The supercritical CO(2) extraction of four PAHs (acenaphthene, phenanthrene, anthracene and fluoranthene) from an artificially contaminated soil has been investigated. The effect of temperature (40-60 degrees C), pressure (300-500 bar) and extraction time (90-150 min) has been assessed by conducting a Box-Behnken experimental design. The results suggest the existence of perturbation variables other than the aforementioned controlled variables leading to a significant dispersion of extraction recoveries. With the exception of anthracene, an optimum in temperature (50 degrees C) is envisaged when extracting the PAHs. Analogously, with the exception of anthracene (positive effect), pressure does not have a significant influence. The recovery yield increases as extraction time is increased to a value of 120 min. No further improvement is experienced thereafter. If a co-solvent is used (H(2)O(2) aqueous solution) a beneficial effect can be noticed. Hydrogen peroxide concentration did exert no significant influence in the process. Methanol used to collect the extracted PAHs could be regenerated by gaseous ozone and reused in several consecutive runs.