Degradation of the azo dye Acid Red 1 by anodic oxidation and indirect electrochemical processes based on Fenton's reaction chemistry. Relationship between decolorization, mineralization and products

Article abstract of DOI:10.1016/j.electacta.2014.07.117

Solutions of 236 mg dm^-3 Acid Red 1 (AR1), an azo dye widely used in textile dying industries, at pH 3.0 have been comparatively treated by anodic oxidation with electrogenerated H₂O₂ (AO-H₂O₂), electro-Fenton (EF) and photoelectro-Fenton (PEF) at constant current density (j). Assays were performed with a stirred tank reactor equipped with a Pt or boron-doped diamond (BDD) anode and an air-diffusion cathode for H₂O₂ generation from O₂ reduction. The main oxidizing agents were hydroxyl radicals produced at the anode from water oxidation in all methods and in the bulk from Fenton's reaction between generated H₂O₂ and 0.5 mmol dm^-3 Fe²⁺ in EF and PEF. For each anode, higher oxidation power was found in the sequence AO-H₂O₂ < EF < PEF. The oxidation ability of the BDD anode was always superior to that of Pt. Faster and similar decolorization efficiency was achieved in EF and PEF owing to the quicker destruction of aromatics with hydroxyl radicals produced in the bulk. The PEF process with BDD was the most potent method yielding almost total mineralization due to the additional rapid photolysis of recalcitrant intermediates like Fe(III)-carboxylate complexes under UVA irradiation. The increase in j always enhanced the decolorization and mineralization processes because of the greater production of hydroxyl radicals, but decreases the mineralization current efficiency. A total of 11 aromatic intermediates, 15 hydroxylated compounds, 13 desulfonated derivatives and 7 short-linear carboxylic acids were identified. NH₄⁺, NO₃^- and SO₄^2- ions were released during azo dye degradation. From the products detected, a comprehensive reaction sequence for AR1 mineralization is proposed. The relationship between decolorization, mineralization and products formed is finally discussed.

Full text of DOI:10.1016/j.electacta.2014.07.117

Electrochimica Acta 142 (2014) 276–288
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Degradation of the azo dye Acid Red 1 by anodic oxidation and indirect
electrochemical processes based on Fenton’s reaction chemistry.
Relationship between decolorization, mineralization and products
a
a,b,1
a
Xavier Florenza , Aline Maria Sales Solano
, Francesc Centellas ,
Carlos Alberto Martínez-Huitleb,1, Enric Brillas , Sergi Garcia-Segura
Laboratori d’Electroquímica dels Materials i del Medi Ambient, Departament de Química Física, Facultat de Química, Universitat de Barcelona,
Martí i Franquès 1-11, 08028 Barcelona Spain
a,1
a,∗
a
b
Universidade Federal do Rio Grande do Norte, Centro de Ciências Exatas e da Terra, Instituto de Química, Lagoa Nova, CEP 59078-970, Natal Brazil
a r t i c l e i n f o
a b s t r a c t
−3
Article history:
Solutions of 236 mg dm Acid Red 1 (AR1), an azo dye widely used in textile dying industries, at pH 3.0
have been comparatively treated by anodic oxidation with electrogenerated H2O2 (AO-H2O2), electro-
Fenton (EF) and photoelectro-Fenton (PEF) at constant current density (j). Assays were performed with a
stirred tank reactor equipped with a Pt or boron-doped diamond (BDD) anode and an air-diffusion cath-
ode for H2O2 generation from O2 reduction. The main oxidizing agents were hydroxyl radicals produced
at the anode from water oxidation in all methods and in the bulk from Fenton’s reaction between gener-
Received 24 June 2014
Received in revised form 25 July 2014
Accepted 29 July 2014
Available online 7 August 2014
Keywords:
Acid Red 1
Anodic oxidation
Boron-doped diamond
Electro-Fenton
Photoelectro-Fenton
−3
2+
ated H2O2 and 0.5 mmol dm Fe in EF and PEF. For each anode, higher oxidation power was found in
the sequence AO-H2O2 < EF < PEF. The oxidation ability of the BDD anode was always superior to that of Pt.
Faster and similar decolorization efficiency was achieved in EF and PEF owing to the quicker destruction
of aromatics with hydroxyl radicals produced in the bulk. The PEF process with BDD was the most potent
method yielding almost total mineralization due to the additional rapid photolysis of recalcitrant inter-
mediates like Fe(III)-carboxylate complexes under UVA irradiation. The increase in j always enhanced
the decolorization and mineralization processes because of the greater production of hydroxyl radicals,
but decreases the mineralization current efficiency. A total of 11 aromatic intermediates, 15 hydroxyla-
+
ted compounds, 13 desulfonated derivatives and 7 short-linear carboxylic acids were identified. NH4 ,
−
2−
NO3 and SO4 ions were released during azo dye degradation. From the products detected, a compre-
hensive reaction sequence for AR1 mineralization is proposed. The relationship between decolorization,
mineralization and products formed is finally discussed.
©
2014 Elsevier Ltd. All rights reserved.
1
. Introduction
azo group (−N = N − ) as chromophore, associated with aromatic
systems containing groups such as −OH and −SO H. Azo dyes are
3
Recently, the UNESCO’s World Water Development Report [1]
highly recalcitrant and are hardly removed by conventional bio-
logical and physicochemical methods, thus persisting in the aquatic
environment [2,3]. They are toxic to aquatic organisms and humans
[4,5] and possess carcinogenic, mutagenic and bactericide prop-
erties [1]. The development of powerful and effective oxidation
processes to remove azo dyes and their by-products from wastew-
aters are required to avoid their adverse environmental impact and
the danger they imply to public health.
manifested the growing concern about dying industrial effluents.
Large volumes of industrial effluents with high dye contents are
discharged into water bodies causing esthetic problems derived
from their color and their worrying hazardous effects on living
beings. Azo dyes are the most used organic dyes, representing
for over 70% of annual dyes consumption [2]. They have the
Electrochemical advanced oxidation processes (EAOPs) are
•
based on the in situ generation of hydroxyl radical ( OH) to min-
∗ Corresponding author. Tel.: +34 934021223; fax: +34 934021231.
E-mail addresses: sergigarcia@ub.edu, sergigs 87@hotmail.com
eralize organic matter in waters [6,7]. The high standard redox
potential of this radical (E ( OH/H O) = 2.80 V/SHE) favors its non-
selective reaction with most organics giving dehydrogenated or
◦
•
(
S. Garcia-Segura).
2
1
ISE Active Member.
http://dx.doi.org/10.1016/j.electacta.2014.07.117
013-4686/© 2014 Elsevier Ltd. All rights reserved.
0

