membranes [Hinds 1982]. Thus, particles smaller than the pore size may be efficiently collected. Sampling filter media may have pore sizes of 0.01 to 10 µm. The efficiency of removing particles from the air depends on the face velocity (i.e., the cross sectional air velocity of the filter holder). For particles less than 1 µm, the overall efficiency decreases with increasing face velocity [Liu et al. 1983; Lippmann 1995]. For particles greater than approximately 1 µm, the filter collection efficiency is greater than 99%. The overall efficiency of membrane filters is approximately 100% for particles larger than the pore size [Lippmann 1995]. Membrane filters are manufactured in a variety of pore sizes from polymers such as cellulose ester, polyvinyl chloride, and polycarbonate. Polymeric membrane filters lack rigidity and must be used with a support pad. The choice of a filter medium depends on the contaminant of interest and the requirements of the analytical technique. For gravimetric analysis, nonhygroscopic materials such as glass fibers, silver, or polyvinyl chloride membranes are selected. For analysis by microscopy, cellulose ester or polycarbonate membranes are the usual choices. Filters are often held in disposable plastic filter cassettes during bioaerosol sampling. The three-piece cassette may be used either in open- or closed-face modes. Open-face sampling is performed by removing the end plug and the plastic cover from the three-piece cassette and is used when the particulate must be uniformly deposited (i.e., for microscopic analysis). If a three-piece cassette is used in the open-face arrangement, the plastic cover is retained to protect the filter after sampling is concluded. All plastic cassettes are securely assembled and sealed with a cellulose shrink band or tape around the seams of the cassette to prevent leakage past the filter. Membrane filters for use in sampling are usually supplied as disks of 37- or 47-mm diameter. Because the pressure drop across a filter increases with the air velocity through the filter, the use of a larger (47-mm) filter results in a lower pressure drop for a given volumetric flow rate. The use of the smaller (37-mm) filter concentrates the deposit of the contaminant onto a smaller total area, thus increasing the density of particles per unit area of filter. This may be helpful for direct microscopic examination of low concentrations of organisms. In areas of high concentration, the microorganisms may have to be eluted, diluted, and then refiltered for microscopic analysis. Filtration techniques are used for the collection of certain fungi and endospore-forming bacteria that are desiccation-resistant. The sampled organisms are washed from the surface of smooth-surface polycarbonate filters. The microorganisms in the wash solution are either cultured or refiltered to distribute the microorganisms uniformly on the membrane filter. In the latter case, the microorganisms are stained and examined microscopically [Wolf et al. 1959; Fields et al. 1974; Lundholm 1982; Palmgren et al. 1986a]. To culture the organisms, the membrane filter from each sampling cassette is washed with a 0.02% Tween™ 20 (J.T. Baker Chemical Co., Phillipsburg, NJ) in aqueous solution (three 2-mL washes), with agitation. Some of the recovered wash volume is serially diluted from full strength (1:10, 1:100, 1:1000) and 0.1 mL of each dilution is inoculated onto duplicate 100-mm x 15-mm petri dishes containing the appropriate medium. Residual culturable microorganisms on the membrane filter from each sampling cassette are counted by placing the filter on a medium 1/15/98
87
NIOSH Manual of Analytical Methods
