persulfate, a third set of batch experiments was conducted. Different persulfate
concentrations (1, 5, 10, 20, 50, and 100 g L
À1 ) were tested on an n-decane/
menthol mixture (50/50 v). Experiments were conducted in 75 mL vials filled with
20 g of fine sand, 30 mL of demineralized water, and 100 μL of n-decane/menthol
mixture. Vials were then shaken for 8 days. A blank was realized for each experimental set.
For Diesel fuel batches, the water–sand–pollutant mixture was centrifuged
(Sigma, Germany) for 10 min at 4000 rpm. Contaminant was extracted from the
aqueous phase two times by 7 mL of n-hexane. Analysis was done by the injection of
1 μL of extraction solvent into a gas chromatograph (CP 3800, VARIAN, France).
This GC is equipped with a flame ionization detector and capillary column
(Rxi-5silMS, 30 m, 0.25 mm, 0.25 μm, RESTEK, USA). The oven temperature
was maintained at 50
C for 30 s then increased at 25
C min
À1 until a temperature of
290
C held for 30 s. Then, oven temperature increased at 10
C min
À1 to a final
temperature of 320
C held for 16 min. The injector and detector temperature were
kept at 280
C and 320
C, respectively. Concerning BTX and n-decane/menthol
batches, samples were directly analyzed by gas chromatography in Head Space
mode. The sample was shaken and heated to 40
C. A volume of 500 μL of the
gas phase was injected into the gas chromatograph. The oven temperature was
maintained at 50
C for 2 min and then increased at 5
C min
À1 to a temperature
of 100
C. Finally, the oven temperature increased at 20
C min
À1 to a final
temperature of 250
C held for 30 min. The injector and detector temperatures
were both kept at 250
C.
4.2.2 Scale 2: Column Experiments
To assess the efficiency of the different remediation techniques (surfactant flushing,
oxidation, sparging, and thermal treatment) and determine the main parameters
influencing contaminants’ removal, several column experiments were conducted.
The experimental details are given in Jousse et al. (2017). All columns were filled
with a 10
À4 m s
À1 sand, except in some columns used for thermal treatment where
the sand was mixed with clay (Kaolinite, Sibelco, France) leading to hydraulic
conductivity equal to 10
À5 and 10
À6 m s
À1 for mixtures containing 10 and 20% of
clay (in mass), respectively.
The contamination and treatment parameters are described in Table 4.3. Each
treatment was triplicated. For persulfate injection, two frequencies were used:
1 injection per week over 4 weeks and 1 injection per month over 2 months. For
all experiments using Fenton, 2 PV of Fenton solutions were injected each day over
7 days. During the first experiment, H 2 O 2 and iron sulfate solutions were injected at
a flow rate of 1 mL min
À1 in three periods of 1 h: 15 min H 2 O 2 , 15 min iron, and
30 min without injection. In the second set of experiments, the injected volumes
were the same but the time slots were changed and lasted 3 min each and were
separated by 9 min without injection. Concerning ozone, it was injected into
4 Comparing the Efficiency of Oxidation, Sparging, Surfactant Flushing, and. . .
217
concentrations (1, 5, 10, 20, 50, and 100 g L
À1 ) were tested on an n-decane/
menthol mixture (50/50 v). Experiments were conducted in 75 mL vials filled with
20 g of fine sand, 30 mL of demineralized water, and 100 μL of n-decane/menthol
mixture. Vials were then shaken for 8 days. A blank was realized for each experimental set.
For Diesel fuel batches, the water–sand–pollutant mixture was centrifuged
(Sigma, Germany) for 10 min at 4000 rpm. Contaminant was extracted from the
aqueous phase two times by 7 mL of n-hexane. Analysis was done by the injection of
1 μL of extraction solvent into a gas chromatograph (CP 3800, VARIAN, France).
This GC is equipped with a flame ionization detector and capillary column
(Rxi-5silMS, 30 m, 0.25 mm, 0.25 μm, RESTEK, USA). The oven temperature
was maintained at 50
C for 30 s then increased at 25
C min
À1 until a temperature of
290
C held for 30 s. Then, oven temperature increased at 10
C min
À1 to a final
temperature of 320
C held for 16 min. The injector and detector temperature were
kept at 280
C and 320
C, respectively. Concerning BTX and n-decane/menthol
batches, samples were directly analyzed by gas chromatography in Head Space
mode. The sample was shaken and heated to 40
C. A volume of 500 μL of the
gas phase was injected into the gas chromatograph. The oven temperature was
maintained at 50
C for 2 min and then increased at 5
C min
À1 to a temperature
of 100
C. Finally, the oven temperature increased at 20
C min
À1 to a final
temperature of 250
C held for 30 min. The injector and detector temperatures
were both kept at 250
C.
4.2.2 Scale 2: Column Experiments
To assess the efficiency of the different remediation techniques (surfactant flushing,
oxidation, sparging, and thermal treatment) and determine the main parameters
influencing contaminants’ removal, several column experiments were conducted.
The experimental details are given in Jousse et al. (2017). All columns were filled
with a 10
À4 m s
À1 sand, except in some columns used for thermal treatment where
the sand was mixed with clay (Kaolinite, Sibelco, France) leading to hydraulic
conductivity equal to 10
À5 and 10
À6 m s
À1 for mixtures containing 10 and 20% of
clay (in mass), respectively.
The contamination and treatment parameters are described in Table 4.3. Each
treatment was triplicated. For persulfate injection, two frequencies were used:
1 injection per week over 4 weeks and 1 injection per month over 2 months. For
all experiments using Fenton, 2 PV of Fenton solutions were injected each day over
7 days. During the first experiment, H 2 O 2 and iron sulfate solutions were injected at
a flow rate of 1 mL min
À1 in three periods of 1 h: 15 min H 2 O 2 , 15 min iron, and
30 min without injection. In the second set of experiments, the injected volumes
were the same but the time slots were changed and lasted 3 min each and were
separated by 9 min without injection. Concerning ozone, it was injected into
4 Comparing the Efficiency of Oxidation, Sparging, Surfactant Flushing, and. . .
217