Вы находитесь на странице: 1из 9

Chemical Engineering Journal 233 (2013) 4755

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

Electrochemical treatment of fresh, brine and saline produced water

generated by petrochemical industry using Ti/IrO2Ta2O5 and BDD
in ow reactor
Andr Jailson Cabral da Silva a, Elisama Vieira dos Santos b, Crislnia Carla de Oliveira Morais a,
Carlos Alberto Martnez-Huitle b, Suely Souza Leal Castro a,

Faculty of Natural and Exact Sciences, University of State of Rio Grande do Norte, CP 70, 59620-625 Mossor, RN, Brazil
Institute of Chemistry, Federal University of Rio Grande do Norte, Campus Universitrio s/n, CEP 59078-970 Natal, RN, Brazil

h i g h l i g h t s

 Electro-oxidation using DSA

and BDD removes efciently the dissolved organic matter from PW.

 BDD electrode shows better performance for removal COD.


electrode operates with lower energy consumption and operating cost.

and BDD electrodes is indirect oxidation.

 Cl favors the indirect electro-oxidation by formation of chlorine/hypochlorite.

 The dominant process on DSA

a r t i c l e

i n f o

Article history:
Received 26 February 2013
Received in revised form 6 August 2013
Accepted 9 August 2013
Available online 16 August 2013
Electrochemical treatment
Boron-doped diamond
Dimensionally stable anode
Produced waters
Direct oxidation and indirect oxidation

a b s t r a c t
This study compares the anodic oxidation of three classes of produced water (PW) (fresh, brine and saline) generated by petrochemical industry using Ti/IrO2Ta2O5 and BDD electrodes in a ow reactor. During electrochemical treatment, various operating parameters were investigated, such as temperature, pH,
conductivity, current density, total organic carbon (TOC), chemical oxygen demand (COD) as well as
energy consumption and cost. When both electrodes materials are compared under the same operating
conditions, higher TOC and COD removal efciencies were achieved for BDD anode, nevertheless, the
energy consumption and cost were higher when compared with the values estimated for Ti/IrO2
Ta2O5. Different Cl concentrations in the efuents promoted the electrogeneration of strong oxidant species, such as chlorine and hypochlorite, increasing the efciency of treatment. These results encourage
the applicability of this method as a pre-treatment process for the petrochemical industry, reducing depuration time.
2013 Elsevier B.V. All rights reserved.

1. Introduction
The petrochemical companies deal with problems of difcult
solution in relation to activities aimed at environmental protection, due to the generation of large amounts of waste during oil
prospecting and exploitation, such as produced water (PW). The
amount of this kind of waste varies depending on the oil reservoir:
a new basin produces 515 vol.%, while at the end of its lifetime it
reaches 7590 vol.% [1].
PW contains with different pollutants, such as heavy metals
(Cd, Cr, Cu, Pb, Hg, Ag, Ni, Zn), organic compounds, and dissolved/suspended solids. Benzene, toluene, xylene, phenol, halogenated aromatic compounds, chloroform and trichloroethylene are
Corresponding author. Tel.: +55 8433153011.
E-mail address: suelycastro@uern.br (S.S.L. Castro).
1385-8947/$ - see front matter 2013 Elsevier B.V. All rights reserved.

the major organic pollutants present in PW, commonly known as

BTEX [1]. Also, the composition of PW depends on chloride concentration, determining if it is fresh, brine or saline PW.
Due to their toxicity and their potential carcinogenic effect,
these efuents can cause irreparable damage to human health
and the environment [25]. Several alternative treatment of PW
has been studied by several research groups around the world
[2]. The physicalchemical and bioremediation methods utilized
for the degradation of these compounds have shown various operational problems, such as: generation of toxic gases, phase transfer
of pollutants, residual sludge production, demand for large territorial areas and the impossibility of destroying refractory compounds
In the last years, the electrochemical technology has been proposed as an alternative process for elimination of contaminants in
efuents, showing excellent results in various matrices such as


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

textile dyes [713], dairy [14,15], herbicides/pesticides [1620],

heavy metals [21], aquaculture [22], landll leachate [23], pharmaceutical residues [2428], pulp and paper [29], among others.
Moreover, a wide variety of electrode materials have been suggested, such as dimensionally stable anodes, noble metals (e.g.
platinum), carbon-based anodes, PbO2 and BDD; obtaining different removal organic matter efciencies [3033] considering that
non-active anodes, such as BDD, are useful for direct oxidation of
organic material via hydroxyl radicals, while DSA, such as Ti/
IrO2Ta2O5, are effective for promoting hypochlorite mediated
chemistry when chloride is present.
In the case of applicability of electrochemical technology for
treating petrochemical wastewaters, dimensionally stable anodes,
platinum and BDD have been preferentially used as electrocatalytic
materials [34]. However, in some cases, Pt anodes are very expensive, and also subject to fouling. Recently, Santos et al. [35] investigated the electrochemical remediation of oil extraction industry
wastewater using Ti/Ru0.34Ti0.66O2 anode. The authors obtained
the best COD reduction (57%) of an oily sample for 70 h at 50 C
with a current density of 100 mA/cm2 [35], but the slow rate of
COD reduction was attributed to the occurrence of secondary reactions involving O2, Cl2 and H2 evolution. Also, 24%, 48% and 57%
COD reduction after 70 h of electrolysis at 10 C, 25 C and 50 C
were achieved, respectively. On the other hand, using Pt/Ir electrodes, the electrochemical purication of bilge water
(COD0 = 3080 mg dm3, 50/50% seawater/fresh water composition)
were investigated by Krbahti and Artut [36], where COD removals
ranging from 85100% were achieved by applying 12.8 mA cm2 of
current density at 32 C.
The anodic oxidation of PW generated by petroleum exploration
of the Petrobras plantBrazil using an electrochemical reactor with
a Ti/RuO2TiO2SnO2 was recently studied [3]. Under galvanostatic
conditions (j = 89 mA cm2), it was found that the organic pollutants degradation using different ow rates (0.25, 0.5, 0.8 and
1.3 dm3 h) achieved distinct removal efciencies (98%, 97%, 95%
and 84% were achieved, respectively).
More recently, Rocha et al. [5] studied the electrochemical oxidation of brine PW in galvanostatic conditions using platinum supported on titanium (Ti/Pt) and BDD anodes, employing a batch
reactor. The results showed that complete COD removal was
achieved using BDD electrode due to the production of high
amounts of hydroxyl radicals (OH) and oxidizing species (Cl2,
HClO, ClO).
The use of these electrode materials have been proposed due to
electrocatalytic features to produce in situ-strong oxidant species,
principally active chlorine. However, studies reported in literature
relative to the treatment of petrochemical efuents, especially PW,
have not contemplated the different types of PW found in oilelds:
fresh, brine and saline, which are assigned to the direct inuence of
the characteristics of the soil where they are conned. Therefore,
the objective of this study was to evaluate the performance of
Ti/IrO2Ta2O5 and BDD electrodes during treatment of real PW
(fresh, brine and saline) using an electrolytic cell in continuous
ow, focusing our attention on the electrochemical conditions that
provide greater efciency of current with lower power requirement to scale up the electrochemical treatment in order to employ
in petrochemical platforms.

