during surfactant addition is much smaller than for persulfate, corresponding to a
slower removal and thus, the surfactant washing is less efficient than oxidation with
persulfate. The variable treatment efficiency in the different lenses can possibly
explain this result. Indeed, persulfate treatment showed a removal yield close to 55%
in all lenses while for surfactant a higher removal was observed in the mid-lens
(78%) but not in the low permeability lenses (33% on average). During the conditioning period, it was observed that during the removal of the first 32% of mass,
regardless of the tank considered, the concentration was close to solubility, while
between 32 and 40% it could decrease rapidly. Applying this empirical approach to
persulfate treatment where all lenses showed removals higher than 50%, they may all
release low concentrations. On the other hand, for surfactant, the released concentration may be negligible for the mid-lens, but close to the equilibrium value for the
top and bottom lenses, leading to an average concentration much higher than for the
persulfate treatment.
Several authors (Adamson et al. 2011; Liu and Ball 2002) showed that low
hydraulic conductivity layers can provide a long-term release of contaminants.
Thus, the most efficient techniques shall be able to treat these layers. This is why
dense persulfate solutions were used to enter the low-permeability lenses and were
successful. On the other hand, surfactant displacement is still largely limited by low
hydraulic conductivity zones. The average efficiency of the surfactant treatment, as
compared to other studies (Ramsburg et al. 2005), could be linked to the poor
increase of toluene solubility (2.5) compared to chlorinated solvents (Pennell et al.
1994).
Fig. 4.10 Toluene concentrations at the tanks’ outlet as a function of the mass removed from the
tanks, during the conditioning (filled symbols) and treatment (open symbols) periods
4 Comparing the Efficiency of Oxidation, Sparging, Surfactant Flushing, and. . .
229
slower removal and thus, the surfactant washing is less efficient than oxidation with
persulfate. The variable treatment efficiency in the different lenses can possibly
explain this result. Indeed, persulfate treatment showed a removal yield close to 55%
in all lenses while for surfactant a higher removal was observed in the mid-lens
(78%) but not in the low permeability lenses (33% on average). During the conditioning period, it was observed that during the removal of the first 32% of mass,
regardless of the tank considered, the concentration was close to solubility, while
between 32 and 40% it could decrease rapidly. Applying this empirical approach to
persulfate treatment where all lenses showed removals higher than 50%, they may all
release low concentrations. On the other hand, for surfactant, the released concentration may be negligible for the mid-lens, but close to the equilibrium value for the
top and bottom lenses, leading to an average concentration much higher than for the
persulfate treatment.
Several authors (Adamson et al. 2011; Liu and Ball 2002) showed that low
hydraulic conductivity layers can provide a long-term release of contaminants.
Thus, the most efficient techniques shall be able to treat these layers. This is why
dense persulfate solutions were used to enter the low-permeability lenses and were
successful. On the other hand, surfactant displacement is still largely limited by low
hydraulic conductivity zones. The average efficiency of the surfactant treatment, as
compared to other studies (Ramsburg et al. 2005), could be linked to the poor
increase of toluene solubility (2.5) compared to chlorinated solvents (Pennell et al.
1994).
Fig. 4.10 Toluene concentrations at the tanks’ outlet as a function of the mass removed from the
tanks, during the conditioning (filled symbols) and treatment (open symbols) periods
4 Comparing the Efficiency of Oxidation, Sparging, Surfactant Flushing, and. . .
229