Laboratory Evaluation of Passive Colorimetric Dosimeter Tubes for Carbon Monoxide

Abstract

The recently developed passive colorimetric dosimeter tubes potentially offer great advantages of compactness, convenience, and simplicity in environmental evaluation of a wide variety of toxic gases and vapors. These devices do not require pumps but operate on diffusion principles to produce stain lengths proportional to time-weighted average concentrations that can be read immediately at the site. The purpose of this study was to evaluate the performance of colorimetric passive dosimeter tubes for carbon monoxide under dry and humid conditions. A flow dilution system using 2.58 percent carbon monoxide in air from a cylinder was used to provide 25, 50, 100, and 400 ppm of carbon monoxide at relative humidities of 20 percent and 82 percent and at a temperature of 26°C. Tubes from three manufacturers (Gastec/Sensidyne, Draeger, and MSA) were exposed for periods up to six hours, and the stain lengths were read (independently) by two observers at quarter- and half-hour intervals, depending upon the test concentrations. The relationships between the doses determined with a continuous monitor and those estimated by reading the stain lengths on the ppm-hr scales printed on the tubes by the manufacturers were determined by statistical regressions. The test results showed that only one of the three commercial carbon monoxide dosimeter tubes had a good correlation between true and estimated concentrations. The slopes and correlation coefficients for these tubes were in the range of 0.99–1.15 and 0.995–0.998, respectively. However, for tubes from the other two manufacturers, the respective slope ranges were 0.18–0.47 and 2.13–2.31, and the respective correlation ranges were 0.482–0.955 and 0.974–0.986. The results also showed that the effects of humidity on the performance of these tubes were statistically significant, considering the level of significance at p < 0.05, and a higher response was noted for dry conditions. The difference between observer readings for stain lengths to determine the estimated concentrations, however, was found not to be statistically significant. The authors suggest that some manufacturers must make further improvements in the design and calibration of each type of tube to achieve more accurate results.