2. Experimental
2.1. Produced water samples and chemicals
The PW samples (fresh, brine and saline) were supplied by
Petrobras Plant in Rio Grande do Norte, northeastern region of
Brazil. The characteristics of the efuents samples were analyzed

Table 1
Results of physical and chemical analyzes of PW samples.
Analytical parameters

PW fresh

PW brine

PW saline

Temperature (C)
Conductivity (mS cm1)
COD (mg dm3)
Salinity (mg dm3)
Chloride (mg dm3)
TOC (mg dm3)
Benzene (lg dm3)
Toluene (lg dm3)
Ethylbenzene (lg dm3)
o-Xylene (lg dm3)




Limits permitted by Brazilian legislation (mg dm3): benzene = 1.2; toluene = 1.2;
ethylbenzene = 0.84 and xylene = 1.6.

using American Public Health Association (APHA) procedures

[37] and given in Table 1. Chemicals were of the highest quality
commercially available, and were used without further purication. The classication of PW, as fresh, brine and saline, was performed according to the salinity of the samples (see Table 1).
2.2. Instrumentation
The analysis of pH, conductivity and temperature were performed using a multiparameter instrument model 5 Star Orion
(USA). The analysis of COD were performed using a multiphotometer HANNA model HI 83,099 (Brazil), after digestion of samples
in a Term Reactor HANNA model HI 839,800 (Brazil); and the
electrolysis were carried out using a power supply MINIPA MLP3303 (Brazil). COD reects the amount of organic matter (biodegradable and refractory) removed during process [38] and the
COD decay was used to evaluate the performance of the electrooxidation experiments. It is important to remark that, chloride is
an interference of COD method; however, the Cl limit for obtaining accurate results using commercially available digestion vials is
2000 mg dm3, allowing the COD determinations for fresh and
brine PW efuents. In the case of saline PW, precipitation of Cl before COD determination was performed; allowing to obtain accurate data. Total organic carbon (TOC) content is also an indicator
of environmental pollution and it is important for environmental
control of water and wastewater. TOC analyzer used in the present
study was Analytik Jena Multi N/C 3100 (Germany).
2.3. Electrolytic systems
The electrochemical oxidation was conducted in galvanostatic
conditions using an electrolytic ow cell [11]. PW efuent was
stored in a reservoir of 1.5 dm3 and it was recirculated through
the electrolytic cell by means of a centrifugal pump, working in
the ow of 160 dm3 h1. The cell contained a Ti/IrO2Ta2O5,
Ti/PbO2 or BDD electrode as anode and a Ti plate as cathode. The
electrodes were circular and each of them had a geometric area
of 63.5 cm2, and with a separation of 1.0 cm between them.
Ti/IrO2Ta2O5 electrode was supplied by Industrie De Nora Elettrodi (Milan, Italy). BDD electrode was supplied by Adamant Technologies with doping level of boron in the diamond layer of 1000 ppm.
Electrochemical oxidation experiments were maintained at 25 C,
to assess the role of applied current density to investigate the
treatment of fresh PW (1, 2.5 and 5 mA cm2); brine PW (10, 20
and 30 mA cm2) and saline PW (10 and 20 mA cm2). Furthermore, the temperature effect was investigated by carrying out
experiments at 40 C and applying a current density of 5 and
20 mA cm2 to fresh and brine PWs, respectively; the temperature
of the efuents was controlled using a thermostatic bath. The cell


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

voltage was periodically registered and the samples were collected

before electrolysis and periodically during electrolyses from reservoir and subjected to COD analysis. For TOC decay evaluation, initial and nal measurements were performed.
Regarding the use of PbO2 (active anode), some bulk experiments were performed; however, the most important problem of
this electrode at higher Cl concentrations (brine and saline PWs)
was the Pb-pollution. After few experiments, PbO2 anode releases
Pb-ions in solution, including very toxic specie. For this reason,
PbO2 anode was disqualied as a good anode material for this electrochemical application.
2.4. Analytical methods
The main parameters indicative of the progress of electrochemical treatment are related to the quantity of degraded matter, energy efciency, time and operational cost of treatment.
Decontamination of PW efuents was monitored from the abatement of their COD and TOC. For TOC removal was used the follow

% TOC removal TOC i  TOC f =TOC i   100

where TOCi and TOCf are the initial and nal TOC measurements,
The energy consumption per volume of treated efuent was
estimated and expressed in kW h dm3. The average cell voltage
during the electrolysis was taken into consideration for estimation
of the energy consumption the following equation.

