UHP DI-water steam has thermal, cleaning applications in chipmaking, says start-up
Sometimes words of inspiration come out of the mouths of babes. Just ask Jeff Spiegelman, whose 3-year-old son gave him a big idea for his new venture. “Jacob has been a train fanatic since birth,” he says. “I started to think about steam and how it was being used presently and [will be used] in the future in semiconductor processing.”
Spiegelman, who owned Aeronex before selling the gas purification business to Mykrolis, had been perusing different technologies since leaving the company in August 2004. He started looking seriously at steam purification in the beginning of 2005. His new company, RASIRC, was established in March 2005. The San Diego–based venture came out of stealth mode recently, with a press announcement as well as a presentation at the Semiconductor Pure Water and Chemicals Conference (SPWCC) in mid-February.
Existing steam systems used in diffusion, rapid thermal processing, and other semiconductor manufacturing applications rely on pyrolytic and catalytic hydrogen-oxygen combustion approaches, with annual costs ranging between $15,000 and $75,000, according to Spiegelman. “The Intaeger UHP steam purification system (our soon-to-be-introduced product) generates ultra-high-purity steam from deionized water. For the same usage described earlier, it would cost about $24 annually.”
Spiegelman lists some advantages of his nascent product. “We have eliminated the need to burn oxygen and hydrogen by replacing it with water. The cost should be under $25,000 per system. The only materials needed are a little compressed air, $24 of [semiconductor-grade] DI water, and about 800 W to boil the water.”
The system is based on a membrane purification technology that, as Spiegelman puts it, “has the ability to remove dissolved gases, hydrocarbons, particles, entrained microdroplets, and ionic contaminants. Data recently presented showed the removal of 66 of 67 metals to below state-of-the-art detection limits.” The test data shared at SPWCC also showed a 94% reduction in total oxidizable carbon content, while the mass transfer rate across the membrane was found to be directly proportional to the upstream pressure and was high enough to be commercially viable for future semiconductor requirements.
Art Hochberg, a semiconductor equipment industry veteran now working on chemical vapor infiltration processes at ATK/COIC, has seen working models of the steam generator as well as chemical analyses of the effluent. He believes the RASIRC technology has good potential.
“I think it is an alternate path for steam oxidation that is much safer than the H2/O2 torch and does not have particle problems with the torch injector,” he explains. “It should be easily adaptable to atomic-layer deposition and other platforms now using direct liquid injection of water. RASIRC steam should be much purer than the output of liquid-flow controllers and vaporizers. If steam cleaning takes off, this will be an easy tool to use.”
Spiegelman thinks wafer cleaning could be a future application for the UHP steam technology. “Supercritical CO2 is dead. Megasonics is too destructive to move much farther down the ITRS roadmap, so a new cleaning technology is needed. Dr. Ohmi has just published an article on the improved cleaning ability of using a water-vapor jet to remove photoresist from wafers.
“I believe that ultrapure steam will be part of the next generation in wafer cleaning…. UHP steam is extremely effective in dissolving particles and removing ionic contaminants. Steam can easily be used in either wet or dry conditions. It has the ability to deliver energy very effectively…and to rapidly penetrate high-aspect-ratio structures that cannot be reached by liquid water. In addition, the pressure and temperatures can be easily integrated into present tool technologies. The chemistries could be readily adapted from wet cleaning, since IPA, HF, NH3, or other chemistries can be easily added to the steam.”
Hochberg advises that “at this early stage, they need to get these systems into user beta sites. Collaborating with an OEM might be interesting, but I’ve found, and told many salesmen, that the user can push for a radical change in a tool component, but it is very hard to sell a user on a significant change.” He would like to see “an absolute measurement of the steam delivery, lifetime data on the internal components, and some metric to ensure purity of the delivery.”
Like any new and unique approach, Spiegelman acknowledges that “customers always take a while to get used to a new technology. Based on the most recent input, we have stepped back the technology to make it more of an incremental technology improvement instead of a big leap forward.”—TC
“The BIG Idea” is a new feature that will report on interesting start-up and early-stage companies and technologies in the semiconductor and related micro/nanoelectronics industries.