You are on page 1of 8

w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

Efficient degradation of atrazine by magnetic


porous copper ferrite catalyzed peroxymonosulfate
oxidation via the formation of hydroxyl and sulfate
radicals

Ying-Hong Guan a,c, Jun Ma a,*, Yue-Ming Ren b,**, Yu-Lei Liu a,
Jia-Yue Xiao a, Ling-qiang Lin b, Chen Zhang b
a
State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090,
People’s Republic of China
b
Key Laboratory of Superlight Materials & Surface Technology, Ministry of Education College of Materials Science
and Chemical Engineering, Harbin Engineering University, Harbin 150008, People’s Republic of China
c
School of Water Conservancy and Construction Northeast Agriculture University, Harbin 150030, People’s Republic
of China

article info abstract

Article history: Magnetic porous copper ferrite (CuFe2O4) showed a notable catalytic activity to perox-
Received 3 February 2013 ymonosulfate (PMS). More than 98% of atrazine was degraded within 15 min at 1 mM PMS
Received in revised form and 0.1 g/L CuFe2O4. In contrast, CuFe2O4 exhibited no obvious catalytic activity to per-
6 June 2013 oxodisulfate or H2O2. Several factors affecting the catalytic performance of PMS/CuFe2O4
Accepted 12 June 2013 were investigated. Results showed that the catalytic degradation efficiency of atrazine
Available online 20 June 2013 increased with PMS and CuFe2O4 doses, but decreased with the increase of natural organic
matters concentration. The catalytic oxidation also showed a dependence on initial pH.
Keywords: The presence of bicarbonate stimulated atrazine degradation by PMS/CuFe2O4 at low
Sulfate radical concentrations but inhibited the degradation at high concentrations. Furthermore, the
Hydroxyl radical reactive species for atrazine degradation in PMS/CuFe2O4 system were identified as hy-
droxyl radical (HO ) and sulfate radical ðSO$
4 Þ through competition reactions of atrazine

Peroxymonosulfate
Copper ferrite and nitrobenzene, instead of commonly used alcohol scavenging, which was not a reliable
Atrazine method in metal oxide catalyzed oxidation. Surface hydroxyl groups of CuFe2O4 were a
Oxone critical part in radical generation and the copper on CuFe2O4 surface was an active site to
catalyze PMS. The catalytic degradation of atrazine by PMS/CuFe2O4 was also effective
under the background of actual waters.
ª 2013 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ86 451 86283010; fax: þ86 451 86282292.
** Corresponding author.
E-mail address: majun@hit.edu.cn (J. Ma).
0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.watres.2013.06.023
5432 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8

1. Introduction mechanism on radical formation in the PMS/CuFe2O4 system,


3) to investigate the influences of water quality parameters on
Atrazine is a kind of triazine herbicides widely used for crops the catalytic performance of PMS/CuFe2O4 including pH value,
and weed control. Photolysis of atrazine by sunlight is slow PMS and CuFe2O4 doses, natural organic matters (NOM), and
due to the weak absorption at the wavelength above 220 nm. inorganic salts, as well as the influences of actual water
Atrazine is also resistant to microorganism degradation and backgrounds.
therefore persistent in water and soil. Various studies were
conducted to generate HO for the destruction of atrazine,
such as photolysis of hydrogen peroxide (De Laat et al., 1999), 2. Experimental
Fenton and photo/electrochemical-Fenton (De Laat et al.,
1999; Ventura et al., 2002), ozone and catalytic ozonation 2.1. Materials
(Acero et al., 2000; Ma and Graham, 2000).
Sulfate radical ðSO$4 Þ is a strong oxidant with the redox Potassium peroxymonosulfate (available as Oxone, PMS),
potential of 2.5e3.1 V (Neta et al., 1988), similar to HO. It is potassium peroxodisulfate (PDS), atrazine, sodium phosphate
usually generated from peroxymonosulfate (PMS) and perox- monobasic monohydrate, sodium phosphate dibasic were of
odisulfate (PDS) by UV, heat, base, transition metals, or qui- ACS reagent grade and purchased from SigmaeAldrich.
nones activation (Anipsitakis and Dionysiou, 2004; Fang et al., Hydrogen peroxide (35% w/w) was obtained from Alfa Aesar.
2013b; Furman et al., 2010; Guan et al., 2011; Waldemer et al., Methanol (Thermo Fisher Scientific) and acetic aid (Dima-
2007). Comparing with PDS, PMS is easy to be activated by Tech) were of HPLC grade. Suwannee River natural organic
transition metals due to the asymmetric molecular structure. matter (SRNOM) was purchased from the International Humic
Homogeneous catalytic oxidation by PMS coupled with tran- Substances Society. Tert-butyl-alcohol (TBA) was of guaran-
sition metals showed efficient decontamination of organic teed reagent grade and purchased from Tianjin Chemical
pollutants, among which Co(II) was found to be a best acti- Reagent Co., Ltd., China. Copper nitrate, ferric nitrate, sodium
vator (Anipsitakis and Dionysiou, 2004). However, cobalt has hydroxide, perchloric acid, sodium bicarbonate were all of
been recognized as a possible human carcinogen by Interna- analytical-reagent grade and purchased from Sinopharm
tional Agency for Research on Cancer (IARC). In order to Chemical Reagent Co., Ltd., China. All solutions were prepared
reduce the concentration of free cobalt ions, attentions have in Milli-Q water produced on a Milli-Q Biocel water system.
been focused on heterogeneous catalysis of PMS by supported
cobalt, cobalt oxides, and metal-cobalt mixed oxides (Chan 2.2. Preparation and characterization of CuFe2O4
and Chu, 2009; Qi et al., 2013; Yang et al., 2009).
Spinel ferrites, expressed as MeFe2O4 (Me ¼ Mn, Co, Ni, Cu, Magnetic porous CuFe2O4 was prepared by a solegel process
Mg, etc.), are kinds of magnetic materials widely used in with egg white at the copper and ferric molar ratio of 1:2,
electronic applications, which could be separated easily from referring to the preparation of NiFe2O4 reported previously
water under magnetic field. The ferrites were used to remove (Ren et al., 2012). Detailed information sees Text S1.
heavy metals from water bodies due to the abundant surface The surface morphologies were obtained on a Hitachi S-
hydroxyl groups and magnetic property (Tang et al., 2013). 3000N scanning electron microscopy (SEM). The X-ray powder
Studies were also conducted on the photocatalytic and cata- diffraction (XRD) was performed on a TTR-III diffractometer
lytic activities of spinel ferrites. ZnFe2O4 exhibited notable (Rigaku, Japan) with Cu Ka radiation. The magnetic property
photocatalytic activity for H2O2 under visible light irradiation was measured by a vibrating sample magnetometer (JDM-14D,
(Su et al., 2012). Catalytic ozonation by NiFe2O4 was also re- China). Fourier transform infrared (FTIR) spectra were per-
ported to destruct di-n-butyl phthalate (Ren et al., 2012). Fer- formed on an IS10 FTIR spectrometer (Nicolet, America) in the
rites (Co, Cu, Mn) catalyzed H2O2 oxidation could decolorize range of 500e4000 cm1. Raman spectra were taken on a

