ArticlePDF AvailableLiterature Review

Electrochemical Advanced Oxidation Processes: Today and Tomorrow. A Review

Authors:

Abstract and Figures

In recent years, new advanced oxidation processes based on the electrochemical technology, the so-called electrochemical advanced oxidation processes (EAOPs), have been developed for the prevention and remediation of environmental pollution, especially focusing on water streams. These methods are based on the electrochemical generation of a very powerful oxidizing agent, such as the hydroxyl radical ((•)OH) in solution, which is then able to destroy organics up to their mineralization. EAOPs include heterogeneous processes like anodic oxidation and photoelectrocatalysis methods, in which (•)OH are generated at the anode surface either electrochemically or photochemically, and homogeneous processes like electro-Fenton, photoelectro-Fenton, and sonoelectrolysis, in which (•)OH are produced in the bulk solution. This paper presents a general overview of the application of EAOPs on the removal of aqueous organic pollutants, first reviewing the most recent works and then looking to the future. A global perspective on the fundamentals and experimental setups is offered, and laboratory-scale and pilot-scale experiments are examined and discussed.
Content may be subject to copyright.
ELECTROCHEMICAL ADVANCED OXIDATION PROCESSES FOR REMOVAL OF TOXIC/PERSISTENT ORGANIC POLLUTANTS FROM WATER
Electrochemical advanced oxidation processes: today
and tomorrow. A review
Ignasi Sirés &Enric Brillas &Mehmet A. Oturan &
Manuel A. Rodrigo &Marco Panizza
Received: 19 February 2014 /Accepted: 10 March 2014 /Published online: 2 April 2014
#Springer-Verlag Berlin Heidelberg 2014
Abstract In recent years, new advanced oxidation processes
based on the electrochemical technology, the so-called electro-
chemical advanced oxidation processes (EAOPs), have been
developed for the prevention and remediation of environmen-
tal pollution, especially focusing on water streams. These
methods are based on the electrochemical generation of a very
powerful oxidizing agent, such as the hydroxyl radical (
OH)
in solution, which is then able to destroy organics up to their
mineralization. EAOPs include heterogeneous processes like
anodic oxidation and photoelectrocatalysis methods, in which
OH are generated at the anode surface either electrochemically
or photochemically, and homogeneous processes like electro-
Fenton, photoelectro-Fenton, and sonoelectrolysis, in which
OH are produced in the bulk solution. This paper presents a
general overview of the application of EAOPs on the removal
of aqueous organic pollutants, first reviewing the most recent
works and then looking to the future. A global perspective on
the fundamentals and experimental setups is offered, and
laboratory-scale and pilot-scale experiments are examined
and discussed.
Keywords EAOPs .Anodic oxidation .Electro-Fenton .
Photoelectrocatalysis .Photoelectro-Fenton .
Sonoelectrochemistry .Water treatment
Abbreviations
ACP 3-Amino-6-chloropyridazine
ADE Air diffusion electrode
AMI 3-Amino-5-methylisoxazole
AO Anodic oxidation
AOP Advanced oxidation process
BDD Boron-doped diamond
BZQ p-Benzoquinone
CF Carbon felt
CNT Carbon nanotube
COD Chemical oxygen demand (mg of oxygen L
1
)
DSA Dimensionally stable anode
e
Electron
e
CB
Electron in the conduction band
E
anod
Anodic potential (V)
EAOP Electrochemical advanced oxidation process
E
cat
Cathodic potential (V)
EF Electro-Fenton
GC-
MS
Gas chromatography coupled to mass spectrometry
hPlanck constant (6.626×10
34
m
2
kg/s)
HPLC High-performance liquid chromatography
h
+VB
Positively charged vacancy or hole in the valence
band
MMO Mixed metal oxides
PEC Photoelectrocatalysis
PEF Photoelectro-Fenton
R Organic compound
ROS Reactive oxygen species
Responsible editor: Philippe Garrigues
I. Sirés :E. Brillas
Laboratori dElectroquímica dels Materials i del Medi Ambient,
Departament de Química Física, Facultat de Química, Universitat de
Barcelona, Martí i Franquès 1-11, 08028 Barcelona, Spain
M. A. Oturan
Laboratoire Géomatériaux et Environnement (LGE), Université
Paris-Est, EA 4508, 5 bd Descartes, 77454 Marne-la-Vallée Cedex 2,
France
M. A. Rodrigo
Department of Chemical Engineering, Faculty of Chemical Sciences
and Technologies, Universidad de Castilla La Mancha, Edificio
Enrique Costa, Campus Universitario s/n, 13071 Ciudad Real, Spain
M. Panizza (*)
Department of Civil, Chemical and Environmental Engineering,
University of Genoa, P.le J.F. Kennedy 1, 16129 Genoa, Italy
e-mail: marco.panizza@unige.it
Environ Sci Pollut Res (2014) 21:83368367
DOI 10.1007/s11356-014-2783-1
RVC Reticulated vitreous carbon
SE Sonoelectrochemistry
SPEF Solar photoelectro-Fenton
TOC Total organic carbon (mg of carbon L
1
)
US Ultrasounds
))) Ultrasounds
Greek symbols
λWavelength (nm)
νFrequency (Hz)
Introduction
In recent decades, the rapid growth of public awareness about
environmental problems has induced many governments to
introduce legislation that prescribes and limits the emission of
pollutants. This has been reflected in a notable increase in both
research and the number of businesses concerned with the
treatment of industrial effluents. Because of the extremely
diverse features of industrial waste that usually contains a
mixture of organic and inorganic compounds, no universal
strategy of reclamation is feasible and it mainly depends on
the nature and concentration of pollutants (Fig. 1). As to the
treatment of effluents polluted with organic compounds, bio-
logical oxidation is certainly the cheapest process, but the
presence of toxic or biorefractory molecules may hinder this
approach. The traditional incineration method poses problems
of emission if the treatment conditions are not perfectly con-
trolled, and above all, it can be conveniently applied only for
concentrated solutions (Fig. 1). Chemical oxidation using
chlorine, ozone, or hydrogen peroxide is currently used for
the treatment of biorefractory contaminants or at least to
decompose them into harmless or biodegradable products.
However, in some reactions, the intermediate products remain
in the solution and they may entail a similar or even higher
toxicity than the initial compounds. In these cases, the pollut-
ants can be removed using a special class of oxidation tech-
nique known as advanced oxidation processes (AOPs).
Within the framework of liquid polluted streams, AOPs can
be broadly defined as aqueous phase oxidation methods based
on the intermediacy of highly reactive species (primarily but
not exclusively) in the mechanisms leading to the destruction
of the target pollutant. The hydroxyl radical (
OH) is a pow-
erful oxidant (Table 1) which is able to nonselectively destroy
most organic and organometallic contaminants until their
complete mineralization into CO
2
, water, and inorganic ions.
These radicals react rapidly with organics (R) mainly by the
abstraction of a hydrogen atom (aliphatics) or the addition on
an unsaturated bond (aromatics) to initiate a radical oxidation
chain:
RHþOHH2OþRð1Þ
RþO2ROOð2Þ
ROOþRHROOHþRð3Þ
ArHþOHArH OHðÞ
ð4Þ
ArH OHðÞ
þO2ArH OHðÞOO½
ð5Þ
ArH OHðÞOO½
ArH OHðÞþHO2
ð6Þ
As summarized in Table 2,alargenumberofAOPshas
been developed, including nonphotochemical and photo-
chemical methods. The AOPs are successfully applied mainly
for the treatment of wastewaters, but they are also used in
many fields including groundwater treatment, soil remedia-
tion, municipal wastewater sludge conditioning, as well as
odor and taste removal from drinking water.
In recent years, new AOPs based on the electrochemical
technology, i.e., the so-called electrochemical advanced
0.001 0.01 0.1 1 10 100 1000
COD (
g
/L)
Biological Treatment
Advanced Oxidation Processes
Electrochemical Advanced Oxidation Processes
Chemical oxidation
Incineration
Fig. 1 Applicability of water treatment technologies based on the
amount of organic load. Adapted from Fryda et al. (2003)
Table 1 Standard potential of some oxidizing species
Oxidizing agent Standard potential (V vs. SHE)
Oxygen (molecular) 1.23
Chlorine dioxide 1.27
Chlorine 1.36
Ozone 2.08
Oxygen (atomic) 2.42
Hydroxyl-radical 2.80
Fluorine 3.06
Positively charged hole on TiO
2
3.2
Environ Sci Pollut Res (2014) 21:83368367 8337
oxidation processes (EAOPs), have been developed (Fryda
et al. 2003; Martínez-Huitle and Ferro 2006; Brillas et al.
