Development and Evaluation of a Denuder-Filter System Designed for Sampling Diisocyanate Aerosols

Sammanfattning: There are many occupational environments in which isocyanates or PUR-products are used, produced or processed. During these operations there is always a risk of isocyanate exposure, which may lead to severe health problems. Isocyanates principally affect the respiratory system and they are the main cause of occupational asthma. The prevalence of asthma among exposed workers is generally 5-10%, according to most studies, but figures as high as 30% have been reported.The most commonly used isocyanates may occur in both vapour and particulate forms. Their physical state may affect both their deposition in the airways and their impact on health. Nevertheless, the occupational exposure limit for isocyanates applies to the total amount of isocyanates present, regardless of their physical state. Setting different exposure limits for vapour and particles would require the air sampling step to be capable of separating and independently measuring the phases. This is not straightforward, since every sampling method may be affected by a number of possible sampling artefacts. To minimize such biases, a profound knowledge of the sampling system and its behaviour is required.In this thesis a chemosorptive cylindrical denuder and filter, connected to a sampling pump, has been developed for personal exposure measurements of isocyanates. The model compounds for the investigation and evaluation were some of the most commonly used diisocyanates in industry: toluene diisocyanate (TDI), hexamethylene diisocyanate (HDI), isophorone diisocyanate (IPDI) and methylene diphenyl diisocyanate (MDI). Using a combination of dimethylpolysiloxane (SE-30) and dibutylamine (DBA) as a chemosorptive coating on both the inner wall of the denuder tube and the filter, the derivatisation was shown to be fast enough for efficient collection of the isocyanates. The repeatability of the air measurements was high, with variabilities generally between 5-10% in the measured concentration range. 15-min samplings were performed of both vapour and aerosols of nm-size, from below the OEL (5ppbv) up to 235 ppbv, without exceeding the coating capacity or causing breakthrough due to chromatographic movement. At the applied airflow rates, between 50 and 500 ml/min, the breakthrough, i.e. the percentage of the TDI vapour that passes through the denuder tube ending up on the filter, was shown to be consistent with amounts predicted by the Gormley-Kennedy equation. For isocyanate aerosols, extensive gas-stripping was shown to occur inside the denuder tube at the lower end of the investigated flow range. This phenomenon was minimized when a flow rate between 300 and 500 ml/min was applied (Papers III-IV).The great advantage of a denuder sampler compared to wet methods, such as the commonly used impinger method, is its suitability for personal exposure measurements. Due to its small size and absence of solvent it is more convenient to carry during a work shift. Furthermore, the results in Paper II indicate that 8-hour denuder measurements can be performed without detectable breakthrough of the vapour phase. The derivatives formed during the sampling of isocyanates were shown to be stable in the samplers for several days after the measurements were finished, provided both the denuder tubes and filters were stored in the freezer and in solvent containing excess reagent. Furthermore, filters and tubes could be prepared at least 4 d prior to sampling, without loss of performance. This is highly relevant to the applicability of the method for field measurements.Both LC-ESI-MS (Paper II-IV) and LC-APCI-MS (Paper I) were used to determine the analytes in the air samples. The former method exhibited the lowest detection limit for the isocyanate-DBA derivatives. Using selected reaction monitoring (SRM) the obtained method detection limit for the investigated compounds was below 10 ng/m3.