synthetic dyes via the generation of HO (Baldrian et al., 2006). LabRAM HR800 confocal microscopic Raman spectrometer
PDS catalyzed by magnetite nanoparticles were investigated (HORIBA Jobin Yvon, France) with 488 nm laser irradiation. X-
to degrade 2,4,40 -trichlorobiphenyl and sulfamonomethoxine ray photoelectron spectroscopy (XPS) analysis was performed
(Fang et al., 2013a; Yan et al., 2011). CoFe2O4 was used to with PHI 5700 ESCA System (Physic Electronics, America)
catalyze PMS to reduce the concentration of free cobalt ion using an Al Ka excitation source (1486.6 eV). All lines recorded
(Yang et al., 2009). Our preliminary study found that CuFe2O4 were calibrated to the C 1 s line at 284.6 eV. The pHPZC of
was also effective to activate PMS among the ferrites (Mn, Co, CuFe2O4 was determined by mass titration.
Ni, Cu, Zn), although not as effective as CoFe2O4. Furthermore, The CuFe2O4 was characterized as a magnetic porous ma-
copper is not recognized as potential carcinogen by IARC. terial of tetragonal-type structure, with the average particle
Considering the catalytic activity and the risk of metal diameter and pore size being 50e100 mm and 1e5 mm (SI Text
leaching of solid catalysts, CuFe2O4 might be applicable to S2 and Fig. S1).
activate PMS in water treatment.
The purpose of this research was 1) to investigate the cat- 2.3. Experimental procedure
alytic activities of porous CuFe2O4 on peroxides (PMS, PDS and
H2O2), which was convenient to be separated from water due Experiments were performed in 250 mL conical flask at
to its large size compared with nanoparticles, 2) to identify the 25  2  C, mixed at 150 rpm. Desired amount of PMS and
reactive species for atrazine degradation and possible atrazine stock solutions were added to solution to a final
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8 5433

volume of 100 mL. Solution pH was adjusted by perchloric acid


and sodium hydroxide. Then the reaction was initiated once
the catalyst was added. Samples were withdrawn and filtrated
by Whatman GF/F glass fiber membranes, and then quenched
by sodium nitrite or ascorbic acid before analysis. In the case
of actual water backgrounds, effluents from different stages of
a water treatment plant in Huainan, China, were used. Raw
surface water was treated in sequence by coagulation, sedi-
mentation, dissolved air flotation (DAF), biological activated
carbon (BAC), and sand filtration (SF) before disinfection. The
effluents from DAF, BAC, and SF were filtrated through
Whatman GF/F glass fiber membranes and used as the water
backgrounds in comparison tests. All the experiments were
conducted in triplicate and error bar represented the standard
deviation of the replicate experimental data.
Fig. 1 e Degradation efficiency of atrazine and
2.4. Analysis concentrations of leached metal ions. Conditions:
[PMS] [ 1.0 mM, [CuFe2O4] [ 0.1 g/L, [atrazine] [ 2.0 mM,
Atrazine and nitrobenzene were quantified by high perfor- initial pH [ 7.5 ± 0.2.
mance liquid chromatography described previously (Guan
et al., 2011), at l ¼ 221 nm and 263 nm with an eluent of
water (0.25& acetic acid) and methanol with ratio of 35:65 and
40:60 (v/v) at a flow rate of 1.0 mL/min. Method detection while the concentration of free copper ion increased signifi-
limits (MDL) were 6.5 mg/L (0.03 mM) and 9.8 mg/L (0.08 mM). Free cantly with reaction time and reached about 620 mg/L by
copper and iron ions were quantified by an ICP-MS (AB Sciex 15 min (Fig. 1). Meanwhile, PMS still remained 92% of the
Qtrap 5500) with MDL of 0.2 mg/L and 6 mg/L. Dissolved organic initial concentration after 15 min. Then atrazine degradation
carbon (DOC) was measured by TOC analyzer (Analytik Jena by the filtrate of PMS and CuFe2O4 suspension (reaction for
Multi N/C 3100). Excitation emission matrix (EEM) fluores- 15 min) was investigated. But the filtrate showed almost no
cence spectra were recorded on a Jasco FP-6500 fluorescence degradation of atrazine (Fig. 1), indicating that the formation
spectrometer. The solutions of PMS and PDS were standard- of reactive species for atrazine degradation would occur at the
ized using iodometric titration (Ball et al., 1967; Güpta, 1961). surface of CuFe2O4.
H2O2 was quantified by its absorbance at 240 nm
(ε ¼ 38.1 M1 cm1) (Morgan et al., 1988). 3.2. Influences of several factors

