Inhalation study on exposure to bitumen fumes. Part 2: Analytical results at two exposure levels

During the hot application of bitumen-containing materials, e.g. in road paving or roofing, fumes are emitted that contain traces of polycyclic aromatic compounds (PACs). Although worker's exposure to these fumes is low, it might lead to health problems. For studying DNA adduct formation as a consequence of inhalation of bitumen fumes we developed and validated an inhalation system (a dynamic fume generator plus a nose only inhalation chamber). This paper presents and discusses the analytical results from the different laboratories involved in this study on the fumes sampled in the inhalation chamber during three series of experiments where the animals were exposed to fumes at the 5 mg/m3 and 50 mg/m3 level, coming from bitumen heated at 200 degrees C and, as a positive control, fumes from coal tar, heated to 110 degrees C at the 5 mg/m3 level. The following parameters were controlled: temperatures at different key places in the generator; humidity of the chamber; the bitumen or coal tar flow rate; and Total Particulate Matter (TPM). Analyses were performed for Benzene Soluble Matter (BSM), the EPA polycyclic aromatic hydrocarbon (PAH) mixture and for a number of heteroatom-containing PACs. The data show that the coal tar fumes produced at 110 degrees C were very volatile and that most of the differences in particulate matter found between the laboratories can be attributed to evaporative losses. The bitumen fumes boil 25-50 degrees C higher and contain higher boiling compounds. A comparison is made between the PAC exposure profiles for bitumen experiments aimed at 5 and 50 mg/m3. Although the same molecules are found in both fumes their proportion is dramatically different. This effect is largest with the 2- and 3-ring PACs, the ratio of the concentrations found in the 50 mg/m3 TPM concentration to that in the 5 mg/m3 experiment gradually declines from 5500 for acenaphthene to 500 for pyrene, for the 5-ring PACs this ratio is 20-30. As function of their vapour pressure, the ratios of the concentrations of the hetero PACs follow the same trend as that of the 16 EPA PAHs and are of the same order of magnitude. In conclusion, for the compounds investigated, the equipment delivers a fume atmosphere in a reproducible manner. The 50 mg/m3 bitumen fumes are not representatives of field fumes. The reason for these quantitative differences is unclear and further work would be needed to clarify this. Nevertheless it was felt that these fumes at 50 mg/m3 might be a useful tool for qualitative detection of DNA adducts in an animal exposure study.