Electrochimica Acta 52 (2007) 4161–4166

Indium tin oxide coated conducting glass electrode for

electrochemical destruction of textile colorants
J. Bandara ∗ , P.T. Wansapura, S.P.B. Jayathilaka
Institute of Fundamental Studies, Photochemistry Group, Hantana Road, Kandy CP 20000, Sri Lanka
Received 2 October 2006; received in revised form 17 November 2006; accepted 27 November 2006
Available online 18 December 2006

Abstract
Galvanostatic oxidation of 5.0 × 10−2 mM textile dyes such as Eosin Y (EY) and Orange II (Or II) was carried out on an indium tin oxide (ITO)
coated glass anode in the presence of 1.0 × 10−2 mM KCl solution at pH 4.0 and 6.0. The degradation results of EY were compared with that of
highly stable azo dyes (Or II). EY dye solution with a concentration of 5.0 × 10−2 mM is totally decolorized in 30 min at an electrical charge (Q)
0.067 A h dm−3 while 5.0 × 10−2 mM Or II degraded in a little less than an hour at the same electrical charge density. The decay kinetics of dyes
follows a pseudo first-order reaction. The degradation of dyes is faster in acidic pH values than in basic pH values. Electrochemical degradation
results show significant decrease in chemical oxygen demand (COD) values after electrodegradation of textile dyes. The key advantage of the ITO
conducting glass anode is that the deposition of polymeric materials on the anode surface during electro-degradation of textile dyes is absent and
therefore the electrode fouling is not observed. Hence, the ITO anodes can be employed an extended period without loss of activity.
© 2006 Elsevier Ltd. All rights reserved.

Keywords: Orange II; Eosin; Electro-degradation; Azo dyes; Electrochemical destruction

1. Introduction process. They include physical or chemical flocculation, elec-

trofilteration, membrane filtration, electrokinetic coagulation
A large amounts of colored wastewater containing unfixed [3–5]. Over the past few years, alternative methods for dye treat-
dyestuff are released from dye-houses as a by-product from tex- ment have been investigated, including chemical oxidation with
tile dyeing operations. Synthetic textile dyes of the azo family reagents such as: ozone, hydrogen peroxide, ozone/UV, hydro-
represent an important part of the world production of synthetic gen peroxide/UV and Fenton’s reagent [6–10]. Other alternative
dyes and are characterized by the presence of azo group. Azo techniques such as biological treatment [11–13], adsorption
dyes released without proper treatment represents about 15% of and precipitation [14–16] and reductive treatment [17–19] have
the total world production and azo dyes are resistant to bacterial applied successfully. For all the proposed techniques, great care
activity [1,2]. The other most commonly used textile dye today has to be taken to keep the costs of such processes at reasonable
is Eosin dyes, which are supposed to be resistant to bacterial levels and to avoid the formation of byproducts, which cause
activity [1]. Thus, the accumulation of those dyestuffs in nature additional costs for secondary treatment or disposal.
is a growing concern related to the health impact. Therefore, the Among the remediation techniques described above, electro-
development of effective methods for their removal from water chemical treatment is a very promising method for the reduction
bodies is warranted. or decolorizing of toxic pollutants by decomposing the organic
To solve the problem of highly colored wastewater, numerous substances dissolved in wastewater. In recent years, anodic
attempts to decolorize textile effluents have been proposed. Dye oxidation has been the focus of much research attention in
wastewater is usually treated by physical or chemical treatment wastewater treatment because of its high oxidation efficiency,
fast reaction rate and ease of operation. It is important to select
the proper electrode material, because the electrolytic products
∗ Corresponding author. Tel.: +94 81 2232002; fax: +94 81 2232131.
strongly depend on those materials as well as the operating con-
E-mail addresses: [email protected], ditions [20,21]. Selection of a proper electrode material is vital
[email protected] (J. Bandara). for an efficient and enduring operation of an electrode. It is well

0013-4686/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2006.11.036
4162 J. Bandara et al. / Electrochimica Acta 52 (2007) 4161–4166