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
277
Fe(OOCR)2 + hꢀ→ Fe2+ + CO + R
+
•
(6)
hydroxylated derivatives up to complete combustion to CO , inor-
ganic ions and water. EAOPs are promising and environmentally
2
2
The aim of this paper is to gain a better knowledge on the decol-
orization and mineralization processes of azo dyes during their
EAOP treatments. To do this, the AO-H O , EF and PEF degrada-
friendly technologies that are very effective for organics degrada-
tion and easily usable and scalable [6–10]. Although some papers
have been recently published dealing with the decolorization and
removal of several azo dyes by these methods [11–15], much
deeper information about the influence of operating parameters
and generated products on the degradation processes of a high
number of these pollutants is needed to clarify the possible viability
of EAOPs for wastewater remediation.
2
2
tions of an azo dye widely used in the textile industry like Acid
2
−
Red 1 (AR1, C18H13N O S
8 2
, see characteristics in Table 1) were
comparatively studied at pH 3.0 using Pt/air-diffusion and BDD/air-
3
−
3
−3
diffusion cells. Solutions with 236 mg dm
AR1 (100 mg dm
of dissolved organic carbon (DOC)) were checked, since it corre-
sponds to typical azo dye contents found in industrial effluents
[34]. The influence of current density (j) on the decolorization effi-
ciency and mineralization degree and rate was examined to better
explain the role of generated oxidizing agents and/or UVA radia-
tion in all the EAOPs tested. Aromatic intermediates were identified
by liquid chromatography-mass spectrometry (LC-MS). Generated
carboxylic acids and released inorganic ions were identified and
quantified by high-performance liquid chromatography (HPLC). A
reaction sequence for AR1 mineralization involving all the products
detected is proposed. Finally, the relationship between decoloriza-
tion, mineralization and products formed is discussed.
The most typical EAOP is anodic oxidation (AO), where adsorbed
•
hydroxyl radical (M( OH)) is generated from water discharge by
reaction (1) at an anode (M) with high O -overpotential by applying
2
a high current [6,13,16–20].
M + H O → M( OH) + H + e−
•
+
(1)
2
The non-active boron-doped diamond (BDD) anode is currently
considered the best anodic material for AO since it generates very
•
high amounts of reactive physisorbed BDD( OH) radicals as a result
•
of its very weak BDD- OH interaction and great O -overpotential
2
[
21]. These properties confer BDD the ability of being more effective
to remove aromatics including azo dyes than other common anodes
such as Pt [22,23] and PbO2 [24,25].
2. Experimental
Carbonaceous cathodes like graphite [26], carbon or graphite
felts [12,26–30], activated carbon fiber [31], carbon sponge [32],
carbon-polytetrafluoroethylene (PTFE) gas (O2 or air) diffusion
2
.1. Chemicals
Reagent grade Acid Red 1 was supplied by Sigma-Aldrich
[
11,15,33–36], carbon nanotubes [37,38] and BDD [14,39,40] have
and used as received. Analytical grade anhydrous sodium sulfate
and iron(II) sulfate heptahydrate were purchased from Fluka and
Sigma-Aldrich, respectively. The solution pH was initially adjusted
with analytical grade sulfuric acid supplied by Across Organics.
Sodium perchlorate and perchloric acid, used in some trials, were
analytical grade purchased from Merck. Carboxylic acids, other
chemicals and solvents used in chromatographic techniques were
of HPLC, LC-MS and analytical grade purchased from Sigma-Aldrich,
Lancaster, Merck and Panreac. All solutions were prepared with
high-purity water obtained from a Millipore Milli-Q system with
shown a high efficiency for H O2 electrogeneration from the two-
electron reduction of injected O2 from reaction (2):
2
+
−
O (g) + H + 2e → H O
(2)
2
2
2
In an undivided cell for AO, the use of a carbon-PTFE air-diffusion
cathode minimizes the possible cathodic reduction of organic pol-
lutants [15,23]. In this EAOP, so-called AO with electrogenerated
H O (AO-H O ), other weaker reactive oxygen species (ROS) than
2
2
2
2
•
•
)
OH can be produced at the anode, like hydroperoxyl radical (HO2
from H O oxidation by reaction (3):
2
2
◦
resistivity > 18 Mꢁ cm at 25 C.
M + H O → M(HO •) + H+ + e−
(3)
2
2
2
2.2. Electrolytic system
Cathodes with ability for H O electrogeneration have also been
2
2
extensively used in indirect EAOPs based on Fenton’s reaction
Electrolytic experiments were performed in an open and undi-
chemistry [7–9]. The most common method is electro-Fenton (EF)
3
•
vided two-electrode cell of 150 cm capacity with an external jacket
allowing circulation of thermostated water regulated by a Selecta
Digiterm 3000524 thermostat. The anode was a BDD thin film sup-
plied by Adamant Technologies (La-Chaux-de-Fonds, Switzerland)
or a Pt sheet of 99.99% purity supplied by SEMPSA (Barcelona,
Spain). The cathode was a carbon-PTFE air-diffusion electrode from
E-TEK (Somerset, NJ, USA), mounted as described elsewhere [46],
[
11,12,15,26–30,32,35,36], where OH is produced in the bulk from
the reaction between generated H O and low amounts of added
2
2
2+
Fe ion by Fenton’s reaction (4) with optimum pH 2.8:
Fe2 + H O + H → Fe + OH + H O
+
+
3+
•
(4)
2
2
2
Reaction (4) is catalytic and can be propagated from Fe2+ regen-
eration, pre-eminently by Fe3+ reduction at the cathode [27,32].
allowing H O generation from air injection at a flow rate of 300
2
2
When an undivided cell is used in EF, organic molecules are
3
−1
2
•
cm min . All electrodes had a geometric area of 3 cm and were
separated about 1 cm. The assays were made at constant j by con-
necting the electrodes to an Amel 2053 potentiostat-galvanostat
and the potential difference of the cell was directly measured with
a Demestres 601BR digital multimeter. To remove the impurities of
the BDD anode surface and activate the air-diffusion cathode before
the degradation trials, they were polarized in 100 cm of a 0.05 mol
dm Na SO4 solution at 100 mA cm for 180 min.
All experiments were carried out with 100 cm of solutions
containing 236 mg dm AR1 in 0.05 mol dm Na SO4 as back-
mainly destroyed by the combined action of M( OH) formed at the
•
anode from reaction (1) and OH produced in the bulk from reac-
tion (4), along with a slower destruction by other weaker ROS such
•
as H O and HO [7]. Unfortunately, the EF treatment of aromatics
2
2
2
with an air-diffusion cathode does not allow total mineralization
of aromatic solutions due to the formation of Fe(III)-carboxylate
3
•
complexes that are only slowly removed by BDD( OH) [7,41]. This
−
3
−2
2
problem can be solved performing the photoelectro-Fenton (PEF)
process, in which the solution treated by EF is irradiated with UVA
light [15,31,33,42–45] and the degradation process is enhanced by:
3
−
3
−3
2
−
3
2+
2
+
3+
ground electrolyte at pH 3.0. In EF and PEF, 0.5 mmol dm Fe
(
i) the photoreduction of Fe(OH) , the pre-eminent Fe species at
were added to the solution as catalyst. The solution pH and the Fe2+
concentration were chosen since they have been found optimal for
similar treatments of other aromatics [11,15,33–36]. The solution
2+
•
pH near 3, to regenerate Fe producing more OH from reaction (5)
and (ii) the photolysis of generated Fe(III)-carboxylate complexes
by the general reaction (6):
◦
was maintained at 35.0 C, which is the maximum temperature
Fe(OH)2 + hꢀ→ Fe2+ + OH
+
•
(5)
that our system can be thermostated without significant water

2
78
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
Table 1
Chemical structure and characteristics of Acid Red 1 azo dye.
Chemical structure
Chemical formula
IUPAC name
C18H13N3Na2O8S2
Sodium 5-(acetylamino)-4-hydroxy-3-(2-phenyldiazenyl)-2,7-naphthalenedisulfonate
Colour Index name
CAS number
Acid Red 1
3734-67-6
18050
509.44
506
Colour Index number
−
1
M [g mol
max [nm]
]
␭
evaporation during prolonged electrolysis time. The solution was
stirred by a magnetic bar at 800 rpm to ensure homogeneity in
solution and the mass transport of reactants toward/from the
electrodes. A Philips 6-W black light blue tube lamp was used
for UVA irradiation in PEF, providing wavelengths of 320-400 nm
From the DOC decay, the mineralization current efficiency (MCE)
for each trial was estimated by Eq. (8) [34]:
nFVsꢃ(DOC)exp
MCE (%) =
100
(8)
7
4
.32 × 10 mIt
−2
with ␭max = 360 nm with an energy of 5 W m , as detected with a
Kipp&Zonen CUV 5 global UV radiometer.
−1
where
F
is the Faraday constant (96,487 C mol ), Vs is the
3
solution volume (dm ), ꢄ(DOC)exp is the experimental DOC
−
3
7
decay (mg dm ), 4.32 × 10 is a conversion factor (3600 s
2.3. Instruments and analytical procedures
−1
−1
h
× 12000 mg C mol ), m is the number of carbon atoms of AR1
(
(
18 C atoms), I is the applied current (A), t is the electrolysis time
h) and n refers to the number of electrons involved in the total
The solution pH was determined using a Crison GLP 22 pH-
meter. Samples were collected at regular time intervals, alkalinized
to stop the mineralization process and filtered with 0.45 ␮m PTFE
filters purchased from Whatman before analysis. The decoloriza-
tion of AR1 solutions was followed by monitoring their absorbance
combustion of AR1 assuming the main formation of ammonium
and sulfate ions from reaction (9):
2
−
2−
4
C18H13N O S
+ 36H O → 18CO + 2SO
3
8
2
2
2
(
A) decrease at the maximum visible wavelength of ꢂmax = 506 nm
3NH4 + 73H+ + 74e−
+
+
(9)
(
see Table 1) using a Shimadzu 1800 UV-vis spectrophotometer
◦
thermostated at 35 C to record the solution spectra between 200
and 800 nm. The percentage of color removal or decolorization effi-
ciency was then calculated by Eq. (7) [2]:
Total nitrogen (TN) referred to all N-species present in solution
was determined with a Shimadzu TNM-1 module coupled to the
above TOC analyzer.
A − At
0
%
Color removal =
100
(7)
The evolution of generated carboxylic acids was followed by
ion-exclusion HPLC using a Waters 600 LC fitted with a Bio-Rad
A
0
◦
where A0 and At are the absorbance at initial time and time t at
Aminex HPX-87H, 300 mm × 7.8 mm (i.d.), column at 35 C and cou-
␭max = 506 nm, respectively.
pled to a Waters 996 photodiode array detector set at ␭ = 210 nm.
For this analysis, 20 ␮L aliquots were injected to the LC and a
The mineralization of AR1 solutions was monitored from their
− −1
4 mmol dm H SO at 0.6 cm min was used as mobile phase.
2 4
3
3
DOC abatement, determined on a Shimadzu VCSN Total Organic
Carbon (TOC) analyzer. Reproducible DOC values with an accuracy
of ± 1% were obtained by injecting 50 ␮L aliquots into the analyzer.
Ammonium, nitrate and sulfate ions released during mineralization
were analyzed by ion chromatography by injecting 25 ␮L aliquots

