Application of magnetite-activated persulfate oxidation for the degradation of PAHs in contaminated soils.

Abstract

In this study, feasibility of magnetite-activated persulfate oxidation (AP) was evaluated for the degradation of polycyclic aromatic hydrocarbons (PAHs) in batch slurry system. Persulfate oxidation activated with soluble Fe(II) (FP) or without activation (SP) was also tested. Kinetic oxidation of PAHs was tracked in spiked sand and in aged PAH contaminated soils at circumneutral pH. Quartz sand was spiked with: (i) single model pollutant (fluorenone) and (ii) organic extract isolated from two PAH contaminated soils (H and NM sampled from ancient coking plants) and was subjected to oxidation. Oxidation was also performed on real H and NM soils with and without an extraction pretreatment. Results indicate that oxidation of fluorenone resulted in its complete degradation by AP while abatement was very low (<20%) by SP or FP. In soil extracts spiked on sand, significant degradation of 16 PAHs was observed by AP (70-80%) in 1 week as compared to only 15% by SP or FP systems. But no PAH abatement was observed in real soils whatever the treatment used (AP, FP or SP). Then soils were subjected to an extraction pretreatment but without isolation of organic extract from soil. Oxidation of this pretreated soil showed significant abatement of PAHs by AP. On the other hand, very low degradation was achieved by FP or SP. Selective degradation of PAHs was observed by AP with lower degradation efficiency towards high molecular weight PAHs. Analyses revealed that no by-products were formed during oxidation. The results of this study demonstrate that magnetite can activate persulfate at circumneutral pH for an effective degradation of PAHs in soils. However, availability of PAHs and soil matrix were found to be the most critical factors for degradation efficiency.

Full text

Page 1

Application of magnetite-activated persulfate oxidation for the degradation
of PAHs in contaminated soils
M. Usmana,b, P. Faureb, C. Rubya, K. Hannac,d,⇑
aLaboratoire de Chimie Physique et Microbiologie pour l’Environnement, LCPME, UMR 7564, CNRS-Université de Lorraine, 405 rue de Vandoeuvre, 54600 Villers Les Nancy, France
bGéologie et Gestion des Ressources minérales et énergétiques, G2R, UMR 7566, CNRS-Université de Lorraine, 54506 Vandoeuvre Les Nancy, France
cEcole Nationale Supérieure de Chimie de Rennes, UMR CNRS 6226 ‘‘Sciences Chimiques de Rennes’’, Avenue du Général Leclerc, 35708 Rennes Cedex 7, France
dUniversité Européenne de Bretagne, 35000 Rennes, France
a r t i c l ei n f o
Article history:
Received 11 December 2011
Received in revised form 29 December 2011
Accepted 2 January 2012
Available online 23 January 2012
Keywords:
Soil
Oxidation
Polycyclic aromatic hydrocarbons
Persulfate
Magnetite
a b s t r a c t
In this study, feasibility of magnetite-activated persulfate oxidation (AP) was evaluated for the degrada-
tion of polycyclic aromatic hydrocarbons (PAHs) in batch slurry system. Persulfate oxidation activated
with soluble FeII(FP) or without activation (SP) was also tested. Kinetic oxidation of PAHs was tracked
in spiked sand and in aged PAH contaminated soils at circumneutral pH. Quartz sand was spiked with:
(i) single model pollutant (fluorenone) and (ii) organic extract isolated from two PAH contaminated soils
(H and NM sampled from ancient coking plants) and was subjected to oxidation. Oxidation was also per-
formed on real H and NM soils with and without an extraction pretreatment. Results indicate that oxida-
