by Robert P. Donovan
|
By my definition, contaminants are any impurities or undesired materials that degrade product performance or reliability or adversely effect product manufacturability.
Particles of any composition larger than a given size, or of most any size of the wrong composition, are a well-recognized, prime example of a contaminant by my definition. However, non-particle contaminants with mass also exist.
In the working sense, molecular contaminants can be considered any contaminants not detected by a particle counter. And herein lies a problem, because ISO Global Cleanroom Standards for classifying the quality of a cleanroom are based totally on particle counter measurements of aerosol particle concentrations, not total contaminant concentrations. These standards completely ignore molecular contaminants. Any contaminant not counted by a particle counter has no impact on the classification assigned to a cleanroom by these standards.
In this respect, contemporary cleanroom classifications lag the needs and practices of the semiconductor industry. The 1999 International Technology Roadmap for Semiconductors,1 for example, contains numerous citations of processes and process areas within cleanrooms for which measurement and control of molecular contaminants are already recognized as being critical. A prime illustration of this sensitivity to molecular contamination is the well-known degradation of certain photoresists exposed to airborne amines and other volatile organic bases.
Photolithographic areas in most semiconductor manufacturing areas are already protected by chemical filters of some kind that physically adsorb or chemically react with amines. Typically these chemical filters do not efficiently control particle concentration, sometimes they add particles to the airflow. They are introduced strictly for control of molecular contaminants. At present the ISO Standards do not consider this need and in this sense do not adequately discriminate among cleanrooms designed for photolithographic areas.
Indeed, outgassing from the very HEPA or ULPA filters used to control particles in cleanrooms can be a source of molecular contaminants, including dioctyl phthalate and boron compounds. Even more alarming are the early data published by Muller et al2 showing that on a mass basis the flux of molecular contaminants to a wafer surface in an ISO Class 2 (Fed-Std-209E Class 10) cleanroom exceeds that of the particle flux by more than six orders of magnitude.
The deleterious effects of molecular contaminants have been more subtle than those of particles and have taken longer to recognize. Perhaps somewhat belatedly, that “oversight” is now being remedied by the people who prepare cleanroom standard, [ISO 14644-8] (see “Saving the most complex for last,” CleanRooms, Dec. 2000, p. 50).
In contrast to the delayed recognition of the need to include molecular contaminants in air standards, guidelines for the quality of semiconductor ultrapure water (UPW), as in documents published by ASTM3 or in preparation by SEMI,4 have long specified measurement and control of molecular contaminants. In UPW, particle contamination typically receives less emphasis than molecular contaminants.
The importance of molecular contaminants in air is only now becoming widely recognized.
Robert P. Donovan is a process engineer assigned to the Sandia National Laboratories as a contract employee by L & M Technologies Inc., Albuquerque, NM.
References
- International Technology Roadmap for Semiconductors, 1999 Edition, Semiconductor Industry Association.
- Muller, A.J., L.A. Psota-Kelty, J.D. Sinclair and P.W. Morrison, “Concentrations of Organic Vapors and their Surface Arrival Rates at Surrogate Wafers during Processing in Cleanrooms”, PV 90-9, pp. 204-211, J. Ruzyllo and R. E. Novak (eds), 1990 (The Electrochemical Society).
- ASTM D 5127-99, “Standard Guide for Ultra Pure Water Used in the Electronics and Semiconductor Industry (ASTM).”
- Dougherty, D., “Development and Application of High Purity Water Standards”, presentation at Watertech 2000, November 2000 (Tall Oaks Publishing Inc.).