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
279
into a Shimadzu 10Avp LC coupled to a Shimadzu CDD 10Avp
120
conductivity detector. The NH4+ concentration was determined
a
using a Shodex IC YK-421, 125 mm × 4.6 mm (i.d.), cationic col-
1
00
◦
−3
−3
umn at 40 C and a 24.2 mmol dm
boric acid, 5.0 mmol dm
−
3
−3
tartaric acid, 1.5 mmol dm 18-crown-6 and 2.0 mmol dm 2,6-
80
3
−1
pyridinedicarboxylic solution at 1.0 cm min
as mobile phase.
−
2−
The NO3 and SO4
contents were quantified with a Shim-Pack
◦
60
40
20
0
IC-A1S, 100 mm × 4.6 mm (i.d.), anionic column at 40 C and circu-
−
3
lating a mobile phase composed of 2.6 mmol dm phthalic acid
−3
and 2.4 mmol dm tris(hydroxymethyl)aminomethane (pH = 4.0)
at 1.5 cm min
Aromatic intermediates formed between 2 and 40 min of EF
treatment of 100 cm of 236 mg dm AR1 solution in a BDD/air-
diffusion cell at 33.3 mA cm
3
−1
.
3
−3
−
2
were identified by LC-MS using a
Shimadzu SIL-20AC LC filled with a Teknokroma Mediterranea Sea
0
60
120
180
240
300
360
420
◦
C-18 3 ␮m, 15 mm × 0.46 mm (i.d.), column at 30 C and coupled
1
20
00
80
to a Shimadzu LCMS-2020 MS. The MS operated in negative mode
with electrospray source ionization (ESI), by applying 4.5 kV inter-
b
1
face voltage and 60 V Q-array RF voltage. The DL temperature was
◦
2
50 C and pure N was used as nebulizing and dryer gas. Mass
2
spectra were collected in the m/z range of 50-600 using both, total
ion current (TIC) and selected ion (SIM) acquisition. This analy-
sis was made by injecting 20 ␮L aliquots into the LC, previously
filtered with a Millipore filter of 0.22 ␮m, using a 75:25 (v/v) ace-
1
00
60
40
20
0
8
0
−3
60
40
tonitrile/water (5 mmol dm ammonium acetate) mixture at 0.2
3
−1
cm min as mobile phase.
2
0
0
0
2
4
6
8
10 12 14 16
3
. Results and discussion
time / min
0
60
120
180
240
300
3.1. Comparative degradation of Acid Red 1 by electrochemical
time / min
advanced oxidation processes with a Pt or BDD anode
Fig. 1. Percentage of (a) DOC abatement and (b) color removal at ␭max = 506 nm vs.
electrolysis time for the degradation of 100 cm3 of 236 mg dm−3 Acid Red 1 (AR1)
To assess the effectiveness of the EAOPs to degrade AR1, a first
series of trials was made by treating 236 mg dm azo dye solutions
−
3
−3
−2
◦
solutions in 0.05 mol dm Na2SO4 at pH 3.0, 100 mA cm and 35 C using a (open
symbols) Pt/air-diffusion or a (solid symbols) BDD/air-diffusion cell. Method: (ꢀ,᭹)
Anodic oxidation with electrogenerated H2O2 (AO-H2O2), (ꢀ,ꢁ) electro-Fenton (EF)
−
3
in 0.05 mol dm Na SO4 at pH 3.0 by AO-H O , EF and PEF using
2
2
2
−
2
Pt/air-diffusion and BDD/air-diffusion cells at 100 mA cm
and
5 C for 360 min. In these experiments, the solution pH remained
practically unchanged, slowly decreasing up to final values of pH
.7-2.8, probably due to the formation of acidic products like car-
−
3
2+
−3
with 0.5 mmol dm Fe and (ꢁ,ꢂ) photoelectro-Fenton (PEF) with 0.5 mmol dm
◦
3
2+
Fe and 6 W UVA light of ␭max = 360 nm.
2
boxylic acids [7,18].
reaching 92% mineralization at 360 min, a value quite similar to 93%
Fig. 1a shows a very poor DOC removal of the AR1 solution by
AO-H O with Pt, only reaching 16% mineralization at the end of
found for AO-H O2 with BDD. These findings suggest the produc-
2
tion of very recalcitrant products like Fe(III)-carboxylate complexes
2
2
•
•
this treatment as a result of the low oxidation ability of Pt( OH)
formed from reaction (1) at the Pt surface, along with the poor
in EF, which are slowly removed mainly by BDD( OH), thereby
largely inhibiting the mineralization process at long electrolysis
time [7,29,32]. When the PEF process was applied, less influence
of the anode material was found due to the potent action of UVA
irradiation, as can be seen in Fig. 1a. This EAOP yielded a very fast
mineralization of the AR1 solution to yield almost total mineraliza-
tion (> 96% DOC removal) after 300 min using Pt, and in a shorter
time of 240 min for BDD. The large acceleration of the degradation
in PEF compared to EF can then be related to the production of more
•
oxidative action of other ROS like H O and HO2 generated from
2
2
reactions (2) and (3). In contrast, the alternative use of a BDD anode
causes a much faster DOC decay, up to attain 93% mineralization at
3
60 min. This positive degradation behavior can be related to the
•
expected greater oxidation ability of generated BDD( OH) at the
BDD surface [6,24,25], indicating that it is the best anode to remove
more rapidly and in larger extent the DOC of the AR1 solution com-
paratively with AO-H O
•
amounts of OH from reaction (5) along with the quick photolysis
2
2.
Fig. 1a also highlights a remarkable effect of the anode mate-
rial in the EF process. Thus, when Pt was used, DOC was rapidly
reduced by 52% in 180 min, whereupon it was slightly removed.
This enhancement in dye degradation compared to the AO-H O
of intermediates like Fe(III)-carboxylate complexes via reaction (6)
[35,36,41,42]. These phenomena are more significant when a Pt
anode was used because the latter complexes are very slowly min-
•
•
eralized by Pt( OH) and OH, as stated above. The quicker removal
2
2
•
•
process is due to the additional oxidation of organics by OH pro-
duced in the bulk from Fenton’s reaction (4). The very slow DOC
removal found for EF with Pt at long electrolysis time suggests
the formation of quite recalcitrant products that are difficultly
of these compounds by BDD( OH) can explain the superiority of
PEF using BDD instead of Pt.
The role of the oxidizing agents generated in the comparative
assays mentioned above to directly remove the azo dye was clar-
ified by determining the decolorization efficiency of the treated
solutions at ␭max = 506 nm. Note that the decolorization process
could involve not only the cleavage of AR1 but also the destruction
of other possible conjugated aromatic products that absorb at sim-
ilar wavelength region, as found for other azo dyes [34,35]. Fig. 1b
•
•
destroyed by Pt( OH) and OH. The faster degradation process
•
under the action of OH in the bulk can also be observed in Fig. 1a
for EF with BDD, primordially during the first 180 min of treat-
ment where 80% DOC reduction was already achieved. At longer
time, however, the degradation was progressively decelerated up to