2009; Panizza and Cerisola 2009a; Sirés and Brillas 2012).
The EAOPs provide several advantages for the prevention and
remediation of pollution problems because the electron is a
clean reagent. Other advantages include high energy efficien-
cy, amenability to automation, easy handling because of the
simple equipment required, safety because they operate under
mild conditions (room temperature and pressure), and versa-
tility because they can be applied to effluents with chemical
oxygen demand (COD) in the range of 0.1 to 100 g L
1
(Fig. 1). The main drawbacks of some of these technologies
include the costs related to the electrical supply, the low
conductance of many wastewaters that require the addition
of electrolytes, and the loss of activity and shortening of the
electrode lifetime by fouling due to the deposition of organic
material on their surface. More specific advantages and dis-
advantages of the technologies will be discussed later.
Key EAOPs include anodic oxidation (AO), in which
heterogeneous
OH are generated at the anode surface, as well
as electro-Fenton (EF), photoelectro-Fenton (PEF), and
sonoelectrochemistry (SE), in which homogeneous
OH are
produced in the bulk solution. It is also possible to couple
various EAOPs such as the AO with EF, PEF, or SE to
produce both heterogeneous and homogeneous
OH.
The growing interest of academic and industrial communi-
ties in EAOPs is reflected in the high number of publications
in peer-reviewed journals, patents, and international confer-
ences. Figure 2a illustrates that more than 50 % of the papers
published in the last 3 years are devoted to the AO, in partic-
ular using the innovative boron-doped diamond (BDD) anode.
Many papers studied the EF and PEF processes, while only
few researches are focused in the less conventional but evolv-
ing SE processes.
The efficiency and flexibility of the EAOPs have been
proven by the wide diversity of effluents treated, as shown
in Fig. 2b, including either synthetic solutions containing
phenols (Cañizares et al. 2003,2004; Polcaro et al. 2003;
Panizza and Cerisola 2009b), dyes (Panizza and Cerisola
2008; Martínez-Huitle and Brillas 2009; Rodriguez et al.
2009; Moreira et al. 2013), pesticides (Polcaro et al. 2005;
Flox et al. 2007;Oturanetal.2008; Panizza et al. 2008; Borràs
et al. 2013), and drugs (Sirés et al. 2006a,2007a;Isarain-
Chávez et al. 2010) or real/industrial effluents (Panizza et al.
2006; Cañizares et al. 2006,2007a; Malpass et al. 2008;
Panizza and Cerisola 2010). Despite the large number of
publications on the EAOPs and the very good results obtained
in laboratory-scale tests, their practical application for the
treatment of organic pollutants is still insufficient. But nowa-
days, given the intensive investigations that have improved
Table 2 Main AOPs and related reactions involving the production of
OH
Reactions
Dark AOPs
Ozone at elevated pH 3O
3
+OH
+H
+
2
OH+ 4O
2
Ozone+hydrogen peroxide 2O
3
+H
2
O
2
2
OH+ 3O
2
Ozone+catalyst O
3
+Fe
2+
+H
2
OFe
3+
+OH
+
OH+ O
2
Fenton Fe
2+
+H
2
O
2
Fe
3+
+OH
+
OH
Photo-assisted AOPs
Ozone/UV O
3
+H
2
O+hνO
2
+H
2
O
2
Hydrogen peroxide/UV H
2
O
2
+hν2
OH
Ozone/H
2
O
2
/UV The addition of H
2
O
2
to the O
3
/UV process
accelerates the decomposition of ozone,
which results in an increased rate of
OH generation
Photo-Fenton Fe
2+
+H
2
O
2
+hνFe
3+
+OH
+
OH
Fe(OH)
2+
+hνFe
2+
+
OH
Fe(OOCR)
2+
+hνFe
2+
+CO
2
+R
Heterogeneous
photocatalysis
(TiO
2
/UV)
TiO
2
+hνTiO
2
(e
+h
+
)
h
+
+H
2
O
OH+H
+
e
+O
2
O
2
0
10
20
30
40
50
60
AO EF PEC SE
snoitacilbupforebmuN%
0
5
10
15
20
25
30
% Number of publications
a
b
Fig. 2 Percentage of publications devoted to the EAOPs in the last
3 years. Distribution by aelectrochemical techniques: anodic oxidation
(AO), electro-Fenton (EF), photoelectrocatalysis (PEC), and
sonoelectrochemistry (SE)andbtype of residue
8338 Environ Sci Pollut Res (2014) 21:83368367
the electrocatalytic activity and stability of electrode materials,
optimized reactor geometry, and deepened knowledge about
reactor hydrodynamics, the EAOPs have reached an advanced
stage of development and, recently, some pilot-scale or full-
scale plants have been effectively commercialized for the
disinfection and purification of wastewater polluted with or-
ganic compounds.
To date, one of the mostdeveloped large-scale applications
of EAOPs is the automated disinfection of swimming pool
water using BDD anodes. In this field, dedicated products
such as Oxineo® and Sysneo® have been developed for
private and public pools. Compared with the other disinfection
methods, these systems have the advantages that there is no
chlorine smell, no accumulation of chemicals in the pool, no
need of anti-algae, and there is a residual action to avoid
nonregular or jagged disinfections. Many of these systems
have been already installed in private pools all over the world
and several public pools and spas in Europe.
CONDIAS and Advanced Diamond Technologies Inc. de-
velop and supply equipment for EAOPs, sold with the trade-
mark of CONDIACELL® and Diamonox®, respectively,
which are based on AO with BDD anode. Typical applications
of these cells are (a) water disinfection and (b) industrial
wastewater treatment. Some details of the Diamonox® system
are reported in Fig. 3. For water disinfection, these cells
produce a mixture of oxygen-based agents, such as
OH and
ozone, directly by water electrolysis, providing high disinfec-
tion rate with low energy consumption, without the addition of
chemicals and they can either be used as a firewall or for
volume disinfection.
The treatment of industrial wastewater is based on the
production of
OH and other oxidants, such as chlorine,
(per)bromate, persulfate, ozone, hydrogen peroxide,
percarbonate, and others, directly on-site using only water,
salt, and energy. The advantage of industrial wastewater treat-
ment using these EAOPs is the possibility to degrade COD/
total organic carbon (TOC) from a value of several hundred
grams O
2
per liter to a minimum of a few milligrams O
2
per
liter or even micrograms O
2
per liter, with the reduction of all
organic water components by approximately 99 %. Some
other advantages related to these processes are the possibility
of combining EAOPs with common methods for wastewater
treatment to achieve an optimal cost-effective operation and
their easy modular adaptation and scale-up.
Another full-scale application of EAOPs is the EctoSys®,
which is an extremely efficient system that provides a reliable
and sustainable disinfection of the ballast water in an econom-
ical and ecological manner. By applying electricity to the
special electrodes, disinfectants are produced from the water
directly in the piping to eliminate bacteria and organisms. In
water with low salinity, the EctoSys® unit produces only
OH
as active substances, while in brackish water or seawater, it
produces short-living
OH and chlorine/bromine.
In 2007, a BDD electro-oxidation pilot plant (Fig. 4)was
installed in Marelo (Cantabria, Spain) for the treatment of
landfill leachates using traditional and advanced oxidation
technologies (Anglada et al. 2009,2010,2011; Urtiaga et al.
2009). The plant was constituted by an aerobic treatment
followed by chemical Fenton oxidation and a final AO treat-
ment. The latter consists of an electrochemical reactor with
BDD anodes of 1.05 m
2
. The raw leachate contained approx-
imately 2.8 g L
1
of TOC and 1.2 g L
1
of ammonia, and the
overall efficiency in the combined system was 99 % of organic
matter mineralization: 50 % of the initial TOC was degraded
in the aerobic treatment, 35 % in the Fenton process, and the
remaining 15 % in the final electro-oxidation step. The am-
monia removal efficiency was greater than 90 %, with 50 %
being due to the electrochemical treatment, since the Fenton
process was unable to reduce the ammonia concentration.
This paper presents a general overview of the application of
EAOPs on the removal of organic compounds, starting each
section with a revision of the very last years and then giving a
Fig. 3 Some details of the
Diamonox® system. Reprinted
with permission from Advanced
Diamond Technologies Inc.