known that the current efficiency (CE) of an oxidation process

depends strongly on the electrode properties. In general, the
anodes must have high O2 evolution over potential. Otherwise,
a large amount of the supplied current will be wasted in pro-
ducing O2 , leading to a low CE. In this respect, several anode
materials such as graphite and platinum which do not have a suf-
ficient O2 evolution over potential have been tested, but some
of them presented a rapid loss of activity due to surface foul-
ing and others showed a limited service life [22,23]. In fact,
there are very few electrodes currently available that have high
O2 evolution over potential. These few are mainly limited to
PbO2 , SnO2 and TiO2 [24–26]. Several researchers have been
investigating electrically conductive diamond as an electrode in
various electrochemical fields and recently, attempts have been
made to fabricate new types of diamond electrodes because they
exhibit high over potentials for the oxygen and hydrogen evolu-
tion resulting in a wide electrochemical potential window, which
makes them better electrodes [27–31].
Therefore, it is imperative to find alternative cheap anode
Fig. 1. CVs of EY at different initial concentration in 1.0 × 10−2 M KCl and
materials, which can be used for longer period without any foul-
50 mV S−1 scan rate in the potential region 0–2.0 V (vs. SCE).
ing. ITO coated glass refers to glass with has an ITO layer coated
through physical vapor deposition in vacuum conditions and ITO
absorbance measurements were done using Shimadzu UV-160A
coated glass maintains high conductivity and high transparency.
spectrophotometer operated from 200–600 nm in quartz cell.
ITO conductive glasses have been used successfully in light
The total organic carbon (TOC) of the solution was measured
emitting devices [32,33], gas sensors [34,35], hetero-junction
using the Total Organic Carbon analyzer (Shimadzu 5000A) and
solar cells [36,37] and electrochemical degradation of nitric
the chemical oxygen demand (COD) was carried out with Pal-
oxides [38–40]. However, ITO anodes have not been tested for
intest digital Tubetest Heater. For COD analysis, 2 mL of test
electrochemical degradation of textile dyes and other common
solution were pipetted into a dichromate reagent and digested
pollutants effectively. Therefore, in this study, we investigated
at 150 ◦ C for 2 h. The color developed during the oxidation was
the cheap ITO coated conducting glasses as anodic materials for
measured against a blank using Palintest Photometer 5000 col-
the degradation of textile dyes and found that the surface fouling
orimeter. The changes in the optical density in the dichromate
is absent on these anode materials.
solution were determined at λ 570 nm. Peroxide concentration of
the solutions was measured using Merckoquant® paper indicator
2. Experimental
obtained from Merck Schweiz AG. The limit for H2 O2 detection
by Merckoquant paper indicator is 25 mg/L with a precision of
All aqueous solutions were prepared with deionized water.
(0.5 mg/L). A triiodide method described by Klassen et al. [41]
Eosin and Or II were from Fluka. Potassium chloride (Fluka,
was also used to confirm the peroxide concentration in which
p.a.) and Oxygen were used as received. ITO coated conducting
detection of I3 − peak at 357 nm was followed in a Shimadzu
glasses with a sheet resistance of 8–10  was purchased from
UV-160A spectrophotometer. All pH measurements were made
Solaronix. All electrochemical measurements were performed
with a Jenco Model 1671 pH meter with a glass electrode, which
using Hokuto Denko potentiostat/galvanostat model 301. The
had been calibrated previously. pH was adjusted by using HCl
cell used for voltammetric experiment was of a three-electrode
or NaOH solutions.
type. The working electrode was ITO conducting glass and
Ag|AgCl, KCl (saturated) and Pt electrodes were used as ref-
3. Results and discussions
erence and counter electrodes, respectively. The surface area
of working electrode was 1 cm × 2 cm. The electro-degradation
3.1. Voltammetric behavior of EY dye
was carried out in one-compartment cell under galvanostatic
conditions with various current densities and substrate concen-
The anodic oxidation of Eosin was carried out under differ-
trations. Linear sweep voltammetry was performed on Eosin Y
ent current densities, initial dye concentrations and pH values.
(EY) using platinum working and counter electrodes in unstirred
Fig. 1 shows cyclic voltammograms (CVs) of EY with the vari-
bulk solutions with 1.0 × 10−2 M KCl as supporting electrolyte
ation of dye concentration which were obtained at a scan rate
at sweeping rate of 50 mV s−1 .
of 50 mV s−1 in the potential region 0–2.0 V (versus SCE) in
Electrolysis cell contained 40 mL of dye solution and
1.0 × 10−2 M KCl1 solution. An anodic current correspond-
supporting electrolyte (1.0 × 10−2 M KCl). Two millilitre of
samples were withdrawn at intervals for analysis and correc-
tions for the change in volume were made. Throughout the 1 In this investigation, KCl was used as an electrolyte and if electrolysis is

experiments, O2 gas was bubbled at a rate of 10 mL min−1 . The carried out at negative potentials (∼1.4 V), this may generate Cl2 , which is
 J. Bandara et al. / Electrochimica Acta 52 (2007) 4161–4166 4163

Fig. 3. Potentiostatic i–t curves recorded at 3 V vs. SCE in a solution of EY (a)

0.5 mM and (b) 50 ␮M. (c) Potentiostatic i–t curve recorded at 7 V vs. SCE in a
solution of 0.5 mM EY.

electrolysis experiments without reactivation of the electrode.

