The effect of vapor polarity and boiling point on breakthrough for binary mixtures on respirator carbon

This research evaluated the effect of the polarity of a second vapor on the adsorption of a polar and a nonpolar vapor using the Wheeler model. To examine the effect of polarity, it was also necessary to observe the effect of component boiling point. The 1% breakthrough time (1% t(b)), kinetic adsorption capacity (W(e)), and rate constant (k(v)) of the Wheeler model were determined for vapor challenges on carbon beds for both p-xylene and pyrrole (referred to as test vapors) individually, and in equimolar binary mixtures with the polar and nonpolar vapors toluene, p-fluorotoluene, o- dichlorobenzene, and p-dichlorobenzene (referred to as probe vapors). Probe vapor polarity (0 to 2.5 Debye) did not systematically alter the 1% t(b), W(e) or k(v) of the test vapors. The 1% t(b) and W(e) for test vapors in binary mixtures can be estimated reasonably well, using the Wheeler model, from single-vapor data (1% t(b) ± 30%, W(e) ± 20%). The test vapor 1% t(b) depended mainly on total vapor concentration in both single and binary systems. W(e) was proportional to test vapor fractional molar concentration (mole fraction) in mixtures. The k(v) for p-xylene was significantly different (p≤0.001) when compared according to probe boiling point; however, these differences were apparently of limited importance in estimating 1% t(b) for the range of boiling points tested (111 to 180°C). Although the polarity and boiling point of chemicals in the range tested are not practically important in predicting 1% t(b) with the Wheeler model, an effect due to probe boiling point is suggested, and tests with chemicals of more widely ranging boiling point are warranted. Since the 1% t(b), and thus, respirator service life, depends mainly on total vapor concentration, these data underscore the importance of taking into account the presence of other vapors when estimating respirator service life for a vapor in a mixture.