as on various trends.

Israel is a recognized center for metrology and inspection. This is because of three reasons. The first dates back to before Israel's beginning in 1948, when before, during and after World War II, huge numbers of academically trained people left Europe. Since then, Israel has consistently boasted the highest number of Ph.D.s, engineers, physicians, etc., per capita in the world. This has resulted in world-class universities. Chief among these are the Technion, Israel's scientific and technical institute in Haifa; the Weissman Institute; Tel Aviv University; and the Hebrew University.

Much in Israel is familiar, although often with subtle differences, giving the feeling of having stepped into a parallel dimension.

When Israel announced its independence, it was attacked by its neighbors and has spent most of its national life either at war or with the expectation of war. This led to a strong military establishment requiring high-tech weaponry. Restrictions by the United States and Europe on the technology and sale of things, such as infrared equipment and detectors — all of it solid-state-based — led to the national development of needed technology.

Transitioning this technology to metrology was simple, recalled Bob Buckwald, CEO of CI Systems (Migdal Haemek). Looking to the future of metrology, he is optimistic. “Many present technologies are extendable, provided that the metrology, chemistry and temperature are closely monitored.”

Buckwald, like others, views shrinks as a challenge. “When you get to the point where thousands or tens of thousands of atoms are involved in features, or the smoothness of features, or whether quantum effects become evident, then every process parameter becomes increasingly critical. If you don't control temperature, chemistry, pH, particle size, some of the most exotic or unexpected things can affect everything. Whether it is producing the features or part of your optics or vibration support, these things have now become absolutely critical — perhaps even technology stoppers.”

Arnon Gat, president and CEO of Negevtech (Rehovot), believes that, with the proper technology, it is possible to perform inspection with overall lower cost of ownership and still fulfill requirements for the 90 and 65 nm nodes. “Right now, we're doing well in R&D with 65 nm. There really is not production per se, but we're either in fab or pilot lines. At 45 and 32 nm, major equipment changes will become necessary.”

Gat sees a boom in the memory market over the next two or three years, driven by many requirements — 4 Gbit of NAND, flash, cell phones, iPods, etc. “We expect a tremendous investment in fab growth on the memory side. We expect this trend to continue driving over half the equipment purchases over the next three years.”

He added that inspection equipment technology is driven by statistics. “One doesn't have to find all defects — 90% is enough. Users want a go/no-go indication to proceed to the next process step. If you repeat the measurement, you get roughly the same defect count, but slightly different defects because it is statistics-based, not deterministic in the way a CVD platform is, where you require a specific layer thickness with a certain composition and uniformity, period.”

“From an inspection perspective, we see lithography reaching the point where they're writing things that optically we cannot see,” said Gilad Almogy, corporate vice president and general manager of the PDC Business Group at Applied Materials (Rehovot). “Then, when you think you've caught up, they bring in immersion. Inspection is forced to change, requiring considerable R&D. You almost must take every aspect of inspection that has evolved over the years — darkfield, brightfield, etc. — and reconsider it from its fundamentals.”

In the device area, Almogy sees as the device manufacturer's biggest problem the fact that they try to cram as much memory as possible per a given area to be cost-competitive. “Lingering defects that remain as they squeeze the process to its limits are the main problem. Detection has become difficult and slow because of defect size. If shrinks are to continue every half year, as is the case with logic, things will not get easy. It isn't just the issue of finding defects; it's a matter of random vs. systematic defect — defects that arise from lithography vs. those caused by other processes that, because of chip density, are very small. Sense must be made out of these, and the need for DFM is obvious.”

Acknowledgement

The author is grateful to Michael Admon, the executive in charge of business development and electronics at IEICI, for his many courtesies in arranging some of these interviews, and facilitating his travel within Israel.