Environ Sci Pollut Res (2014) 21:83368367 8339
look to the future. A global perspective on the fundamentals
and experimental setups is offered, and laboratory and pilot
plant experiments are examined and discussed.
Production of oxidants by electrolysis and their role
in mediated anodic oxidation
One of the key points to explain the high efficiencies reached
by EAOPs in the removal of organic pollutants is the under-
standing of the role depicted by mediated oxidation processes
in the overall oxidation carried out during the treatment.
Mediated oxidation in EAOPs can be understood as the oxi-
dation of pollutants contained in wastewater by the chemical
reaction between these compounds and the oxidants produced
previously on the electrode surfaces. Thus, AO does not only
lead to the direct oxidation of organic pollutants on the anode
surface but it also promotes the formation of huge amounts of
oxidants which can act not only on the surface of the elec-
trodes but extend the oxidation process to the bulk solution of
the treated waste (Panizza and Cerisola 2009a). The type and
extension of the production of oxidants depend on many
inputs, being the most relevant the electrode material and the
occurrence of suitable raw matter for the production of oxi-
dants in the wastewater. Their influence on the efficiency of
EAOPs is very important because the oxidation of pollutants
is extended from the vicinity of the electrode surface to the
bulk of the electrolyte. However, it should be taken into
account that these oxidants largely affect the mechanisms of
the oxidation of pollutants and, occasionally, they can lead to
the formation of unwanted intermediates or final stable
products. Sometimes, the species that promote the formation
of oxidants are not contained in the wastewater but added as
reagents, which results in well-known and very effective
processes. One of the most interesting examples is the treat-
ment of wastes with Ag(II), whose formation was demonstrat-
ed to be very effective with conductive diamond electrodes
(Panizza et al. 2000).
However, the most referenced example of mediated elec-
trochemical oxidation arises from the effect of chlorides on the
oxidation of organics. Chlorides are commonly contained in
most wastewater flowstreams and they are known to be easily
oxidized to chlorine by many types of anode materials (Eq. 7).
This gaseous oxidant diffuses into the wastewater and forms
hypochlorite and chloride in the reaction medium by dispro-
portionation (Eqs. 8and 9). Deprotonation of hypochlorous
acid produces hypochlorite (Eq. 10). Since hypochlorite is the
primary final product, in literature it is common to find the
direct transformation of chloride into hypochlorite instead of
the complete set of reactions (Eq. 11). However, the oxidation
in that media is carried out by a mixture of reagents and the
particular concentration of each species depends on the con-
centration and pH.
2ClCl2þ2eð7Þ
Cl2þH2OHClO þHþþClacidic mediumðÞð8Þ
Cl2þ2OH
ClOþClþH2OalkalinemediumðÞð9Þ
HClO þOHClOþH2Oð10Þ
ClþH2OClOþ2H
þþ2eð11Þ
The resulting mixture (chlorine, hypochlorite, and
hypochlorous acid) is highly reactive with many organics,
being efficient for their mineralization (Comninellis and
Nerini 1995; Panizza and Cerisola 2003). However, it is also
known to form many organochlorinated species as intermedi-
ates and final products that can be even more harmful than the
raw pollutant (Comninellis and Nerini 1995). Total depletion
of these species is frequently very difficult and even the
formation of low-molecular-weight products such as chloro-
form becomes a verysignificant problem because it could lead
to additional treatments, increasing significantly the total cost
of the remediation (Cañizares et al. 2003).
This is a negative and common example of the action of
oxidants that does not exclude the promotion of the mediated
oxidation processes in EAOPs, but alerts about some draw-
backs and limitations of use. Thus, even with chlorides, when
no organochlorinated by-products can be formed or when the
oxidation of pollutants such as cyanide is aimed, the formation
Fig. 4 EAOP pilot plant for the treatment of landfill leachate in Marelo
(Cantabria, Spain) (Anglada et al. 2009,2010,2011; Urtiaga et al. 2009)
8340 Environ Sci Pollut Res (2014) 21:83368367
of chlorinated oxidants is a significant advantage and allows
increasing the effectiveness of EAOPs. This is clearly ob-
served in Fig. 5(obtained from the data of Cañizares et al.
2005a), where the electrochemical oxidation of cyanide syn-
thetic wastes using sulfate-supporting and chlorine-supporting
electrolytes and three different anode materials is compared in
terms of COD removal. As it can be clearly observed, degra-
dation of cyanide is much faster when chloride is contained in
the synthetic waste. Likewise, it can be observed that anode
material does not behave as a simple sink of electrons but it
has a clear role in the reactivity of the system.
Nevertheless, and despite the fact that chlorine-mediated
oxidation is very well known, it is not the only case of
mediated oxidation processes and, of course, it is not the most
significant one. Thus, when the objective is focused on the
promotion of mediated oxidation, three important aspects
should be taken into account:
Direct electrochemical production of oxidants on the
anode surface from non-oxidant species contained in the
waste and the transport of these species toward the bulk
(wastewater). The raw matter for the production of oxi-
dants should be contained in the wastewater or dosed, and
typically, it can be an ion (i.e., chloride, sulfate, etc.), an
organic pollutant (acetic acid), dissolved gasses (oxygen),
or even water.
Effect of the raw oxidants produced electrodically on the
organic pollutants.
Activationof oxidants in the bulk, that is, the formation of
highly reactive species from poorly reactive oxidants.
These three points will be studied in the following sections.
Figure 6shows a comprehensive summary of the main pro-
cesses occurring during the oxidation of a pollutant contained
in wastewater. It includes the mass transport of species from
the bulk of the waste to the electrode surface and vice versa,
and the main oxidation mechanisms which will be explained
in this section, including direct oxidation and different types
of mediated oxidation that typically occur during EAOPs.
Direct electrochemical production of oxidants
For the formation of oxidants in electrochemical wastewater
treatment processes, three main points should be considered:
Direct oxidation of species on the anode surface, involv-
ing the formation of radical species that combine to
produce stable oxidants.
Oxidation of water to
OH and further attack of this
powerful oxidant to species promoting the formation of
radicals. Then, the combination of radicals leads to the
production of stable oxidants.
Reduction of oxygen to produce hydrogen peroxide on
the cathode surface.
In the following subsections, a detailed description of these
three mechanisms is going to be carried out.
Production of oxidants from direct oxidation processes
The first process pointed out in this subsection is the direct
oxidation of species on the anode surface with the subsequent
formation of radical species that combine to produce stable
oxidants. This is known to occur for many species present in
wastewater, in particular for chlorine and peroxo species, and
also for ferrates. With some anode materials like diamond or
PbO
2
coatings, the formation of radicals from anions such as
sulfate (Eq. 12), phosphate (Eq. 13), carbonate (Eq. 14), and
even acetic acid is promoted. By this mechanism, the forma-
tion of chloride radicals (Eq. 15) can also be explained
(Bergmann 2010). These processes are also known to occur
with other electrode materials such as platinum, but the effi-
ciency observed is much worse and concentrations produced
are quite insignificant toproduce an effect on the results of the
treatment process (Cañizares et al. 2009).
SO4
2SO4
ðÞ
þeð12Þ
PO4
3PO42

þeð13Þ
CO3
2CO3
ðÞ
þeð14Þ
ClClþeð15Þ
These radicals can combine based on the reactions shown
in Eqs. 16,17,18,and19, which explains the occurrence of
0
50
100
150
200
250
300
0 5 10 15
Q /Ah dm-3
COD / mg O2 dm-3
Fig. 5 Changes in the COD concentration during the electrochemical
oxidationof synthetic wastes polluted with cyanide (375 mg NaCN L
1
).
Supporting electrolyte: 0.05 M Na
2
SO
4
(black triangle DSA, black
diamond Pb/PbO
2
,black square p-Si-BDD) and 0.05 M NaCl (white
triangle DSA, white diamond Pb/PbO
2
,white square p-Si-BDD).
Adapted from Cañizares et al. (2005a)
Environ Sci Pollut Res (2014) 21:83368367 8341
the stable oxidants in the reaction media, including
peroxosulfates (Serrano et al. 2002), peroxophosphates
(Cañizares et al. 2005b), peroxocarbonates (Ruiz et al.
2009), and chlorine (Bergmann 2010).
