This study investigates the diol epoxide pathway of phenanthrene (PHE) together with phenolic metabolites of PHE and pyrene (PYR) in workers with and without exposure to bitumen fumes. The metabolite concentrations were determined in urine samples collected from 91 mastic asphalt workers and 42 construction workers as reference group before and after shift. During shift, vapours and aerosols of bitumen were measured according to a German protocol in the workers' breathing zone. The median concentration of vapours and aerosols of bitumen in mastic asphalt workers was 6.3mg/m3. Metabolite concentrations were highest in post-shift urines of smokers with bitumen exposure and showed an increase during shift. The Spearman correlations between the creatinine-adjusted concentrations of metabolites and vapours and aerosols of bitumen in non-smokers were weak (e.g. sum of Di-OH-PYR: 0.28) or negligible (e.g. 1,2-PHE-diol: 0.08; PHE-tetrol: 0.12). Metabolites from the diol epoxide pathway of PHE were excreted in higher concentrations than phenolic metabolites (post-shift, non-smoking asphalt workers: 1,2-PHE-diol 2.59µg/gcrea vs. sum of all OH-PHE 1.87µg/gcrea). 1,2-PHE-diol was weakly correlated with PHE-tetrol (Spearman coefficient 0.30), an endpoint of the diol epoxide pathway. By contrast, we found a close correlation between the sum of 1,6-DiOH-PYR and 1,8-DiOH-PYR with 1-OH-PYR (Spearman coefficient 0.76). Most urinary PAH metabolites were higher after shift in bitumen-exposed workers, although the association with bitumen was weak or negligible likely due to the small PAH content. The additional metabolites of PHE and PYR complete the picture of the complex metabolic pathways. Nevertheless, none of the PAH metabolites can be considered to be a specific biomarker for bitumen exposure.