of variation (CV) of 1.0 and 0.33 when sampling at 2 L/min and 10 L/min, respectively, while the button sampler gave a CV of 0.10 under the same conditions [77]. The button sampler has few internal surfaces for wall deposition, suggesting that elimination of this type of loss would improve the precision of the 37-mm cassette. Demange et al. found improved precision for the 37-mm cassette data when the wall deposits were added to the analyte [23]. In spite of some of its drawbacks, the 37-mm cassette is likely to be used for some time. It appears that from the standpoint of improving agreement with the inhalable convention and improving precision, the inhalable sampler wall losses should be minimized through sampler design (use of a cartridge such as the AccuCap or in the MSA coal mine cassette) or the wall deposits should be added to the analyte. 10. CONCLUSIONS Clearly, when proper features are incorporated in the sampler design, significant improvements in bias and precision can be achieved for some currently used aerosol samplers. Several recommendations regarding the application of these samplers are listed: a. Classifiers used to select respirable, thoracic, or other fractions should be evaluated based on bias maps obtained from experimental data and combined with size distributions from workplace measurements to evaluate their applicability. b. Further research and development is needed to improve sampler design to better match ACGIH/ISO conventions and reduce inter-laboratory variability in conducting aerosol sampling. It is important to report the sampler and flow rate used to allow evaluation of potential biases due to sampling. c. The filter cassette and fittings should be air-tight and have no by-pass leakage. A pneumatic or mechanical press should be used to assemble the cassette and a leak test should be used to establish appropriate pressure and proper assembly procedures. See chapter entitled: Aerosol Sampling Cassette Assembly: Leak Testing d. The sampler should be made of conductive or static-dissipative materials. e. Internal deposits in sampling cassettes should be included in the analysis. One approach to improving the closed-face cassette is to use an internal cartridge that collects all the sampled dust entering the cassette. The cartridge must be compatible with the analytical method. Another approach is to wipe or wash the internal surfaces of the cassette and add this material to the filter analyte. 11.
DISCLAIMER Mention of company names or products does not constitute endorsement by the Centers for Disease Control and Prevention.
3/15/03
195
NIOSH Manual of Analytical Methods
