Because of their toxic and carcinogenic properties, 16 PAHs are regulated and measured for polluted site diagnosis and monitoring. Associated to PAHs, polar PACs (O-PACs and N-PACs) occur and could induce a risk for water resources. Moreover, recent studies revealed that polar PACs (i) are leached in higher proportions than PAHs and (ii) can be as toxic as corresponding PAHs. Consequently, the mechanisms involved in polar PAC mobilization and transfer to water need to be elucidated. In this project, the mobilization and transfer of polar PACs and PAHs were assessed via leaching experiments of aged contaminated soils. A multi-scale approach was used: batch (50 g), laboratory column (2 kg) and pilot scale column (2 m 3 lysimeter). In a first step, PAC mobilization was investigated in laboratory batch experiments under various conditions (temperature, ionic strength, PAC availability). For laboratory column, different nature (ionic strength, organic matter concentration) and flow of the percolating solutions were tested on two different columns (raw and contaminant availability increased soils). On the field scale, leaching tests were performed with artificial rain applied on a lysimeter column filled with a PAC contaminated soil. Batch experiments show a release of low molecular weight PAHs and polar PACs mainly by dissolution (Raoult's law), while higher molecular weight PAHs were mainly released in association with colloids. Aging mainly controls the former mechanism and ionic strength mainly controls the latter one. Column experiments confirm a preferential release of polar PACs compared with PAHs and satisfactory predictions are obtained for polar PAC release when the pollution is available, it is not the case for the raw soil. Leaching tests performed on the lysimeter column confirm a release of PACs according to a coaltar dissolution mechanism, although PAC concentrations in water are much lower compared to those predicted by the Raoult's law.