3.2.1. Effect of PMS and catalyst doses


3. Results and discussions Atrazine degradation was enhanced with the increase of PMS
dose. The degradation efficiency was 16% by 2 min at 0.1 mM
3.1. Catalytic oxidation of atrazine PMS, and kept increasing to 49% at 1 mM PMS and 57% at
2.5 mM PMS (Fig. 2a). It could be deduced that the active sites
Atrazine was fast degraded by PMS/CuFe2O4 and more than on CuFe2O4 surface were still not totally occupied by PMS even
98% of atrazine was destructed within 15 min at 1 mM PMS at the concentration of 1 mM with CuFe2O4 dose of 0.1 g/L,
and 0.1 g/L CuFe2O4. Meanwhile, PMS alone showed no since the degradation efficiency kept increasing with the in-
degradation of atrazine and less than 5% of atrazine was crease of PMS concentration although the radical-scavenging
adsorbed onto CuFe2O4 (Fig. 1). Ferrites were effective photo- ability of PMS also increased.
catalysts in visible and UV region, capable of producing reac- Atrazine degradation also showed a positive dependence
tive oxygen species (Su et al., 2012). In order to exclude the role on CuFe2O4 dose (Fig. 2b). Degradation efficiency increased
of light in catalytic oxidation, experiments were further con- continuously from 8% to 67% by 2 min as CuFe2O4 dose
ducted under dark condition (covered by aluminum foil) and increased from 0.05 g/L to 1 g/L. Previous study showed that
normal interior lighting condition (Fig. 1). Results showed that the decontamination rate reached a maximum and then
there was almost no difference of atrazine degradation effi- decreased in catalytic ozonation due to radical scavenging or
ciencies under the two conditions, which indicated the role of ineffective ozone consumption at higher catalyst dose (Zhang
light being negligible at the intensity of normal interior et al., 2012). Fe3O4 nanoparticles showed the maximal cata-
lighting. So the catalytic oxidation of atrazine would be due to lytic activity to PDS at a dose of 0.55 g/L (2.4 mM) (Yan et al.,
the synergistic effect of PMS and CuFe2O4. 2011). In the present study, the continuous increase of atra-
Concentrations of leached free metal ions from CuFe2O4 zine degradation efficiency at the initial stage (0e5 min) with

were monitored since they might activate PMS to produce HO CuFe2O4 dose, reflected that the detrimental effect of high
and/or SO$4 . In the absence of PMS, the concentrations of iron dose of CuFe2O4 was overwhelmed by its stimulation effect.
and copper ions in the filtrate of CuFe2O4 suspension were This would be a solution to reduce costs in actual application
within 50 mg/L and 30 mg/L (data not shown). In the presence of by high dose of CuFe2O4 and low dose of PMS. Whilst, the
PMS, the concentration of free iron ion kept below 40 mg/L, lower degradation efficiency for higher catalyst dose (0.5 g/L)
5434 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8

Fig. 3 e Influence of initial pH on atrazine degradation by


PMS/CuFe2O4. Conditions: [PMS] [ 1.0 mM,
[atrazine] [ 5.2 mM, [CuFe2O4] [ 0.1 g/L.