2
80
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
3
3
2
2
1
1
0
0
3
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
.5
.0
.0
.5
.0
.5
.0
.5
.0
suggesting near 50% of each tautomeric form. In addition, the UV
a
region exhibited three bands, the weakest one at 364 nm corre-
sponding to the amide group linked to the naphthalene ring, other
band centered at 312 nm related to the naphthalenic ring and the
most intense one associated with the benzenic ring at 236 nm. As
can be seen in Fig. 2a, the intensity of all the bands of AR1decayed at
similar rate with prolonging the AO-H O treatment up to 240 min.
0 min
min
5
20 min
3
4
0 min
0 min
2
2
6
0 min
This indicates that the azo dye is the principal species present in the
solution during the decolorization process, suggesting that its aro-
9
1
0 min
•
matics products are rapidly removed by BDD( OH). Note also the
20 min
very low absorption intensity between 200 and 400 nm at 240 min
of electrolysis, as expected if small quantities of aliphatic products
like carboxylic acids are accumulated in the medium because they
are destroyed at similar rate as produced. For the EF treatment,
however, Fig. 2b shows a rapid drop in intensity of all the bands dis-
appearing at 6 min of electrolysis, although a wide undefined band
remained in the visible region up to 10 min. This behavior can be
accounted for the generation of some colored aromatic intermedi-
ates while solution is decolorized, which are completely destroyed
1
80 min
2
40 min
b
0 min
•
by OH in 10 min. At this time, an intense absorption in the UV
region of 200-400 nm can be observed, suggesting the accumu-
lation of large quantities of aliphatic products that are destroyed
2
4
min
min
•
more slowly and pre-eminently by BDD( OH), in agreement with
the slow DOC fall found in EF (see Fig. 1a).
6
8
min
min
3.2. Effect of current density on the electrochemical advanced
oxidation processes with BDD
2
00
300
400
500
nm
600
700
800
The current density is an operating variable that influences the
amount of oxidizing hydroxyl radicals generated in EAOPs. To clar-
ify the effect of this parameter on AR1 degradation, the above
/
3
Fig. 2. Evolution of the UV-vis spectra recorded during the treatment of 100 cm
−
3
−
3
−3
◦
236 mg dm azo dye solutions were treated by the methods tested
of 236 mg dm AR1 solutions in 0.05 mol dm Na2SO4 at pH 3.0 and 35 C by
−
3
−2
(
1
a) AO-H2O2 and (b) EF with 0.5 mmol dm Fe2+ using a BDD/air-diffusion cell at
between 16.7 and 100 mA cm using the BDD/air-diffusion cell,
−
2
00 mA cm
.
which comparatively yielded the most potent procedures.
Figs. 3a, 3b and 3c depict a gradual increase in DOC removal with
increasing j for AO-H O , EF and PEF, respectively. After 360 min
shows that AO-H O with BDD led to overall decolorization in about
2
2
2
2
2
40 min, time in which only 86% color removal was achieved using
of the former treatment, Fig. 3a shows that mineralization rose
−
2
−2
the Pt anode. This corroborates the existence of a fast attack of
BDD( OH) on AR1, which becomes much slower under the action of
Pt( OH) as expected from its lower oxidation ability. In contrast, the
from 54% for 16.7 mA cm to 93% for 100 mA cm , as expected
from the concomitant increase in rate of reaction (1) yielding more
•
•
•
amounts of BDD( OH) that accelerate the degradation process. For
inset panel of Fig. 1b depicts a much quicker decolorization under
EF conditions, which was completely attained after ca. 14 min for
Pt and only 10 min for BDD. This behavior is indicative of the pre-
the comparative EF processes, Fig. 3b depicts an enhancement
of the mineralization rate due to the acceleration of reaction (2)
•
yielding more H O2 and hence higher quantities of OH from Fen-
2
•
dominant destruction of AR1 by OH formed from Fenton’s reaction
ton’s reaction (4). Thus, DOC was reduced by 68% and 92% at the
−
2
(
4), suggesting the pre-eminent destruction of aromatic intermedi-
end of the electrolyses at 16.7 and 100 mA cm , respectively. The
faster DOC removal was always achieved by the most potent PEF
ates by this oxidant. The slightly higher color removal reached in EF
using BDD instead of Pt can be justified again by the quicker paral-
process, where the time needed for reaching almost total min-
•
•
−2
lel attack of BDD( OH) than Pt( OH), although the reactivity of both
eralization decreased from 360 min at 16.7 mA cm
to 240 min
•
−2
radicals over the azo dye is much smaller than OH. These results
at 100 mA cm (see Fig. 3c). This trend can be explained by the
quicker photolysis of intermediates by UVA light since they are
evidence the different reactivity of hydroxyl radicals generated in
the bulk by Fenton’s reaction and those formed at the anode surface
from water oxidation. It is noteworthy that the percentage of color
removal-time plots for the PEF processes with Pt and BDD (data not
shown) overlapped with the corresponding EF ones. This evidences
•
•
more rapidly formed by the greater amounts of BDD( OH) and OH
produced at higher j.
Figs. 4a, 4b and 4c present the MCE values calculated from Eq.
(8) for the trials of Figs. 3a,3b and 3c, respectively. The efficiency
increased progressively with decreasing j in all cases, becoming
•
low production of additional OH induced from photolytic reaction
(
5) and high stability of AR1 under UVA illumination.
greater in the sequence AO-H O < EF < PEF, in agreement with their
2 2
Figs. 2a and 2b show the evolution of the UV-vis spectra recorded
relative oxidation ability over AR1. Thus, these processes yielded
in the above AO-H O and EF treatments, both with BDD, respec-
maximum MCE values of 12%, 32% and 39% at the lowest j of
16.7 mA cm . Note that the efficiency did not vary significantly
2
2
−
2
tively. The spectrum of the starting solution displayed a visible band
with two strong peaks centered at 506 nm (characteristic of AR1,
see Table 1) and 530 nm. According to the behavior of other azo
dyes [47], these bands can be associated with two tautomeric forms
in equilibrium given in Table 1, the azo form with ␭max = 506 nm
and the hydrazone one with ␭ = 530 nm, where the hydroxyl group
appears as carbonyl group and its hydrogen is linked to the azo
group. A similar absorbance can be observed for both peaks,
during each AO-H O2 treatment, suggesting a practically constant
2
•
destruction rate of intermediates with BDD( OH), according to the
evolution of the UV-vis spectra shown in Fig. 2a. In contrast, the
MCE values underwent a dramatic decay at long electrolysis time of
EF and PEF, corroborating the formation of more recalcitrant prod-
ucts like Fe(III)-carboxylate complexes at the beginning of both
treatments, in agreement to the UV-vis spectra of Fig. 2b. These

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
281
1
1
20
00
14
a
a
1
1
2
0
8
6
4
2
0
8
6
4
2
0
0
0
0
0
b
b
1
00
30
8
6
4
2
0
0
0
0
0
2
1
0
0
0
c
c
4
3
2
1
0
0
0
0
0
1
00
8
6
4
2
0
0
0
0
0
0
60
120
180
240
300
360
420
0
60
120
180
240
300
360
420
time / min
time / min
Fig. 3. Effect of current density on the change of per cent of DOC abatement with
3
−3
−3
Fig. 4. Mineralization current efficiency vs. electrolysis time for the trials of Fig. 3.
electrolysis time for 100 cm of 236 mg dm AR1 solutions in 0.05 mol dm Na2SO4
◦
at pH 3.0 and 35 C treated by (a) AO-H2O2, (b) EF and (c) PEF using a BDD/air-
−
3
2+
diffusion cell. In the two latter methods, 0.5 mmol dm Fe was added to the
Fe + OH → Fe + OH−
2+
•
3+
(12)
−
2
−2
−2
solution. Applied j: (ꢃ) 16.7 mA cm , (ꢁ) 33.3 mA cm , (ꢀ) 66.7 mA cm and (ꢀ)
−
2
1
00 mA cm
.
The positive effect of increasing j on the decolorization efficiency
of the above trials was also found. As can be seen in Fig. 5a, a rise in j
−
2
•
from 33.3 to 100 mA cm promoted the production of BDD( OH) in
AO-H O , enhancing the destruction of AR1 and hence, favoring the
species are then more slowly destroyed by the oxidizing species
generated and/or photolyzed by UVA light, thereby making the pro-
cess less efficient. On the other hand, the rise in MCE at lower j for
all the EAOPs seems contradictory to the decay in DOC abatement
shown in Fig. 3. This opposite tendency can be related to a grad-
ual loss in the relative production of hydroxyl radicals due to the
acceleration of their non-oxidizing reactions with the increase of j,
which causes fewer attacks on organic compounds, thus reducing
2
2
decolorization process. Total decolorization was achieved at about
−
2
3
00 and 240 min for 66.7 and 100 mA cm , respectively, needing a
much longer time for lower j values. Fig. 5b highlights that times of
4
3
0, 20 and 10 min were needed for reaching total decolorization at
−
2
3.3, 66.7 and 100 mA cm , respectively, of EF. This trend can be
•
related to the enhancement of OH formed from Fenton’s reaction
(
4) because of the acceleration of H O2 production from reaction
2
the efficiency. These parasitic reactions involve primordially the
•
(2). Similar results were obtained for PEF, corroborating the small
oxidation of BDD( OH) to O at the anode by reaction (10), as well
2
•
•
participation of reaction (5) in OH generation.
as the reaction of OH in the bulk with generated H O to form
2
2
HO2 via reaction (11) and with Fe2+ by reaction (12) [6–8]:
•
The decolorization rate related to the change of the absorbance
at ␭ = 506 nm with time was analyzed for each trial from kinetic
equations with simple reaction orders. Excellent straight lines were
obtained for a pseudo-first-order reaction, as presented in the inset
panels of Fig. 5a and 5b. Note that the decolorization rate cannot
BDD( OH) → 2BDD + O + 2H + 2e−
•
+
(10)
(11)
2
•
•
H O + OH → HO + H O
2
2
2
2

