by Robert P. Donovan
 |
The remarkable success of the fibrous filters used to create near particle-free cleanroom environments for manufacturing precision products easily overshadows their darker side. While the particle-removal capabilities of these fibrous filters are undeniable, recognition of some of their other, less desirable properties has been slower in coming.
One of the early indications that all was not perfect in the HEPA filter world was a 1981 publication reporting particle shedding from the filter itself.1 Particle shedding refers to particle emissions originating from the filter media, its housing or the filter mounting hardware. Particle shedding from media is undoubtedly aided by exposure of the filter fibers to some of the reactive chemicals used within a cleanroom. But V.K. Bhola2 reported measurable particle emissions from both newly installed HEPA filters as well as from “aging” effects that characterized some HEPA filters.
Another property not desired in semiconductor manufacturing is outgassing of boron from the borosilicate glass from which the fibers are made and phosphorus from sealants used in filter construction. Both boron and phosphorus are standard dopants for controlling the electrical properties of silicon semiconductors. That exposure of silicon wafers to uncontrolled sources and concentrations of these impurities can be catastrophic to product yield.
One of my most disappointing projects in the distant past was assessing the magnitude of dioctylphthalate (DOP) emissions from cleanroom HEPA filters that had been challenged with DOP-based aerosols during manufacturing. An early test method for confirming the performance of HEPA filters consisted of generating a challenge aerosol from a liquid source of DOP. Apparatus existed for reproducibly generating an aerosol of appropriate known size distribution for measuring the collection efficiency of these filters.
Many filter manufacturers proudly proclaimed their practice of 100% filter testing. To assess the magnitude of DOP emissions from such filters, I obtained HEPA filters that had been deliberately exposed to varying concentrations of DOP aerosol. I also included a blank—a filter not exposed to any DOP.
I mounted each test filter inside a sealed enclosure which drew outside laboratory air into a duct that provided the input air flow to the test filter. In the outlet duct downstream of the test filter was a vapor absorption column through which a small pump drew an air sample at a steady controlled flow rate. This column was subsequently desorbed in an analyzer which measured the quantity of DOP degassed and, knowing the volume of air from which this mass had been collected, computed the concentration of DOP in the air sample.
I found that reproducibility from one week to the next was unacceptably poor. Even from my baseline, blank filters varied wildly, the analyses sometimes reporting that the DOP concentration from one week's test blanks exceeded the DOP concentration measured downstream of the most heavily exposed filter the previous week.
In retrospect, this maddening observation proved to be the clue. The DOP concentration in the laboratory air, assumed to be negligible, varied significantly from week to week. Subtracting this baseline concentration of each test period greatly improved reproducibility.
My point is to emphasize that DOP is ubiquitous in the ambient air and has been detected even out at sea. It too can be a molecular contaminant in a cleanroom even without being introduced by HEPA filters. And indeed, DOP challenging of HEPA filters was abandoned long ago as a test method for filters being delivered to semiconductor manufacturers.
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
- Davis, C. M., G. Bergeron, R. LaCourse and G. Trombley, “HEPA Filters as a Contamination Source”, J. Environ. Sci., Mar/Apr 1981, pp. 27-28, 33-35.
- Bhola, V. K., “Future Issues in Design and Construction of HEPA Filters”, 1993 Microcontamination Conference Proceedings, pp 162-169 (Canon Communications, Inc, Santa Monica, CA).