tion of fluorenone resulted in its complete degradation by AP while abatement was very low (<20%) by SP
or FP. In soil extracts spiked on sand, significant degradation of 16 PAHs was observed by AP (70–80%) in
1 week as compared to only 15% by SP or FP systems. But no PAH abatement was observed in real soils
whatever the treatment used (AP, FP or SP). Then soils were subjected to an extraction pretreatment but
without isolation of organic extract from soil. Oxidation of this pretreated soil showed significant abate-
ment of PAHs by AP. On the other hand, very low degradation was achieved by FP or SP. Selective deg-
radation of PAHs was observed by AP with lower degradation efficiency towards high molecular
weight PAHs. Analyses revealed that no by-products were formed during oxidation. The results of this
study demonstrate that magnetite can activate persulfate at circumneutral pH for an effective degrada-
tion of PAHs in soils. However, availability of PAHs and soil matrix were found to be the most critical fac-
tors for degradation efficiency.
? 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Chemical oxidation treatments are showing great potential for
the remediation of contaminated soils in which strong oxidants
are injected into the subsurface. Recently, persulfate oxidation
has emerged as an optionfor chemicaloxidation of organic contam-
inants in soils and sediments (Liang et al., 2004, 2007; Ferrarese
et al., 2008; Yen et al., 2011). Persulfate salts dissociate in aqueous
solution to the persulfate anion S2O2?
(E? = 2.01 V). But its reaction kinetics is slow in destroying most
of the recalcitrant organic contaminants (Osgerby, 2006). However,
persulfate activation can be initiated by thermal or chemical means
to form sulfate radical ðSO??
stronger oxidant (E? = 2.6 V) than persulfate anion (Latimer, 1952).
8
which is a strong oxidant
4Þ (House, 1962). The sulfate radical is a
Iron is a commonly used transition metal for chemical activation of
persulfate anion (Liang et al., 2004).
Due to their high toxicity, environmental persistence and car-
cinogenic effects, polycyclic aromatic hydrocarbons (PAHs) are of
great environmental and health concern and thus are considered
as priority pollutants by US EPA and European community (Wild
and Jones, 1995). In situ chemical oxidation (ISCO) is an increas-
ingly popular method for the remediation of contaminated soils
and groundwater in which various oxidants are injected to degrade
contaminants (ITRC, 2005). There are few studies on the treatment
of PAHs using persulfate oxidation activated by soluble FeII(Nadim
et al., 2006; Ferrarese et al., 2008; Gryzenia et al., 2009). As rapid
oxidation and precipitation of ferrous ion could make FeIIinactive,
chelating agents were previously used to overcome this limitation
and to maintain FeIIin solution (Liang et al., 2004). Another possi-
bility is the use of minerals like ferrihydrite, goethite, manganese
oxide and clay that could activate persulfate oxidation for the
degradation of trichloroethylene or diesel compounds in recent
studies (Ahmad et al., 2010; Do et al., 2010). Recently, FeII-bearing
0045-6535/$ - see front matter ? 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2012.01.001
⇑Corresponding author at: Ecole Nationale Supérieure de Chimie de Rennes, UMR
CNRS 6226 ‘‘Sciences Chimiques de Rennes’’, Avenue du Général Leclerc, 35708
Rennes Cedex 7, France. Tel.: +33 2 23 23 80 27; fax: +33 2 23 23 81 20.
E-mail address: khalil.hanna@ensc-rennes.fr (K. Hanna).
Chemosphere 87 (2012) 234–240
Contents lists available at SciVerse ScienceDirect
Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Page 2