SO4
ðÞ
þSO4
ðÞ
S2O82ð16Þ
PO4
2

þPO42

P2O84ð17Þ
CO3
ðÞ
þCO3
ðÞ
C2O62ð18Þ
ClþClCl2ð19Þ
Regarding chlorine, it is important to keep in mind
that the efficiency is particularly high in electrolysis
with some mixed metal oxide (MMO) anodes in which
this process is known to be promoted with respect to
the water oxidation (dimensionally stable anode [DSA]-
type electrodes). As mentioned previously, this process
produces a very active oxidation mixture, although it is
not always a good way to remove organic pollutants
because it promotes the formation of organochlorinated
intermediates and final products. In addition, this reac-
tion mixture can promote the formation of chlorates.
This process is not always electrochemically based, but
it is also chemically activated by a well-known dispro-
portionation reaction (Eq. 20) and it is stimulated with
the aging of the reaction mixture (Bolyard et al. 1992).
Chlorate is usually an unwanted product in the effluent
from an EAOP, and its formation could also prevent the
use of the EAOP technology in various applications.
3ClOClO3þ2Cl
ð20Þ
Regarding direct oxidation, an unresolved case is the for-
mation of ferrates, which have been used to explain the better
efficiencies of some EAOPs when iron is present in the treated
wastewater, even in electrocoagulation processes
(Phutdhawong et al. 2000). However, conditions used in
EAOPs are far from those required to produce them efficiently
from Eq. 21, and it is very difficult to explain this observation
in light of the present knowledge (Sáez et al. 2008).
Fe3þþ4H
2OFeO42þ8H
þþ3eð21Þ
Production of oxidants from hydroxyl radical mediated
processes
For the second process under discussion in this subsection, the
mediated production of oxidants by the action of
OH formed
electrodically, it is important to know more about the produc-
tion of such radicals in the reaction media.
OH is an interme-
diate in the AO of water to oxygen (Eq. 22)thatisrarely
detected inthe reaction media because it combineschemically
to components of the anode surface before forming oxygen. In
addition, it may be rapidly transformed into hydrogen
pollutant
e-
e-
pollutant
product
pollutant
DIRECT
ELECTROLYSIS
Oxidant
precursor
Radical
Stable oxidants
product
Oxidant
precursor
product
pollutant
Activated
oxidan ts
product
pollutant
product
product
MEDIATED
ELECTROLYSIS
BULK
ANODE LIGHT,
ULTRASOUND
OR CHEMICAL
ACTIVATION
Fig. 6 A conceptual approach to
mediated oxidation in EAOPs
8342 Environ Sci Pollut Res (2014) 21:83368367
peroxide (Eq. 23) and into hydroperoxyl radical (Eq. 24)
(Oturan et al. 2012).
H2OOH þHþþeð22Þ
2OHH2O2ð23Þ
OH þH2O2HO2
þH2Oð24Þ
The oxidation of water to oxygen is typically considered an
undesired side reaction in the electrochemical treatment of
pollutants because it seriously affects the efficiency of the
process, increasing significantly the operation costs (it leads
to a nonvaluable product). Nevertheless, in various works
(Kapałka et al. 2007,2008), it has been stated that this oxygen
may have a positive effect on the oxidation of organics be-
cause it may help induce the mineralization of organics,
contributing to explain the high efficiencies of the electrolysis
of organics. Anyway, there is a consensus in the role of
OH as
primary responsible for the high efficiencies of electrolyses
with some types of electrodes. At this point, the electrodes in
which
OH are not effective because they are not free on their
surface were defined as active electrodes in a pioneering work
of the group of Comninellis (Gandini et al. 1999). However,
for some electrodes classified as non-active, it has been pro-
posed that
OH cannot combine with the components of their
surface and then, during a veryshort time, they are available to
oxidize organics or other species such as anions contained in
waste (Marselli et al. 2003). This explains the formation of
radical species and the increased efficiency in the production
of oxidants when these anode materials are used, which even
push some research not only for the treatment of wastewater
(Panizza et al. 2001; Weiss et al. 2008a) but also for the
industrial production of these oxidants. Some of the reactions
promoted by
OH are summarized in Eqs. 25,26,and27 and
some of the products formed after the combination of radicals
can be explained by formerly stated Eqs. 16,17,18,and19
(Cañizares et al. 2009).
O2þ2OHO3þH2Oð25Þ
HSO4
þOHSO4
ðÞ
þH2Oð26Þ
HPO4
2þOHPO42

þH2Oð27Þ
In addition, some new oxidants can be formed by the
combination of
OH and the new radicals formed by their
action (Weiss et al. 2008b). Eqs. 28,29,and30 show some
examples.
SO4
ðÞ
þ
˙OHHSO5
ð28Þ
H2PO4
ðÞ
þOHH3PO5ð29Þ
CO3
þOHHCO4ð30Þ
This explains the formation of many new types of oxidants
with this non-active materials, and this also justifies the higher
concentration measured. It also allows explaining the forma-
tion of some undesirable species such as chlorates and per-
chlorates in the electrolysis of wastes containing chlorides
with non-active electrodes, as shown in Eqs. 31,32,33,and
34 (Sánchez-Carretero et al. 2011), and it may also shed light
on the formation of rare species such as perbromate, as shown
in Eq. 35 (Sáez et al. 2010a).
ClþOHClOþHþþeð31Þ
ClOþOHClO2þHþþeð32Þ
ClO2
þOHClO3þHþþeð33Þ
ClO3
þOHClO4þHþþeð34Þ
BrO3
þOHBrO4þHþþeð35Þ
Production of hydrogen peroxide on the cathode
The third way to produce oxidants in the reaction media is
very interesting because it complements very efficiently the
two approaches described previously. It consists in the pro-
duction of hydrogen peroxide by reduction of oxygen on the
cathode surface (Zhou et al. 2012). From the thermodynamic
point of view, hydrogen peroxide is less powerful than oxygen
but, kinetically, at room temperature, it is much more efficient.
This means that a way to enhance the efficiency of an EAOP is
promoting the formation of hydrogen peroxide by the other-
wise unproductive reaction at the cathode. Production of
hydrogen peroxide by the reduction of oxygen develops in
most cathode materials, but to increase efficiency, some three
boundary points are required, that is, points in which cathode,
water, and oxygen are in contact. For this reason, a special
type of porous cathode, known as gas diffusion cathode, is
employed for this application. Additional information about
this process is going to be given in other sections of this
manuscript where Fenton processes are described.
Environ Sci Pollut Res (2014) 21:83368367 8343
Effect of the raw oxidants produced electrochemically
on reaction performance
There are many works in the literature in which the effects of
the reaction media or small changes in the reaction media on
the treatment results are assessed. These works demonstrate
that reaction media have a large influence on results and that
unexpected results were sometimes obtained (Panizza and
Cerisola 2005; Sirés et al. 2006b; Martínez-Huitle and
Brillas 2009; Sirés and Brillas 2012). Some illustrative exam-
ples are highlighted, as follows:
(i) There are no significant differences in the oxidation of
organics in sulfate-supporting and phosphate-supporting
electrolytes using BDD anodes, as can be clearly ob-
served in Fig. 7, in which the influence of COD and
supporting electrolyte composition on the instantaneous
current efficiency is shown (Cañizares et al. 2005c). For
many years, phosphate media were used as an inert media
because peroxophosphate production was not expected.
The absence of significant differences between electrol-
yses carried out with organic solutions containing sulfates
and phosphates allow realizing that peroxophosphates
were produced efficiently, leading to the proposal of a
method for their manufacture (Cañizares et al. 2007b).
(ii) The effect of current density is much smaller than
expected based on predictions of electrochemical mass
transport models (Polcaro et al. 2002,2009). This is also
observed in Fig. 7in which no significant differences are
observed between results obtained for the removal of 4-
chlorophenol (4-CP) by AO with BDD anode at current
densities within the range of 150600 A m
2
.
(iii) The comparative AO of the same organic pollutant in
chloride-supporting media with DSA and BDD anodes
reveals a better performance of DSA electrodes in spite of
the expected best characteristics of BDD. With BDD
anode, chlorides are not only oxidized to chlorine but
also to chlorate and perchlorate. At room temperature,
these latter oxidants are not very effective and this ex-
plains the best performance of DSA, in which only chlo-
rine and hypochlorite formation is promoted.
(iv) With non-active electrodes, such as conductive dia-
mond coatings, it is very difficult to find an inert supporting
electrolyte and merely perchlorate seems to be the only
supporting electrolyte in which no reactivity is obtained.