During long-term use of the electrode, the deactivation or the
decrease of electro-reactivity of the electrode was not observed
which will be explained later. By consecutive CV analysis it
was observed that deactivation of electrode was not observed
even with other compounds such as 2-chlorophenol (2CP), i.e.
preliminary experiments were done with 0.5 mM of 2CP and it
was noticed stable anodic peak intensity during consecutive CV
analysis of 2CP (Fig. 2c). Stability of electrode was further stud-
ied by potentiostatic i–t curves recorded for 0.5 M and 50 ␮M
EY solutions at 1.5–3.0 V (SCE) which was shown in Fig. 3.
Only a slight decrease in current density was observed initially
but relatively stable current was observed in both cases indicat-
ing that the electrode does not get deactivated with the reaction
time by deposition of polymeric by product materials. The sta-
Fig. 2. Consecutive CVs of EY and 2CP in 0.1 M KCl at scan rate 50 mV, (a) bility of the anode was further tested at higher voltage and the
50 ␮M of EY, (b) 0.5 mM of EY and (c) 0.5 mM of 2CP. i–t curve at 7 V is shown in Fig. 3, curve c. The time dependence
of current at higher voltage is similar to that of lower voltage
ing to the oxidation of EY is observed at about 0.9 V. With indicating that the electrode can be employed as anode for the
the increase of dye concentration, the anodic oxidation peak electrolysis of pollutant at high voltage as well.
of EY shifts slightly to positive potential while peak intensity
increases linearly with the increase of dye concentration. Fig. 2a 3.2. Electrolyses results of EY dye
and b shows consecutive CVs of solution containing 50 ␮M
and 0.5 mM of EY in KCl solution, respectively. As shown in A series of galvanostatic runs were carried out to asses the
Fig. 2a and b, the anodic peaks intensities do not decrease with effect of operating parameters such as current density, pH, tem-
the number of CV scans indicating that the electrode activity perature and initial EY concentration on the electrochemical
does not decrease or deteriorate due to deposition of polymeric kinetics. EY underwent gradual decolorization with increasing
by-products on the electrode surface. In fact, the stability of the electrolysis time. Fig. 4 shows the anodic oxidation of EY as a
electrode was confirmed by the use of same electrode for several function of specific charge consumed (Q, Ah dm−3 ) for a 50 ␮M
EY solution treated at current densities of 7 mA cm−2 (curve a),
5 mA cm−2 (curve b) and 3 mA cm−2 (curve c). As shown in
also another strong oxidant. To avoid Cl2 generation, other electrolytes such as Fig. 4, decolorization is rapid when current raises from 3 to
Na2 SO4 or HClO4 equally can be employed. 7 mA cm−2 and the total decolorization of EY was observed
4164 J. Bandara et al. / Electrochimica Acta 52 (2007) 4161–4166

Fig. 4. Variation of concentration of EY during electrolysis on ITO in 0.1 M KCl Fig. 6. H2 O2 formation during electrolysis of 0.1 M KCl in water under (a) O2
at a scan rate of 50 mV with the variation of current density: (a) 7 mA cm−2 , (b) saturated and pH 3.0, (b) O2 saturated and pH 9.0 and (c) Ar saturated and pH
5 mA cm−2 and (c) 3 mA cm−2 . 3.0 on ITO electrode.