EC DEC It=1000V

where t is the time of electrolysis (h), DEC is the average cell voltage
(V), I is the electrolysis current (A) and V is the sample volume
Taking into consideration the electrical energy cost of about US$
0.15 (U.S. currency) per kW h, the monetary value required to
degrade a unit volume of efuent is given by the following


CostUS$dm ECkW h dm  0:15US$=kW h

3. Results and discussion

3.1. Characteristics of the efuent samples
The results in Table 1 shows that the PW efuents are completely
different in their nature, particularly in relation to the parameters
such as conductivity, COD, TOC, salinity as well as BTEX and chloride

concentrations. These differences are attributed to the different oil

extraction level in the basin where the water was injected. Table 1
also shows that the three classes of PW contain different concentrations of benzene, toluene, ethylbenzene and xylene, above of the
limits permitted by Brazilian legislation for disposal in the environment [39]. Table 2 illustrates the concentration of the most important inorganic species determined for PW efuents.
3.2. Electrochemical treatment of fresh PW using Ti/IrO2Ta2O5 and
BDD anodes
To understand the performance of the electrochemical treatment, mainly due to different concentrations of chloride ions,
different applied current densities were used for each type of PW
(fresh, brine and saline), evaluating the efciency of the anode
materials (Ti/IrO2Ta2O5 and BDD). Bulk electrolyses for treating
fresh PW were performed at different applied current densities
(1, 2.5, 5, 10, 20, and 30 mA cm2), however, higher current densities (10, 20, and 30 mA cm2) caused a considerable increase on
the cell potential due to lower conductivity of fresh PW. For this
reason, lower current densities were employed (1, 2.5, 5 mA cm2)
in order to decrease the nal energy requirements. Fig. 1 shows the
inuence of applied current density (1, 2.5 and 5 mA cm2) on COD
removal, as a function of electrolysis time, at Ti/IrO2Ta2O5
(Fig. 1a) and BDD (Fig. 1b) electrodes, during the electro-oxidation
of fresh PW. The rate of COD decay was dependent on the applied
current density, and only complete COD removal was achieved
when 5 mA cm2 was applied, at BDD anode after 80 min
(Fig. 1b). Although higher COD removal was obtained for both
anodes by applying 5 mA cm2, the COD decay rate was faster at
BDD electrode. In order to strengthen this statement, supplementary electronic material regarding the COD decay, as a function of
charge passed (Q = Ah dm3), during the electrochemical treatment of fresh PW, was included to this work. From these results
it was conrmed that the better performances, on the COD decay,
were achieved at BDD anode when 5 mA cm2 was applied
(Fig. 1S). Results clearly demonstrated that when lower current
densities were applied, lower Q was employed, avoiding a
complete elimination of COD, principally when Ti/IrO2Ta2O5 was
used as electrocatalytic material (Fig. 1S(a)). Conversely, when
BDD anode was employed, lower Q was necessary to achieve a
signicant COD removal by applying 2.5 and 5 mA cm2
(Fig. 1S(b)).
TOC measurements were carried out in order to conrm the
elimination of organic matter. Higher TOC removal efciencies
were achieved on BDD (ranging from 40% to 90%, see inset in
Fig. 1b) respect to the %TOC removals obtained at Ti/IrO2Ta2O5
(Fig. 1a).

Table 2
Average concentration of inorganic species in the PW samples before and after electrochemical treatment.


Inorganic species
(mg dm3)

Fresh PW

After (Ti/IrO2



After (Ti/IrO2



After (Ti/IrO2












5 mA cm2.
20 mA cm2.
Below detection limit of technique.

Brine PW

Saline PW


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

BDD is a non-active anode at which the principal reactive species is hydroxyl radicals, whereas Ti/IrO2Ta2O5 is an active anode
that depends mostly on the higher oxide mechanism. At Ti/IrO2
Ta2O5 electrode, the oxidation process depends on direct and mediated oxidation [3,5,40] for removing organic matter. During direct
oxidation, adsorption of organic compounds and by-products on
the anode surface plays an important role in the electrochemical
oxidation process, and these chemical mechanisms can reduce
the anode efciency [3,10].
The surface of active anode, as Ti/IrO2Ta2O5, strongly interacts
with OH produced during water electrolysis (H2O ? OH + H+ + e).
Metal cations in the oxide lattice may reach higher oxidation
states, so-called superoxide (MOx), participating as a mediator in
the oxidation of organics, which competes with the side reaction
of oxygen evolution via chemical decomposition of the higher
oxide species. Consequently, both hydroxyl radicals and higher
oxide formed on anode surface (Ti/IrO2Ta2O5(OH/MOx)) promote
the electrochemical conversion of organic matter dissolved in fresh
PW to other oxidized compounds [41,42].
Another interesting feature of the electrochemical treatment of
fresh PW is that this petrochemical efuent contains with
47.8 mg dm3 of chloride, and it probably participates in the electrogeneration of active chlorine. In fact, analyzing the solution pH,
originally around 6.87, it moderately increases to values of 7.1.
Thus, ClO will prevail during the electrolysis, but HClO can be active at the electrode surface, as a consequence of the large acidity
(pH is slightly acidic) due to the concomitant oxygen evolution, as
indicated previously by other authors [3]. Based on results
reported in Fig. 1and 1S, we can assume that, using

Ti/IrO2Ta2O5, direct anodic oxidation is favored at lower current

densities, achieving modest organic removal efciencies (COD
and TOC). However, at 30 mA cm2, both direct and Cl-mediated
oxidations (even the efuent contains with lower Cl concentrations) promote a faster abatement of COD, increasing modestly
the TOC removal (Fig. 1a and 1S(a)).
Conversely, electrochemical treatment of fresh PW using BDD
anode depends completely on applied current density because
when current density was increased (Fig. 1b) or lower Q was
passed (1S (b)), a signicant increase on COD decay was achieved.
At the same time, more efcient TOC removal was achieved at BDD
than that obtained at Ti/IrO2Ta2O5 (inset in Fig. 1b). It is due to
thank to the weak adsorption properties of BDD anode (nonactive
anode) and higher production of hydroxyl radicals, avoiding
adsorption of organic compounds and by-products dissolved on
fresh PW [7,1012]. Nevertheless, Cl-mediated oxidation by active
chlorine was not satisfactorily favored (even when the pH solution
remains around 66.5) due to lower concentration of Cl in solution, predominating the production of reactive oxygen species
(ROS, such as, hydroxyl radicals), that oxidize organic matter dissolved in real efuent [10]. As consequence, higher TOC removal
efciencies were obtained using BDD anode.

Fig. 1. Inuence of applied current density on COD removal, as a function of time,

during fresh PW electrolysis at 25 C using Ti/IrO2Ta2O5 (a) and BDD (b) anodes.
Insets: %TOC removal, as a function of applied current density. Operational
conditions: ow rate = 160 dm3 h1 and different applied current densities:
1 mA cm2 (circle), 2.5 mA cm2 (square), and 5 mA cm2 (triangle).