2
82
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
1
20
00
140
a
a
1
1
20
00
1
8
6
4
2
0
0
0
0
0
2
.5
8
6
4
2
0
0
0
0
0
2.0
1
1
0
.5
.0
.5
0
0
60 120 180 240 300 360
time / min
0
60
120
180
240
300
360
420
b
120
b
1
1
5
0
5
0
100
8
6
4
2
0
0
0
0
0
6
5
4
3
2
1
0
0
10
20
30
40
time / min
0
60
120
180
240
300
360
420
time / min
0
10
20
30
40
50
Fig. 6. Evolution of (a) oxalic and (b) oxamic acids detected during the treatment
time / min
3
−3
−3
AR1 solutions in 0.05 mol dm Na2SO4 at pH 3.0
of 100 cm of 236 mg dm
◦
−3
2+
and 35 C by (᭹) AO-H2O2, (ꢀ,ꢁ) EF with 0.5 mmol dm Fe and (ꢁ,ꢂ) PEF with
Fig. 5. Influence of current density on the variation of decolorization efficiency with
−3
2+
0
.5 mmol dm Fe using a: (open symbols) Pt/air-diffusion or (solid symbols)
3
−3
electrolysis time for the degradation of 100 cm of 236 mg dm AR1 solutions in
−2
BDD/air-diffusion cell at 100 mA cm
.
−
3
◦
0
.05 mol dm Na2SO4 at pH 3.0 and 35 C by (a) AO-H2O2 and (b) EF with 0.5 mmol
−
3
−2
−2
dm using a BDD/air-diffusion cell. Applied j: (ꢁ) 33.3 mA cm , (ꢀ) 66.7 mA cm
−
2
and (ꢀ) 100 mA cm . The inset panels show the corresponding kinetic analysis
considering a pseudo-first-order decay for the dye absorbance at ␭max = 506 nm.
derivatives coming from the loss or oxidation of functional groups.
Acetamide (13) and aniline (14), expected to be released from these
degradation reactions, were not detected by LC-MS.
be strictly associated with the decay kinetics of AR1 due to the
formation of colored aromatic intermediates, as deduced from the
UV-vis spectra recorded in EF (see Fig. 2b). Thus, the apparent rate
constant (kdec) derived from the above analysis has only experi-
The further destruction of the above aromatic products is
expected to yield short-chain carboxylic acids [7,8]. This point was
−2
confirmed by analyzing the solutions treated at 100 mA cm by
ion-exclusion HPLC. These chromatograms revealed the formation
−
2
4
−1
2
mental validity. Increasing k values of 1.3 × 10
s
(R = 0.997),
dec
of succinic (15, t = 11.75 min), maleic (16, t = 8.24 min), acetic (17,
r
r
−
4
−1
2
−4 −1
1
.8 × 10
s
(R = 0.991) and 2.9 × 10
s
(R = 0.993) for AO-
tr = 14.9 min), glycolic (18, tr =12.2), oxamic (19, tr = 9.4 min), oxalic
(20, t = 6.9 min) and formic (21, t = 13.7 min) acids. Acids 15-18 are
−
3
−1
2
−3 −1
2
H O and 2.6 × 10
s
(R = 0.992), 6.1 × 10
s
(R = 0.994) and
2
2
r
r
−
3
−1
2
8
.8 × 10
s
(R = 0.999) for EF were determined at 33.3, 66.7
expected to be formed from the cleavage of the benzene moieties,
which are then oxidized to acids 20 and 21 [7,15,23,32–36]. Acid
19 could be formed from the degradation of 13 [48]. Note that acids
19-21 are ultimate intermediates that are directly transformed into
CO2 [41,44].
−
2
and 100 mA cm , respectively. These findings demonstrate that
the kdec values for EF and PEF are about 20-30 times higher than
those of AO-H O , corroborating the predominant role of generated
OH in the bulk to remove aromatics in the indirect electrochemical
processes based on Fenton’s reaction chemistry.
2
2
•
Ion-exclusion chromatograms of the solutions electrolyzed by
AO-H O with Pt did not exhibit any carboxylic acid due to the low
2
2
•
3
.3. Identification of aromatic intermediates and evolution of
oxidation ability of Pt( OH), giving small mineralization at 360 min
(see Fig. 1a). In contrast, when a BDD anode, with much higher oxi-
dation ability, was used maximum contents of 10.5 mg dm for 18,
8.0 mg dm for 17, 4.6 mg dm for 20 (see Fig. 6a), 4.1 mg dm for
19 (see Fig. 6b) and 1.2 mg dm for 21, along with traces (< 1.0 mg
generated carboxylic acids
−
3
−
3
−3
−3
−3
Table 2 collects the aromatic products detected by LC-MS during
the first 40 min of EF treatment of a 236 mg dm azo dye solu-
−
3
−
2
−3
tion at 33.3 mA cm , i.e., while it is decolorized. Since organics are
mainly attacked by hydroxyl radicals, the same kinds of products
are expected in AO-H O and PEF. As can be seen, apart from the
starting compound (1), 11 aromatic intermediates, 15 hydroxyla-
ted derivatives and 13 desulfonated compounds were identified
by this technique. The aromatic intermediates include molecules
with similar structure to the azo dye (compounds 2-4), as well
as naphthalenic (compounds 5-8) and benzenic (compounds 9-12)
dm ) of 15 and 16, were found. All these acids were practically
removed at the end of the treatment, only corresponding to about
−
3
10% of the 7.0 mg dm DOC of the final solution (see Fig. 1a). This
means that small amounts of other persistent products that react
2
2
•
even more hardly with BDD( OH) than carboxylic acids are formed
during AR1 degradation.
A different pattern was found for the EF and PEF treatments,
where acids 15-18 and 21 were accumulated at short electrolysis

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
283
Table 2
Aromatic compounds along with their hydroxylated and desulfonated derivatives identified by LC-MS during the EF treatment of 100 cm of 236 mg dm AR1 solution using
3
−3
−
2
a BDD/air-diffusion cell at 33.3 mA cm
.
Hydroxylated derivatives
Desulfonated
derivatives
Number Name
Molecular
formula
Number of
-OH added
m/z
Number of
-SO3 lost
m/z
−
1
Acid Red 1
5-Acetylamino-4-
hydroxy-3-
phenylazo-
naphthalene-2,7-
disulfonate)
-
232b
1
1
1 400a
(
1
-
240b
2
5-Amino-4-
hydroxy-3-
211b
phenylazo-
naphthalene-2,7-
disulfonate
b
1
2
219
b
227
3
4-Hydroxy-5-
-
226b
234b
407a
1 388a
nitro-3-phenylazo-
naphthalene-2,7-
disulfonate
1
-
1
1
4
4-Hydroxy-3-
phenylazo-
1 327a
naphthalene-2,7-
disulfonate
b
1
2
211
b
219
5
5-Acetylamino-3-
amino-4-hydroxy-
naphthalene-2,7-
disulfonate
-
187b
1
1 295a
b
1 311a
1
2
3
195
203
1
1
1
b
a
1 327
b
1 343a
211

2
84
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
Table 2 (Continued)
Hydroxylated derivatives
Desulfonated
derivatives
Number Name
Molecular
formula
Number of
-OH added
m/z
Number of
-SO3 lost
m/z
−
6
4-Acetylamino-5-
-
180b
2 232a
hydroxy-
naphthalene-2,7-
disulfonate
a
1 328a
1
2
3
376
a
392
2
1
a
408
7
4-Hydroxy-3,5-
dinitro-
naphthalene-2,7-
disulfonate
-
-c
2
1
2 233a
1 329a
1
2
3
409a
425
441
a
a
8
4-Amino-5-
hydroxy-
-
159b
2 190a
naphthalene-2,7-
disulfonate
2
-c
2
9
3-Hydroxy-5-
sulfophthalic
acid
-
261a
138b
292a
1 197a
1
1
1
10
4-Amino-3,6-
dihydroxy-5-
sulfophthalic
acid
1
1
1
2
Hydroxy-3-
acetylamino-
benzenesulfonate
1
232a
Aminophenol
3
156
Negative ion with z = a 1 and b
Not detected
2
c

