A personal direct-reading instrument employing a surface-acoustic-wave microsensor array for the measurement of organic vapor exposures

A prototype personal monitoring instrument capable of selective, multi-vapor monitoring of organic vapors in workplace environments was developed and tested in the laboratory. The instrument employs an array of three polymer-coated surface-acoustic-wave (SAW) sensors and a thermally desorbable adsorbent preconcentrator. Two identical instruments containing different polymer-coated sensors were calibrated against 16 organic vapors and selected mixtures to establish a response library. Limits of detection were ≤0.1 x TLV for most vapors.;Monte Carlo simulations coupled with extended disjoint principal component regression analyses were used to evaluate the performance of the instrument as a function of a number of critical variables. The ability to recognize a vapor was shown to decrease with decreasing concentration. A limit of recognition (LOR) was defined as the concentration below which a vapor can no longer be reliably recognized from its response pattern. The LOR is suggested as an additional criterion for evaluating the performance of multi-sensor arrays. Vapor recognition was also shown to depend strongly on the polymer coatings on the sensors in the array and the nature and complexity of the vapor mixtures being analyzed. Results indicate that six or fewer sensors provide adequate vapor recognition capabilities and that mixtures of ≥5 vapors cannot be analyzed with polymer-coated sensor arrays. For mixtures of <5 vapors, the minimum number of sensors required was equal to the number of possible vapors in the mixture. Increasing the number of sensors in the array did not improve vapor recognition rates significantly.;Humidity changes did not significantly affect the sensitivities of the sensors due to the use of hydrophobic adsorbent and incorporation of a dry-air purge step prior to thermal desorption of samples. The sensitivities of the sensors were affected considerably by temperature changes. However, sensors exhibiting similar temperature dependencies for all vapors were identified such that vapor recognition rates were unaffected by temperature.;The instrument was adapted to the problem of determining solvent and solvent-mixture permeation through chemical protective clothing materials. Results demonstrate the feasibility of the instrument as an effective semi-automated tool for field permeation screening.