Fig. 2. Inuence of applied current density on COD removal, as a function of time,

during brine PW electrolysis at 25 C for Ti/IrO2Ta2O5 (a) and BDD (b) anodes.
Insets: %TOC removal, as a function of applied current density. Operational
conditions: Flow rate = 160 dm3 h1 and different applied current densities:
10 mA cm2 (circle), 20 mA cm2 (square) and 30 mA cm2 (triangle).

3.3. Electrochemical treatment of brine PW using Ti/IrO2Ta2O5 and

BDD anodes
Electrochemical bulk experiments were performed using brine
PW samples ([Cl] = 1573 mg dm3) to increase the understanding
of real efuent treatment by involving the electrochemical

A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

production of reactive oxidant species (hydroxyl radicals or active

chlorine) at Ti/IrO2Ta2O5 and BDD anodes. To provide faster and
more efcient removal rate, as a method economically viable, higher current densities (10, 20, and 30 mA cm2) are required, since
the brine PW presents good conductivity and higher COD load
(700 mg dm3) than that determined for fresh PW (see Table 1).
Fig. 2 shows the COD decay, as a function of electrolysis time,
at different applied current density (10, 20, and 30 mA cm2) for
Ti/IrO2Ta2O5 (Fig. 2a) and BDD (Fig. 2b) anodes. As can be seen
in Fig. 2a, COD decay depends on applied current density as well
as the Q (Fig. 2S(a)), using Ti/IrO2Ta2O5 electrode. However, only
complete COD removal was achieved by applying 30 mA cm2,
after approximately 240 min of electrolysis (6 Ah dm3, see
Fig. 2S(a)). Whereas, at 10 and 20 mA cm2, a partial COD removal
was achieved, about 71.5% (2 Ah dm3) and 78.6% (4.2 Ah dm3),
respectively. This behavior can be justied because an increase
on the applied current density promotes the mediated oxidation
by hydroxyl radicals as well as active chlorine species [3,34
36,40]. Considering that, the pH solution remained between 7.5
and 7.9 and DSA-electrodes are good electrocatalysts for active
chlorine production [43], we can assume that the production of
species like hypochlorite and hypochlorous acid favors a faster oxidation of organic compounds dissolved in the efuent and byproducts generated. However, TOC results indicated that the efuent seems to have some recalcitrant compounds (or degradation
products) that are not oxidized under these experimental conditions. This suggests that recalcitrant intermediates are being
formed in the chloride medium, most probably organochloride
species that are not completely oxidized.
For BDD anode, under the same experimental conditions, complete COD removal was achieved at all current densities (after
1.7 Ah dm3 (10 mA cm2), 3 Ah dm3 (20 mA cm2) and
4 Ah dm3 (30 mA cm2), see Fig. 2S(b)), decreasing remarkably
the electrolysis time (Fig 2b). These results conrm the great ability of the BDD has to remove organic pollutants, even at chloride
medium. Theoretically, based on the existing literature, the electrochemical treatment can be carried out at lower potentials in
the presence of chlorides, compared with those required for the direct anodic oxidation [43,44]. Considering that, the efuent conductivity was elevated due to higher concentrations of Cl in
solution (see Table 1), affecting the cell potentials; it promotes
the production of active chlorine species and consequently, favoring a faster COD abatement. Also, TOC removals ranged from 92%
to 99%, indicating that the degradation process occurs with no signicant formation of recalcitrant intermediates.
In consideration of the fact that, in the case of chloride mediation, incineration reactions should be mainly a set of volume rather
surface reactions, the change from DSA-electrode to BDD should
not involve dramatic changes in the incineration mechanism.
The, chloro and oxychlororadicals, co-generated at the electrode
surface and reacting in solution, have to be considered in the
mechanism of the electrochemical destruction [10]. Indeed,
changes in the pH solution indicated that ClO prevailed during
the brine PW electrolysis, but the co-existence of HClO can be
hypothesized at the electrode surface, as a consequence of the
large acidity (pH is slightly acidic) due to the concomitant oxygen
evolution [10]. Also, a minor proportion of Cl can be eliminated
from solution as Cl2.
3.4. Effect of temperature on electrochemical treatment of fresh and
brine PWs using Ti/IrO2Ta2O5 e BDD anodes
The effect of temperature was also studied, at 40 C, for treating
fresh and brine PWs by applying 5 mA cm2 and 20 mA cm2,
respectively. The temperature selected is mimicking the real temperature utilized to discharge efuents by the petrochemical


Fig. 3. Effect of temperature on COD removal, as a function of time, during

electrochemical treatment of fresh (a) and brine (b) PWs using Ti/IrO2Ta2O5 and
BDD anodes. Insets: %TOC removal, at 40 C. Operational conditions: Flow
rate = 160 dm3 h1, applied current densities: 5 mA cm2 (fresh PW (a)) and
20 mA cm2 (brine PW (b)).

industry. Assuming that the rate of the each reaction must be

dependent on temperature, as already demonstrated by other
authors [1013], we decided to study the effect of an increase on
the temperature during PW oxidation.
Fig. 3a shows to the inuence of temperature on COD removal,
as a function of time, during treatment of fresh PW by applying
5 mA cm2. It was observed that temperature has a signicant impact on the kinetic of PW electrochemical oxidation, especially
when Ti/IrO2Ta2O5 electrode was used, because the COD removal
rate was considerably enhanced upon rising temperature. An elevation in temperature from 25 to 40 C reduced the electrolysis
time required to remove 90% of COD from 90 to 15 min (Fig. 3a)
as well as the electrical charge passed (Q) (Fig. 3S(a)). Conversely,
electrolysis times (Fig. 3b) as well as Q (Fig. 3S(b)) required were
modestly reduced when the BDD anode was used, but mass transport limitations were observed (Fig. 3a). This behavior may be due
to the fact that the increase in temperature had little inuence on
the electro-oxidation via hydroxyl radicals [10] and active chlorine
at BDD anode.
It is important to remark that, this enhance on the rate of chemical reactions cannot be attributed to an increased activity of the
anodes but to an increase in indirect reactions of organic compounds with electrogenerated oxidizing agents from PW efuent
oxidation [7,8,10,11]. In fact, is well known that electrolysis in
aqueous medium containing chloride ions generates active chlorine at the anode surface [10]. Also, these powerful oxidizing
agents can rapidly oxidize organic materials by a chemical reaction
whose rate increases with temperature. In this context, restricting
now our analysis to the behavior at both anodes, Ti/IrO2Ta2O5
promotes the production of chlorine gas more than BDD anodes,


