Cleanroom classifications in both Fed-Std-209E and the ISO 14644 standards are based strictly on aerosol particle concentration. Contamination not detected or counted by a particle counter makes no difference in the classification.
Yet, on a mass basis, the deposition of nonparticulate contamination (a.k.a. airborne molecular contamination [AMC]) in a semiconductor manufacturing facility often exceeds that deposited by aerosol particles by a factor of 104 to 106 [Muller et. al., 1990].1
It would seem something's been overlooked and that the cleanroom classifications certified by the procedures of the Standards are at best somewhat incomplete and perhaps even out and out misnomers. To be fair, both standards explicitly note their restricted definitions of cleanliness in their introductory sections. But after that warning, it's caveat emptor and often the limited cleanliness criterion used in classification by both standards are cursorily noted and then forgotten or ignored. “A Class 1 cleanroom is all I (you) need” has been a common perception in the semiconductor industry. This cavalier usage of a cleanroom classification has been widespread in the semiconductor industry where the cleanroom classification alone has traditionally been the sole specification for cleanroom air quality.
Part of the explanation for this past lack of concern with nonparticulate contamination is that nonparticulate contamination simply hasn't caused the production or yield problems in semiconductor production that particulate contamination has or at least has not yet been widely recognized as doing so.
Conventional cleaning processes have been more successful at removing AMC than particles. Indeed, experiments have shown that conventional cleaning processes eliminate any electrically detectable effects of organic molecular contaminants deliberately added during gate oxide processing (although omitting the cleaning steps before the oxidation [no pre-gate clean] does degrade capacitance-voltage and spreading resistance properties) [Shaneyfelt et al, 1996].2 Organic AMCs, however, are generally not as threatening as the acids, bases and dopants that make up other types of AMCs [Kinkead et. al., 1995].3
The evidence of AMC impact is perhaps strongest in lithography where airborne amines can interfere with deep ultraviolet photolithographic processing. Other interferences from other AMCs are expected to become equally important in coming device generations with the smaller critical dimensions that will characterize these structures. The semiconductor industry's National Technology Roadmap for Semiconductors (NTRS),4 a consensus guideline document prepared by a cross section of industry experts, now addresses the need for control of nonparticulate contaminants and confirms the growing importance the industry now accords AMCs.
Already the NTRS includes target guidelines for ambient concentrations of AMCs in the manufacturing environment. These guideline concentrations are projected to become more stringent over the next decade. Meeting these target values implies more widespread use of AMC measurement methods and control strategies. AMC control needs will no doubt promote the already growing popularity of minienvironments. Controlling contaminants in a small, localized enclosure is simpler and more effective than controlling contaminants in a large ballroom area, whether the contaminants are particles or AMCs. In addition, a minienvironment enclosure can be customized for control of those AMCs most inimical to the specific process being carried out in that dedicated minienvironment. The AMC control advantages of minienvironments have been recognized by the semiconductor industry [English-Seaton, 1998]5 and the status and impact of minienvironments in the semiconductor industry will be discussed in a future column.
What does this need to characterize and control nonparticulate contaminants mean for the current standards for classifying cleanrooms and clean zones? Only that they in and of themselves are no longer, if they ever were, a complete and adequate specification for cleanrooms or clean zones used in semiconductor manufacturing. Recognition of the limited scopes of these standards is not a new development, as the standards have already been considered only a partial specification in the past in many applications, notably pharmaceutical manufacturing. The existing standards for classifying cleanroom cleanliness will not be discarded or abandoned but more and more they will be supplemented as required on an industry by industry basis, including, eventually, standards for AMCs in semiconductor manufacturing.
Robert P. Donovan is a process engineer assigned to the Sandia National Laboratories as a contract employee by L & M Technologies Inc., Albuquerque, NM. His Sandia project work is developing technology for recycling spent rinse waters from semiconductor wet benches.
References
1. Miller, A. J., L. A. Psota-Kelty and J. D. Sinclair, “Concentrations of Organic Vapors and Their Surface Arrival Rates at Surrogate Wafers during Processing in Cleanrooms”, pp. 204-211 in Semiconductor Cleaning Technology/1989, Proceedings Volume 90-9, edited by J. Ruzyllo and R. E. Novak, The Electrochemical Society, Inc., 10 South Main St., Pennington, NJ 08534-2896 (1990)
2. Shaneyfelt, M. et al, “Sandia/SEMATECH Contamination Free Manufacturing (CFM) Research center Report on Organic Contamination in Wafer Environments: Surface Analysis and Impact Studies”, SEMATECH Technology Transfer #96013080A-ENG, February 15, 1996
3. Kinkead, D. et. al., “Forecast of Airborne Molecular Contamination Limits for the 0.25 Micron High Performance Logic Process”, SEMATECH Technology Transfer # 95052812A-TR, May 31, 1995 (available from the SEMATECH website [www.sematech.org] )
4. The National Technology Roadmap for Semiconductors, 1997 Edition, Semiconductor Industry Association, 181 Metro Drive, Ste. 450, San Jose, CA 95110
5. English-Seaton, S., “Airborne Molecular Contamination in Cleanrooms”, Cleanrooms, Volume 12, No. 1, January, 1998, pp. 21-25