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
285
times up to 4 mg dm 3 as maximal and their Fe(III) complexes
−
10
8
•
disappeared in about 40-60 min owing to the action of BDD( OH),
OH and/or UVA light. However, these species acted much more
slowly over the Fe(III) complexes of acids 19 and 20, which were
accumulated in large extent and persisted during long electrolysis
times. Fig. 6a shows a quick accumulation of acid 20 up to near
in both EF and PEF processes. This acid was
slowly reduced to 67 mg dm after 360 min of EF with Pt because
of the low reactivity of Pt( OH) and OH to destroy Fe(III)-oxalate
species, whereas it was removed up to 7 mg dm in EF with BDD
a
b
c
•
6
−3
8
0-120 mg dm
−3
•
•
4
−
3
•
as a result of the quicker attack of BDD( OH) on such complexes.
The predominant photolysis of Fe(III)-oxalate complexes under
UVA irradiation via reaction (6) explains the total disappearance
2
of acid 20 at 300 min of PEF with Pt, which was even shortened to
0
•
1
80 min using BDD by their parallel destruction with BDD( OH).
A similar behavior can be observed in Fig. 6b for the evolution of
acid 19, accumulated in much lesser extent. For EF, the maximum
1
1
5
0
5
0
−3
content of 5.4 mg dm attained by this acid at 90-120 min using Pt
−
3
decayed to a final value of 2.8 mg dm because of the low removal
•
•
power of Pt( OH) and OH on Fe(III)-oxamate complexes, whereas
using BDD these species were totally destroyed by the more potent
action of BDD( OH). In PEF with Pt much more quantity of acid
•
−3
1
9 was accumulated (16 mg dm ), suggesting that UVA light
promotes the degradation of N-derivatives, and it was rapidly
−
3
reduced to a final content of 3.0 mg dm due to the photolysis of
•
Fe(III)-oxamate complexes. The combined action of BDD( OH) and
UV light caused the total disappearance of these species in only
1
80 min using PEF with BDD.
A mass balance of the final content of both acids 19 and 20
3
2
0
5
−
3
revealed their contribution in only 18.7, 1.9 and 0.8 mg dm to
−3
the 42.3, 8.3 and 4.0 mg dm DOC of the final solutions of EF with
Pt, EF with BDD and PEF with Pt, respectively (see Fig. 1a). These
findings, along with the absence of carboxylic acids in the final solu-
tion of PEF with BDD with 3.3 mg dm DOC, indicate the formation
of other persistent products that are hardly removed by hydroxyl
radicals and UVA light.
20
15
−3
1
0
5
0
3.4. Time-course of released inorganic ions
To assess the evolution of inorganic ions like NH4+, NO3− and
0
60
120
180
240
300
360
420
SO42 expected from the degradation of AR1, 236 mg dm
−
−3
azo
dye solutions were treated by AO-H O , EF and PEF with BDD
time / min
2
2
−3
using 0.05 mol dm
NaClO4 at pH 3.0 regulated with HClO4 at
Fig. 7. Evolution of (a) ammonium, (b) nitrate and (c) sulfate ions detected during
−2
1
00 mA cm for 360 min. The corresponding DOC-time plots found
3
−3
−3
the degradation of 100 cm of 236 mg dm AR1 solutions in 0.05 mol dm NaClO4
in this perchlorate medium were quite similar to those previously
described for 0.05 M Na SO in Fig. 1a; the former medium was then
at pH 3.0 and 35 ◦C using a BDD/air-diffusion cell at 100 mA cm−2. Method: (᭹) AO-
−
3
2+
−3
2+
H2O2, (ꢁ) EF with 0.5 mmol dm Fe and (ꢂ) PEF with 0.5 mmol dm Fe
.
2
4
preferred for a better measurement of the concentration of released
2
−
SO4 ion. Moreover, the TN of final solutions was determined to
quantify the remaining soluble N.
a large proportion of unidentified N-recalcitrant products in their
final solutions. It is remarkable that the continuous oxidation with
After 360 min of AO-H O , Figs. 7a and 7b highlight the accu-
2
2
• −
BDD( OH) favors the preponderance of N lost as NO in AO-H O ,
3 2 2
−3
+
−3
mulation of 1.9 mg dm NH4 (7.6% of initial N) and 19.1 mg dm
NO3 (22.0% of initial N). The sum of both species and the N of acid
•
+
−
whereas OH and UVA light promote the release of N as NH4 in
•
EF and PEF. Moreover, the action of OH and UVA light in the latter
1
9 contained in the remaining solution (see Fig. 6b) only yielded
two processes leads to the loss of a larger proportion of initial N
(about 40%) from the solution, probably in the form of N and N O
2
x y
2
9.8% of initial N, very far from 93% determined from TN. This
−
3
means that the final solution with 7.0 mg dm
DOC is primor-
species.
dially composed of other unidentified N-products of low molecular
mass and low carbon content. In contrast, Figs. 7a and 7b reveal
Fig. 7c shows the evolution of SO 2 ion released from the
−
4
the superiority of N loss as NH4+ compared to NO in EF and PEF.
−
sulfonic groups of the azo dye. While this ion was continuously
3
At the end of the former process, 8.3 mg dm NH4+ (32.0% of ini-
−
3
accumulated for 300 min in AO-H O , it was released throughout
2 2
−3
−
less time, in about 120-180 min, and in a similar way by EF and
tial N) and 14.1 mg dm
NO3 (16.5% of initial N) were found,
•
•
whereas the latter one yielded 5.5 mg dm NH4+ (22.0% of ini-
−
3
PEF, suggesting that OH rather than BDD( OH) favors the desul-
fonation reactions of intermediates. In all these treatments, a final
SO4
−
3
−
tial N) and 9.2 mg dm NO3 (10.5% of initial N). The TN of both
final solutions corresponded to 60.0% of initial N, a value superior
to 48.5% and 32.5% obtained as sum of the released inorganic ions
in EF and PEF, respectively. This indicates again the presence of
2−
−3
content of 29.6 mg dm corresponding to 100% of initial S
was obtained, thereby corroborating that the two sulfonic groups
of AR1 are completely released as SO42 ion.
−

2
86
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
Fig. 8. Proposed reaction sequence for the mineralization of Acid Red 1 in acidic medium by AO-H2O2, EF and PEF with a BDD anode. The compounds in parentheses are
hypothesized.