maximal decontamination efficiency around pHPZC of the


catalyst (Ren et al., 2012; Zhang et al., 2008). However, the
pHPZC of CuFe2O4 was measured as 8.5, which was not
consistent with the optimal initial pH (Fig. 3). Metal oxides
usually acted as buffer and might change solution pH. So pH
variation during the reaction was monitored roughly (Fig. S3).
pH increased gradually with reaction time at pH 4 and pH 5.5
due to the buffer effect of CuFe2O4. Meanwhile, it reached
around or below pH 8.5 (pHPZC) within 2 min at initial pH range
Fig. 2 e Influences of PMS dose (a) and CuFe2O4 dose (b) on of 6.5e9.5. At initial pH 10.5, pH decreased with time and
atrazine degradation Conditions: [atrazine] [ 2.0 mM, finally kept above pH 9.5, which made the surface of CuFe2O4
[NaHCO3] [ 2 mM, initial pH [ 7.5 ± 0.2, [CuFe2O4] [ 0.1 g/L negatively charged. Meanwhile, PMS existed mostly in the
for (a), [PMS] [ 0.1 mM for (b). form of SO2 5 at pH range of 9.5e10.5. Repulsive force from
electrostatic charge reduced the interaction chance of PMS
and CuFe2O4, resulting in the decline of atrazine degradation
at late stage of reaction (10e15 min), might be due to the fast when pH ranging from 9.5 to 10.5. At 4 < pH < 8.5, PMS mostly
decomposition of PMS. existed as HSO 5 . With the decrease of pH, the surfaces of
CuFe2O4 would have a net positive charge and facilitate the
3.2.2. Role of pH contact of PMS and CuFe2O4 by attractive force, but atrazine
Metal oxides catalysts have amphoteric characteristics in degradation was declined. This indicated that the surface
water suspension as depicted in Eqs. (1) and (2) (Ma et al., hydroxyl groups in neutral state would be responsible for the
2005). The surface charge property of catalyst depends on catalysis of PMS. Furthermore, the relative higher degradation
the difference between solution pH and pHPZC of the catalyst efficiency of atrazine at pH 10.5 than that at pH  8.5, indicated
(pH at the point of zero charge). The surfaces of catalyst have a SO2 
5 was easier to be activated than HSO5 in heterogeneous
net positive (negative) charge at pH lower (higher) than pHPZC, catalytic oxidation.
while the net particle charge is zero at pH around pHPZC.
3.2.3. Influence of bicarbonate
hMeOHþ
2 #hMeOH þ H
þ
(1)
Bicarbonate was the representative of inorganic carbon existed
in natural water at the range of 50e200 mg/L (Ma and Graham,
hMeOH#hMeO þ Hþ (2)
2000), which usually acted as radical scavengers in advanced
Besides, the pKa of PMS (second pKa of its parent acid) was oxidation process (De Laat et al., 2011). In the PMS/CuFe2O4
9.4 (Rani et al., 2009) and its parent acid (H2SO5) was a strong system, bicarbonate showed enhancement on atrazine degra-
acid as sulfuric acid (Lawrance and Ward, 1985). Thus, in the dation at lower concentration (2 mM) and detrimental effect at
investigated pH range of 4e10.5, the existed form of PMS higher concentration (4 mM) as shown in Fig. 4. Final solution
depended mainly on solution pH and the second pKa. pH was measured for its obvious influence on atrazine degra-
Catalytic oxidation of atrazine by PMS/CuFe2O4 showed a dation. It was around 6.2 in the absence of bicarbonate (Fig. S3)
dependence on initial pH (Fig. 3 and Fig. S2). Atrazine degra- and 7.3 in the presence of 2 mM bicarbonate (not shown). So the
dation was enhanced as the initial pH increased from 4 to 9.5, higher solution pH during reaction might be one reason for the
and then declined from pH 9.5 to pH 10.5. Heterogeneous enhancement of bicarbonate at low concentration. Bicarbonate
catalytic ozonation by a-FeOOH or NiFe2O4 showed the

reacts with HO or SO$ 4 to form carbonate radical. At high
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8 5435

Fig. 4 e Influence of bicarbonate on atrazine degradation Fig. 5 e Influence of NOM on atrazine degradation by PMS/
by PMS/CuFe2O4. Conditions: [PMS] [ 1.0 mM, CuFe2O4. Conditions: [PMS] [ 1.0 mM, [CuFe2O4] [ 0.1 g/L,
[CuFe2O4] [ 0.1 g/L, [atrazine] [ 2.0 mM, initial [atrazine] [ 2.0 mM, initial pH [ 7.5 ± 0.2.
pH [ 7.5 ± 0.1.


Therefore, TBA was used as scavenger for HO and methanol


concentration (4 mM), bicarbonate would mostly act as HO


 as scavenger for HO and SO$ 4 .

and SO$ As shown in Fig. 6, TBA and methanol indeed inhibited


4 scavengers and therefore inhibit atrazine degradation.
At lower concentration, carbonate radical might act as a pro- atrazine degradation and the inhibition were enhanced with
moter stimulating PMS decomposition into HO or SO$
 the increased of their dosages. At low concentration (1 mM),
4 , similar
to the enhancement of bicarbonate in catalytic ozonation (Zhao methanol did not inhibit atrazine degradation as much as TBA
et al., 2010). Furthermore, carbonate could form complexes on while methanol inhibited atrazine degradation more signifi-
hematite surface (Bargar et al., 2005) and bicarbonate enhanced cantly than TBA at high concentration (100 mM). Methanol


the oxidation of ferrous by oxygen (King, 1998). The complexes reacts with HO and SO$ 4 to form hydroxylmethyl radical
of bicarbonate on CuFe2O4 surface might also be a reason for the (CH2OH), which might stimulate the decomposition of PMS
enhancement at low concentration. into radicals (Guan et al., 2011), and further alleviate the in-
hibition of methanol on atrazine degradation at low concen-
3.2.4. Influence of NOM tration. The different inhibition effects of the two alcohols at
NOM usually act as radical scavenger via competing for HO high concentration (Fig. 6), seemed due to the formation of
and SO$ 4 . However, TBA inhibited the non-radical interface
SO$
4 . Phenolic hydroxyl and carboxyl groups of NOM
might also be adsorbed onto catalyst surface and block reac- oxidation of oxalate possibly due to its adsorption onto cata-
tive sites. Whilst, semiquinone radicals, formed from hydro- lyst (Zhang et al., 2012). The different inhibition effect of al-
quinones, quinones and phenols in NOM, might also cohols might not only be due to their different abilities for
stimulate the decomposition of PMS into HO or SO$
 scavenging radicals, but also due to their different adsorption
4 , similar

