Electrochemical oxidation of paraquat in neutral medium

Article abstract of DOI:10.1016/j.electacta.2015.07.099

Abstract Steel, Pt and pelleted Co₂FeO₄ electrodes were used for the electrochemical oxidation of paraquat in aqueous solutions at room temperature. The oxide electrodes were characterized by cyclic voltammetry. Paraquat electrochemical oxidation was carried out by electrolysis at constant current and monitored by UV-vis absorbance measurements. Different anode/cathode pairs were used. After 1.5 h of electrolysis the highest removal (≈79%), was obtained with the electrode pair Pt/steel followed by Co₂FeO₄/Pt (≈55%), corresponding to the paraquat oxidation by a conversion mechanism. Removals of ≈64% were obtained with Co₂FeO₄ / Co₂FeO₄ after 3 h of electrolysis. Mass spectrometry analysis indicates that the main intermediate oxidation products were monopyridone and dipyridone derivatives.

Full text of DOI:10.1016/j.electacta.2015.07.099

Electrochimica Acta 176 (2015) 1010–1018
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrochemical oxidation of paraquat in neutral medium
M.A.M. Cartaxoa,b, C.M. Borgesb, M.I.S. Pereirac, M.H. Mendonçab,
*
a
Unidade Departamental de Engenharias, Escola Superior de Tecnologia, Instituto Politécnico de Tomar, Quinta do Contador, Estrada da Serra, 2300-313
Tomar, Portugal
b
Centro de Química e Bioquímica, Faculdade de Ciências, Universidade de Lisboa, 1749-019 Lisboa, Portugal
Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade de Lisboa, 1749-019 Lisboa, Portugal
c
A R T I C L E I N F O
A B S T R A C T
Article history:
Steel, Pt and pelleted Co2FeO4 electrodes were used for the electrochemical oxidation of paraquat in
aqueous solutions at room temperature. The oxide electrodes were characterized by cyclic voltammetry.
Paraquat electrochemical oxidation was carried out by electrolysis at constant current and monitored by
UV–vis absorbance measurements. Different anode/cathode pairs were used. After 1.5 h of electrolysis
the highest removal (ꢀ79%), was obtained with the electrode pair Pt/steel followed by Co2FeO4/Pt
(ꢀ55%), corresponding to the paraquat oxidation by a conversion mechanism. Removals of ꢀ64% were
obtained with Co2FeO4 / Co2FeO4 after 3 h of electrolysis. Mass spectrometry analysis indicates that the
main intermediate oxidation products were monopyridone and dipyridone derivatives.
ã 2015 Elsevier Ltd. All rights reserved.
Received 27 February 2015
Received in revised form 3 June 2015
Accepted 18 July 2015
Available online 26 July 2015
Keywords:
paraquat
anodic oxidation
Co2FeO4 electrode
electrochemical conversion
mass spectrometry
1. Introduction
photochemical and photocatalytic systems. Several authors have
comprehensively reviewed the development in the field [3–5].
Paraquat is one of the most toxic and widely used herbicides in
the world. Although it is prohibited by European Union, it is still
used in developing countries in other continents. As a non-
selective contact herbicide, paraquat has been available to farmers
for over 40 years and it is present as an environmental pollutant
both in soil and in surface waters. It represents a threat to human
health because even at very low doses, this herbicide can pass
some treatment steps in a water treatment plant and reach the
water distribution systems [1].
Electrochemical oxidation methods, which present environ-
mental compatibility, are promising alternatives for the destruc-
tion of toxic organic waste [6]. They can be used as a pre-treatment
technology in detoxifying ahead of bio-treatment, rather than
mineralizing them completely. Due to its effectiveness and ease of
operation, the application of electrochemical oxidation to treat
effluents containing pesticides was studied with promising results
[7]. The overall performance of electrochemical processes is
determined by the complex interplay of different parameters,
namely, electrode material and electrolysis conditions [8,9]. In the
last decade there was a successful development of low cost mixed
oxide electrodes with high stability and relevant activity to
pollution abatement. During the past years our efforts have been
devoted to study mixed valence oxides with spinel structures, with
focus on the search of novel electrode materials, for oxygen
reactions [10–12]. Recent work has used successfully Fe-Co3O4 thin
film electrodes (Fe = 0%, 5% and 10%), for the electrooxidation of
phenol in alkaline medium. An enhancement of the phenol
removal was observed with the presence of Fe in the oxide. For the
electrodes with 10% of Fe, around 30% of the initial phenol has been
removed at ca. 3 h and the complete removal was achieved after
54 h of electrolysis [13]. The phenol oxidation was accomplished by
a conversion mechanism giving rise to different species in solution
and as final by-products organic acids. The use of iron-containing
materials was also addressed through the so called heterogeneous
Fenton like processes [14–16].
Paraquat [(C12H14N2)2+ 2Clꢂ] is a bipyridinium herbicide very
ꢁ
soluble in water (625 mg/L at 25 ꢃC), and its IUPAC name is 1, 10-
dimethyl-4, 40- bipyridinium dichloride, also known as methyl
viologen dichloride hydrate (MV). Fig. 1 presents the chemical
structure of the 1, 1-dimethyl-4, 40- bipyridinium dication (MV2+).