One important point regarding the oxidants produced dur-
ing the electrochemical treatment of a particular wastewater is
that they are not always detected in the reaction media al-
though their effect is clearly observed by comparison of
efficiencies when the composition of the raw wastewater is
modified. In this context, the detection and quantification of
oxidants during an electrolytic treatment can be understood as
an indication of the low reactivity of the oxidant with the
pollutants contained in the wastewater and not as an improve-
ment of the process performance. The best way to obtain a
highly efficient process is to promote the activation of the
oxidants produced electrochemically, either by chemical,
sonochemical, or photochemical methods.
Activation of oxidants produced electrochemically
As it has been described in the previous section, the reactivity
of many of the raw oxidants produced in EAOPs with or-
ganics is not very high and some sort of activation is often
required to obtain a clear improvement of the process. As an
example of the improvement that such an activation can yield,
an illustrative example can be considered: the transformation
0.01
0.1
1
0 500 1000 1500 2000 2500 3000 3500 4000
COD / mg O
2
dm
-3
ICE /
0
/
1
Fig. 7 Changes in the
instantaneous current efficiency
(ICE) with COD in the
electrochemical oxidation of
wastes containing 4-CP
(C
0
= 15 mM). Black circle pH 2,
5,000 mg Na
2
SO
4
L
1
,2C,
30 mA cm
2
;white circle pH 2,
5,000 mg Na
2
SO
4
L
1
,2C,
15 mA cm
2
;black triangle pH 2,
5,000 mg Na
2
SO
4
L
1
,2C,
60 mA cm
2
;white diamond pH
12, 5,000 mg Na
2
SO
4
L
1
,2C,
30 mA cm
2
;white triangle pH 2,
5,000 mg Na
2
SO
4
L
1
,6C,
30 mA cm
2
;white square pH 2,
3,333 mg Na
3
PO
4
L
1
,2C,
30 mA cm
2
.Adaptedfrom
Cañizares et al. (2005c)
8344 Environ Sci Pollut Res (2014) 21:83368367
of peroxosulfate into sulfate radicals that may increase the
process performance very significantly because the sulfate
radical typically reacts 10
3
10
5
times faster than the persulfate
ions (Tsitonaki et al. 2010). At this point, activation means the
formation of highly reactive species from the oxidants
contained in the wastewater and, as it has been pointed out
in Fig. 6, there are three different modes:
Chemical activation;
Activation by light irradiation;
Activation by ultrasound (US) irradiation.
Chemical activation of oxidants
Chemical activation is one of the most important ways to
enhance the effectiveness of an oxidant. It involves the com-
bination of the oxidant produced electrochemically with an-
other species (not necessary an oxidant), which leads to the
production of a third, very reactive species. This is the case of
the well-known Fenton processes (Brillas et al. 2009) which
will be described afterward in this manuscript. In these pro-
cesses, a metal ion (most likely iron(II), but also other transi-
tion metal cations) catalyzes the formation of
OH in the bulk
from the decomposition of hydrogen peroxide. As it is known,
hydrogen peroxide is a weak oxidant, while
OH is one of the
most active oxidants known.
Another example of chemical activation is the synergistic
combination of oxidants that results when ozone and hydro-
gen peroxide are combined (Table 2). This mixture also results
in the production of important concentrations of
OH that
explains the better efficiency of the processes in which the
formation of both oxidants is promoted.
The activation of hydrogen peroxide is very important in
EAOPs because this species is produced on the cathode of the
electrochemical cell, and then, it can double the efficiency of
the oxidation processes if properly activated (raw hydrogen
peroxide is not very active).
Activation by light irradiation
Light irradiation activation means the promotion in the
formation of highly active species by UV light irradia-
tion. This irradiation can be applied naturally (solar
driven) or artificially (using UV lamps). Excluding het-
erogeneous photoelectrocatalytic processes on the sur-
face of the anodes (typically based on the use of
MMO anodes with titanium dioxide as one of the com-
ponents) because they will be reviewed afterward, in
this subsection, light irradiation stands only for the
decomposition of oxidants in the bulk upon the action
of light. Thus, the photoactivation (or light-assisted de-
composition) of electrochemically generated reactive
species, such as H
2
O
2
or O
3
, by the reactions proposed
in Eqs. 36 and 37 (Pelegrini et al. 2000)iswellknown.
H2O2þhν2OH ð36Þ
H2OþO3þhν2OH þO2ð37Þ
However, there are more processes with relevance in EAOPs
(Bergmann et al. 2002). Radical species are expected to be
produced by light decomposition of peroxo compounds such
as peroxophosphates, peroxosulfates, and peroxocarbonates.
As an example, the production of sulfate radical from persulfate
is shown in Eq. 38 (Lin et al. 2011;Shihetal.2012).
S2O8
2þhν2SO
4
ðÞ
ð38Þ
The production of radicals from chlorine has also been
assessed in the literature (Oliver and Carey 1977;Chanetal.
2012), and it has been demonstrated that, under non-extreme
pH, hydroxyl and chlorine radicals are the main products
resulting from the light-assisted degradation of hypochlorite
(Eqs. 39 and 40).
ClOþþhνOþClð39Þ
OþH2OOHþ2OH ð40Þ
Activation by ultrasound irradiation
US irradiation as a treatment technology consists in the pro-
duction of a cyclic sound pressure with a frequency greater
than the upper limit of human hearing (20,000 Hz) in waste-
water. Unlike what one could expect, the main effect of US
irradiation on chemicals is not based on the direct interaction of
the mechanical acoustic field with chemical bonds of mole-
cules, but it is supported by the formation, growth, and implo-
sive collapse of bubbles irradiated with US (ultrasonic cavita-
tion). This phenomenon takes place in a very short moment
and space and it can be considered as adiabatic (Hiller et al.
1992). As a consequence, high temperatures and pressures are
reached inside the bubble due to gas compression. This in-
crease in temperature and pressure generates a huge concen-
tration of energy in a very small place known as a hot spot
(Flannigan and Suslick 2005). This energy is dispersed to the
surroundings so that the gas temperature in the hot spot quickly
returns to the ambient value. However, during a very short
time, it can produce significant changes in chemical composi-
tion of the hot spot and can form new radical species and
components, and so, it can increase the reactivity of the system
(Rooze et al. 2013). Hence, the formation of
OH is known
and, based on what has been described previously, this will
account for the formation of many other oxidizing species.
Environ Sci Pollut Res (2014) 21:83368367 8345
In addition, a further advantage in the electrochemical
system comes from the increase in the mass transport pro-
duced by the mechanical acoustic field which improves the
efficiency of processes in which the diffusion of pollutants
limits the rate of direct AO. The most representative and
studied process that demonstrates such assertion is
sonoelectro-Fenton (SEF), which will be discussed in the
Sonoelectro-Fentonsection.
Prospects
Perspectives of enhancing mediated oxidation for future
EAOP developments seem favorable. As it has been pointed
out along this section, mediated oxidation processes become
the key point to attain an enhancement in the efficiency of
EAOPs. Actually, it is not the production of oxidants, but their
activation in the reaction media, which is the most interesting
topic of research. Costs of US technology are very high and
improvements in efficiency are not always as good as to
propose their coupling with EAOPs. This is a direct conse-
quence of the huge amounts of energy dispersed as heat or
mechanical energy with this technology. Any novelty in this
topic has to come from a more efficient use of energy to
promote the formation of many hot spots in which radical
reactions could be started up. In contrast, light irradiation
already seems to be a very promising alternative with good
perspectives to be used in the near future. The synergistic
effect of the activation of oxidants has been clearly demon-
strated and the energy irradiated is much lower than that
applied in US irradiation.
Chemical and electrochemical generation of hydroxyl
radicals based on Fentonschemistry
The Fentons reagent, a mixture of H
2
O
2
and Fe(II), consti-
tutes the basis of the chemical generationof the strong oxidant
OH. A pioneering work was reported by Fenton in 1894 on
the oxidation of tartaric acid (Fenton 1894). Fenton observed
the enhancement of the oxidation power of H
2
O
2
in the
presence of iron(II) ions. Later, in the 1930s, Haber and
Weis undertook a detailed work to clarify the mechanism of
the reaction between H
2
O
2
and Fe
2+
(Eq. 41) and showed that
Fentons reagent led to the formation of
OH (Haber and Weiss
1934). They concluded that the catalytic decomposition of
H
2
O
2
by ferrous ion through a radical and chain mechanism
constitutes the origin of the oxidizing power of the Fentons
system. Eq. 41 was then named Fentons reaction.