to the way of reducing H2O2 to HO (Zhu et al., 2002). In the properties onto CuFe2O4 and subsequent occupation for active
PMS/CuFe2O4 system, atrazine degradation was inhibited by sites. Hence, it was not a reliable method to use TBA and
the addition of SRNOM. The degradation efficiency at 15 min
decreased from 98% to 77% and 23% when SRNOM concen-
tration increased from 0 mg/L to 1.0 mg/L and 3.2 mg/L (Fig. 5).
It indicated that the stimulated effect of NOM was over-
whelmed by its detrimental effect in the range of
1.0e3.2 mg/L.

3.3. Reactive species and possible mechanism

3.3.1. Reactive species


Atrazine was rarely adsorbed on CuFe2O4 (Fig. 1) and the
structure of atrazine (Fig. S4) did not facilitate itself to undergo
non-radical interface reaction. Furthermore, catalytic ozona-
tion of atrazine by metal oxide was proved mainly via radical
attack rather than non-radical interface reaction (Zhang et al.,
Fig. 6 e Inhibition of tert-butyl alcohol (TBA) and methanol

2012). PMS could generate HO and/or SO$ 4 by homogeneous or
heterogeneous activation. Peroxomonosulfate radical ðSO$ 5 Þ (Me) on atrazine degradation by PMS/CuFe2O4. Conditions:
might also be formed, but would not contribute to atrazine [PMS] [ 1.0 mM, [CuFe2O4] [ 0.1 g/L, [atrazine] [ 2 mM,
degradation for its lower redox potential (Neta et al., 1988). [phosphate] [ 5 mM, initial pH [ 7.5 ± 0.2.
5436 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8


methanol to identify HO and SO$ 4 in heterogeneous metal around pHPZC of CuFe2O4 (Fig. 3), indicating that the surface
oxides catalyzed oxidation. Further experiments were con- hydroxyl groups in neutral state would be major part
ducted to identify the reactive species in PMS/CuFe2O4 system. responsible for the activation of PMS.
A red-shift of 10 cm1 of the peak at 1200 cm1 in the IR

Nitrobenzene was selected as HO probe compound due to
its low reactivity with SO$4 (Guan et al., 2011). Nitrobenzene spectra was observed when CuFe2O4 was added to PMS solution
and atrazine were rarely removed by PMS or CuFe2O4 alone (see SI Text S3 and Fig. S5), indicating the change of SeO bond.
(data not shown). However, nitrobenzene was degraded The newly emerging peak of the OeO bond in Raman spectrum
obviously by PMS/CuFe2O4 (Fig. 7), which verified the forma- of PMS/CuFe2O4, might indicate the complexes of PMS on

tion of HO in PMS/CuFe2O4 system. CuFe2O4 surface (see SI Text S4 and Fig. S6). Furthermore, the
If the reactive species for atrazine degradation were both increased concentration of leached copper ion during reaction

HO and SO$ 4 , the kinetic expressions of atrazine and nitro- (Fig. 1), indicated that Cu on the surface of CuFe2O4 might be the
benzene degradation could be expressed as follows. active sites to catalyze PMS. The analysis of XPS spectra of
CuFe2O4 before and after catalytic reaction (Fig. S7), showed that
dCNB
 ¼ k1 ½HO$ CNB ¼ k0;NB CNB (3) the valence of Fe on the surface of CuFe2O4 did not change, while
dt
the valence of Cu changed from Cu2þ to the mixture of Cu2þ and
Cuþ (see SI Text S5). SO$


  4 and HO might be generated during the


dCAT
 ¼ k2 ½HO$ CAT þ k3 SO$
4 CAT ¼ ko;AT CAT (4) cycle of Cu2þ/Cuþ via the electron transfer from PMS to Cu2þ and
dt

the reduction of PMS by Cuþ (Ding et al., 2013; Gilbert and Stell,
where k1 (k2) was the rate constant between HO and nitro- 1990). Although one electron transfer from PMS to Cu2þ
benzene (atrazine), k3 was the rate constant between SO$4 and seemed unfavorable based on the standard reduction potential
atrazine. k0,NB and k0,AT were the pseudo first order rate con- of Cu2þ/Cuþ (0.17 V) and SO2  2
5 /HSO5 (1.1 V) or SO5 /SO5 (0.81 V)
$

stants of nitrobenzene and atrazine degradation. CNB and CAT (Das et al., 1999; Ding et al., 2013; Neta et al., 1988), the low con-
were the concentrations of nitrobenzene and atrazine. centration of SO$5 would render the actual reduction potential of
k1 and k2 were in the ranges of 3.2e4.7 109 M1 s1 and SO2  2
5 /HSO5 or SO5 /SO5 to a relative low value in the investi-
$

2.5e3 109 M1 s1 (Balci et al., 2009; Buxton et al., 1988; gated system. CuFe2O4 exhibited significant catalytic activity on
Gallard et al., 1998). And k2/k1 would be in the range of PMS while leached or free copper ions showed almost no cata-
0.5e1. However, the value of k0,AT/k0,NB, obtained from linear lytic activity on PMS as shown in Fig. 1 and the results reported
fit of atrazine and nitrobenzene degradation (Inset of Fig. 7), previously (Zhang et al., 2013). It indicated that the heteroge-
was 6.4. It was much larger than the value of k2/k1, indicating neous catalytic activity of CuFe2O4 on PMS might not only be
the contribution of SO$4 to atrazine degradation. Thus, both decided by the standard reduction potential of redox couple of