The MV2+/MV+ redox couple has the lowest reversible standard
redox potential known for an organic species in aqueous solution:
u
E = ꢂ0.689 V vs. SCE [2].
Advanced methods are in demand for effective treatment of
herbicide-polluted ground and surface waters. Relevant results
were obtained with advanced oxidation processes (AOPs) in which
OHꢄ radicals are produced by chemical, electrochemical,
* Corresponding author.
E-mail addresses: mamcartaxo@ipt.pt (M.A.M. Cartaxo),
cmborges@ciencias.ulisboa.pt (C.M. Borges), mipereira@ciencias.ulisboa.pt
(M.I.S. Pereira), mhmendonca@ciencias.ulisboa.pt (M.H. Mendonça).
http://dx.doi.org/10.1016/j.electacta.2015.07.099
0013-4686/ã 2015 Elsevier Ltd. All rights reserved.

M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
1011
scanning rate of 0.02ꢃ sꢂ1, using Cu K
the incident radiation.
Morphological characterization was performed by scanning
electron microscopy (SEM) using a JEOL (JSM-35C) unit.
a
radiation (
l
= 1.5406 Å) as
+
+
H C
N
N
CH
3
3
Fig. 1. Chemical structure of paraquat dication.
2.3. Electrochemical studies
When using active mixed oxides, oxidation reactions occur
through a higher oxidation state of the metal oxide surface sites,
resulting in a partial oxidation of the organic compounds, due to
the accumulation of intermediates, which are quite stable against
further attack at these electrodes [8,17]. Accordingly, the metal
cations in the oxide lattice may reach higher oxidation states under
anodic polarization and a stabilization of adsorbed OHꢄ radicals
takes place, favoring the oxygen evolution at the expense of the
electrochemical incineration reaction. As the best of our knowl-
edge, only one study was published on the anodic oxidation of
paraquat aqueous solutions. Pt cathodes and carbon felt anodes
were used [18]. It was proposed the formation of demethylation
products of the parent compound (4, 40-bipyridine and monop-
yridone) and hydroxylation or oxidative ring cleavage products
(4-carboxy-1-methyl-pyridinium ion, 4-picolinic acid and hydrox-
yl-4-picolinicacid), that subsequently undertake minor degrada-
tion, leading to incomplete mineralization of the herbicide.
The present work focuses on the electrochemical partial
oxidation of paraquat in aqueous solution, using Pt, steel and a
novel active spinel oxide Co2FeO4 as electrode materials. Cyclic
voltammetric studies were undertaken to characterize the
Co2FeO4/MV system. Subsequently several electrolysis operating
factors were assessed. Attention was paid to simultaneous anodic
and cathodic reactions. The paraquat removal and its oxidation
intermediates were monitored by UV–vis absorbance measure-
ments and identified by mass spectrometry analysis (MS).
All voltammetric experiments were undertaken in a conven-
tional three-compartment glass cell at room temperature. Before
experiments, the solution was saturated with either air or N2 or O2,
using vigorous bubbling for ꢀ 30 min. Platinum foil was used as
counter electrode and Hg/Hg2SO4/K2SO4 sat. (+0.656 V vs. SHE) as
reference (M.S.E.).
Bulk electrolysis preliminary studies were performed in a
three-compartment glass cell, under various constant potentials
and/or current intensities, from which the best conditions were
selected. The subsequent bulk electrolysis were then accomplished
in a one-compartment glass cell at constant current (0.1 A) and
room temperature, under magnetic stirring with several duration
times. The cell capacity was 100 mL. Co2FeO4 and Pt were used as
anodes, Co2FeO4, Pt and steel as cathodes and M.S.E as reference
electrode. The solutions were not deaerated.
Paraquat dichloride 10ꢂ4 M (Sigma1) in 0.05 M K2SO4 or KCl
(Merck pro analysis) solutions, with pH ꢀ 7, were prepared using
Millipore Milli-Q ultrapure water (18 M
Vcm).
The electrochemical measurements were carried out either
using a low noise operational amplifier potentiostat programmed
by a PPR1Hi-TEK Instruments wave generator, and a Kipp & Zonen
Pro-1 recorder or
consisting of DEA-I Digital Electrochemical Analyzer, and
a
VoltaLabTM 32 System (RadiometerTM),
a
comprising the IMT102 Electrochemical Interface and the
DEA332 33 V/2A potentiostat with the VoltaMaster 2 software.
The present study represents a continuing attempt in the search
for more active electrode materials and we believe that the results
can provide a direction guide for an exploratory study of other
spinel oxides of interest for environmental applications.
2.4. Analysis methods
The paraquat oxidation was followed by UV–vis absorbance
measurements. At regular time intervals, small volume samples
were withdrawn from the electrolytic cell and analyzed to follow
the paraquat removal. Semi– micro quartz cells with 1.4 mL of
capacity and 10 mm of optical length were used.
