tical Instrument Development Inc., Avondale, PA) was used for generating test atmospheres of chlorine. The system is further described in reference 9.6. Two chlorine permeation vials were used as the chlorine source. The permeation vials were maintained at a constant temperature of 30 0.1 C. The chlorine concentration was determined from the weight loss of the permeation vial over a measured time period and was calculated according to the equation listed in Appendix A. After the chlorine was produced it was then diluted with tempered air so that a controlled concentration at 80% RH and 25 C was achieved. Flow rates for the generation system were measured with a soap bubble flow meter to accurately determine dilution and the final concentration of the gas. Samples were taken from a glass sampling manifold which was attached to the generation system. The majority of samples were collected in sets of six at one of three concentrations, about 0.5, 1, or 2 ppm chlorine. 2. Analysis Procedure: Samples containing 0.1% sulfamic acid were spiked with standardized chlorine solutions. This test determined the precision and accuracy of the analysis portion of the method. Results: The results are shown in Table 1. Average analytical recovery was 96.7% and the coefficient of variation (CV1) was 0.03 for the three sets of spiked samples. 3. Sampling and Analysis Procedure: Samples were collected in 0.1% sulfamic acid solutions at three different chlorine concentrations using the generation system described in Section 1. Midget fritted glass bubblers (MFGB) were used to disperse the air samples in the sulfamic acid solutions. This test determined the precision and accuracy of the sampling and analysis portion of the method. Results: Sampling and analysis data are presented in Table 2. The NIOSH Statistical Protocol (9.7.) developed for evaluation of methods under the Standards Completion Program (SCP) was used as a guide for the determination ...