in less than 25 min and in 60 min for current densities 7 and

3 mA cm−2 , respectively, which shows that decolorization of
Fig. 5 shows the effect of the initial EY concentration on
EY is not proportional to current density. At current density of
decolorization during electrolysis at 25 ◦ C by applying current
7 mA cm−2 , a potential of ∼3.0 V was being applied and as cur-
density of 6 mA cm−2 . At the beginning of the electrolysis, the
rent density increases, the electrolysis cell is operated at higher
concentration of EY decreases linearly with specific charge,
voltage.
which indicates that under these conditions, the decolorization of
EY is under current control. Also at lower EY concentration, i.e.
30–70 ␮M, the decolorization rate is proportional to the charged
consumed but it deviates for higher EY concentration.
It was observed that the decolorization of EY is highly
depended on solution pH and EY under goes decoloriation either
with Ar or O2 saturated conditions. The fact the EY undergoes
decolorization under Ar saturated condition confirms that rup-
ture of EY structure before it undergoes electro-oxidation. The
TOC measurements during electrolysis of EY under O2 and Ar
saturated indicate that under Ar saturated condition, a decrease
in TOC value was not observed while under O2 saturated con-
dition a decrease in TOC was observed. These results confirm
that under O2 saturation conditions, oxidative species could be
generated via the following reaction and the generated oxidative
species are responsible for mineralization of EY.
O2 + 2H+ + 2e− = H2 O2 , E◦ = 0.695 V
To confirm the generation of oxidative species, we inves-
tigated the formation of H2 O2 under similar electrolysis
conditions without EY in the solution. Fig. 6 shows the H2 O2
amount as a function of electrolysis time at two different pH
values in O2 and Ar saturated conditions2 . As shown in Fig. 6,
Fig. 5. Variation of concentration of EY during electrolysis on ITO in 0.1 M
KCl at a scan rate of 50 mV at current density of 6 mA cm−2 with the variation
2 It should be noted here that the precious measurement of peroxide concen-
of initial EY concentration: (a) 80 ␮M, (b) 70 ␮M, (c) 60 ␮M and (d) 40 ␮M of
EY. tration is a difficult task and therefore we used both iodometirc method [41]
 J. Bandara et al. / Electrochimica Acta 52 (2007) 4161–4166 4165

Fig. 7. Effect of solution pH during electrolysis of 50 ␮M of EY on ITO in 0.1 M Fig. 8. COD change of 50 ␮M of EY during electrolysis on ITO electrode in
KCl at current density of 6 mA cm−2 (a) pH 6.0, (b) pH 4.0, (c) pH 3.0 and (d) 0.1 M KCl at current density of 6 mA cm−2 and pH 4.0 under O2 saturated
pH 1.0. Inset: The decolorization rate of 50 ␮M of EY at pH 4.0 under O2 and condition.
Ar saturated conditions during electrolysis at current density of 6 mA cm−2 .

the formation of H2 O2 was not observed when the solution was

saturated with Ar (curve c). Under O2 saturated condition, at
acidic pH (curve a), formation of a higher amount of H2 O2 was
observed compared with the H2 O2 amount produced at basic
pH values (curve b). Theses oxidative species could react with
EY and the by-products of EY and mineralize EY and hence a
decrease in TOC under O2 saturated can be understood.
Fig. 7 shows the decolorization/degradation of EY at differ-
ent pH values. Decrease in solution pH from 6.0 to 3.0 results in
increase in EY decolorization rate by a factor of 2. The effect of
pH on the decolorization can be understood based on H2 O2 for-
mation where a higher oxidative H2 O2 formation was observed
at acidic pH than in basic pH values as described above. There-
fore, a higher decolorization/degradation rate was expected at
acidic pH than in basic pH values. Inset in Fig. 7 shows the decol-
orization of EY at pH 4.0 under O2 and Ar saturated conditions.
According to decolorization rates shown, a higher decoloriza-
tion was observed under O2 than under Ar indicating that under
O2 both decolorization and degradation of EY occurs simulta-
neously. On the other hand, under Ar saturated condition, only
decolorization of EY was observed.
Chemical oxygen demand of EY with the electrolysis time
is given in Fig. 8. COD was determined after treating the EY
solution at pH 4.0 under current density of 6 mA cm−2 . Under
the above reaction condition, decolorization of EY is completed
in 30–40 min. However, from the results shown in Fig. 8, total
mineralization of EY does not happen in 30–40 min. Therefore,
it can be concluded that the decolorization of EY is followed by
Fig. 9. Absorption spectra of dye solutions during electrolysis on ITO in 0.1 M
KCl at pH 4.0 and current density of 6 mA cm−2 (A) 50 ␮M of EY: (a) 0 min,
and commercially available peroxide sensitive Merckoquant paper indicator to (b) 10 min, (c) 20 min and (d) 30 min; (B) 50 ␮M of Or II: (a) 0 min, (b) 10 min,
assay the peroxide amount in the solution. (c) 20 min, (d) 30 min, (e) 45 min and (f) 60 min.
4166 J. Bandara et al. / Electrochimica Acta 52 (2007) 4161–4166