2. Experimental
2.1. Electrode preparation
The UV–vis spectra of electrolyzed and non-electrolyzed
solutions were acquired with a Jasco V 560 spectrophotometer.
In order to identify paraquat oxidation products, MS experi-
ments were carried out on a LCQ Duo quadrupole ion trap mass
spectrometer from ThermoFinnigan (San Jose, CA, USA) equipped
with an electrospray ionization source (ESI). A positive potential of
+4.5 kV was applied to the electrospray source (electrospray
needle) working therefore in positive ion mode. A stream of
nitrogen, carrier gas and a counter-flow gas, also nitrogen, 20-
40 psi were applied, depending on the sample behavior under the
ESI-MS operating conditions. The metallic capillary, located in the
system interface, was maintained at 200 ꢃC. The pressure in the
region capillary/skimmer of the mass spectrometer recorded a
value of 0.92 Torr, while the base pressure in the region of the
analyzer and mass detector recorded a value of 1.12ꢅ10ꢂ5 Torr. The
samples of analyte solutions were directly introduced into the
Co2FeO4 powder samples were obtained by a previously
described coprecipitation method, from the addition of aqueous
solutions of FeCl3
with molar ratio Co/Fe = 2/1, added to a KOH boiling solution under
vigorous stirring. black precipitate was formed and was
ꢁ
6H2O and CoCl2 6H2O, both Merck pro-analysis,
ꢁ
A
subsequently filtered and dried on a sand bath at approximately
473 K [19,20]. The dry product was finally heated in a muffle
furnace at 1173 K for 6 h.
The electrodes were prepared as parallelepiped pellets with
dimensions 1.5 cm ꢅ 0.77 cm ꢅ 0.19 cm by pressing the powders
onto an inserted platinum mesh, which served as current collector
and subsequently heated at 1173 K for 6 h. The electrical contact
was made by welding the Pt mesh to a silver wire. The samples
were then mounted in a glass tube with epoxy resin (Araldite1) so
that the electrolyte could only make contact with the oxide.
Pt and steel were used as foils, with an immersed geometric
area similar to the oxide electrode.
electrospray source at a flow rate of 5 mL/min. Full scan mass
spectra were recorded in the range mass/charge (m/z) ratio
80–285. The mass spectrometer was maintained in operation,
registering three microscans with a maximum ion injection time of
50 ms (reference values), whereas the mass spectra obtained were
based on a set of microscans equivalent to an operation time of the
mass spectrometer of 1 minute. This procedure was adapted from
the previously used on paraquat photodegradation studies
performed in the same equipment [21].
2.2. Structural and morphological characterization
The pellets were characterized by X-ray powder diffraction,
using a Philips PW 1730 diffractometer with automatic data
acquisition (APD Phillips v.3.6B) and the ICDD files for X-ray data
indexation. All scans were recorded between 15ꢃ and 80ꢃ
2u at a

1012
M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
3. Results and discussion
3.1. Electrode characterization
3.1.2. Voltametric studies
The voltammetric behavior of the Co2FeO4/0.05 M K2SO4 system
was studied between 0 and ꢂ1.2 V vs. M.S.E. in air, N2 and O2
saturated solutions (Fig. 3). The negative potential limit has been
established, according to the literature, in order to detect the MV2
+/MV+ꢄ pair [2]. The recorded voltammograms (CV) present broad
peaks, indicating a large heterogeneity in the surface sites and
superposition of the redox processes regarding the metal oxide
transitions [22]. When the solution is saturated with air or N2 the
CVs are quite similar, presenting a progressive increase of the
cathodic intensity current between ꢂ0.7 and ꢂ1.0 V. On the anodic
scan the development of peak A1 followed by hump A2 are
observed between ꢂ0.9 and ꢂ0.6 V. Afterwards the current
approaches zero and subsequently increases slowly up to the
positive potential limit. In O2-saturated solution, the voltammo-
gram exhibits a different profile: on the cathodic sweep, peak C4
goes with hump C3, between ꢂ0.5 and ꢂ0.7 V, followed by a
current increase until the negative potential limit is reached. On
the anodic scan two new peaks (A3 and A4) turn up at potentials
greater than ꢂ0.5 V. These peaks are the anodic counterpart of
peaks C3 and C4. Moreover an increase of current intensity is
observed between ꢂ0.9 and ꢂ1.1 V. The voltammetric profiles
contrast with those already reported by us for the same oxide
electrode in alkaline medium. In that case the redox processes take
3.1.1. Structural and morphological characterization
The Co2FeO4 electrodes, fresh and after electrochemical tests,
were characterized by X-ray powder diffraction. X-ray patterns
show the formation of a single spinel phase indexed by comparison
with the Powder Diffraction Files of Co3O4 (ICDD 9-418), CoFe2O4
(ICDD 3-0864) and Fe3O4 (ICDD 19-629). Comparing the diffracto-
grams for fresh and used electrodes no structural changes were
observed in the oxide after use, as Fig. 2 shows.