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

as reported by other studies [43], and consequently, this limits the

oxidation of organic matter at 25 C (Fig. 3a). Conversely, strong
oxidants of active chlorine are produced efciently at both electrodes at 40 C, favoring COD removal under these experimental
conditions; even when the oxygen evolution reaction is also favored. From the TOC removals obtained for each anode (inset in
Fig. 3a), it can be inferred that the electrochemical treatment of
fresh PW, at 40 C, occurs with signicant elimination of organic
matter, more than that eliminated at 25 C at Ti/IrO2Ta2O5 (Inset
in Fig. 1a), but a slight decrease in efciency (from 89% (at 25 C) to
79% (at 40 C)) at BDD anode (inset in Fig. 1b) was attained. This
means that, after 20 min of electrolysis using BDD anode at
40 C, a signicant amount of more simplex organic compounds
(probably aliphatic acids) is produced, complicating their complete
elimination by mass transport limitations and consequently,
decreasing the TOC elimination performances.
In the case of electrochemical treatment of brine PW, the temperature effect on COD abatement was signicant, using BDD and
Ti/IrO2Ta2O5 electrodes by applying 20 mA cm2. It was observed
that the kinetics of the indirect reactions were favored; achieving a
relevant COD removal (93%) in a minor time (Fig. 3b) as well as Q
(Fig. 3S(b)) than those necessary to 25 C. While, 89% of COD was
removed employing Ti/IrO2Ta2O5 (Fig. 3b), after 1.5 Ah dm3.
Taking into account the results obtained for TOC removal for
brine PW oxidation, at both anodes, at 40 C (inset in Fig. 3b), it
can be inferred that better performances were obtained at BDD anode than those obtained at Ti/IrO2Ta2O5. Using BDD anode, an increase on TOC removal efciencies was obtained than that
obtained at 25 C, indicating that, electrogenerated active chlorine
on BDD surface [10] favors a complete elimination of organic matter in brine PW when an increase on temperature was attained,
even when a little amount of organochlorinated compounds would
be formed (as commented in section 3.3). Conversely, at Ti/IrO2
Ta2O5, similar TOC removals were obtained at both temperatures
(ranging from 71% (at 25 C) to 65% (at 40 C)), even when a significant increase on COD decay was accomplished. This behavior indicates that the recalcitrant compounds are formed.
At this point, the results obtained in the electrochemical treatment of brine PW suggest a promising alternative for the treatment
of this kind of efuent because its natural characteristics (pH,
salinity, concentration of Cl and temperature) allow its treatment
when this efuent is extracted from oil wells, not requiring any
pre-treatment. However, particular attention and experimental
observations must be taken into consideration before or during
its use, due to the formation of organochlorinated compounds
using DSA-electrode.
3.5. Electrochemical treatment of saline PW using Ti/IrO2Ta2O5 e BDD
Several studies in the literature show that the addition of chloride in the electrolytic solution increases the current efciency,
improving electrical conductivity and generation of active chlorine
species [40,45], favoring the oxidation of organic matter. However,
when the electrochemical treatment was applied to samples of saline PW, which have higher COD and chloride concentrations (see
Table 1), the results showed a slower reaction rate and lower
COD removal.
Fig. 4 shows COD removal, as a function of time, during the electrochemical treatment of saline PW using Ti/IrO2Ta2O5 (Fig. 4a)
and BDD (Fig. 4b) electrodes at 25 C. It was observed that at both
materials, COD decay was slight after 8 h of electrochemical treatment, independent on applied current density (10 mA cm2 or
20 mA cm2). Results clearly show that a modest anodic oxidation
of saline PW was performed, achieving 50% of COD removal
(Fig. 4a) by applying 10 and 20 mA cm2, at Ti/IrO2Ta2O5 elec-

Fig. 4. Inuence of applied current density on COD removal, as a function of time,

during saline PW electrolysis at 25 C for Ti/IrO2Ta2O5 (a) and BDD (b) anodes.
Insets: %TOC removal, as a function of applied current density. Operational
conditions: ow rate = 160 dm3 h1 and different applied current densities:
10 mA cm2 (circle) and 20 mA cm2 (square).

trode after 3.5 and 7 Ah dm3, respectively (Fig. 4S(a)). Nevertheless, an important observation from inset in Fig. 4a is that the
mineralization rate (TOC removal) seems not to be signicantly
inuenced by the higher concentrations of chloride in the petrochemical efuent, suggesting that the production of Cl2 is the preferential reaction (based on literature [43], DSA-electrodes are good
electrocatalysts for Cl2 production) instead the electrochemical
oxidation (direct or/and indirect). At the same time, the Cl-mediated oxidation does not lead to the strong oxidation conditions
necessary to eliminate completely the organic matter, but forming
recalcitrant compounds giving a false COD abatement. This outcome is conrmed by TOC removal efciencies, below 37%.
Likewise, for BDD anode, the COD removal, under similar conditions, occurs with similar efciencies than those achieved for Ti/
IrO2Ta2O5. As can be observed from Fig. 4b, 44% of COD removal
was achieved for BDD independent on applied current density
due to two features: higher Cl concentrations (86,875 mg dm3)
and higher initial COD (11,541 mg dm3).
From the information discussed above, it can be inferred that
the degradation process when higher Cl concentrations are present in the efuent, it occurs with signicant formation of Cl2 gas
in concomitance with higher production of O2, complicating the
complete oxidation of organic matter (see inset in Fig. 4b, TOC removal) and increasing the Q (3.5 Ah dm3 (10 mA cm2) and
7 Ah dm3 (20 mA cm2), see Fig. 4S(b)). Also, higher initial COD
is traduced in the production of several by-products decreasing
the efciency process [5]. This assertion is in agreement with the
TOC efciencies obtained during electrochemical treatment of
saline PW by applying 10 and 20 mA cm2 of current density.
These statements are important observations, because these demonstrate that for some kind of petrochemical efuents, no efcient


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

elimination of organic matter can be achieved, independent on nature of electrode material and experimental conditions used. But, as
pointed by Sales et al. [10], the electrochemical oxidation of industrial efuents in presence of higher concentrations of chloride may
lead to more toxic pollutants (organochloride by-products). In this
frame, the treatment of saline PW can be performed by using combined advanced oxidation processes, such as ltration, adsorption,
electrochemical and UV-radiation [46].

and conrmed by the decrease of conductivity. These results indicated that electrochemical treatment technology could be also
used for removing inorganic species from petrochemical wastewaters (Table 2). However, an integrate treatment (anodic oxidation
and electrocoagulation) could be planned in order to increase the
efciency and decrease the energy consumption.