X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
287
•
3
.5. Proposed reaction pathway for Acid Red 1 mineralization
OH over the aromatic products generated. However, persistent N-
derivatives of low molecular mass are always formed avoiding the
total mineralization of AR1.
Based on the main intermediates detected in the BDD/air-
diffusion cell, Fig. 8 presents a reaction sequence proposed for
AR1 mineralization in acidic medium by all the EAOPs with BDD.
While all compounds can be always oxidized by BDD( OH), OH is
considered as the main oxidant of aromatics in EF and PEF. Their
hydroxylated and desulfonated derivatives are not specified for
simplification. In EF and PEF, Fe(III)-carboxylate species are always
4
. Conclusions
•
•
The effectiveness of EAOPs using a Pt or BDD anode and an air-
diffusion cathode to degrade AR1 has been corroborated in terms
of the decolorization efficiency and mineralization rate at pH 3.0.
The oxidation power of these methods increased in the sequence
AO-H O < EF < PEF regardless of the anode used. This behavior is
explained by the quicker oxidation of aromatic pollutants with OH
in the bulk in the two latter methods than with M( OH) formed at
the corresponding anode surface. The use of a BDD anode compared
to Pt always yielded faster decolorization and higher mineralization
rate, because of the superior oxidation power of BDD( OH) than
formed, only being detailed for acids 19-21. Parallel slower reac-
•
tions with other weaker ROS (H O2 and HO2 ) are also feasible.
2
2
2
The sequence is initiated by the oxidation of 1 to yield either
the compound 2 with release of 13 or the cleavage of the azo
bond to produce the naphthalenic compound 5 losing 14. Further
degradation of 2 leads to the nitro-compound 3 or its deaminated
compound 4, whereas the subsequent deamination of 5 gives the
compound 6. The oxidation of 3, 4 and 6 yields the naphthalenic
derivatives 7 and 8, the phthalic acid compounds 9 and 10 and the
benzenesulfonic acid 11, which is converted into the aminophenol
•
•
•
•
Pt( OH). Faster and similar decolorization efficiency was found for
both EF and PEF processes due to the high photostability of AR1
under UVA irradiation. The UV-vis spectra of their electrolyzed
solutions with BDD showed the formation of colored aromatic
intermediates, which were detected by LC-MS. Almost total min-
eralization was only achieved in PEF due to the rapid photolysis
of persistent intermediates like Fe(III)-carboxylate complexes. The
increase in j always enhanced the decolorization and mineraliza-
1
9
2 with loss of 13. The cleavage of the benzenic ring of compounds
, 10, 12 and 14 leads to a mixture of acids 15-18, which are finally
converted into the acids 20 and 21. Parallel oxidation of 13 gives
the acid 19. The ultimate acids 19-21 can then be slowly mineral-
•
ized with BDD( OH) in AO-H O , whereas their Fe(III) complexes
2
2
•
can be slowly destroyed primordially with BDD( OH) in EF, but
quickly photolyzed by UVA light in PEF. NH4 , NO3 and SO4
•
tion processes because of the greater production of BDD( OH) and
+
−
2−
ions
•
OH, but the concomitant increase in rate of the parasitic reactions
are released during the deamination, denitration and desulfonation
reactions, respectively, involved in the degradation process.
of these radicals decreased the mineralization current efficiency.
Under all experimental conditions, the PEF process with BDD was
the most powerful EAOP for AR1 degradation. However, the degra-
dation kinetics of real dying effluents could be enhanced and be
more cost-effective by using sunlight as inexpensive and renew-
able energy source, with high irradiation intensity and a wider
applicable range of wavelengths than artificial UVA light [15]. A
total of 11 aromatic intermediates, 15 hydroxylated compounds
and 13 desulfonated derivatives were identified by LC-MS and 7
short-linear carboxylic acids were detected and quantified by ion-
exclusion HPLC. While all the initial S of AR1 was converted into
3.6. Relationship between decolorization, mineralization and
generated products
The above findings indicate a close relationship between decol-
orization, mineralization and accumulation of products for AR1
degradation. The slow decolorization found by AO-H O due to the
2
2
slow cleavage of the azo bond suggests a higher recalcitrant charac-
ter of AR1 than its aromatic products, which are poorly accumulated
as deduced from the UV-vis spectra shown in Fig. 2a. Results of
2−
−
SO4 ion, the initial N was mineralized preferentially as NO3 ion
Fig. 1a highlights that AR1 mineralization in AO-H O2 is enhanced
in AO-H O2 and as NH4+ ion in EF and PEF. A large loss of N from
2
2
when decolorization and, consequently, destruction of aromatic
intermediates are already significant, which makes more effective
solution was found for the two latter methods and the formation
of very persistent N-derivatives of low molecular mass prevented
reaching total mineralization in PEF. A reaction sequence involving
all the products detected is proposed for AR1 mineralization by the
EAOPs with BDD. The relationship between decolorization, miner-
alization and products formed during azo dye degradation is finally
discussed.
•
the attack of BDD( OH) to mineralize the generated short-linear
aliphatic compounds like carboxylic acids. In the indirect electro-
chemical methods based on the Fenton’s reaction chemistry, the
•
more potent action of OH in the bulk favors both the decoloriza-
tion and mineralization at the starting steps, as can be seen in
Figs. 1a and 1b, leading to a much higher accumulation of inter-
mediates during their degradation processes. The UV-vis spectra
recorded in Fig. 2b during the EF treatment suggests the accu-
mulation of colored aromatic products, identified as compounds
Acknowledgements
The authors thank financial support from MICINN (Mini-
2-4 together their hydroxylated and desulfonated derivatives by
sterio de Ciencia
CTQ2010-16164/BQU, co-financed with FEDER funds. A.M.S. Solano
acknowledges her Doctoral fellowships from CAPES and CNPq
e Innovación, Spain) under the project
LC-MS (see Table 2). These colored oxidation products are also
expected in AO-H O , but their low concentration in the medium
2
2
does not affect significantly the shape of the peaks in its UV-vis
spectra. A significant decolorization is also needed in EF and PEF
to begin an effective mineralization, primordially at the lowest j
(
doutorado sanduíche under the program Ciências sem fronteiras).
S. Garcia-Segura thanks the Doctoral grant awarded from MEC
Ministerio de Educación y Ciencia, Spain).
(
•
•
values where less amounts of oxidant BDD( OH) and OH are gen-
erated, as deduced from Figs. 3b and 5b. The recalcitrant character
of final Fe(III)-carboxylate complexes can be clearly observed at
long electrolysis time in EF, which caused a dramatic fall in MCE.
The quick photodecarboxylacion of such complexes, along with the
photolysis of other intermediates, explains that PEF is the most
powerful EAOP for AR1 mineralization. Inorganic ions are usually
released at the beginning of all process, with pre-eminent loss of
References
[
1] UNESCO, The United Nations World Water Decelopment Report 4, Volume 1:
Managing Water Report under Uncertainty and Risk, 2012.
[2] C.A. Martínez-Huitle, E. Brillas, Decontamination of wastewaters containing
synthetic organic dyes by electrochemical methods: A general review, Appl.
Catal. B: Environ. 87 (2009) 105.
[
3] T. Robinson, G. McMullan, R. Marchant, P. Nigam, Remediation of dyes in textile
effluent: a critical review on current treatment technologies with a proposed
alternative, Biores. Technol. 77 (2001) 247.
−
+
the initial N as NO3 ion in AO-H O and as NH ion in EF and PEF.
2
2
4
•
This points again to the lower oxidation ability of BDD( OH) than

