The present work deals with the treatment of a former metallurgical wasteland contaminated with PAHs and metals by chem. oxidn. in flow system. It is a part of the OXYSOL project (www.oxysolanr. org) aiming at the conception of a global treatment pathway including in situ chem. oxidn., bioremediation and soil construction. Chem. oxidn. is investigated from the bench scale to the lysimeter scale at the exptl. site of the GISFI (French Scientific Group on Industrial Wastelands, www.gisfi.fr). At the bench scale, expts. were run to select the most effective oxidant for PAH degrdn.: hydrogen peroxide via Fenton reaction and sodium persulfate activated by hydrogen peroxide. However degrdn. rates with reasonable oxidant amt. kept below 25%. Therefore, as a first step, we investigated to what extent PAHs availability affected chem. oxidn. effectiveness with the two selected oxidants. Thus, we conducted flow expts. in a stainless column (i.d.: 6 cm, height: 16 cm), in unsatd. conditions, to treat different types of aged and artificially PAH-contaminated soil. Degrdn. rates were calcd. from PAH anal. before and after soil treatment. PAHs availability in different soils has been evaluated from extn. tests with concd. cyclodextrin soln. This work showed that PAHs availability is a major limiting factor for soil treatment by chem. oxidn. pH condition can also become a drawback: under acidic conditions, persulfate is inefficient. Thus, degrdn. rates with aged contaminated soil were low (below 10%) when it reached nearly 60% with artificially contaminated soil and Fenton's reagent. These results highlight main limiting factors affecting chem. oxidn. efficiency for PAHs-contaminated soil treatment and the need to combine different techniques. [on SciFinder(R)]