SEM images show a fairly homogeneous surface with small
grain size and some porosity, which was expected considering the
preparation method used. After the electrochemical tests no
significant changes were observed on the electrodes surface
morphology (Inset Fig. 2).
The electrodes roughness factor was estimated from double
layer pseudo-capacitance measurements (SD-1). A roughness
factor of ꢀ 2 was obtained, indicative of fine and compact
powders, which is in accordance with SEM results.
Fig. 2. X-ray powder diffractograms of (a) oxide electrode after being used in the electrochemical studies (b) Co2FeO4 powder. * peaks due to the sample holder.

M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
1013
discarded, considering that for reaction (2) at pH = 7, Eeq = ꢂ1.070 V
vs. M.S.E.
2H+ + 2eꢂ ! H2
(2)
On the other hand the cathodic current between ꢂ1.1 and
ꢂ0.9 V is due to the oxygen reduction.
CVs recorded in the presence of MV show a similar profile
although with lower current densities, owing to the MV adsorption
on the oxide surface, which slows down the redox processes. In
order to confirm the presence of the MV2+/MV+ꢄ redox pair, the CV
presented in Fig. 4, was recorded after polarization at ꢂ0.4 V vs. M.
S.E for 1 h. Clearly the reduction and oxidation peaks are rather
broad and asymmetrical, as expected for adsorbed species
undergoing facile electron exchange with substrate [26]. The
development of the (A/C) pair of peaks, centered at about ꢂ1.0 V vs.
MMS, is attributed to MV2+/MV+ꢄ [2]. The higher cathodic peak
current intensity is due to the reaction between MV+ꢄ radical cation
with oxygen present in solution, producing hydrogen peroxide,
according to equations (3) and (4):
Fig. 3. Co2FeO4 voltammograms in 0.05 M K2SO4 in solution under air, nitrogen and
to ꢂ1.250 V vs. M.S.E. at 100 mV sꢂ1
. Electrode
MV2+ + eꢂ ! MV+ꢄ (3)
oxygen atmospheres, from
0
geometric area: 3 cm2.
2MV+ꢄ + O2 + 2H+ ! 2MV2+ + H2O2 (4)
place on two well-defined regions, separated by a large potential
window (400 mV) where no Faradaic currents flow [12,20]. Using
voltammetric and literature data [23–25] it was possible to find out
the redox reactions occurring at the electrode surface by a method
that has been used previously [12,20]. Table 1 presents values of
The reduction is coupled with a subsequent homogeneous
chemical reaction and consequently the cathodic current of the
first reaction is increased by the second reaction, while the anodic
current is decreased by the loss of MV+ꢄ [27].
The first reduction of the colorless viologen dication is highly
reversible and leads to the formation of the intensely colored
radical cation. However, the blue color was not detected in this
assay. In order to ascertain the color absence, tests were performed
under the same conditions replacing the Co2FeO4 oxide by a gold
electrode. In this case a blue color was observed. For a direct
comparison, the inset, in Fig. 4 presents the CV obtained with a
gold electrode where the characteristic paraquat pair of peaks,
appears in the same potential zone, confirming the assignment of
the A/C peaks. The absence of color on the experiments with the
oxide electrode indicates that the reduction occurs inside the oxide
pores. This assumption is confirmed by a family of CVs recorded in
regular time intervals of 10 min up to 2 h (not shown). The peak
current associated with the MV2+/ MV+ꢄ redox process increases
initially and then tend towards constant value owing to the slow
diffusion of electroactive species into the electrode pores. This
u
peak potential, formal potential (E ) taken as the midpoint
between the anodic and cathodic peaks and the calculated
thermodynamic equilibrium potential (Eeq) for the redox couples
of the oxide component species. Literature data on Fe3O4 and
Co3O4 oxide electrodes were used for comparison. Peaks A1 and A2
occur in the potential region of the oxidation of Fe (II) to Fe (III)
surface sites and are assigned to the oxidation of Fe(OH)2 to Fe3O4
and Fe3O4 to Fe2O3 respectively [24,25]. The CVs obtained either in
air or N2-saturated solutions indicate passivation of the electrodes
surface, probably due to the formation of Fe2O3 film which
presents low conductivity [23,25]. These results explain the low
anodic current intensity in the less negative potentials as well as
the absence of peaks on the cathodic scan [24]. On the other hand,
the shape of the voltammogram recorded under O2-saturated
solutions suggests that the presence of high amount of oxygen
dissolved in solution inhibits the electrode passivation. This fact
could be due to the occurrence of reaction (1) that competes with
the oxidation of Fe3O4 to Fe2O3, preventing the formation of Fe2O3
3
4
9
2
3Fe3O4 þ O2 þ H2O ! 9FeOOH
(1)
The appearance of pairs of peaks A3/C3 and A4/C4 were assigned
to the formation of cobalt oxides/hydroxides, due to the oxidation
of weakly hydrated Co3+ species [24]. The negative current
observed at the switching potential is due to oxide reduction
although the occurrence of hydrogen evolution cannot be
Table 1
u
Peak potential (Ep), formal potential (E ) and possible solid-state surface redox
processes associated to the voltammetric peaks.