Fe2þþH2O2Fe3þþOH þOHð41Þ
More recent studies have demonstrated that Fentonsreac-
tion could be applied to the degradation/destruction of
different types of organic pollutants (Sun and Pignatello
1993a; Gallard et al. 1998; Pignatello et al. 2006), and because
of its significant development during the twentieth century in
the treatment of wastewater, several review papers have fo-
cused on this process (Walling 1998;Bautistaetal.2008).
Fentons reaction as a source of hydroxyl radicals
Sun and Pignatello (1993b) showed that Fentons reaction can
be applied in acidic pH of 2.83.0 to efficiently produce
OH.
At this pH, Fentons reaction (Eq. 41) can be propagated by
the catalytic behavior of the Fe
3+
/Fe
2+
couple. Indeed, under
excess of H
2
O
2
, ferrous ions can be generated based on the
following reactions (Eqs. 42 and 43) to catalyze Fentons
reaction (Haber and Weiss 1934):
Fe3þþH2O2Fe2þþHO2
þHþð42Þ
HO2
þFe3þFe2þþO2þHþð43Þ
However, the HO
2
radical has a lower oxidation power
compared with
OH, and consequently, it is less reactive
toward organic pollutants. In addition, these reactions are
much slower than Fentons reaction and lead to the accumu-
lation of Fe
3+
in the medium, causing the formation of sludge
in the form of Fe(OH)
3
. In addition to pH value, the concen-
trations of H
2
O
2
and Fe
2+
and their ratio ([H
2
O
2
]/[Fe
2+
]) have
a significant role regarding the practical efficiency of the
Fenton process and have to be optimized for each specific
case (Bouafia-Chergui et al. 2010), since for high concentra-
tions, the reagents H
2
O
2
and Fe
2+
react with
OH through the
following wasting reactions, Eqs. 44 and 45,thatsignificantly
impair process efficiency:
H2O2þOHH2OþHO2
ð44Þ
Fe2þþOHFe3þþOHð45Þ
The Fenton process was applied to the oxidation of or-
ganics and treatment of wastewaters starting from the 1960s
(Brown et al. 1964), and many applications were developed in
the 1990s (Gogate and Pandit 2004; Pignatello et al. 2006).
However, several studies have shown the limitations of this
process in several cases, and the following drawbacks have
been highlighted: (i) high cost and risks due to the provision,
storage, and transport of H
2
O
2
, (ii) accumulation of iron
sludge that must be removed at the end of the treatment, and
(iii) lower mineralization efficiency due to the presence of
wasting reactions and, as a consequence, the potential forma-
tion of intermediates that are more toxic than raw pollutants.
Therefore, to improve the practical application for the
8346 Environ Sci Pollut Res (2014) 21:83368367
treatment of wastewaters, the Fenton process has been
coupled to other physicochemical processes like coagula-
tionflocculation, membrane filtration, and biological oxida-
tion to eliminate organic pollutants more effectively (Lucas
et al. 2007).
Electro-Fenton process: principles and running
The EF process is among the mostknown and popular EAOPs
and constitutes an indirect electrochemical manner to generate
OH in aqueous solutions. It was developed and extensively
applied over the last decade, particularly by Brillasand
Oturans groups since the 2000s (Brillas et al. 2000,2009;
Oturan 2000).This process has been developed to achieve the
implementation of a new and powerful advanced oxidation
method by avoiding the drawbacks of the chemical Fenton
process. Indeed, it can be defined as an electrochemically
assisted Fenton process.
OH is produced via the Fentons
reaction (Eq. 41), in which Fentons reagent is electrochemi-
cally generated in situ, avoiding the use of high quantities of
H
2
O
2
and iron(II) salt.
The H
2
O
2
production rate is one of the crucial parameters
of process efficiency, since the rate of Fentonsreactionis
predominantly controlled by this parameter. It can be contin-
uously supplied to the wastewater solution to be treated in an
electrochemical reactor from the two-electron cathodic reduc-
tion of oxygen gas, directly injected as compressed air, as
expressed in Eq. 46:
O2gðÞþ2H
þþ2eH2O2ð46Þ
The current efficiency of H
2
O
2
production is generally not
very high and depends on some factors such as operating
conditions (O
2
solubility, temperature, and pH) and cathode
properties. It can be destroyed by parasitic chemical decom-
position (Eq. 47), cathodic reduction (divided cell) (Eq. 48), or
AO (undivided cell) (Eqs. 49 and 50), resulting in a slower
accumulation in the bulk. Therefore, the use of optimal oper-
ating conditions (acidic pH, ambient temperature, etc.) and an
appropriate cathode material are important to obtain better
production rates.
2H2O2O2gðÞþ2H
2Oð47Þ
H2O2þ2H
þþ2e2H
2Oð48Þ
H2O2HO2
þHþþeð49Þ
HO2
O2gðÞþHþþeð50Þ
Several cathode materials such as mercury, graphite, car-
bonpolytetrafluoroethylene (PTFE) O
2
diffusion, and three-
dimensional electrodes like carbon felt (CF), activated carbon
fiber, reticulated vitreous carbon (RVC), carbon sponge, and
carbon nanotubes (CNT) (Oturan et al. 1992; Brillas et al.
1995; Alverez-Gallegos and Pletcher 1999; Oturan 1999)
were tested for H
2
O
2
production. Based on the results pub-
lished nowadays, 3D CF and carbonPTFE O
2
-fed cathodes
seem to constitute better cathode material for efficient H
2
O
2
generation; the use of Hg has been disregarded owing to its
potential toxicity.
The second component of Fentons reagent, i.e., the Fe
2+
ion, is initially introduced in a catalytic amount (typically
0.1 mM) in the form of ferrous (or ferric) salts and is regen-
erated electrocatalytically (Eq. 51) from the reduction of Fe
3+
formed by Fentonsreaction.
Fe3þþeFe2þð51Þ
Thus, Fentons reagent is continuously produced in the
solution to be treated in a catalytic manner, producing
OH
via Fentons reaction to ensure the destruction of organic
pollutants in aqueous medium. Formed
OH quickly reacts
in the bulk with organics, leading to their oxidation/
mineralization following Eqs. 52 and 53.
Organic pollutants þOHoxidation intermediates ð52Þ
Intermediates þOH→→→CO2þH2Oþinorganic ions
ð53Þ
Compared with the classical Fenton process, the main
advantages of the EF process are (i) in situ and controlled
generation of Fentons reagent (cost-effectiveness), thus
avoiding the risks related to transport, storage, and handling
of H
2
O
2
, (ii) elimination of parasitic reactions that waste
OH
(very low Fentons reagent concentration), (iii) total master-
ship of the processing by current or potential control, (iv)
possibility of controlling the degradation kinetics and
performing mechanistic studies, and (v) almost total mineral-
ization of organics including the intermediates.
Influence of applied current on the oxidation/mineralization
efficiency
A number of operating parameters, such as solution pH,
applied current, catalyst (Fe
2+
) concentration, supporting elec-
trolyte, organic load, etc., influence process efficiency.
Although the optimal value of pH is well known to be 2.8
(Sun and Pignatello 1993b), the process can effectively occur
within the range 2.5<pH<3.5.
The nature and concentration of the used catalyst have a
significant role in the EF process. To be used in EF, the catalyst
should be one of the forms of the redox couple, both forms
being soluble in water to allow the electrogeneration of the
Environ Sci Pollut Res (2014) 21:83368367 8347
reduced form in homogeneous medium. To clarify the effect of
the nature of the catalyst, a number of M
z+
/M
(z1)+
couples
were investigated (Brillas et al. 2004; Pimentel et al. 2008;
Balci et al. 2009;Oturanetal.2010; Salazar et al. 2012). Cu
2+
showed good catalytic characteristics in combination with Fe
2+
or Fe
3+
(Brillas et al. 2004; Salazar et al. 2012), but when used
alone, high concentrations are needed for obtaining the same
efficiency than Fe
2+
(Oturan et al. 2010). Mn
2+
was found to be
a good candidate in place of iron ions when their use is
compromised (Balci et al. 2009). Co
3+
and Ag
+
exhibited
catalytic behavior similar to that of Fe
2+
(Pimentel et al.