minerals like magnetite (Fe3O4) were found to be the most effec-
tive catalyst as compared to the only FeIIIoxides for heterogeneous
catalytic oxidation of organic pollutants (Kong et al., 1998; Matta
et al., 2007; Hanna et al., 2008; Xue et al., 2009a). In addition, mag-
netite exhibited excellent structural and catalytic stabilities and
can be used for several oxidation cycles (Xue et al., 2009a,b). The
objective of this study was to evaluate the ability of magnetite to
activate persulfate at circumneutral pH for the oxidation of PAH-
contaminated soils. To date, the use of magnetite, the most stable
mixed valence oxide instead of soluble FeIIto activate persulfate
oxidation has not been tested for PAH degradation in aged contam-
inated soils.
To achieve this goal, magnetite-activated persulfate oxidation
was studied in spiked sand and two PAH contaminated soils in
batch slurry system. These soils were sampled from two former
coking plants situated in the Northeast of France. Single model pol-
lutant (fluorenone) and organic extract of two soils was spiked on
sand followed by oxidation. Kinetic degradation of 16 PAHs was
also tracked in both real contaminated soils with or without an
extraction pretreatment. To compare with magnetite, oxidation
treatments were also evaluated under the same conditions with
persulfate activated by soluble FeIIor without iron activation.
The oxidation was studied versus time and organic analyses were
performed by GC–MS and lFTIR.
2. Experimental section
2.1. Chemical reactants
Pure fluorenone 98%, ferrous sulfate heptahydrate (FeSO4.
7H2O) and sodium persulfate (Na2S2O8) were purchased from Sig-
ma–Aldrich Co. Dichloromethane (DCM) and chloroform were pur-
chased from VWR and used as received.
Magnetite (FeIIFeIII2O4) used in this study was synthesized and
characterized by X-ray powder diffraction and Mössbauer spec-
troscopy in the context of our previous work (Usman et al.,
2012). Fontainebleau sand, with a grain size range of 150–
300 lm (mean diameter 257 lm) obtained from Prolabo was used
as support for oxidation experiments. The sand was cleaned with
1 M HCl to remove metal impurities. Rinsing with oxygenated
water was done to remove organic matter. The mineralogy of the
sand was characterized by X-ray diffraction and was found to be
exclusively quartz. Deionized water was produced with a Milli-Q
system from Millipore.
2.2. Soil samples
In this study, 16 PAHs in two different soils, Homécourt (H) and
Neuves-Maisons (NM) were selected as target compounds. These
soils were sampled from two former coking plant sites that are lo-
cated in the Northeast of France. The properties of both soils are
presented in the Supplementary materials (Table S1). Mineral size
fraction distributions were similar for both soils that are domi-
nated by sand mineral fractions (more than 60%) with pH from
neutral (7.20 in NM) to basic (8.35 in H). This pH could be related
to 10 times higher carbonate contents in H than NM soil. Extract-
able organic matter (EOM) contents were also higher in H than
NM. The major element concentrations were approximately the
same in both untreated soils. Samples were crushed to 500 lm
and freeze-dried. Oxidation was performed on these two soil sam-
ples. Isolation of organic extracts of both soils was done through
automatic extractor Dionex ASE 200 (Accelerated Solvent Extrac-
tor) at 100 ?C and 130 bar with DCM (Biache et al., 2008). Soil or-
ganic extract is composed of EOM which was separated by
solvent extraction from rest of soil (mineral fractions and insoluble
organic matter (IOM)). Soil organic matter (SOM) is composed of
EOM and IOM. Solvent was unable to extract IOM from soil.
2.3. Oxidation experiments
Before oxidation, all vessels were rinsed with DCM and then
several times with deionized milli-Q water. The persulfate medi-
ated degradation of PAHs was evaluated in the following different
systems: (1) model pollutant, fluorenone spiked in sand, (2) H and
NM organic extracts spiked in sand, (3) H and NM soil samples and
(4) H and NM organic extracts in soil.
In first two systems, fluorenone and organic extracts dissolved
in DCM was added in sand (4 mg g?1w/w). The DCM was allowed
to evaporate to dryness followed by oxidation. In the third system,
both soils H and NM were subjected to oxidation. In the last sys-
tem, soil samples were subjected to an extraction pretreatment
by agitating in chloroform during 45 min at 60 ?C. Normally, this
method is used for organic extraction of soils. But here, chloroform
was not withdrawn from the system; instead it was evaporated to
dryness. Our purpose here was to increase PAH availability.
Oxidation treatments at circumneutral pH were performed as
following: (i) sodium persulfate alone without iron activation
(SP), or (ii) sodium persulfate activated with soluble FeII(FP) or so-
dium persulfate activated with magnetite (10% w/w) (AP). Equiva-
lent molar amount of Fe was used to compare the efficiency of both
catalysts to activate persulfate. Blank experiments were performed
by using magnetite alone without oxidant to study possible
desorption or degradation of the PAHs. Kinetic degradation of PAHs
was studied for 1 week in batch series by assigning one batch for
each time point (1 h, 6 h, 24 h, 48 h and 1 week). All batch experi-
ments were performed in triplicates. All results were expressed as
a mean value of the 3 experiments and relative standard deviation
(RSD) of the three replicates was less than 5%. The RSD was calcu-
lated as described in Supplementary material. The kinetic oxida-
tion procedure is also detailed in Supplementary materials (SM).
2.4. Extraction and analysis
The freeze dried samples were extracted in chloroform during
45 min at 60 ?C. The volume was reduced to 20 mL under nitrogen
flow and 5 mL of the solution was dried and weighed to determine
the amount of EOM. The organic analyses were performed by GC–
MS and lFTIR. All details of analysis procedures are presented in
Supplementary materials (SM).
3. Results and discussion
3.1. Oxidation of fluorenone
Magnetite and soluble FeIIwere investigated for their potential
to activate persulfate oxidation for a single PAH pollutant, fluore-
none. It was chosen as a model pollutant due to its higher solubil-
ity, mobility and abundance in PAH-contaminated soils (Benhabib
et al., 2010). GC–MS quantification and EOM evolution were used
to monitor its abatement, both of which are in agreement. The
EOM recovered from sand at t = 0 was almost similar to the initially
added amount of fluorenone (i.e. ?4 mg g?1) which states that no
fluorenone was retained in sand. The degradation of fluorenone
is represented by plotting Ct/C0versus time (Fig. 1). Its concentra-
tion was determined by GC–MS where Ctis concentration at spe-
cific time point and C0 is the concentration at t = 0 before
oxidation. Blank experiments did not cause any degradation
(<3%) of fluorenone. AP treatment showed rapid degradation of flu-
orenone resulting in almost 90% of fluorenone abatement within
48 h and complete removal after 1 week of oxidation. Slight abate-
M. Usman et al./Chemosphere 87 (2012) 234–240
235