Peak
A1
Ep/V
E/V
Solid-state surface redox process
Eeq/V
ꢂ1.10ꢆ0.03
–
2Fe(OH)2 $ Fe2O3 + H2O + 2H+ + 2eꢂ
3Fe(OH)2 $ Fe3O4 + 2H2O + 2H+ + 2eꢂ
Fe(OH)2 + H2O $ Fe(OH)3 + H+ + eꢂ
2Fe3O4 + H2O $ 3Fe2O3 + 2H+ + 2eꢂ
ꢂ1.127
ꢂ1.267
ꢂ0.799
ꢂ0.850
A2
ꢂ0.81ꢆ0.08
–
Fig. 4. Voltammogram for
a
Co2FeO4 electrode in 10ꢂ4 M MV + 0.05 M K2SO4,
recorded at 100 mV sꢂ1 after polarization at ꢂ0.400 V vs. M.S.E. for 1 h. Inset:
A3/C3
A4/C4
–
–
ꢂ0.53 3Co2O3 + H2O + 2eꢂ $ 2Co3O4 + 2 OHꢂ ꢂ0.565
ꢂ0.31 3CoO + H2O $ Co3O4 + 2H+ + 2eꢂ
ꢂ0.293
voltammogram recorded under the same conditions with
Electrodes geometric area: 3 cm2.
a gold electrode.

1014
M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
result shows that the interior of the porous electrodes makes an
important contribution to the process.
A more efficient removal was found by using a one-compart-
ment cell. With these experimental conditions the paraquat
oxidation could profit from the oxidative ability of both anodic
and cathodic reactions [31]. In view of that, all subsequent
electrolysis were carried out using a one-compartment cell with
magnetic stirring. The MV concentration was also increased to
10ꢂ4 M.
3.2. Paraquat electrochemical oxidation
3.2.1. Preliminary tests
Preliminary bulk electrolysis were performed to evaluate the
influence of supporting electrolyte and type of cell. Electrolysis of
10ꢂ5 M MV + 0.05 M K2SO4 aqueous solutions were done in a three
compartments glass cell using as anode materials Co2FeO4 or Pt
and an applied constant current I = 1.5 A. The paraquat removal was
3.2.2. UV-vis absorption spectroscopy
The effect of the electrode material (Co2FeO4, Pt and steel) on
the MV removal from aqueous solution was studied using Co2FeO4
as anode. Fig. 5a presents UV–vis spectra of 10ꢂ4 M MV + 0.05 M KCl
electrolyzed solutions during 1.5 and 3 h at 0.1 A.
monitored by UV–vis spectroscopy using as reference
l
ꢀ 257 nm,
the typical paraquat maximum absorption peak [28]. The electro-
lyzed solutions spectra did not show any significant differences to
the initial spectrum. This could be due either to high ohmic drop or
formation of an adherent film on the anode surface, which poisons
the electrode surface [29]. The supporting electrolyte was then
replaced by KCl in order to avoid film formation and to check the
possibility of electrogenerated chloride/hypochlorite ions pair
acting as mediator on the MV oxidation. Accordingly, at the anode
the chloride ion is oxidized to hypochlorite ion, which could react
with the MV and subsequently reduced back to chloride ion. On the
other hand, at the cathode the dissolved oxygen is reduced, with
the formation of hydrogen peroxide, which could also oxidize the
MV [30]. In accordance an increase on the removal rate was
expected. However, no significant changes were observed.
Spectra show similar variation with the maximum absorbance
at
l
ꢀ 257 nm characteristic of paraquat in aqueous solution. A
clear decrease in the absorbance maximum is observed depending
on the cathode used, indicating that the cathodic processes
contribute for the paraquat disappearance. Probably at the cathode
hydrogen peroxide is electrogenerated, by the reduction of O2
produced at the anode, generating hydroxyl radicals. Considering
the assays with the duration of 1.5 h, the Co2FeO4/steel pair
exhibits the best performance (ꢀ 30%) followed by Co2FeO4/
Co2FeO4 (ꢀ 20%). When the electrolysis was performed during 3 h
and using Co2FeO4 as both anode and cathode, the electrolyzed
solution spectrum shows a reduction of ꢀ 65% on the absorbance
maximum, more than twice the obtained after 1.5 h, indicating that
the paraquat removal rate increases with the progress of the
Fig. 5. UV–vis spectra of 10ꢂ4 M MV + 0.05 M K2SO4 solution after 0, 1.5 and 3 h of electrolysis at 0.1 A, undertaken in a one-compartment cell, with magnetic stirring, using as
anodes Co2FeO4 (a) and Pt (b) and different cathode materials. Relative absorbance for the different anodes/cathodes (c). * t = 3 h.

M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
1015
electrolysis. The paraquat peak is still visible after 3 h, accompanied
by a hump for >270 nm, already noticeable for the spectrum
A comparison between our results and selected studies on the
paraquat degradation is presented in SD-2. The results obtained in
this work are similar to those reported for paraquat anodic
oxidation [18].