2008), but their use should be disregarded due to their
ecotoxicity. Usually, Fe
2+
(or Fe
3+
) behaves as the best catalyst
in the EF process particularly because it acts efficiently at
lower concentrations, typically around 0.1 mM. In this case,
the oxidation/mineralization of organic pollutants occurs effi-
ciently at low concentrations, but the effectiveness of the
process decreases with increasing Fe
2+
concentration, in par-
ticular at long treatment times (Abdessalem et al. 2008), due to
the enhancement of the rate of the parasitic reaction (Eq. 45).
The applied current (or current density) is the most impor-
tant operating parameter of the EF process, since it governs
the rate of generation of H
2
O
2
(Eq. 46), as well as the regen-
eration rate of Fe
2+
(Eq. 51) and, consequently, the rate of
generation of
OH from Fentonsreaction(Eq.41). In general,
the rate of the process increases with applied current since
more
OH are formed at a given time. By contrast, the applied
current cannot be increased indefinitely, since high current
values promote parasitic reactions, leading to the decrease in
current and process efficiencies. In particular, the applied
current should not reach the reduction potential of H
2
O
2
(Eqs. 49 and 50) to preserve it in the solution. Another wasting
reaction that becomes enhanced when increasing the applied
current is the evolution of H
2
at the cathode. Figure 8clearly
shows the effect of the applied current in the case of the
oxidation of 0.125 mM of the herbicide picloram in aqueous
medium at pH 3.0 by the EF process (Özcan et al. 2008). We
observed that the oxidation kinetics was enhanced with ap-
plied current from 50 to 300 mA, although the increase in
decay kinetics was not proportional to current due to the
gradual enhancement of the parasitic reactions. A further
increase in applied current did not yield a positive effect on
the oxidation kinetics, as shown for the case of 500 mA. We
noted that the oxidation process was very fast, and the total
disappearance of picloramwas reached within less than 5 min
for applied current values of 200, 300, and 500 mA.
Therefore, the value of 200 mA can be considered as the
optimal value to minimize the energy consumption at practi-
cally the same reaction time.
The classical EF process has been initiated by using Pt as
the anode. In this case, the process occurs mainly in the bulk
solution by
OH generated homogeneously through Fentons
reaction. Recently, a significant enhancement has been
attained by replacing the Pt anode with the more effective
BDD anode. The use of the BDD anode enables the EF
process to become more potent, since this anode allows gen-
erating supplementary heterogeneous
OH (BDD(
OH)) at its
surface (Eq. 54), in addition to those produced in bulk solution
from Fentons reaction. The use of the BDD anode in the EF
process also has three other advantages, as follows: (i) the
oxidizing power of BDD(
OH) is higher than other anodes due
to a larger O
2
overvoltage, (ii) BDD(
OH) is physisorbed at
the surface and thereby more easily available (compared with
the Pt anode), and (iii) the high oxidation window of the BDD
anode (approximately 2.5 V) allows the direct oxidation of
organic pollutants (Brillas and Martínez-Huitle 2011).
BDD þH2OBDD OHðÞþHþþeð54Þ
A recent and interesting study clearly showing the im-
provement of the EF process by using the BDD anode reports
the mineralization of the refractory herbicide atrazine (Oturan
et al. 2012). Indeed, a large variety of AOPs have been already
applied to the oxidative degradation and/or mineralization of
atrazine. However, in all cases, they yielded the persistent end
product cyanuric acid as predominant by-product, with 40
60 % mineralization yields, corresponding to the mineraliza-
tion of the side chains of the molecule. In contrast, when a
BDD anode was used in the EF process, an almost total
mineralization (97 % TOC removal) of atrazine aqueous so-
lutions was obtained (Fig. 9). The significant mineralization
power of EF with BDD anode relative to the classical process
with Pt anode can be clearly appreciated. In addition, the
authors showed in the same study that cyanuric acid, which
was already reported as recalcitrant to
OH, can also be almost
completely mineralized because of the action of BDD(
OH)
that is more potent than
OH in the mineralization of some
recalcitrant organics like carboxylic acids.
0.05
0.15
036912151821
[Picloram] / mM
Time / min
0.10
0.00
Fig. 8 Effect of applied current on the degradationkinetics of 0.125 mM
picloram in aqueous medium at pH 3 and room temperature by the EF
process: multiplication sign 30 mA, white square 60 mA, asterisk
100 mA, white diamond 200 mA, white triangle 300 mA, plus sign
500 mA. [Fe
3+
] = 0.1 mM, [Na
2
SO
4
] = 50 mM. Reprinted with permis-
sion from Özcan et al. (2008). Copyright 2008 Elsevier
8348 Environ Sci Pollut Res (2014) 21:83368367
Some recent applications
Since the publication of the first reports on the treatment of
wastewaters by the EF process (Brillas et al. 2000; Oturan
2000), it has been significantly developed and applied to the
treatment of a large variety of wastewaters polluted by toxic
and/or persistent organic pollutants such as pesticides (Oturan
et al. 2011;Zhaoetal.2012), synthetic dyes (Lahkimi et al.
2007; Khataee et al. 2009; Martínez-Huitle and Brillas 2009;
Panizza and Oturan 2011), industrial pollutants (Panizza and
Cerisola 2001; Bellakhal et al. 2006), pharmaceuticals and
personal care products (Sirés et al. 2007b,2010), landfill
leachates (Zhang et al. 2006), reverse osmosis concentrates
(Zhou et al. 2012), and many others. Among all these appli-
cations, three applications are especially detailed in the fol-
lowing paragraphs.
The first application deals with the effect of the chlorine
atom substituent on the oxidation efficiency of the process
(Oturan et al. 2009). This study reports the comparative ki-
netics of the degradation of several chlorophenols such as
monochlorophenols (2-chlorophenol (2-CP) and 4-CP),
dichlorophenols (2,4-dichlorophenol (2,4-DCP) and 2,6-di-
chlorophenol (2,6-DCP)), trichlorophenols (2,3,5-
trichlorophenol (2,3,5-TCP) and 2,4,5-trichlorophenol
(2,4,5-TCP)), 2,3,5,6-tetrachlorophenol (2,3,5,6-TeCP), and
pentachlorophenol (PCP), using a CF cathode and a Pt anode.
It was demonstrated that the number and the position of the
chlorine atoms in the aromatic ring influence significantly the
oxidation and mineralization kinetics of chlorophenols. This
effect was evaluated in terms of apparent and absolute rate
constants of the reaction between
OH and chlorophenols.
Apparent rate constants were obtained following the pseudo-
first-order kinetics and have been found to decrease with the
increase in the number of chlorine atoms, in the following
sequence: 4-CP>2-CP>2,4-DCP>2,6-DCP>2,3,5-TCP>
2,4,5-TCP>2,3,5,6-TeCP>PCP. Then, the absolute rate con-
stants of the second-order reaction between chlorophenols and
OH were determined by the competition kinetics method.
The values of the absolute rate constants (k
abs
)wereinthe
3.567.75×10
9
M
1
s
1
range, following the same sequence
of the apparent rate constants (Table 3). The mineralization of
several chlorophenols and of their mixture was also carried
out by monitoring with TOC removal percentage. Results
showed that more highly chlorinated phenols were more dif-
ficult to mineralize, with the mineralization rate decreasing
when the number of chlorine atoms increased.
The second application focuses on the assessment of solu-
tion toxicity when treated by the EF process. Indeed, impor-
tant efforts have been devoted to studies on the removal of a
new class of emerging pollutants, the pharmaceuticals and
personal care products, because of their occurrence in natural
waters and their potentially toxic effects on aquatic species
(Sirés et al. 2007b). The removal of many of these substances
was studied, including treatment efficiency, determination of
apparent and absolute rate constants, mechanistic assess-
ments, and mineralization pathways (Oturan et al. 2009;
Sirés et al. 2010; Dirany et al. 2011). A recent study devoted
to the removal of the antibiotic sulfamethoxazole from water
by the EF process focused on a special issue like the changes
in the solution toxicity during treatment (Dirany et al. 2011).
The evolution of the global toxicity of the treated solution was
monitored by using the Microtox® bioluminescence method,
in which toxicity is expressed as the inhibition percentage of
the luminescence of Vibrio fischeri bacteria. Figure 10 high-
lights an interesting behavior during the EF treatment of
sulfamethoxazole. Inhibition percentages were measured after
the exposure of bacteria to the solution for 15 min. Inhibition
curves obtained during the application of different applied
currents exhibited different peaks appearing as a function of
the treatment time. These inhibition peaks can be related to the
formation of primary and then secondary or tertiary aromatic
(and/or cyclic) intermediates formedduringthe treatment. The
significant increase in global toxicity at the beginning of the
treatment revealed the formation of some oxidation interme-
diates that were more toxic than the parent compound. This is
a behavior often observed during the application of AOPs.
