Forty-five workers at 11 paving sites across the United States were evaluated for exposure to paving asphalt (bitumen) fumes. Traditional measures of exposure such as total particulate matter (TPM) and benzene soluble matter (BSM) were monitored. In addition, total organic matter (TOM), which includes both the BSM residue and the more volatile components that pass through the filter and are collected on sorption material, was quantified and further characterized using a gas chromatography technique and a recently developed fluorescence test. The latter method, which indirectly estimates the content of four- to six-ring polycyclic aromatic compounds, is used as a predictor of carcinogenicity. The correlation between fluorescence emission intensity and carcinogenicity for 36 laboratory generated fume fractions, as measured in a mouse skin-painting bioassay, was then used to estimate the carcinogenic potential of worker monitoring samples. Emission levels, and therefore predicted carcinogenicity, for these samples were at least 17-fold below the value corresponding to a minimal carcinogenic effect. This result was consistent with more extensive chemical analysis (using gas chromatography/mass spectrometry) of two of the samples, which showed the predominant constituents to be alkanes, monocycloparaffins, alkyl-benzenes, alkyl-naphthalenes, and alkyl-benzothiophenes. The geometric mean exposures for all worker studies were 0.21 mg/m3 (TPM), 0.06 mg/m3 (BSM), and 1.23 mg/m3 (TOM).

This study had two fundamental goals. The first was to collect sufficient quantities of bitumen fumes used in both the roofing and paving industry in the United States to do detailed chemical and biological testing in preparation for a two-year animal skin painting study on select samples. These bitumen fumes need to be similar to the fumes to which workers in these industries are actually exposed. In this first phase, bitumen from four geographical regions of the United States were selected, which produced bitumen from four different major crude sources in roofing and four sources in paving. Each source was supplied to projects where workers were monitored for exposure. The industrial hygiene samples from workers were then tested for physical and chemical properties. These results were used to guide collection of fumes from each source using the vapor headspace of tanks containing each bitumen. Bitumen fumes were collected using the same protocol that the Fraunhofer Institute of Technology and Medicine employed in a recent animal inhalation study. The second goal of this study was to better understand previous studies conducted by NIOSH which generated fumes for mouse skin painting studies using a roofing bitumen in a laboratory fume generator. In the current study, bitumen fumes from the four roofing sources were generated using the NIOSH protocol and compared to fumes collected in the roofing work sites and taken from the tank headspace. One of the sources was selected to generate enough fumes for a chronic two-year animal skin painting study. Results from an array of physical, chemical and biological tests were used to characterize the subsequent fumes. Acceptance criteria were designed to match parameters that could be analyzed both on the workplace samples and the bitumen fumes collected from the tank and involved three basic categories: simulated distillation, fluorescence and gas chromatography/mass spectroscopy for polynuclear aromatic hydrocarbon analysis and extracted ion fingerprinting. Methods employed for the collection of these fumes were selected to optimize the capture of the most toxicologically relevant components.Distinctions between paving and roofing fume samples are outlined. Bitumen fumes from the paving sources produced from four different crude sources were quite similar to each other in a number of the tests performed. Bitumen fumes from the roofing sources also were very similar to each other. On the other hand, fumes generated using the NIOSH protocol show results that significantly differ from the corresponding worker exposures.

A novel laboratory asphalt fume generator was developed and validated against fumes collected on personnel monitors from field paving sites and above paving asphalt storage tanks. Once the apparatus was validated, fumes were generated from eight core Strategic Highway Research Program (SHRP) single crude asphalts, as well as from synthetically produced positive and negative control asphalts. The fumes were characterized using a number of analytical techniques, including gas chromatography/mass spectrometry, gas chromatography with flame ionization detection, UV/fluorescence, and a short-term bioassay called the modified Ames test. For comparison purposes, the same methods were applied to fume condensate preparations from the National Institute of Occupational Safety and Health (NIOSH) animal skin-painting studies. On the basis of the chemical and biological tests, these SHRP asphalt fumes were found to have a lower carcinogenic potential than a NIOSH fume fraction found to be noncarcinogenic in the NIOSH studies.

Exposure to asphalt fumes has a threshold limit value (TLV of 0.5 mg m(-3) (benzene extractable inhalable particulate) as recommended by the American Conference of Governmental Industrial Hygienists (ACGIH). This reflects a recent change (2000) whereby two variables are different from the previous recommendation. First is a 10-fold reduction in quantity from 5 mg m(-3) to 0.5 mg m(-3). Secondly, the new TLV specifies the "inhalable" fraction as compared to what is presumed to be total particulate. To assess the impact of these changes, this study compares the differences between measurements of paving asphalt fume exposure in the field using an "inhalable" instrument versus the historically used 'total' sampler. Particle size is also examined to assist in the understanding of the aerodynamic collection differences as related to asphalt fumes and confounders. Results show that when exposures are limited to asphalt fumes, a 1:1 relationship exists between samplers, showing no statistically significant differences in benzene soluble matter (BSM). This means that for the asphalt fume ACGIH TLV, the 'total' 37-mm sampler is an equivalent method to the "inhalable" method, referred to as IOM (Institute of Occupational Medicine), and should be acceptable for use against the TLV. However, the study found that when confounders (dust or old asphalt millings) are present in the workplace, there can be significant differences between the two samplers' reported exposure. The ratio of IOM/Total was 1.37 for milling asphalt sites, 1.41 for asphalt paving over granular base, and 1.02 for asphalt over asphalt pavements.

No abstract