The above findings allow concluding that the conversion of
paraquat on monopyridone, dipyridone and its derivatives takes
place for all the tested electrodes. Further oxidation of the
generated intermediates was not achieved after 3 h of electrolysis.
In order to confirm this proposal, MS analysis was undertaken.
l
recorded after 1.5 h. An increase of absorbance occurs between
220 and 240 nm for the spectra acquired at t = 1.5 h followed by a
decrease, probably due to formation of intermediate species,
which is in accordance with other authors [28,32]. The comparison
of the bands for the paraquat electrolyzed solutions and those for
possible by-products such as paraquat monopyridone: 225, 257,
and 345 nm [33] or 222, 260, 347 nm [34], and paraquat
dipyridone: 224 and 320 nm [33], indicates that these compounds
are the main by-products of paraquat in neutral medium. This
assumption is in accordance with published data [18,21,28,32].
Parallel experiments were carried out using Pt as anode. Fig. 5b
presents UV–vis spectra of electrolyzed solutions, during 1.5 h. The
spectra are symmetrical and show, as expected, the influence of the
cathode material. The most striking feature is the shift of the
3.2.3. MS analysis
Fig. 6a presents MS with electrospray ionization for non-
electrolyzed and electrolyzed solutions, using Co2FeO4 as both
anode and cathode.
The paraquat MS (t = 0 h) shows two peaks at m/z = 171 and
m/z = 186 with a relative intensity of ꢀ 100% and 60% respectively.
absorption peak to higher values of
l, in the region of the hump
The peak at m/z = 171, corresponds to an ion fragment formed from
the intact paraquat ion with loss of the methyl radical [MV+ꢄ-CH3
]
ꢄ
observed when Co2FeO4 is used as anode. This indicates the
formation of a new product, associated with the decrease on the
amount of paraquat in solution, which reaches the maximum of
absorbance when Co2FeO4 is used as cathode. According to other
studies [21], the formation of paraquat iron complexes formed by
ions leached from the oxide electrode into the solution bulk is
possible. This assumption is supported by atomic absorption
spectroscopic data that indicates the presence of Fe (2.64 ppm) and
Co (2.33 ppm) in the electrolyzed solutions. After 1.5 h of
electrolysis, the paraquat conversion is ꢀ 79% for Pt/steel and
ꢀ55% for the other electrodes (Fig. 5c).
and the m/z = 186 to the intact paraquat radical ion (MV+ꢄ). It is
also noted a low intensity peak at m/z = 113 which, as explained
ahead, corresponds to a fragment ion resulting from paraquat.
The appearance of new peaks (m/z = 93, 100, 197, 202 and 216)
on the MS obtained for the electrolyzed paraquat solutions during
1.5 and 3 h indicates the formation of new ionic species. Table 2
presents an attempt to identify the ions corresponding to the most
significant peaks observed on the MS of the electrolyzed solutions.
The peaks at m/z = 93 and 186, both due to paraquat correspond to
the (MV2+) dication and the (MV+ꢄ) radical cation, respectively. To
note that the peak at m/z = 93 is absent on the paraquat spectrum,
+
Fig. 6. Mass spectra with electrospray ionization of 10ꢂ4 M MV + 0.05 M K2SO4 solution after 0, 1.5 and 3 h of electrolysis at 0.1 A, undertaken in a one-compartment cell, with
magnetic stirring, using Co2FeO4 as anode and cathode (a) and variation of the relative abundance of the peaks observed in the MS with electrolysis time (b).

1016
M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
Table 2
List of possible intermediates and reaction products of paraquat electrodegradation based on MS data.
Peak m/z
93
Species
100
113
171
186
197
202
216

M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
1017
probably due to resonance stabilization of the cation radical. The
[3] E. Brillas, I. Sirés, M.A. Oturan, Electro-fenton process and related
electrochemical technologies based on Fenton’s Reaction Chemistry, Chem.
Rev. 109 (2009) 6570–6631.
[4] A.M. Polcaro, M. Mascia, S. Palmas, A. Vacca, Case studies in the
electrochemical treatment of wastewater containing organic pollutants
using BDD, in: C. Comninellis, G. Chen (Eds.), Electrochemistry for the
Environment, Springer, New York, 2010, pp. 205–227.
peak at m/z = 113 with low intensity, corresponds to [MV+ꢄ
-
+
CH2NCH3-CH3NHꢄ] ions.
Furthermore the peaks at m/z = 100 and 202 are assigned to
ionic species associated to monopyridone, while the peaks at
m/z = 197 and 216 correspond to ionic species related with
dipyridone [21].
The variation of the relative intensity of each peak as a function
of electrolysis time is presented in Fig. 6 b. The peak at m/z = 93,
absent in the paraquat solution appears after 1.5 h of electrolysis
with considerable relative intensity ꢀ 30%, decreasing significantly
after 3 h to ꢀ 5%, which suggests that MV2+ species (m/z = 93) is an
intermediate in the degradation process.
