ASM押注新材料，傳統晶片微縮已達極限

A CEO Summit panel session at SEMICON West 2026, held at the Moscone Center in San Francisco on Oct. 14. From left: Dave Anderson, president of NY CREATES; Glen Wilk, vice president of technology at ASM; Eric Byers, vice president of advanced process, mask and laboratory technology at Micron; Debbie Gustafson, CEO of Energetiq Technology; and Mukesh V. Khare, CEO of Anderon. (Photo: THE ELEC)

“Traditional scaling has effectively reached the end of its useful life.”

Glen Wilk, vice president of technology at ASM, made the remark during a CEO Summit panel session at SEMICON West 2026, held at the Moscone Center in San Francisco on Oct. 14 local time.

“Materials will no longer merely enable scaling but will drive semiconductor innovation,” Wilk said. “The future of computing will be determined by materials, beyond semiconductor design and device architecture.”

Wilk explained that process miniaturization alone can no longer meet the computing performance and power efficiency requirements of artificial intelligence (AI).

He cited transistor gates as a prominent example of a past materials transition. Silicon dioxide, which had been used as a gate dielectric for more than 50 years, was replaced by high-k materials such as hafnium oxide. The transition was intended to reduce gate leakage current and improve transistor performance.

“Twenty years ago, the challenge involved a single device and a single thin film. Today, it spans the entire semiconductor industry, including logic, memory and advanced packaging,” Wilk said. “The scope of materials innovation required has become much broader.”

Wilk also emphasized the need for materials innovation in thermal management. Heat dissipation becomes increasingly challenging as logic, memory and optical devices are integrated into smaller spaces and operated at higher speeds.

ASM’s emphasis on materials reflects its equipment business. Introducing new materials also requires new processes to deposit them onto wafers. Atomic layer deposition (ALD), which enables thin, uniform layers of hafnium oxide to be deposited at the atomic scale, became essential with the transition to high-k dielectrics. ASM secured an early position in this market and expanded its business.

ASM’s main revenue sources are ALD and epitaxy equipment. New interconnect metals such as molybdenum can also be deposited using ALD. Silicon and silicon-germanium thin films required for gate-all-around (GAA) transistors are grown using epitaxy equipment. For ASM, the introduction of new materials creates opportunities for revenue growth.

ASM has also changed its research and development (R&D) approach to accelerate the adoption of new materials in mass production. The company places R&D teams close to customers in the United States, Europe and Asia to identify their requirements early and share research findings across regions. Its advanced R&D teams assess the potential of candidate materials even before beginning customer collaborations.

“To move quickly, we need to stay connected,” Wilk said. “We must quickly eliminate materials with limited potential and focus on promising candidates to shorten the time it takes to reach mass production in collaboration with customers.”