Particle transport complexities considered
09/01/2002
Air entering the cleanroom goes through HEPA/ULPA filters, whether this entering air originates as makeup air from outside the building or recirculated air from the exhaust ducts of the cleanroom.
by Robert P. Donovan
Achieving high-quality air in a cleanroom depends on meeting this design feature.
 |
No air can be permitted to seep into the cleanroom through the walls, doors or viewing windows that typically surround the cleanroom workspace.
However, the seals separating the cleanroom from the outside air surrounding it do not necessarily have to be airtight or hermetic, even though many are. Users do need to operate the cleanroom at a slight overpressure with respect to the outside ambient so that convective flow from within the cleanroom to the outside prevents particles from flowing in through any small leaks.
The figure illustrates the need to maintain this positive pressure differential. The top plot shows particle concentration within a clean room as a function of time and, on the same time scale, the bottom graph shows the corresponding plot of pressure differential between the cleanroom interior and the outside.
This pressure differential changed at various times throughout the day on which these data were collected, as the fan power was periodically and inadvertently interrupted, dropping the cleanroom differential pressure from slightly less than 0.07 inch of water to slightly less than 0.02 inch of water.
The impact of these pressure changes on particle concentration within the cleanroom was dramatic. The particle concentration measured in the cleanroom increased by three to four orders of magnitude each time the cleanroom pressure dropped. Admittedly, the particle concentration data reported in the figure were collected by a condensation particle counter located adjacent to an outside cleanroom wall, a position most likely more sensitive to incoming air leakage than locations more remote from the outside wall.
And this particular condensation particle counter detected aerosol particles as small as 0.01 micron, a particle size an order of magnitude smaller than the smallest allowable particle size used to classify cleanroom air quality by any contemporary cleanroom standard.
Nonetheless, the conclusion is clear: Satisfactory cleanroom operation demands that the convective component of airflow through all orifices connecting the cleanroom to the outside exceed that of any inward component, including the diffusive component associated with the large contaminant gradient between the dirty air outside the cleanroom and the relatively particle-free air inside the cleanroom.
Convective airflow through a leak depends on the pressure difference between the inside and outside of the cleanroom. Diffusive transport of particles through a leak depends on the particle concentration gradient, ∂C/∂x, between the outside and the inside of the cleanroom and the particle diffusion coefficient, D, as described by Fick's Law.
J = -D∂C/∂x,
where J = the particle flux crossing a plane perpendicular to the centerline of the leakage path
Diffusion provides a mechanism for transporting particles into the cleanroom through any open-air passageway and, with equal air pressure inside and outside the cleanroom (no convective airflow), particles will enter through these leaks. Not until the convective flow velocity directed outward exceeds the diffusive flow velocity inward will the cleanroom be protected from in-diffusing particles.
For particles of small diameter (<1 micron), diffusion coefficients vary inversely with particle diameter squared. For this reason, diffusion transport is more significant for small particles (~0.01 micron) than for the larger micron-sized particles. However, the net result of particle diffusion and convection transport through a leak strongly depends on the dimensions (diameter and length) and boundary layers of the orifice constituting the leak as well as upon particle size distribution.
Thus, it is no surprise that particle transport problems associated with cleanroom leaks are complex and difficult to solve theoretically. Qualitative understanding is about the best one can do without extensive analyses and verification. Collected data can empirically define the overpressures needed to protect a given cleanroom.
Editor's Note: These data were measured in the cleanroom at the Microelectronics Center of North Carolina (MCNC) at Research Triangle Park, NC.
Reference
- Viner, A. S., "Predicted and Measured Cleanroom Contamination," Chap 8 in Particle Control for Semiconductor Manufacturing, R. P. Donovan, editor, Marcel Dekker, Inc.1990.
Robert P. Donovan is a process engineer assigned to the Sandia National Laboratories as a contract employee by L & M Technologies Inc., Albuquerque, NM.