Regarding the peak m/z = 100, appearing after 1.5 h of electrol-
ysis (ꢀ 38%) and reaching ꢀ 100% after 3 h, indicates the formation
of a new product, probably monopyridone or a derivative (Table 2).
The appearance of this peak is accompanied by the decrease of the
peak with m/z = 171, corresponding to an ion fragment obtained
from the MV+ꢄ ion, by loss of a methyl radical. The peaks at
m/z = 197 and 216 follow a behavior similar to peaks with
m/z = 100 and 202, therefore, these species are consistent with
the formation of intermediates in the electrochemical oxidation
process. It is important to note that, according to the International
Programme on Chemical Safety (IPCS) (1991) [35], the resulting
oxidation species are less toxic than paraquat.
[5] G. Chen, Electrochemical technologies in wastewater treatment, Sep. Purif.
Technol. 38 (2004) 11–41.
[6] C.A. Martínez-Huitle, E. Brillas, Electrochemical alternatives for drinking water
disinfection, Angew. Chem. Int. Ed. 47 (2008) 1998–2005.
[7] C.A. Martínez-Huitle, S. Ferro, Chemical Society Reviews, Electrochemical
oxidation of organic pollutants for the wastewater treatment - direct and
indirect processes, Chem. Soc. Rev. 35 (2006) 1324–1340.
[8] M. Panizza, Importance of electrode material in the electrochemical treatment
of wastewater containing organic pollutants, in: C. Comninellis, G. Chen (Eds.),
Electrochemistry for the Environment, Springer, New York, 2010, pp. 25–54.
[9] S. Rao, Metal oxide-coated anodes in wastewater treatment, Environ. Sci.
Pollut. Res. 21 (2014) 3197–3217.
[10] A.C. Tavares, M.I. da Silva Pereira, M.H. Mendonça, M.R. Nunes, F.M. Costa, C.M.
Sá, XPS and voltammetric studies on Ni1-xCuxCo2O4 spinel oxide electrodes, J.
Electroanal. Chem. 449 (1998) 91–100.
[11] M. Isabel Godinho, M. Alice Catarino, M.I. da Silva Pereira, M.H. Mendonça, F.M.
Costa, Effect of partial replacement of Fe by Ni and/or Mn on the
electrocatalytic activity for oxygen evolution of the CoFe2O4 spinel oxide
electrode, Electrochim. Acta 47 (2002) 4307–4314.
[12] E. Laouini, Y. Berghoute, J. Douch, M.H. Mendonça, M. Hamdani, M.I.S. Pereira,
Electrochemical behaviour of FexCo3-xO4 (with x = 0, 1, 2 and 3) oxides thin
film electrodes in alkaline medium, J. Appl. Electrochem. 39 (2009)
2469–2479.
[13] M.A.M. Cartaxo, K. Ablad, J. Douch, Y. Berghoute, M. Hamdani, M.H. Mendonça,
M.I.S. Pereira, Phenol electrooxidation on Fe-Co3O4 thin film electrodes in
alkaline medium, Chemosphere 86 (2012) 341–347.
[14] M.A. Herney-Ramirez, Heterogeneous photo-Fenton oxidation with pillared
clay-based catalysts for wastewater treatment: a review, Appl. Catal. B 98
(2010) 10–26.
4. Conclusions
[15] C. Oliveira, M.S.F. Santos, F.J. Maldonado-Hódar, G. Schaule, A. Alves, L.M.
Madeira, Use of pipe deposits from water networks as novel catalysts in
paraquat peroxidation, Chem. Eng. J. 210 (2012) 339–349.
[16] C. Oliveira, K. Gruskevica, T. Juhna, K. Tihomirova, A. Alves, L.M. Madeira,
Removal of paraquat pesticide with Fenton reaction in a pilot scale water
system, Drink, Water Eng. Sci. 7 (2014) 11–21.
[17] Ch. Comninellis, Electrocatalysis in the electrochemical conversion/
combustion of organic pollutants for wastewater treatment, Electrochim.
Acta 39 (1994) 1857–1862.
[18] A. Dhaouadi, N. Adhoum, Degradation of paraquat herbicide by
electrochemical advanced oxidation methods, J. Electroanal. Chem. 637
(2009) 33–42.
These results led to conclude that using anodic oxidation, under
the present experimental conditions, paraquat is partially oxidized
by a conversion mechanism due to the fact that the Co2FeO4
electrode has an active behaviour and only permits the partial
oxidation.
Voltammetric techniques allowed to identify the redox pair
MV2+/MV+ꢄ. UV–vis and MS spectroscopic data permitted to
identify as main intermediate oxidation products monopyridone
and dipyridone derivatives.
The paraquat higher conversion rate was obtained when using
Pt/steel (ꢀ 79%) followed by Pt/Co2FeO4 (ꢀ 55%) after 1.5 h of
electrolysis. Conversion of ꢀ 64% was obtained with Co2FeO4/
Co2FeO4 after 3 h of electrolysis.
[19] J.C. Ferreira, M.R. Mendonça, F.M. Costa, Structural and morphological
characterization of FeCo2O4 and CoFe2O4 spinels prepared by
a
coprecipitation method, Solid State Sci. 5 (2003) 383–392.