3.6. BTEX and inorganic species removal

The adoption of an electrochemical technology as an efuent

treatment must consider some aspects in order to make its implementation feasible (performance of anode material, energy consumption and operating cost).
Fig. 5 shows the results of energy consumption, as a function of
%COD removal, during electrochemical treatment of fresh PW
(Fig. 5a) and brine PW (Fig. 5b). It is observed, for all kinds of water,
that the values are proportional to the applied current density during electrochemical treatment, and that there was a reduction in
energy requested when the temperature was increased (insets in
Fig. 5a and b).
For the treatment of fresh PW efuent, the electrode material
that showed the best performances was BDD anode (with average
cell potentials of 5.5 V, 7.7 V and 10.8 V for 1, 2.5, 5 mA cm2,
respectively), which showed lower energy consumptions (Fig. 5a)
with higher %COD removals than those achieved at Ti/IrO2Ta2O5
(with average cell potential of 3.7 V, 5.6 V and 8.8 V for 1, 2.5,
5 mA cm2, respectively) and consequently, a reduction on the
efuent treatment costs must remained (Table 3 illustrate the costs
of electrochemical treatment under different operating conditions). However, a signicant reduction on energy requirements
(due to the decrease on average cell potentials from 8.8 V to
7.5 V) was accomplished at Ti/IrO2Ta2O5, when an increase on
temperature was attained (see inset in Fig. 5a) for treating fresh
For removal of COD from brine PW, similar energy consumptions were estimated at both electrodes (Fig. 5b), achieving up to
79% of COD removal, increasing exponentially with the increase
on% of COD removal. On the other hand, when an increase on the
temperature (from 25 C to 40 C) was attained, energy requirements decreased substantially at both electrodes, increasing modestly the% of COD removal (see inset in Fig. 5b). Obviously, this

Due to the high toxicity of BTEX, which can cause irreparable injury to the central nervous system even in small concentrations,
Brazilian environmental laws have been strongly restrictive
regarding the releases of these contaminants in the aquatic environments [3,5,39]. From the data obtained by GCMS analyses,
electrochemical treatment proved to be very effective for removing
BTEX in operating conditions using Ti/IrO2Ta2O5 and BDD. For all
PW efuents (fresh, brine and saline), these pollutants were completely removed, obtaining results below the detection limit of
the method (0.5 ppb). On the other hand, the variation of inorganic
species concentration dissolved in the PW after electrochemical
process using Ti/IrO2Ta2O5 and BDD anodes, under specic experimental conditions, is present in Table 2. As can be observed, the
initial amount of inorganic species was removed from petrochemical wastewater after the electrochemical treatment. It appears
that the OH species produced during electrolysis of water, favor
the formation of hydroxides with some heavy metals. These inorganic compounds are deposited on the cathode forming a salt deposit, as observed at this electrode after electrolysis treatment;

3.7. Estimation of energy consumption

Table 3
Cost required for electrochemical treatment of fresh, brine and saline PW, as a
function of temperature and applied current density, after 2, 4 and 8 h of treatment,
respectively; according to the energy requirements obtained in Figs. 57.


j (mA cm2)

T (C)
















Fig. 5. Evolution of the energy consumption against% of COD removal during
electrochemical treatment of fresh (a) and brine (b) PWs on Ti/IrO2Ta2O5 and BDD
anodes. Insets: evolution of the energy consumption, as a function of temperature
and anode material.


Cost (US$ m3)


A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

4. Conclusions

Fig. 6. Evolution of the energy consumption against% of COD removal during

electrochemical treatment of saline PW on Ti/IrO2Ta2O5 and BDD anodes.

The study showed that the anodic oxidation can be an alternative treatment for COD and BTEX removal from petrochemical
efuent. Both electrode materials showed satisfactory results in
the oxidation process and it could be carried out by both direct
and mediated oxidation by powerful oxidants agents produced
on anode surface, using specic operating conditions.
The treatment efciency of petrochemical efuents shown to be
strongly dependent on nature of efuent and anode materials as
well as operating conditions, such as current density, chloride concentration and temperature. Consequently, the energy requirements can be traduced on lower costs of electrochemical
treatment, depending strongly on experimental conditions. It is
important to remark that, each efuent is different, for that reason,
when different concentrations of Cl are present; other undesired
by-products could be produced, as already conrmed by other
authors in a recent study [10].
Perhaps the production of undesired by-products, such as
organochloride compounds, can vary and it depends on Cl concentration into the real efuent. Then, particular attention and experimental observations must be taken into consideration before or
during the treatment of brine and saline PWs.
Finally, in the present work have recently allowed to start the
design of a pilot industrialelectrochemical cell to be used in petrochemical industry. These experiments are in progress and their
results will be reported in detail in a separate paper in a near

Fig. 7. Evolution of the energy consumption, as a function of% COD removal, during
electrochemical treatment of fresh, brine and saline PWs on Ti/IrO2Ta2O5 and BDD
anodes, at 25 C.

decrease on power energy will be traduced on a decrease on treatment costs of brine PW (Table 3). These gures are in agreement
with the data reported in the literature, where it was demonstrated
that the electrochemical treatment is performed at lower potentials in the presence of chlorides, compared with those required
for the direct anodic oxidation [47], diminishing the energy consumptions and costs.
For the treatment of saline PW efuent, the results obtained at
both electrode materials demonstrated that, higher energy consumptions were achieved (Fig. 6), independent on anode material
and applied current density, obtaining lower COD removal efciencies (about 50% at both electrode materials). However, Ti/IrO2
Ta2O5 electrode by applying 10 mA cm2 at 25 C showed lower
energy requirements than those obtained at BDD anode. This
behavior is due to lower cell potentials (3.5 V and 4.4 V for 10
and 20 mA cm2, respectively) using Ti/IrO2Ta2O5 anode [43],
favoring principally the production of Cl2 [47]. Conversely to BDD
electrode, where higher cell potentials were attained (7.9 V and
9.9 V for 10 and 20 mA cm2, respectively), producing Cl2 in concomitance with O2, limiting the oxidation of organic matter.
A general comparison between the trend of energy consumption, as a function of% of COD removal, for Ti/IrO2Ta2O5 and
BDD anodes is presented in Fig. 7. This graph clearly evidences that
the brine PW requires more energy consumptions that those used
for treating fresh and saline PWs. Also, in the case of BDD, it allows
to remove a signicant% of COD employing lower energy requirements that those required at Ti/IrO2Ta2O5, at different petrochemical efuents. However, when an increase on the
temperature is attained, Ti/IrO2Ta2O5 anode exhibits better performances (see insets in Fig. 5a and b).