HO and SO$ 4 were the reactive species for atrazine degrada- cycled metal, but also influenced by the heterogeneous reaction
tion in PMS/CuFe2O4 system. system and the complexation of copper on the surface of
CuFe2O4 with PMS, which might make the reaction complicated.
3.3.2. Possible active sites Further investigations are needed to clarify the mechanisms on
The role of surface hydroxyl groups of catalyst was investi- the radical formation pathway and the cycle of metal ions in the
gated via the substitution by phosphate since phosphate had future studies.
stronger affinity to active sites (Zhang et al., 2008). As shown
in Fig. 6, phosphate showed significant inhibition on atrazine 3.4. Comparison of peroxides
degradation, which confirmed the role of surface hydroxyl
groups of CuFe2O4 in the catalysis of PMS. Besides, the The decontamination efficiencies by PDS/CuFe2O4 and H2O2/
maximal efficiency of atrazine degradation was obtained CuFe2O4 were investigated in comparison with PMS/CuFe2O4.
However, PDS/CuFe2O4 and H2O2/CuFe2O4 showed almost no
degradation of atrazine (Fig. 8). Previous study reported that
CuFe2O4 could decolorize synthetic dyes at H2O2 and CuFe2O4
doses of 100 mM and 1.25e25 g/L (Baldrian et al., 2006). The
catalytic activity of CuFe2O4 on H2O2 reported in the literature
might be due to the high doses of H2O2 and CuFe2O4, as well as
the long reaction time. The different reactivity of peroxides in
heterogeneous CuFe2O4 catalytic oxidation, was similar to the
result in homogeneous catalytic oxidation, in which PMS
showed significant reactivity with the majority of metals espe-
cially under neutral pH (Anipsitakis and Dionysiou, 2004). This
indicated that the asymmetric structure of PMS made it also
easy to be activated in the heterogeneous catalytic oxidation.

3.5. Degradation of atrazine under actual water


Fig. 7 e Degradation of nitrobenzene (NB) and atrazine (AT) background
by PMS/CuFe2O4. Conditions: [PMS] [ 1 mM,
[CuFe2O4] [ 0.1 g/L, [nitrobenzene] [ 6.4 mM, Water quality parameters of the effluents from DAF, BAC, and
[atrazine] [ 2.0 mM, initial pH [ 7.5 ± 0.2. SF were given in Tab. S1 with almost the same alkalinity and
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8 5437

the Cu on CuFe2O4 surface was the active sites to catalyze


PMS via the cycle of Cu2þ/Cuþ.
- Catalytic degradation of atrazine by PMS/CuFe2O4 was also
effective under the backgrounds of investigated actual
waters, which provide a reference for application.

Acknowledgments

This research was supported by the Funds for Creative


Research Groups of China (Grant No. 51121062), the Natural
Science Foundation of China (Grant No. 51178134,
No.51108111), the National Science & Technology
Fig. 8 e Atrazine degradation by peroxides/CuFe2O4 in Pillar Program of China (Grant No. 2012BAC05B02), and
Milli-Q water and actual water backgrounds. Conditions: Fundamental Research Funds for the Central Universities
[oxidants] [ 1.0 mM, [CuFe2O4] [ 0.1 g/L, (HEUCF).
[atrazine] [ 2.0 mM, [NaHCO3] [ 2 mM for Milli-Q water,
initial pH [ 7.5 ± 0.2.
Appendix A. Supplementary data