[20] M.A.M. Cartaxo, T.A.S. Ferreira, M.R. Nunes, M.H. Mendonça, M.I. da Silva
Pereira, F.M. Costa, Studies on Fe-Co spinel electrodes, Solid State Sci. 9 (2007)
744–749.
The use of Pt leads to high costs for practical applications.
However a further way to improve paraquat oxidation and lower
the process costs could be the use of Co2FeO4 and Pt nanoparticles
composites.
[21] M.H. Florêncio, E. Pires, A.L. Castro, M.R. Nunes, C. Borges, F.M. Costa,
Photodegradation of Diquat and Paraquat in aqueous solutions by titanium
dioxide: evolution of degradation reactions and characterization of
intermediates, Chemosphere 55 (2004) 345–355.
[22] S. Trasatti, P. Kurzweil, Electrochemical supercapacitors as versatile energy
stores - potential use for Platinum metals, Platinum Met. Rev. 38 (1994) 46–56.
[23] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions,
Pergamon Press, Oxford, 1966.
Acknowledgement
[24] S.I. Cordoba, R.E. Carbonio, M. Lopez Teijelo, V.A. Macagno, The
electrochemical response of binary mixtures of hydrous transition metal
hydroxides co-precipitated on conducting substrates with reference to the
oxygen evolution reaction, Electrochim. Acta 31 (1986) 1321–1332.
[25] E.R. Vago, E.J. Calvo, Electrocatalysis of oxygen reduction at Fe3O4 oxide
electrodes in alkaline solutions, J. Electroanal. Chem. 339 (1992) 41–67.
[26] P. Bonhôtea, E. Gogniata, M. Grätzel, P.V. Ashritb, Novel electrochromic devices
based on complementary nanocrystalline TiO2 and WO3 thin films, Thin solid
films 350 (1999) 269–275.
The authors gratefully acknowledge financial support by
Fundação para a Ciência e Tecnologia – FCT, under contracts UID/
MULTI/00612/2013. M.A.M. Cartaxo also thanks FCT for the grant
SFRH/BD/30500.
Appendix A. Supplementary data
[27] C.W. Lee, J.M. Jang, Electrocatalytic reduction of dioxygen at glassy carbon
electrodes with irreversible self-assembly of N-hexadecyl-N0-methyl viologen,
Bull. Korean Chem. Soc. 15 (1994) 563–567.
[28] M.J. Cantavenera, I. Catanzaro, V. Loddo, L. Palmisano, G. Sciandrello,
Photocatalytic degradation of paraquat and genocity of its intermediate
products, Journal of Photochemistry and Photobiology A: Chemistry 185
(2007) 277–282.
[29] N.A. Ghalwa, H.M. Abu-Shawish, M. Hamada, K. Hartani, A.A.H. Basheer,
Studies on degradation of Diquat pesticide in aqueous solutions using
electrochemical method, Am. J. of Analytical Chem. 3 (2012) 99–105.
[30] J.S. Do, W.-C. Yeh, In situ paired electrooxidative degradation of formaldehyde
with electrogenerated hydrogen peroxide and hypochlorite ion, J. Appl.
Electrochem. 28 (1998) 703–710.
Supplementary data associated with this article can be found, in
the
online
version,
at
http://dx.doi.org/10.1016/j.
electacta.2015.07.099.
References
[1] N. Klamerth, S. Malato, M.I. Maldonado, A. Agüera, A.R. Fernández-Alba,
Application of photo-Fenton as a tertiary treatment of emerging contaminants
in municipal wastewater, Environ. Sci. Technol. 441 (2010) 792–1798.
[2] C. Monk, Electrochemical behaviour of methyl viologen in a matrix of paper,
Electrochim. Acta 44 (1999) 4817–4826.

1018
M.A.M. Cartaxo et al. / Electrochimica Acta 176 (2015) 1010–1018
[31] E. Brillas, I. Sirés, P.L. Cabot, Use of both anode and cathode reactions in
wastewater treatment, in: C. Comninellis, G. Chen (Eds.), Electrochemistry for
the Environment, Springer, New York, 2010, pp. 25–54.
[32] E. Moctezuma, E. Leyva, E. Monreal, N. Villegas, D. Infante, Photocatalytic
degradation of the herbicide paraquat, Chemosphere 39 (1999) 511–517.
[33] A. Calderbank, D.F. Charlton, J.A. Farrington, R. James, Bipyridylium quaternary
salts and related compounds. Part IV. Pyridones derived from Paraquat and
Diquat, J. C. S., Perkin I 1 (1972) 138–142.
[34] D.W. Bahnemann, C.H. Fischer, E. Janata, A. Henglein, The two-electron
oxidation of methyl viologen. Detection and analysis of two fluorescing
products, J. Chem. Soc., Faraday Trans. 1 83 (1987) 2559–2571.
[35] International Programme on Chemical Safety (IPCS) Paraquat, Health and
Safety Guide No 51. Published by World Health Organization (WHO), Geneva,
1991.