Figure 10a shows a significant increase in the luminescence
inhibition between approximately 10 min (for I=300mA) and
0
2
4
6
8
10
02468
TOC / mg L-1
Electrolysis time / h
EF-Pt
AO-BDD
EF-BDD
Fig. 9 TOC removal during the mineralization of 0.1 mM atrazine
aqueous solution by the EF process using Pt and BDD anodes and CF
cathode. EF-Pt electro-Fenton process with Pt anode, AO-BDD anodic
oxidation using BDD anode, EF-BDD electro-Fenton process with BDD
anode. Reprinted with permission from Oturan et al. (2012). Copyright
2012 Springer
Table 3 Absolute rate
constants (k
abs
)forthe
oxidation of
chlorophenols by
OH
generated during the EF
process (reprinted with
permission from Oturan
et al. 2009)
Chlorophenol k
abs
(10
9
)M
1
s
1
4-CP 7.75±0.07
2,6-DCP 6.13± 0.05
2,4,5-TCP 5.72±0.05
2,3,5,6-TeCP 4.95±0.07
PCP 3.56±0.06
Environ Sci Pollut Res (2014) 21:83368367 8349
60 min (for I=30 mA) of treatment. Then, a rapid decrease
occurs with the appearance/disappearance of minor inhibition
peaks at electrolysis times of 30180 min. As indicated by
different curves, the solution toxicity was more quickly elim-
inated with higher current values. The time-related shifts of
luminescence inhibition peaks with the current value can be
explained by different formation rates of
OH, depending on
the applied current. As explained previously in the Influence
of applied current on the oxidation/mineralization efficiency
section, the formation rate of
OH is governed mainly by
applied current and increases from 50 to 300 mA (Fig. 8).
As a consequence, sulfamethoxazole and its by-products were
mineralized more quickly, leading to a rapid decrease in
solution toxicity. The high-performance liquid chromatogra-
phy (HPLC) analyses indicated that 3-amino-5-
methylisoxazole (AMI) and p-benzoquinone (BZQ) were the
primary oxidation by-products of sulfamethoxazole. To
clarify the relative toxicity of these intermediates, their diluted
solutions were treated under the same operating conditions.
Figure 10b was obtained from diluted solutions of AMI and
BZQ effectively attained during the EF oxidation of sulfa-
methoxazole and demonstrated that both aromatic (and/or
cyclic) intermediates were, at least partly, responsible for the
increase in toxicity of sulfamethoxazole solutions because, at
least, one of the major intermediates, like BZQ, is significantly
more toxic than sulfamethoxazole toward V. fischeri bacteria.
The third application concerns the removal ofthe antibiotic
drug sulfachloropyridazine from water and its mineralization
pathway during the treatment of its aqueous solution by the EF
process (Dirany et al. 2012). The suggested mineralization
pathway includes 15 cyclic intermediates (identified by HPLC
and gas chromatography coupled to mass spectrometry [GC-
MS] analyses), 5 aliphatic carboxylic acids (oxalic, maleic,
pyruvic, glyoxylic, and malic acids), and a mixture of released
inorganic ions (Cl
,SO
42
,NH
4+
,andNO
3
). Based on the
action of
OH onto four different sites of
sulfachloropyridazine, a detailed scheme for the complete
mineralization of sulfachloropyridazine was proposed. The
reaction of
OH with this pharmaceutical yielded different
primary cyclic by-products, shown as pathways ADin
Fig. 11. Pathways A and B involve the formation of five
benzenesulfonamides promoted by consecutive hydroxylation
steps with or without Cl
release. Pathways C and D occur
simultaneously and include the oxidative cleavage of the
structure, leading to the formation of a dozen pyridazine and
benzenic derivatives. Among these intermediates, 3-amino-6-
chloropyridazine (ACP) and BZQ were detected as major
intermediates. Successive hydroxylation of the primary inter-
mediates weakens the structure and promotes ring breaking to
yield carboxylic acids, accompanied by the release of inor-
ganic ions like chloride, sulfate, ammonium, and nitrate. In
addition, the time course of some available reaction interme-
diates such as ACP and BZQ was satisfactorily correlated with
the toxicity profiles determined using the Microtox® method
in terms of the inhibition of V. fischeri luminescence. ACP and
BZQ, which are the predominant intermediates, were found
responsible for the significant increase in toxicity during the
first stages of treatment.
Prospects
The EF process emerged as an environmentally friendly AOP
approximately 10 years ago, mainly in its two basic versions
based on the nature of the cathode material: CF and carbon
PTFE gas diffusion electrode. Nowadays, there are several
dozen groups studying and publishing works related to this
process. It is very largely investigated at the laboratory scale,
mainly in its two initial versions. The use of the emergent
BDD anode significantly enhanced the oxidation power and
mineralization efficiency due to the production of
0 30 60 90 120 150 180 210 240
Time / min
300 mA
120 mA
60 mA
30 mA
0 30 60 90 120 150 180 210 240 270
Time / min
AMI
BZQ
120
90
60
30
0
120
90
60
30
0
% Inhibition 15 min% Inhibition 15 min
a
b
Fig. 10 Evolution of the inhibition of the V. f i s c h er i bacteria lumines-
cence during EF treatment, with Pt anode and CF cathode, of asulfa-
methoxazole (SMX) aqueous solutions and bits cyclic derivatives 3-
amino-5-methylisoxazole (AMI)andp-benzoquinone (BZQ)diluted
aqueous solutions, after an exposure time of 15 min. V=250 mL,
[SMX]
0
=0.208 mM, [AMI]
0
=0.016 mM, [BZQ]
0
=0.018 mM, [Fe
2+
]=
0.2 mM, [Na
2
SO
4
]= 50 mM, pH=3, I=30, 60, 120, and 300 mA in a,and
I=60 mA in b. Reprinted with permission from Dirany et al. (2011).
Copyright 2011 Springer
8350 Environ Sci Pollut Res (2014) 21:83368367
supplementary
OH at the anode surface. Therefore, this tech-
nology becomes now mature enough for passing to pilot-scale
reactor design and application to treatment of large volumes of
wastewaters.
The first step will be the conception and design of a pilot-
scale reactor. This conception can include combined processes
to increase the effectiveness of the treatment. Both batch
reactor and continuous (flow) reactor should be considered.
The modeling of the process can be useful to optimize the
operating parameters and predict the behavior of pollutants
and can help in the economical and practical application to
real wastewater treatment.
The design of a tubular reactor can constitute an interesting
way to reach continuous treatment. A joint project focused on
the coupling of EF with nanofiltration, in which a carbona-
ceous material is suggested as both filter and cathode, was
recently applied for funding to the French ANR (National
Research Agency) by three French universities (including
the LGE Laboratory of Université Paris-Est) and a company
of the field. The second step will consist of checking if the
parameters optimized at the laboratory scale could be consid-
ered for the work in the pilot-scale reactor, before the
industrial-scale reactor stage. Otherwise, the key parameters
should be optimized at the pilot scale.
Coupling with a biological process as a pretreatment or
posttreatment unit is another promising way for cost-effective
treatment. Some recent studies have shown the feasibility of
such coupling. Indeed, the EF process is able to transform
toxic and/or biorefractory molecules to biodegradable species
during a short treatment time. The complete mineralization of
solutions thus obtained can then be achieved by biological
treatment.
To develop cost-effective treatments, the use of green and
cheap energy source based on sunlight-driven electrical power
systems such as an EF reactor directly powered by photovol-
taic panels can also be considered.
Photoelectrochemical processes
There is an increasing interest in the use of
photoelectrochemical processes for water and wastewater re-
mediation. These photo-assisted treatments are based on the
irradiation of a contaminated solution or a photoactive elec-
trode with UV or solar light (Brillas et al. 2009; Sirés and
Brillas 2012; Daghrir et al. 2012a; Georgieva et al. 2012).
UVA (λ=315400 nm), UVB (λ=285315 nm), and UVC (λ
<285 nm) lights supplied by UV lamps as energy sources are
m/z =284.01(SCP)
SN
H
O O
H2N
N
NCl
A
B
C
D
m/z =266.05
SN
H
O O