pH value but a little difference in DOC. EEM fluorescence Supplementary data related to this article can be found at
spectra were used to characterize the compound-class http://dx.doi.org/10.1016/j.watres.2013.06.023.
composition of dissolved organic matter in the effluents and
the fluorescence characteristics of the effluents were different
(SI Text S6 and Fig. S8), which might affect the abilities of NOM references
to complex with metals and scavenge radicals. Atrazine
degradation by PMS/CuFe2O4 was inhibited in actual water
compared with in Milli-Q water to some extent (Fig. 8), which Acero, J.L., Stemmler, K., Von Gunten, U., 2000. Degradation
would be due to the presence of NOM and inorganic salts. The kinetics of atrazine and its degradation products with ozone
and OH radicals: a predictive tool for drinking water
difference of the degradation efficiencies under the back-
treatment. Environmental Science & Technology 34 (4),
grounds of three kinds of effluents was not obvious, which
591e597.
indicated that the decontamination efficiency of PMS/CuFe2O4 Anipsitakis, G.P., Dionysiou, D.D., 2004. Radical generation
was hardly influenced by the fluorescence characteristics, but by the interaction of transition metals with common
might be affected mainly by other parameters such as DOC, oxidants. Environmental Science & Technology 38 (13),
pH value, and alkalinity. 3705e3712.
Balci, B., Oturan, N., Cherrier, R., Oturan, M.A., 2009. Degradation
of atrazine in aqueous medium by electrocatalytically
generated hydroxyl radicals. A kinetic and mechanistic study.
4. Conclusions Water Research 43 (7), 1924e1934.
Baldrian, P., Merhautová, V., Gabriel, J., Nerud, F., Stopka, P.,
- The magnetic porous CuFe2O4, synthesized by a solegel Hrubý, M., Bene s, M.J., 2006. Decolorization of synthetic dyes
by hydrogen peroxide with heterogeneous catalysis by mixed
method, was applied to the catalysis of peroxides. CuFe2O4
iron oxides. Applied Catalysis B: Environmental 66 (3e4),
exhibited a notable catalytic activity to PMS for atrazine 258e264.
degradation but almost no catalytic activity to PDS and Ball, R.E., Edwards, J.O., Haggett, M.L., Jones, P., 1967. A kinetic
H2O2. and isotopic study of the decomposition of
- Atrazine degradation by PMS/CuFe2O4 was enhanced with monoperoxyphthalic acid. Journal of the American Chemical
the increase of PMS and CuFe2O4 doses at respective range Society 89 (10), 2331e2333.
of 0.1e2.5 mM and 0.05e1 g/L. It showed a dependence on Bargar, J.R., Kubicki, J.D., Reitmeyer, R., Davis, J.A., 2005. ATR-FTIR
spectroscopic characterization of coexisting carbonate surface
initial pH and the maximal degradation efficiency was
complexes on hematite. Geochimica et Cosmochimica Acta 69
obtained at initial pH 9.5. Bicarbonate stimulated atrazine (6), 1527e1542.
degradation in the catalytic oxidation at the concentration Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 1988.
of 2 mM but inhibited at >4 mM. NOM inhibited the cata- Critical review of rate constants for reactions of hydrated
lytic degradation of atrazine at the range of 1.0e3.2 mg/L. electrons, hydrogen atoms and hydroxyl radicals ($OH/$O) in
- Both HO and SO$

aqueous solution. Journal of Physical and Chemical Reference
4 were identified in PMS/CuFe2O4 system
Data 17 (2), 513e886.
by competition reactions of atrazine and nitrobenzene, a
Chan, K.H., Chu, W., 2009. Degradation of atrazine by cobalt-
more reliable method than alcohol scavenging in hetero-
mediated activation of peroxymonosulfate: different cobalt
geneous catalytic oxidation. counteranions in homogenous process and cobalt oxide
- Surface hydroxyl groups of CuFe2O4 played an important catalysts in photolytic heterogeneous process. Water Research
role in the generation of radicals from PMS/CuFe2O4 and 43 (9), 2513e2521.
5438 w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 5 4 3 1 e5 4 3 8

Das, T.N., Huie, R.E., Neta, P., 1999. Reduction potentials of SO 3, Neta, P., Huie, R.E., Ross, A.B., 1988. Rate constants for reactions