A.J.C. da S. and E.V. dos S. gratefully acknowledge for CAPESMaster and PhD-PETROBRASPFRH22 fellowships, respectively.
The authors thank Industrie De Nora S.p.A. (Milan, Italy) for supply
the electrodes and PETROBRAS for providing the PW samples.
Financial support from National Council for Scientic and Technological Development (CNPq, Brazil) is also gratefully acknowledged. The authors thank to Prof. Djalma Ribeiro da Silva for his
assistance for performing the analyses of petrochemical efuents
at the Ncleo de Processamento Primrio e Reuso de gua Produzida e Residuos (NUPRAR-UFRN).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.cej.2013.08.023.
[1] R.M. Gomes de Lima, G.R.S. Wildhagen, J.W.S. Dias da Cunha, J.C. Afonso,
Removal of ammonium ion from produced waters in petroleum offshore
exploitation by a batch single-stage electrolytic process, J. Hazard Mater. 161
(2009) 15601564.
[2] F.R. Ahmadun, A. Pendashteh, L.C. Abdullah, D.R.A. Biak, S.S. Madaeni, Z.Z.
Abidin, Review of technologies for oil and gas produced water treatment, J.
Hazard Mater. 170 (2009) 530551.
[3] A.M.Z. Ramalho, C.A. Martnez-Huitle, D.R. Silva, Application of
electrochemical technology for removing petroleum hydrocarbons from
produced water using a DSA type anode at different ow rates, Fuel 89
(2010) 531534.
[4] B.H. Diyauddeen, W.M.A.W. Daud, A.R. Abdul Aziz, Treatment technologies for
petroleum renery efuents: a review, Process Saf. Environ. Prot. 89 (2011)
[5] J.H.B. Rocha, M.M.S. Gomes, N.S. Fernandes, D.R. da Silva, C.A. Martnez-Huitle,
Application of electrochemical oxidation as alternative treatment of produced
water generated by Brazilian petrochemical industry, Fuel Process. Technol. 96
(2012) 8087.
[6] P. Caizares, A. Beteta, C. Sez, L. Rodrguez, M.A. Rodrigo, Use of
electrochemical technology to increase the quality of the efuents of biooxidation processes. A case studied, Chemosphere 72 (2011) 10801085.

A.J.C. da Silva et al. / Chemical Engineering Journal 233 (2013) 4755

[7] M. Panizza, G. Cerisola, Electrocatalytic materials for the electrochemical
oxidation of synthetic dyes, Appl. Catal. B 75 (2007) 95101.
[8] M. Panizza, G. Cerisola, Removal of colour and COD from wastewater
containing acid blue 22 by electrochemical oxidation, J. Hazard. Mater. 153
(2008) 8388.
[9] C.A. Martnez-Huitle, E. 515 Brillas, Decontamination of wastewaters
containing synthetic organic dyes by electrochemical methods: a general
review, Appl. Catal. B 87 (2009) 105145.
[10] A.M.S. Sales, C.K.C. Arajo, J.V. Melo, J.M. Peralta-Hernandez, D.R. Silva, C.A.
Martnez-Huitle, Decontamination of real textile industrial efuent by strong
oxidant species electrogenerated on diamond electrode: viability and
disadvantages of this electrochemical technology, Appl. Catal. B 130131
(2013) 112120.
[11] C.A. Martnez-Huitle, E.V. Santos, D.M. Araujo, M. Panizza, Applicability of
diamond electrode-anode to the electrochemical treatment of a real textile
efuent, J. Electroanal. Chem. 674 (2012) 103107.
[12] S. Hammami, N. Bellakhal, N. Oturan, M.A. Oturan, M. Dachraoui, Degradation
of Acid Orange 7 by electrochemically generated OH radicals in acidic aqueous
medium using a boron-doped diamond or platinum anode: a mechanistic
study, Chemosphere 73 (2008) 678684.
[13] J.H.B. Rocha, A.M.S. Solano, N.S. Fernandes, D.R. da Silva, J.M. PeraltaHernandez, C.A. Martnez-Huitle, Electrochemical degradation of remazol red
BR and novacron blue CD dyes using diamond electrode, Electrocatalysis 3
(2012) 112.
[14] K. Bensadok, N. El Hana, F. Lapicque, Electrochemical treatment of dairy
efuent using combined Al and Ti/Pt electrodes system, Desalination 280
(2011) 244251.
[15] Y. Kitazono, I. Ihara, G. Yoshida, K. Toyoda, K. Umetsu, Selective degradation of
tetracycline antibiotics present in raw milk by electrochemical method, J.
Hazard. Mater. 243 (2012) 112116.
[16] M. Panizza, I. Sirs, G. Cerisola, Anodic oxidation of mecoprop herbicide at lead
dioxide, J. Appl. Electrochem. 38 (2008) 923929.
[17] N. Oturan, E. Brillas, M.A. Oturan, Unprecedented total mineralization of
atrazine and cyanuric acid by anodic oxidation and electro-Fenton with a
boron-doped diamond anode, Environ. Chem. Lett. 10 (2) (2012) 165170.
[18] G.R.P. Malpass, D.W. Miwa, S.A.S. Machado, P. Olivi, A.J. Motheo, Oxidation of

