Abstract. Objectives: The aims of this study were to determine external and internal polycyclic aromatic hydrocarbon (PAH) exposure in a coke plant and to contribute to the evaluation of biological limit values. Methods: Personal air sampling was carried out on a study population of 24 coke plant workers. In detail, 16 PAHs were determined, among them phenanthrene, pyrene and benzo[a]pyrene. For the determination of internal exposure, post-shift urine samples from the workers were examined for hydroxylated metabolites of pyrene (1-hydroxypyrene, 1-OHP) and phenanthrene (1-, 2+9-, 3-, 4-hydroxyphenanthrene, 1-, 2+9-, 3-, 4-OHPhe). In both cases, approved HPLC methods were used. Results: The workers were classified into three groups. The group most exposed was that on the topside of the coke oven, with a mean value of 491.2g/m3 for the sum of all sixteen PAHs. The mean value at the bench side was 26.61g/m3, while it was 76.18g/m3 in the group of workers with no dedicated location (=complete area). Both the excretion of 1-OHP and the sum of hydroxyphenanthrenes had the highest mean levels in the topside group of workers (mean: 19.70g/g creatinine (crea.) and 39.18g/g crea.), followed by the bench-side workers (mean: 7.01g/g crea. and 12.95g/g crea.) and the whole-area workers (mean: 3.57g/g crea. and 8.70g/g crea.). The concentrations of all urinary metabolites correlated significantly with benzo[a]pyrene (BaP) in the air. Conclusions: According to these correlations, urinary concentrations of 8 and 11g 1-OHP/g crea., respectively 16 and 23g/g crea. for the sum of hydroxyphenanthrenes, correspond to BaP concentrations in air of 2 and 5g/m3, the German technical guiding concentration (TRK value). Our results are in line with others reported in the literature. That means that it should be possible today to define a standard of occupational hygiene and medicine in coke plants using the elimination of hydroxyphenanthrenes and hydroxypyrene in urine as markers for internal PAH exposure.