2
88
X. Florenza et al. / Electrochimica Acta 142 (2014) 276–288
[
4] K.P. Sharma, S. Sharma, S. Sharma, P.K. Singh, S. Kumar, R. Grover, P.K. Sharma,
A comparative study on characterization of textile wastewaters (untreated and
treated) toxicity by chemical and biological tests, Chemosphere 69 (2007) 48.
5] S.M.A.G. Ulson de Souza, E. Forgiarini, A.A. Ulson de Souza, Toxicity of textile
dyes and their degradation by the enzyme horseradish peroxidase (HRP), J.
Hazard. Mater. 147 (2007) 1073.
[28] A. Dhaouadi, N. Adhoum, Degradation of paraquat herbicide by electrochemical
advanced oxidation methods, J. Electroanal. Chem. 637 (2009) 33.
[29] N. Oturan, E. Brillas, M.A. Oturan, Unprecedent total mineralization of atrazine
and cyanuric acid by anodic oxidation and electro-Fenton with a boron-doped
diamond anode, Environ. Chem. Lett. 10 (2012) 165.
[30] F.E.F. Rêgo, A.M.S. Solano, I.C. da Costa Soares, D.R. da Silva, C.A. Martínez-Huitle,
M. Panizza, Application of electro-Fenton process as alternative for degradation
of Novacron Blue dye, J. Environ. Chem. Eng. 2 (2014) 875.
[31] A. Wang, J. Qu, H. Liu, J. Ru, Mineralization of an azo dye Acid Red 14 by
photoelectro-Fenton process using an activated carbon fiber cathode, Appl.
Catal. B: Environ. 84 (2008) 393.
[
[
[
6] M. Panizza, G. Cerisola, Direct and mediated anodic oxidation of organic pol-
lutants, Chem. Rev. 109 (2009) 6541.
7] E. Brillas, I. Sirés, M.A. Oturan, Electro-Fenton process and related electrochem-
ical technologies based on Fenton’s reaction chemistry, Chem. Rev. 109 (2009)
6
570.
[
[
8] I. Sirés, E. Brillas, Remediation of water pollution caused by pharmaceutical
residues based on electrochemical separation and degradation technologies: A
review, Environ. Int. 40 (2012) 212.
9] L. Feng, E.D. van Hullebusch, M.A. Rodrigo, G. Esposito, M.A. Oturan, Removal of
residual anti-inflammatory and analgesic pharmaceuticals from aqueous sys-
tems by electrochemical advanced oxidation processes. A review, Chem. Eng.
J. 228 (2013) 944.
[32] A. Özcan, Y. Sahin, A.S. Koparal, M.A. Oturan, Carbon sponge as a new cathode
material for the electro-Fenton process. Comparison with carbon felt cath-
ode and application to degradation of synthetic dye Basic Blue 3 in aqueous
medium, J. Electroanal. Chem. 616 (2008) 71.
[33] C. Flox, P.L. Cabot, F. Centellas, J.A. Garrido, R.M. Rodríguez, C. Arias, E. Brillas,
Solar photoelectro-Fenton degradation of cresols using a flow reactor with a
boron-doped diamond anode, Appl. Catal. B: Environ. 75 (2007) 17.
[34] E.J. Ruiz, A. Hernández-Ramírez, J.M. Peralta-Hernández, C. Arias, E. Brillas,
Application of solar photoelectro-Fenton technology to azo dyes mineraliza-
[
[
10] G.R. de Oliveira, C.K.C. de Araújo, C.A. Martínez-Huitle, D.R. da Silva, Comple-
mentary mechanism model for the electrochemical mineralization, Curr. Org.
Chem. 16 (2012) 1957.
11] S. Garcia-Segura, F. Centellas, C. Arias, J.A. Garrido, R.M. Rodríguez, P.L. Cabot,
E. Brillas, Comparative decolorization of monoazo, diazo and triazo dyes by
electro-Fenton process, Electrochim Acta 58 (2011) 303.
2
+
tion: Effect of current density, Fe and dye concentration, Chem. Eng. J. 171
(2011) 385.
[35] S. Garcia-Segura, A. El-Ghenymy, F. Centellas, R.M. Rodríguez, C. Arias, J.A. Gar-
rido, P.L. Cabot, E. Brillas, Comparative degradation of the diazo dye Direct
Yellow 4 by electro-Fenton, photoelectro-Fenton and photo-assisted electro-
Fenton, J. Electroanal. Chem. 681 (2012) 36.
[
[
12] M. Panizza, M.A. Oturan, Degradation of Alizarin Red by electro-Fenton process
using a graphite-felt cathode, Electrochim. Acta 56 (2011) 7084.
13] M.G. Tavares, L.V.A. da Silva, A.M.S. Solano, J. Tonholo, C.A. Martínez-Huitle,
C.L.P.S. Zanta, Electrochemical oxidation of Methyl Red using Ti/Ru0.3Ti0.7O2
and Ti/Pt anodes, Chem. Eng. J. 204-206 (2012) 141.
14] C. Ramírez, A. Salda n˜ a, B. Hernández, R. Acero, R. Guerra, S. Garcia-Segura, E.
Brillas, J.M. Peralta-Hernández, Electrochemical oxidation of methyl orange azo
dye at pilot flow plant using BDD technology, J. Ind. Eng. Chem. 19 (2013) 571.
15] F.C. Moreira, S. Garcia-Segura, V.J.P. Vilar, R.A.R. Boaventura, E. Brillas, Decol-
orization and mineralization of Sunset Yellow FCF azo dye by anodic oxidation,
electro-Fenton, UVA photoelectro-Fenton and solar photoelectro-Fenton pro-
cesses, Appl. Catal. B: Environ. 142-143 (2013) 877.
[36] A. El-Ghenymy, N. Oturan, M.A. Oturan, J.A. Garrido, P.L. Cabot, F. Cen-
tellas, R.M. Rodríguez, E. Brillas, Comparative electro-Fenton and UVA
photoelectro-Fenton degradation of the antibiotic sulfanilamide using a stirred
BDD/air-diffusion tank reactor, Chem. Eng. J. 234 (2013) 115.
[37] M. Zarei, A.R. Khataee, R. Ordikhani-Seyedlar, M. Fathinia, Photoelectro-Fenton
combined with photocatalytic process for degradation of an azo dye using
supported TiO2 nanoparticles and carbon nanotube cathode: Neural network
modeling, Electrochim. Acta 55 (2010) 7259.
[38] A. Khataee, A. Khataee, M. Fathinia, B. Vahid, S.W. Joo, Kinetic modeling of
photoassisted-electrochemical process for degradation of an azo dye using
boron-doped diamond anode and cathode with carbon nanotubes, J. Ind. Eng.
Chem. 19 (2013) 1890.
[39] K. Cruz-González, O.A. Torres-Lopez, M. García-León, E. Brillas, A. Hernández-
Ramírez, J.M. Peralta-Hernández, Optimization of electro-Fenton/BDD process
for decolorization of a model azo dye wastewater by means of response surface
methodology, Desalination 286 (2012) 63.
[40] E. Isarain-Chávez, C. de la Rosa, L.A. Godínez, E. Brillas, J.M. Peralta-Hernández,
Comparative study of electrochemical water treatment processes for a tannery
wastewater effluent, J. Electroanal. Chem. 713 (2014) 62.
[41] S. Garcia-Segura, E. Brillas, Mineralization of the recalcitrant oxalic and oxamic
acids by electrochemical advanced oxidation processes using a boron-doped
diamond anode, Water Res. 45 (2011) 2975.
[42] E. Guinea, J.A. Garrido, R.M. Rodríguez, P.-L. Cabot, C. Arias, F. Centellas, E.
Brillas, Degradation of the fluoroquinolone enrofloxacin by electrochemical
advanced oxidation processes based on hydrogen peroxide electrogeneration,
Electrochim. Acta 55 (2010) 2101.
[43] H. Zhang, Y. Li, X. Zhong, X. Ran, Application of experimental design method-
ology to the decolorization of Orange II using low iron concentration of
photoelectro-Fenton process, Water Sci. Technol. 63 (2011) 1373.
[44] E. Isarain-Chávez, J.A. Garrido, R.M. Rodríguez, F. Centellas, C. Arias, P.-L. Cabot,
E. Brillas, Mineralization of metoprolol by electro-Fenton and photoelectro-
Fenton processes, J. Phys. Chem. A 115 (2011) 1234.
[
[
[
[
16] M. Panizza, A. Kapalka, Ch. Comninellis, Oxidation of organic pollutants on BDD
anodes using modulated current electrolysis, Electrochim. Acta 53 (2008) 2289.
17] M.A. Rodrigo, P. Ca n˜ izares, A. Sánchez-Carretero, C. Sáez, Use of conductive-
diamond electrochemical oxidation for wastewater treatment, Catal. Today 151
(
2010) 173.
[
[
[
[
[
[
18] E. Brillas, S. Garcia-Segura, M. Skoumal, C. Arias, Electrochemical incineration
of diclofenac in neutral aqueous medium by anodic oxidation using Pt and
boron-doped diamond anodes, Chemosphere 79 (2010) 605.
19] E. Tsantaki, T. Velegraki, A. Katsaounis, D. Mantzavinos, Anodic oxidation of
textile dyehouse effluents on boron-doped diamond electrode, J. Hazard. Mater.
2
07-208 (2012) 91.
20] M.B. Ferreira, J.H.B. Rocha, J.V. de Melo, C.A. Martinez-Huitle, M.A. Quiroz Alfaro,
Use of a dual arrangement of flow cells for electrochemical decontamination
of aqueous solutions containing synthetic dyes, Electrocatalysis 4 (2013) 274.
21] A. Kapalka, G. Fóti, C. Comninellis, Kinetic modeling of the electrochemical
mineralization of organic pollutants for wastewater treatment, J. Appl. Elec-
trochem. 38 (2008) 7.
22] M. Hamza, R. Abdelhedi, E. Brillas, I. Sirés, Comparative electrochemical degra-
dation of the triphenylmethane dye Methyl Violet with boron-doped diamond
and Pt anodes, J. Electroanal. Chem. 627 (2009) 41.
23] E.B. Cavalcanti, S. Garcia-Segura, F. Centellas, E. Brillas, Electrochemical incin-
eration of omeprazole in neutral aqueous medium using a platinum or
boron-doped diamond. Degradation kinetics and oxidation products, Water
Res. 47 (2013) 1803.
24] L. Ciríaco, C. Anjo, J. Correia, M.J. Pacheco, A. Lopes, Electrochemical degradation
of ibuprofen on Ti/Pt/PbO2 and Si/BDD electrodes, Electrochim. Acta 54 (2009)
[45] L.C. Almeida, S. Garcia-Segura, C. Arias, N. Bocchi, E. Brillas, Electrochemical
mineralization of the azo dye Acid Red 29 (Chromotrope 2R) by photoelectro-
Fenton process, Chemosphere 89 (2012) 751.
[
[
[
[
[46] E. Brillas, M.A. Ba n˜ os, S. Camps, C. Arias, P.L. Cabot, J.A. Garrido, R.M. Rodríguez,
2
+
2+
1
464.
Catalytic effect of Fe , Cu and UVA light on the electrochemical degradation
of nitrobenzene using an oxygen-diffusion cathode, New J. Chem. 28 (2004)
314.
25] C. Flox, C. Arias, E. Brillas, A. Savall, K. Groenen-Serrano, Electrochemical
incineration of cresols: A comparative study between PbO2 and boron-doped
diamond anodes, Chemosphere 74 (2009) 1340.
26] E. Rosales, M. Pazos, M.A. Longo, M.A. Sanromán, Electro-Fenton decoloration
of dyes in a continuous reactor: A promising technology in colored wastewater
treatment, Chem. Eng. J. 155 (2009) 62.
[47] M. Stylidi, D.I. Kondarides, X.E. Verykios, Pathways of solar light-induced pho-
tocatalytic degradation of azo dyes in aqueous TiO2 suspensions, Appl. Catal.
B: Environ. 40 (2003) 271.
[48] D. Vogna, R. Marotta, A. Napolitano, M. d’Ischia, Advanced oxidation chemistry
of paracetamol. UV/H2O2-induced hydroxylation/degradation pathways and
15 N-aided inventory of nitrogenous breakdown products, J. Org. Chem. 67
(2002) 6143.
27] A. Özcan, M.A. Oturan, N. Oturan, Y. S¸ ahin, Removal of Acid Orange 7 from water
by electrochemically generated Fenton’s reagent, J. Hazard. Mater. 163 (2009)
1
213.