Page 3

ment (20–25%) was, however, obtained by FP or SP treatments.
This similar degradation yield with or without soluble FeIIactiva-
tion suggests that the soluble FeIIwas unable to activate persulfate
oxidation at circumneutral pH. Oxidation and precipitation of FeII
may occur at neutral pH conditions and thus making it unavailable
for activation. These results suggest that magnetite was effective to
activate persulfate oxidation for fluorenone degradation.
3.2. PAHs degradation in organic extracts spiked on sand
In this section, organic extracts previously isolated from both
soils (H or NM) were spiked on sand. After evaporation of solvent,
oxidation treatments AP, FP and SP were applied. Degradation of
16 PAHs by persulfate oxidation is represented in terms of Ct/C0
versus time (Fig. 2). Negligible degradation was observed for blank
experiments. After 1 week of oxidation in the AP system, 16 PAHs
were removed by approximately 83% and 72% in H and NM extract
respectively. Only 15–20% of degradation was achieved by FP or SP
treatments. An iodometric test (Kolthoff and Stenger, 1947)
showed that persulfate was still present in the first four points
(from 1 h to 48 h) and disappeared after 1 week in the AP system.
Thus, as observed for fluorenone, magnetite was found as a poten-
tial activator for persulfate oxidation of PAHs.
To observe the behavior of individual PAHs during oxidation, the
remaining PAHs were quantified for each experiment using GC–MS
at each time point. And 16 PAHs present in H and NM before (t = 0)
and after 1 week of oxidation, are shown in Fig. 3. For t = 0, the total
content of PAHs in H extract was lower (?204 lg g?1) than NM ex-
tract with almost 348 lg g?1of sand. The ratio of low molecular
weight PAH (LMW: sum of naphthalene to pyrene concentrations)
over high molecular weight PAH (HMW: sum of benzo[a]anthra-
cene to benzo[g,h,i]perylene concentrations) LMW/HMW was
different for both soils. The ratio of LMW/HMW suggests the pre-
dominance of LMW-PAHs with a value of almost 3.8 in H extract
as compared to 0.9 in NM extract with higher proportion of
HMW-PAHs. After 1 week of oxidation, ?83% of degradation for
16 PAHs in H extract was achieved by AP (Fig. 2). AP showed better
PAH removal efficiency for LMW-PAHs and lower reactivity to-
wards HMW-PAHs. As HMW-PAHs are abundant in NM extract,
therefore lower degradation yield (?72%) was achieved as com-
pared to H extract (?83%). Over time, the PAH composition was dis-
placed towards HMW PAHs as a result of more extensive
degradation of the smaller PAHs in NM extract. This might be be-
cause the LMW-PAHs are more degradable than HMW-PAHs and
were removed by persulfate oxidation during the early stage. It
was reported that a selective degradation of PAH can be attributed
to the weak oxidation conditions, while vigorous oxidation condi-
tions allowed the equal removal of PAHs (Ferrarese et al., 2008).
They stated that limited PAH removal was achieved by activated
persulfate, but the combined use of activated persulfate and hydro-
gen peroxide led to a better removal of both LMW and HMW PAHs.
Persulfate oxidation also exhibited slower degradation rate of fuel
oil than diesel which was likely to result from more complex com-
ponents in fuel oil (Yen et al., 2011).
During PAH oxidation, degradation by-products can be pro-
duced as a result of incomplete mineralization of PAHs. In this
study, the research of the eventual oxidation by-products was car-
ried out by lFTIR (Fig. 4) and GC–MS (Fig. S1) at different oxidation
times. The findings by lFTIR revealed that the initial EOM charac-
teristics remain unchanged after AP oxidation for both H and NM
organic extract (Fig. 4). The stability of the relative intensity of oxy-
genatedbands(especiallymOH:3700–3100 cm?1andmC@O: 1745–
1705 cm?1) suggests the absence of oxygenated by-products for-
mation. Moreover, the similarity of aliphatic profiles (mCHali:
3000–2800 cm?1and dCHali: 1470–1360 cm?1) and aromatic pro-
files (mCHaro: 3100–3000 cm?1, mC@C: 1620–1590 cm?1
cCHaro: 900–700 cm?1) reveals that no major molecular reorgani-
zation occurs (Fig. 4). Molecular analysis of the PAH distribution
and
Blank
SP
FP
AP
Fig. 1. Degradation of fluorenone (spiked on sand) during persulfate oxidation
experiments by: sodium persulfate alone without iron activation (j-SP), or sodium
persulfate activated with soluble FeII(s-FP) or sodium persulfate activated with
magnetite (d-AP). Blank (N) experiments were conducted by using only magnetite
without any oxidant. This degradation is represented in terms of Ct/C0where Ctis
the fluorenone concentration at specified oxidation time and C0is the concentration
at t = 0 (before oxidation) measured by GC–MS. Lines are only visual guide.
Experimental conditions were: solid matrix = 2 g, Volume of solution = 20 mL.
Oxidant dose was used according to oxidant: Fe molar ratio of 1:1.
Organic extract of H soil spiked on sand
Organic extract of NM soil spiked on sand
Fig. 2. PAH degradation in organic extracts extracted from soils (H and NM) during
persulfate oxidation experiments by: sodium persulfate alone without iron
activation (j-SP), or sodium persulfate activated with soluble FeII(s-FP) or sodium
persulfate activated with magnetite (d-AP). Blank (N) experiments were conducted
by using only magnetite without any oxidant. This degradation is represented in
terms of Ct/C0where Ctis the sum of 16 PAH concentration at specified oxidation
time and C0is their concentration at t = 0 measured by GC–MS. Lines are only visual
guide. Experimentalconditionswere:
tion = 20 mL. Oxidant dose was used according to oxidant: Fe molar ratio of 1:1.
solid matrix = 2 g, Volumeofsolu-
236
M. Usman et al./Chemosphere 87 (2012) 234–240

