Abstract
A performance test is proposed for evaluating passive (pumpless) gas and vapor samplers for use in a time-variable environment. Such a test is necessary because of potentially excessive analyte loss by some passive samplers. Suggested is a measurement of the maximum bias in estimates of the time-weighted average concentration of individual pulses of selected duration and time. This bias is then combined with the bias and random sampling/analytical errors from measurements at constant concentration. Similar to the current NIOSH Standards Completion Program criteria for sampling methods using pumps, the proposed performance criterion is simply a requirement of less than 25 percent overall single-sample inaccuracy at the 95 percent confidence level under both constant and varying conditions. Alternatively, the test given here may be used simply to define a specific sampler performance component.