the pesticide atrazine at DSA electrodes, J. Hazard. Mater. B 137 (2006) 565
[19] Y. Samet, L. Agengui, R. Abdelhdi, Electrochemical degradation of chlorpyrifos
pesticide in aqueous solutions by anodic oxidation at boron-doped diamond
electrodes, Chem. Eng. J. 161 (2010) 167172.
[20] A. Kesraoui Abdessalem, N. Bellakhal, N. Oturan, M. Dachraoui, M.A. Oturan,
Treatment of a mixture of three pesticides by photo- and electro-Fenton
processes, Desalination 250 (2010) 450455.
[21] F. Fu, Q. Wang, Removal of heavy metal ions from wastewaters: a review, J.
Environ. Manage. 92 (2011) 407418.
[22] W.T. Mook, M.H. Chakrabarti, M.K. Aroua, G.M.A. Khan, B.S. Ali, M.S. Islam, M.A.
Abu Hassan, Removal of total ammonia nitrogen (TAN), nitrate and total
organic carbon (TOC) from aquaculture wastewater using electrochemical
technology: a review, Desalination 285 (2012) 113.
[23] M. Panizza, M. Delucchi, I. Sirs, Electrochemical process for the treatment of
landll leachate, J. Appl. Electrochem. 40 (2010) 17211727.
[24] I. Sirs, E. Brillas, Remediation of water pollution caused by pharmaceutical
residues based on electrochemical separation and degradation technologies: a
review, Environ. Int. 40 (2012) 212229.
[25] A. Dirany, I. Sirs, N. Oturan, M.A. Oturan, Electrochemical abatement of the
antibiotic sulfamethoxazole from water, Chemosphere 81 (2010) 594602.
[26] M. Haidar, A. Dirany, I. Sirs, N. Oturan, M.A. Oturan, Electrochemical
degradation of the antibiotic sulfachloropyridazine by hydroxyl radicals
generated at a BDD anode, Chemosphere 91 (2013) 13041309.


[27] J. Wu, H. Zhang, N. Oturan, Y. Wang, L. Chen, M.A. Oturan, Application of

response surface methodology to the removal of the antibiotic tetracycline by
electrochemical process using carbon-felt cathode and DSA (Ti/RuO2IrO2)
anode, Chemosphere 87 (2012) 614620.
[28] N. Oturan, J. Wu, H. Zhang, V.K. Sharma, M.A. Oturan, Electrocatalytic
destruction of the antibiotic tetracycline in aqueous medium by
electrochemical advanced oxidation processes: effect of electrode materials,
Appl. Catal. B 140141 (2013) 9297.
[29] K. Chanworrawoot, M. Hunsom, Treatment of wastewater from pulp and paper
mill industry by electrochemical methods in membrane reactor, J. Environ.
Manage. 113 (2012) 399406.
[30] C.A. Martnez-Huitle, L.S. Andrade, Quim. Nova 34 (2011) 850.
[31] M. Panizza, G. Cerisola, Direct and mediated anodic oxidation of organic
pollutants, Chem. Rev. 109 (2009) 65416569.
[32] C.A. Martnez-Huitle, S. Ferro, Electrochemical oxidation of organic pollutants
for the wastewater treatment: direct and indirect processes, Chem. Soc. Rev.
35 (2006) 13241340.
[33] E. Brillas, I. Sires, M.A. Oturan, Electro-Fenton process and related
electrochemical technologies based on Fentons reaction chemistry, Chem.
Rev. 109 (2009) 65706631.
[34] E.V. dos Santos, J.H.B. Rocha, D.M. Araujo, D.C. Moura, C.A. Martnez-Huitle,
Decontamination of produced water containing petroleum hydrocarbons by
electrochemical methods: a mini-review, Curr. Org. Chem., accepted for
publication, http://dx.doi.org/10.2174/1385272043369421.
[35] M.R.G. Santos, M.O.F. Goulart, J. Tonholo, C.L.P.S. Zanta, The application of
electrochemical technology to the remediation of oily wastewater,
Chemosphere 64 (2006) 393399.
[36] B.K. Krbahti, K. Artut, Electrochemical oil/water demulsication and
purication of bilge water using Pt/Ir electrodes, Desalination 258 (2010)
[37] APHA, AWWA, WEF, Standard methods for the examination of water and
wastewater, 21st ed., APHA/AWWA/WEF, Washington, DC, 2005.
[38] J. Lu, X. Wang, B. Shan, X. Li, W. Wan, Analysis of chemical compositions
contributable to chemical oxygen demand (COD) of oileld produced water,
Chemosphere 62 (2006) 322331.
[39] Environmental Brazilian Laws: Resoluo Ministrio do Meio Ambiente
Conselho Nacional do Meio Ambiente (CONAMA) No. 430, de 13 de maio de.
Dirio Ocial da Unio, Braslia, DF, 2011.
[40] Y. Yavuz, A. Savas Koparal, .B. gtveren, Treatment of petroleum renery
wastewater by electrochemical methods, Desalination 258 (2010) 201205.
[41] T.-N. Wu, Electrochemical removal of MTBE from water using the iridium
dioxide coated electrode, Sep. Purif. Technol. 79 (2011) 216220.
[42] M.G. Tavares, L.V.A. Silva, A.M. Sales Solano, J. Tonholo, C.A. Martnez-Huitle,
C.L.P.S. Zanta, Electrochemical oxidation of Methyl Red using Ti/Ru0,3Ti0,7O2
anodes, Chem. Eng. J. 204206 (2012) 141150.
[43] L. Vazquez-Gomez, S. Ferro, A. De Battisti, Preparation and characterization of
RuO2IrO2SnO2 ternary mixtures for advanced electrochemical technology,
Appl. Catal. B 67 (12) (2006) 3440.
[44] C.A. Martinez-Huitle, E. Brillas, Electrochemical alternatives for drinking water
disinfection, Angew. Chem., Int. Ed. 47 (2008) 19982005.
[45] D. Santos, M. Dezotti, A.J.B. Dutra, Electrochemical treatment of efuents from
petroleum industry using a Ti/RuO2 anode, Chem. Eng. J. 226 (2013) 293299.
[46] G. Li, T. An, J. Chen, G. Sheng, J. Fu, F. Chen, S. Zhang, H. Zhao,
Photoelectrocatalytic decontamination of oileld produced wastewater
containing refractory organic pollutants in the presence of high
concentration of chloride ions, J. Hazard. Mater. 138 (2006) 392400.
[47] F. Bonfatti, S. Ferro, F. Lavezzo, M. Malacarne, G. Lodi, A. De Battisti, J.
Electrochem. Soc. 147 (2000) 592596.