Page 4

LMW-PAHsHMW-PAHs
PAH contents (µg g-1 dw)
Before oxidation (t = 0)
AP oxidation after 1 week
H soil extract spiked on sand
NM soil extract spiked on sand
Fig. 3. Contents of individual PAHs in organic extracts of both soils (H and NM) at t = 0 before oxidation (h) and after 1 week of oxidation by magnetite-activated persulfate
(j-AP) PAH contents are based on the measurements by GC–MS.
ν νCHali
ν νCHaro
ν νC= O
ν νC= C
δ δCHali
γ γCHaro
ν νN-H
ν νC-N
Before oxidation (t = 0)
Wavenumbers (cm-1)
Wavenumbers (cm-1)
ν νCHali
ν νCHaro
ν νC= O
ν νC= C
δ δCHali
γ γCHaro
800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800
800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 3200 3400 3600 3800
Oxidation of H extract spiked on sand
Oxidation of NM extract spiked on sand
AP oxidation (t = 1 week)
Before oxidation (t = 0)
AP oxidation (t = 1 week)
(a)
(b)
Fig. 4. FTIR spectra of: (a) organic extract from H soil and (b) organic extract from NM soil before (t = 0) and after oxidation (1 week) by AP oxidation (sodium
persulfate + magnetite). Organic extracts from H and NM soil were added to sand and after evaporation of the solvent, oxidation was performed. Experimental conditions
were: solid matrix = 2 g, Volume of solution = 20 mL. Oxidant dose was used according to oxidant: Fe molar ratio of 1:1.
M. Usman et al./Chemosphere 87 (2012) 234–240
237

