![]() |
The Relationship between Pore Size and Filter Efficiency |
| Users of membrane filters are often confused by the "pore size" printed in the specifications for a particular catalog number. For filters used for air sampling, pore size has nothing to do with the collection efficiency of a filter. The face velocity governs the efficiency, through the filter and the aerodynamic diameter characteristics of the particles. All membrane filters, with one exception, have collection efficiencies in excess of 99% for particles down to 0.05 µM, diameter. Pore size is an indicator of the pressure drop across a filter, not its collection efficiency. The one exception, are filters which are manufactured by bombarding a solid surface, such as Mylar, with neutrons. An imperfection is created on the surface, which is subsequently acid etched to form a microscopic hole through the membrane. Filters of this type will have significantly reduced efficiencies when a combination of large pore size (3-5 µM) and small diameter (less than 1 µM) is encountered. Etched membranes are not specified for any of the NIOSH or EPA sampling methods.
Historically, filters selected for use in a particular application have had a smaller pore size than was required. This was only a minor problem, prior to the advent of battery operated sampling equipment. Too small a pore size would only result in a too rapid, plugging or blinding, of the filter so that the requisite quantity of air could not be drawn through it. With battery operated equipment, too small a pore size would lead to high power consumption, premature shutdown and with unregulated equipment, and incorrect flow rate.
The physics of the filtration of particulate material from a gas stream is complex. Several mechanisms are acting in concert or opposition. Those desiring an understanding of filtration theory should consult the following source, which also contains a data table. K. Willeke, Baron, P.A., AEROSOL MEASUREMENT, PRINCIPLES, TECHNIQUES AND APPLICATIONS, Van Nostrand Reinhold, New York, 1993. See Chapter 10 starting on page 179. Table 10-5 on pages 196-7 is particularly illuminating. R. Gussman, July 1998 Copyright © 2006 by BGI / Modified:
Wednesday, January 17, 2007
|