SO$5 and S4 O6 radicals in aqueous solution. The Journal of of inorganic radicals in aqueous solution. Journal of Physical
Physical Chemistry A 103, 3581e3588. and Chemical Reference Data 17 (3), 1027e1284.
De Laat, J., Dao, Y.H., Hamdi El Najjar, N., Daou, C., 2011. Effect of Qi, F., Chu, W., Xu, B., 2013. Catalytic degradation of caffeine in
some parameters on the rate of the catalysed decomposition aqueous solutions by cobalt-MCM41 activation of
of hydrogen peroxide by iron(III)-nitrilotriacetate in water. peroxymonosulfate. Applied Catalysis B: Environmental
Water Research 45 (17), 5654e5664. 134e135 (0), 324e332.
De Laat, J., Gallard, H., Ancelin, S., Legube, B., 1999. Comparative Rani, S.K., Easwaramoorthy, D., Bilal, I.M., Palanichamy, M., 2009.
study of the oxidation of atrazine and acetone by H2O2/UV, Studies on Mn(II)-catalyzed oxidation of a-amino acids by
Fe(III)/UV, Fe(III)/H2O2/UV and Fe(II) or Fe(III)/H2O2. peroxomonosulphate in alkaline medium-deamination and
Chemosphere 39 (15), 2693e2706. decarboxylation: a kinetic approach. Applied Catalysis A:
Ding, Y., Zhu, L., Wang, N., Tang, H., 2013. Sulfate radicals General 369 (1e2), 1e7.
induced degradation of tetrabromobisphenol a with Ren, Y.M., Dong, Q., Feng, J., Ma, J., Wen, Q., Zhang, M.L., 2012.
nanoscaled magnetic CuFe2O4 as a heterogeneous catalyst of Magnetic porous ferrospinel NiFe2O4: a novel ozonation
peroxymonosulfate. Applied Catalysis B: Environmental 129, catalyst with strong catalytic property for degradation of di-n-
153e162. butyl phthalate and convenient separation from water.
Fang, G.D., Dionysiou, D.D., Al-Abed, S.R., Zhou, D.M., 2013a. Journal of Colloid and Interface Science 382, 90e96.
Superoxide radical driving the activation of persulfate Su, M.H., He, C., Sharma, V.K., Abou Asi, M., Xia, D., Li, X.Z.,
by magnetite nanoparticles: implications for the Deng, H.Q., Xiong, Y., 2012. Mesoporous zinc ferrite: synthesis,
degradation of pcbs. Applied Catalysis B: Environmental characterization, and photocatalytic activity with H2O2/visible
129, 325e332. light. Journal of Hazardous Materials 211, 95e103.
Fang, G., Gao, J., Dionysiou, D.D., Liu, C., Zhou, D., 2013b. Tang, W., Su, Y., Li, Q., Gao, S., Shang, J.K., 2013.
Activation of persulfate by quinones: free radical reactions Superparamagnetic magnesium ferrite nanoadsorbent for
and implication for the degradation of pcbs. Environmental effective arsenic (III, V) removal and easy magnetic
Science & Technology 47 (9), 4605e4611. separation. Water Research 47 (11), 3624e3634.
Furman, O.S., Teel, A.L., Watts, R.J., 2010. Mechanism of base Ventura, A., Jacquet, G., Camel, V., Bermond, A., 2002.
activation of persulfate. Environmental Science & Technology Comparison between the classical and the electrochemical
44 (16), 6423e6428. Fenton systems for atrazine degradation in acidic medium.
Güpta, Y.K., 1961. Iodometric determination of persulphate in Journal of Advanced Oxidation Technologies 5 (1), 113e120.
sulphuric acid solution. Fresenius’ Journal of Analytical Waldemer, R.H., Tratnyek, P.G., Johnson, R.L., Nurmi, J.T., 2007.
Chemistry 180 (4), 260e265. Oxidation of chlorinated ethenes by heat-activated persulfate:
Gallard, H., de Laat, J., Legube, B., 1998. Effect of pH on the kinetics and products. Environmental Science & Technology
oxidation rate of organic compounds by FeII/H2O2. 41 (3), 1010e1015.
Mechanisms and simulation (Influence du pH sur la vitesse Yan, J.C., Lei, M., Zhu, L.H., Anjum, M.N., Zou, J., Tang, H.Q., 2011.
d’oxydation de compose organiques par FeII/H2O2. Degradation of sulfamonomethoxine with Fe3O4 magnetic
Mecanismes reactionnels et modelisation). New Journal of nanoparticles as heterogeneous activator of persulfate.
Chemistry 22 (3), 263e268. Journal of Hazardous Materials 186 (2e3), 1398e1404.
Gilbert, B.C., Stell, J.K., 1990. Mechanisms of peroxide Yang, Q., Choi, H., Al-Abed, S.R., Dionysiou, D.D., 2009.
decomposition: an electron paramagnetic resonance study of Ironecobalt mixed oxide nanocatalysts: heterogeneous
the reaction of the peroxomonosulphate anion ðHOOSO 3Þ peroxymonosulfate activation, cobalt leaching, and
with CuI. A marked contrast in behaviour with that of TiIII and ferromagnetic properties for environmental applications.
FeII. Journal of the Chemical Society, Faraday Transactions 86 Applied Catalysis B: Environmental 88 (3e4), 462e469.
(19), 3261e3266. Zhang, T., Li, C., Ma, J., Tian, H., Qiang, Z., 2008. Surface hydroxyl
Guan, Y.H., Ma, J., Li, X.C., Fang, J.Y., Chen, L.W., 2011. Influence of groups of synthetic a-FeOOH in promoting oh generation from
pH on the formation of sulfate and hydroxyl radicals in the aqueous ozone: property and activity relationship. Applied
UV/peroxymonosulfate system. Environmental Science & Catalysis B: Environmental 82 (1e2), 131e137.
Technology 45 (21), 9308e9314. Zhang, T., Li, W., Croué, J.P., 2012. A non-acid-assisted and
King, D.W., 1998. Role of carbonate speciation on the oxidation non-hydroxyl-radical-related catalytic ozonation with ceria
rate of Fe(II) in aquatic systems. Environmental Science & supported copper oxide in efficient oxalate degradation in
Technology 32 (19), 2997e3003. water. Applied Catalysis B: Environmental 121-122, 88e94.
Lawrance, G.A., Ward, C.B., 1985. Kinetics of oxidation of Zhang, T., Zhu, H., Croué, J.P., 2013. Production of sulfate radical
manganese(II) by peroxomonosulfuric acid in aqueous acidic from peroxymonosulfate induced by a magnetically separable
solution. Transition Metal Chemistry 10 (7), 258e261. CuFe2O4 spinel in water: efficiency, stability, and mechanism.
Ma, J., Graham, N.J.D., 2000. Degradation of atrazine by Environmental Science & Technology 47 (6), 2784e2791.
manganese-catalysed ozonationdinfluence of radical Zhao, L., Sun, Z., Ma, J., Liu, H., 2010. Influencing mechanism of
scavengers. Water Research 34 (15), 3822e3828. bicarbonate on the catalytic ozonation of nitrobenzene in
Ma, J., Sui, M.H., Zhang, T., Guan, C.Y., 2005. Effect of pH on aqueous solution by ceramic honeycomb supported
MnOx/GAC catalyzed ozonation for degradation of manganese. Journal of Molecular Catalysis A: Chemical 322
nitrobenzene. Water Research 39 (5), 779e786. (1e2), 26e32.
Morgan, M.S., Van Trieste, P.F., Garlick, S.M., Mahon, M.J., Zhu, B.Z., Zhao, H.T., Kalyanaraman, B., Frei, B., 2002. Metal-
Smith, A.L., 1988. Ultraviolet molar absorptivities of aqueous independent production of hydroxyl radicals by halogenated
hydrogen peroxide and hydroperoxyl ion. Analytica Chimica quinones and hydrogen peroxide: an ESR spin trapping study.
Acta 215, 325e329. Free Radical Biology & Medicine 32 (5), 465e473.