Page 5

by GC–MS in organic extracts of H and NM was in agreement with
lFTIR findings (Fig. S1). At different oxidation times, almost the
same chromatograms were found with respect to retention time
of existing molecules. No reaction intermediates or by-products
were detected after oxidation suggesting the complete oxidative
degradation of PAHs.
These results reveal that magnetite was effective to activate
persulfate oxidation of PAHs in spiked sand, while soluble FeII
seems to be unable to activate persulfate. To test if the magnetite
was effective in the presence of soil matrix, oxidation was per-
formed in real soils.
3.3. PAH degradation in soils
Before oxidation (t = 0), target soil samples showed higher con-
centration of 16 PAHs in H soil (?1369 lg g?1) than that in NM soil
(?1279 lg g?1). The value of LMW/HMW PAHs was almost 5.4 and
1.0 in H and NM soil, respectively. The ratio in H soil is higher than
observed for its organic extract (Fig. 3) that could be caused by a
small loss of LMW compounds during the evaporation of organic
extract in spiked sand.
Oxidation treatments applied were AP, FP or SP for the degrada-
tion of the 16 PAHs in both soils (H and NM). No PAH removal was
observed at pH 8.2 in H and 7.2 in NM, whatever the treatment was
used. Same results were obtained even with higher oxidant doses
(oxidant:Fe molar ratio of 2:1) (Fig. S2). Thus magnetite-activated
persulfate oxidation was effective to degrade PAHs in spiked sand
(Section 3.2) but not in real soil. The possible explanations for this
lack of oxidation in real soil could be PAH unavailability or the soil
matrix effect. To determine the possible reason, oxidation was per-
formed in soil after increasing PAH availability by an extraction
pretreatment but without isolating the organic extract from soil.
In other words, soil extraction was performed as explained in Sec-
tion 2.3 but organic extract was not separated from the rest of soil
(mineral + IOM). The chloroform was evaporated from the system
and then persulfate oxidation was performed by using oxidant
dose according to oxidant:Fe molar ratio of 2:1. The 16 PAHs quan-
tified by GC–MS are represented in terms of Ct/C0versus time in
Fig. 5. No PAH degradation was observed in blank experiments. Al-
most 5–10% of degradation was observed with FP or SP. Magnetite
activation resulted in degradation extent of PAHs of almost 60%
and 50% in H and NM soils, respectively. Thus PAH availability
was the crucial factor responsible for the absence of degradation
in real soil observed before.
Selective degradation (as observed for organic extracts in Sec-
tion 3.2) resulted in slightly lower PAH degradation in NM soil than
H soil. In NM soil, HMW-PAHs are abundant which are difficult to
degrade by AP, thus less PAH degradation was observed. Similar to
the oxidation of organic extract, GC–MS and lFTIR analyses re-
vealed the absence of by-products (data not shown).
These results indicate that once availability of PAHs was in-
creased in tested soils, effective PAH degradation was observed.
This highlights the important role of the PAHs availability in the
determination of treatment efficiency in a contaminated soil what-
ever the catalyst used. Degradation extent was, however, de-
creased in the presence of soil, that can be attributed to the soil
matrix effect (mineral and IOM). During persulfate oxidation of tri-
chloroethylene in soil (Liang et al., 2003, 2008) soil constituents
exhibited a considerable influence and appeared to scavenge sul-
fate free radicals. It was stated that the PAH degradation extent
by Fenton oxidation was inversely proportional to the total organic
carbon (TOC) for the soil with TOC above 5% (Bogan and Trbovic,
2003). TOC in both tested soil samples is close to 10% and 7% for
H and NM respectively (Table S.I.1). As the contents of 16-PAHs
(?1369 and ?1279 lg g?1for H and NM respectively) represent
less than 2% of soil TOC and EOM (36 and 13 mg g?1for H and
NM respectively) corresponds to almost 36% and 17% of H and N
soil respectively of the TOC of tested soils. This data represents that
most of the SOM is trapped as IOM which would affect oxidation
efficiency by oxidant depletion and PAH retention capacity by
SOM. The effect of soil matrix is in accordance with previous stud-
ies where different soil factors like SOM or the inorganic mineral
fractions were found to hinder PAH degradation (Bogan and Trbo-
vic, 2003; Goi and Trapido, 2004; Flotron et al., 2005; Jonsson et al.,
2007). In addition to SOM, clay particles contribute to the strong
sorption of PAHs in soils (Kawahara et al., 1995), thus making it
less available for degradation. However, the impact of soil matrix
is less pronounced for persulfate oxidation than Fenton oxidation
system. It was previously reported that persulfate can be activated
at basic pH by organic compounds similar to those present in soil
organic matter (Ahmad et al., 2010). In the same time, some soil
mineral components were unable to activate persulfate oxidation,
rather they could act as scavenging agents (Ahmad et al., 2010).
The interactions between persulfate and FeII–FeIIIoxide surface
(e.g. magnetite) can be explained by heterogeneous reactions anal-
ogous to the solution phase reactions and those proposed for het-
erogeneous Fenton-like oxidation:
BFeIIþ S2O2?
Since ferric oxides like ferrihydrite and goethite were shown to
activate persulfate (Ahmad et al., 2010), FeIII-oxide surface could
react with S2O2?
8
to generate sulfate radical but the interactions
of persulfate anion with the iron mineral surface are not yet well
argued. It was, however, reported that Fe-surface can act as radical
sulfate scavenger as it is known for hydroxyl radical:
8! BFeIIIþ SO??
4þ SO2?
4
ð1Þ
Blank
SP
FP
AP
H soil
Blank
SP
FP
AP
NM soil
Fig. 5. PAH degradation in soils after an extraction treatment (H and NM) during
persulfate oxidation experiments by: sodium persulfate alone without iron
activation (j-SP), or sodium persulfate activated with soluble FeII(s-FP) or sodium
persulfate activated with magnetite (d-AP). Blank (N) experiments were conducted
by using only magnetite without any oxidant. This degradation is represented in
terms of Ct/C0where Ctis the sum of 16 PAH concentration at specified oxidation
time and C0is their concentration at t = 0 measured by GC–MS. Lines are only visual
guide. Experimentalconditionswere:
tion = 20 mL. Oxidant dose was used according to oxidant: Fe molar ratio of 2:1.
solid matrix = 2 g,Volume ofsolu-
238
M. Usman et al./Chemosphere 87 (2012) 234–240

Page 6

BFeIIþ SO??
Other competitive or scavenging reactions may be occurred
such as the recombination of sulfate radicals to generate sulfate
anion (Mora et al., 2009).
4! BFeIIIþ SO2?
4
ð2Þ
2SO??
4! S2O2?
On the other hand, hydroxyl radical can be produced at alkaline
pH conditions (Mora et al., 2009)
via : SO??
8
k ¼ 5 ? 108Ms?1
ð3Þ
4þ HO?! SO2?
and at all pH values via : SO??
4þ?OH
ð4Þ
4þ H2O ! HO?þ HSO?
4
ð5Þ
Liang et al. (2007) suggested that the sulfate radical predomi-
nates under acidic conditions and the hydroxyl radical under alka-
line conditions. Both hydroxyl and sulfate radicals could attack the
target contaminants, but they also react with persulfate anion as:
HO?þ S2O2?
HO?þ S2O2?
8! HSO?
8! S2O??
4þ SO??
8þ HO?
4þ 1=2O2
or
ð6Þ
SO??
4þ S2O2?
Finally, radical sulfate may react with organic matter to give or-
ganic radical, as for instance:
SO??
8! S2O??
8þ SO2?
4
ð7Þ
4þ ROH ! HSO?
In the soil slurry system, there should be many reactions with
synergic and/or antagonistic effect on the persulfate activation
and persulfate oxidation reaction. All the soil matrix factors could
contribute in influencing the persulfate oxidation efficiency.
4þ R?OH
ð8Þ
4. Conclusion
This study reported the use of magnetite to activate persulfate
oxidation for the degradation of PAHs in spiked sand and in aged
contaminated soils at circumneutral pH. All experimental results
indicated the higher efficiency of magnetite-activated persulfate
oxidation (AP) for PAH removal. Experiments performed with sol-
uble FeII(FP) or without iron activation (SP) resulted in only 15–
20% of PAH removal in spiked sand while 5–10% in real soil. Such
similar degradation with or without soluble FeIIexhibits the inabil-
ity of soluble FeIIto activate persulfate oxidation. However, magne-
tite was highly reactive to activate persulfate oxidation as almost 5
times higher PAH removal efficiency was observed in the AP oxida-
tion system. This big discrepancy points out the importance of cat-
alyst type sued to activate persulfate decomposition.
On the other hand, the PAH unavailability as well as soil matrix
effect seems to be the most important factors for persulfate oxida-
tion process. Selective degradation behavior was shown by persul-
fate oxidation with less efficiency towards HMW-PAHs. No by-
products were observed in both treated soils. Application of mag-
netite is therefore effective for significant activation of persulfate
at circumneutral pH and could be a promising way for improving
chemical oxidation of aged contaminated soils. For in situ applica-
tions, further investigations should be done for the transport study
of magnetite and oxidant in soil columns.
Acknowledgements
The authors gratefully acknowledge the financial support of this
work by HEC (Higher Education Commission of Pakistan) and
ADEME ‘‘Agence de l’Environnement et de la Maîtrise de l’Energie’’
(Grant No. 0972C0016). We are also thankful to the Région Lor-
raine and GISFI (Groupement d’Intérêt Scientifique sur les Friches
Industrielles) for support. The authors would like to thank the
anonymous reviewers for their valuable comments and sugges-
tions that have improved the manuscript.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.chemosphere.2012.01.001.
References
Ahmad, M., Teel, A.L., Watts, R.J., 2010. Persulfate activation by subsurface minerals.
J. Contam. Hydrol. 115, 34–45.
Benhabib, K., Faure, P., Sardin, M., Simonnot, M.-O., 2010. Characteristics of a solid
coal tar sampled from a contaminated soil and of the organics transferred into
water. Fuel 89, 352–359.
Biache, C., Mansuy-Huault, L., Faure, P., Munier-Lamy, C., Leyval, C., 2008. Effects of
thermal desorption on the composition of two coking plant soils: impact on
solvent extractable organic compounds and metal bioavailability. Environ.
Pollut. 156, 671–677.
Bogan, B.W., Trbovic, V., 2003. Effect of sequestration on PAH degradability with
Fenton’s reagent: roles of total organic carbon, humin, and soil porosity. J.
Hazard. Mater. 100, 285–300.
Do, S.-H., Kwon, Y.-J., Kong, S.-H., 2010. Effect of metal oxides on the reactivity of
persulfate/Fe(II) in the remediation of diesel-contaminated soil and sand. J.
Hazard. Mater. 182, 933–936.
Ferrarese, E., Andreottola, G., Oprea, I.A., 2008. Remediation of PAH-contaminated
sediments by chemical oxidation. J. Hazard. Mater. 152, 128–139.
Flotron, V., Delteil, C., Padellec, Y., Camel, V., 2005. Removal of sorbed polycyclic
aromatic hydrocarbons from soil, sludge and sediment samples using the
Fenton’s reagent process. Chemosphere 59, 1427–1437.
Goi, A., Trapido, M., 2004. Degradation of polycyclic aromatic hydrocarbons in soil:
The Fenton reagent versus ozonation. Environ. Technol. 25, 155–164.
Gryzenia, J., Cassidy, D., Hampton, D., 2009. Production and accumulation of
surfactants during the chemical oxidation of PAH in soil. Chemosphere 77, 540–
545.
Hanna, K., Kone, T., Medjahdi, G., 2008. Synthesis of the mixed oxides of iron and
quartz and their catalytic activities for the Fenton-like oxidation. Catal.
Commun. 9, 955–959.
House, D.A., 1962. Kinetics and mechanism of oxidations by peroxydisulfate. Chem.
Rev. 62, 185–203.
ITRC, 2005. Technical and Regulatory Guidance for In Situ Chemical Oxidation of
Contaminated Soil and Groundwater. ITRC, ISCO Team, Washington, DC.
Jonsson, S., Persson, Y., Frankki, S., van Bavel, B., Lundstedt, S., Haglund, P., Tysklind,
M., 2007. Degradation of polycyclic aromatic hydrocarbons (PAHs) in
contaminated soils by Fenton’s reagent: a multivariate evaluation of the
importance of soil characteristics and PAH properties. J. Hazard. Mater. 149, 86–
96.
Kawahara, F.K., Davila, B., Al-Abed, S.R., Vesper, S.J., Ireland, J.C., Rock, S., 1995.
Polynuclear aromatic hydrocarbon (PAH) release from soil during treatment
with Fenton’s Reagent. Chemosphere 31, 4131–4142.
Kolthoff, I.M., Stenger, V.A., 1947. Volumetric analysis. Titration Methods: Acid-
Base, Precipitation, and Complex Reactions, vol. II Second Revised Edition.
Kong, S.H., Watts, R.J., Choi, J.H., 1998. Treatment of petroleum-contaminated
soils using iron mineral catalyzed hydrogen peroxide. Chemosphere 37,
1473–1482.
Latimer, W.M., 1952. Oxidation Potentials. Prentice-Hall, Inc., Englewood Cliffs, NJ.
Liang, C.J., Bruell, C.J., Marley, M.C., Sperry, K.L., 2003. Thermally activated persulfate
oxidation of trichloroethylene (TCE) and 1, 1, 1-trichloroethane (TCA) in
aqueous systems and soil slurries. Soil Sediment Contam. 12, 207–228.
Liang, C., Bruell, C.J., Marley, M.C., Sperry, K.L., 2004. Persulfate oxidation for in situ
remediation of TCE. II. Activated by chelated ferrous ion. Chemosphere 55,
1225–1233.
Liang, C., Wang, Z.-S., Bruell, C.J., 2007. Influence of pH on persulfate oxidation of
TCE at ambient temperatures. Chemosphere 66, 106–113.
Liang, C., Lee, I.L., Hsu, I.Y., Liang, C.-P., Lin, Y.-L., 2008. Persulfate oxidation of
trichloroethylenewithand without
Chemosphere 70, 426–435.
Matta, R., Hanna, K., Chiron, S., 2007. Fenton-like oxidation of 2, 4, 6-trinitrotoluene
using different iron minerals. Sci. Total Environ. 385, 242–251.
Mora, V.C., Rosso, J.A., Carrillo Le Roux, G., Mártire, D.O., Gonzalez, M.C., 2009.
Thermally activated peroxydisulfate in the presence of additives: a clean
method for the degradation of pollutants. Chemosphere 75, 1405–1409.
Nadim, F., Huang, K.C., Dahmani, A.M., 2006. Remediation of soil and ground water
contaminated with PAH using heat and Fe(II)-EDTA catalyzed persulfate
oxidation. Water Air Soil Poll. Focus 6, 227–232.
Osgerby, I.T., 2006. ISCO technology overview: do you really understand the
chemistry? In: Calabrese, E.J., Kostecki, P.T., Dragun, J., Osgerby, I.T. (Eds.),
Contaminated Soils, Sediments and Water. Springer, US, pp. 287–308.
Usman, M., Faure, P., Hanna, K., Abdelmoula, M., Ruby, C., 2012. Application of
magnetite catalyzed chemical oxidation (Fenton-like and persulfate) for the
remediation of oil hydrocarbon contamination. Fuel, inpress, doi:10.1016/
j.fuel.2012.01.017.
ironactivation inporousmedia.
M. Usman et al./Chemosphere 87 (2012) 234–240
239

Page 7

Wild, S.R., Jones, K.C., 1995. Polynuclear aromatic hydrocarbons in the United
Kingdom environment: a preliminary source inventory and budget. Environ.
Pollut. 88, 91–108.
Xue, X., Hanna, K., Abdelmoula, M., Deng, N., 2009a. Adsorption and oxidation of
PCP on the surface of magnetite: kinetic experiments and spectroscopic
investigations. Appl. Catal. B – Environ. 89, 432–440.
Xue, X., Hanna, K., Despas, C., Wu, F., Deng, N., 2009b. Effect of chelating agent on
the oxidation rate of PCP in the magnetite/H2O2 system at neutral pH. J. Mol.
Catal. A – Chem. 311, 29–35.
Yen, C.-H., Chen, K.-F., Kao, C.-M., Liang, S.-H., Chen, T.-Y., 2011. Application of
persulfate to remediate petroleum hydrocarbon-contaminated soil: feasibility
and comparison with common oxidants. J. Hazard. Mater. 186, 2097–2102.
240
M. Usman et al./Chemosphere 87 (2012) 234–240