mineralization of EY according to the following reaction: [2] J. Bandara, J.A. Mielczarski, J. Kiwi, Appl. Catal. B: Environ. 34 (2001)
307.
C20 H6 Br 4 Na2 O5 + 35H2 O [3] S.J. Judd, P. Hillis, Water Res. 35 (2001) 2895.
[4] A.A. Chan, Environ. Prog. 25 (2006) 152.
→ 20CO2 + 4Br− + 2Na+ + 76H+ + 74e− [5] J. Bandara, J.A. Mielczarski, J. Kiwi, Appl. Catal. B: Environ. 34 (2001)
321.
As mentioned above, for total decolorization of EY it takes only [6] W.S. Perkins, W.K. Walsh, I.E. Reed, C.G. Namboodi, Text. Chem. Color.
30 min while corresponding COD decrease is only by 75% indi- 28 (1995) 31.
cating that mineralization of by products takes longer time. One [7] E. Brillas, B. Boye, I. Sires, J.A. Garrido, R.M. Rodrigues, C. Arias, P.-L.
of the important factors of electrode material is the stability of Cabot, C. Comninellis, Electrochim. Acta 49 (2004) 325.
the electrode. In fact, the stability of the electrode was confirmed [8] Y.W. Kang, K.Y. Hwang, Eater Res. 34 (2000) 4487.
[9] M. Bader, J. Hoigne, Water Res. 15 (1981) 449.
by the use of same electrode for several electrolysis experiments [10] M. Faouzi, P. Canizares, A. Gadri, J. Lobato, B. Nase, R. Paz, M.A. Rodrigo,
and the decrease of activity was not observed. C. Saez, Electrochim. Acta 52 (2006) 325.
The ITO coated anode was tested for the electrochemical [11] A.A. Oxpring, G. McMullan, W.F. Smyth, R. Merchant, Biotechnol. Lett.
degradation of other textile dye such as Or II. Changes of absorp- 18 (1996) 527.
tion spectra during electrolyses of EY and Or II are given in [12] C.T. Matos, S. Velizarov, J.G. Crespo, M.A.M. Reis, Water Res. 40 (2006)
231.
Fig. 9. As shown in Fig. 9, EY decolorizes in 30 min (Fig. 9A) [13] H. Zilouei, B. Guieysse, B. Mattiasson, Process Biochem. 41 (2006)
while Or II takes 60 min for total decolorization under similar 1083.
electrolysis conditions (Fig. 9B). The variation in decoloriza- [14] K.R. Ramakrishna, T. Viraraghavan, Am. Dyestuff Rep. 81 (1996) 15.
tion kinetics of different dyes may arise due to differences [15] S. Rio, L. Le Coq, C. Faur, P. Le Cloirec, Water Sci. Technol. 53 (2006)
in molecular structure and electrochemical properties of dye 237.
[16] G. Crini, Bioresour. Technol. 97 (2006) 1061.
molecules. The anode was also further tested for the electro- [17] J. Tokuda, R. Ohura, T. Iwasaki, Y. Takeuchi, A. Kawashida, M. Nango,
chemical degradation of chlorophenol (CP) and noted that CPs Text. Res. J. 69 (1999) 956.
can be degraded electrochemically effectively. These results will [18] B. Nasr, G. Abdellatif, J. Electrochem. Soc. 152 (2005) D113.
be presented in another publication. Therefore, it can be con- [19] M. Panizza, P.A. Michaud, G. Cerisola, Ch. Comninellis, Electrochem.
cluded that several types of textile dyes and other pollutants can Commun. 3 (2001) 336.
[20] D. Johnson, J. Feng, J. Houkm, Electchim. Acta 46 (2000) 323.
be decolorized/degraded using ITO electrode. [21] W. Zhang, T. An, X. Xiao, J. Fu, H. Sheng, M. Cui, G. Li, Appl. Catal. A:
Gen. 255 (2003) 221.
4. Conclusion [22] N.B. Tahar, A. Savall, J. Electrochem. Soc. 145 (1998) 3427.
[23] J. Zielonka, R. Podsiadly, M. Czerwinska, A. Sikora, J. Sokolowaska, A.
ITO conducting glass anode was employed successfully for Marcinek, J. Photochem. Photobiol. A: Chem. 163 (2004) 373.
[24] J. Feng, D.C. Johnson, J. Electochem. Soc. 138 (1991) 3329.
the degradation/decolorization of textile colorant. Under Ar sat- [25] R. Kotz, S. Stucki, B. Carcer, J. Appl. Electrochem. 21 (1991) 1420.
urated conditions, only decolorization of dye was observed while [26] J.M. Kesselman, O. Weres, N.S. Lewis, M.R. Hoffmann, J. Phys. Chem. B
under O2 saturated condition, dyes under go both decoloriza- 101 (1997) 2637.
tion and mineralization. The decolorization is highly depended [27] M. Granger, G. Swain, J. Electochem. Soc. 146 (1999) 4551.
[28] M. Hupert, A. Muck, J. Wang, J. Stotter, Z. Cvackova, S. Haymond, Y.
on the solution pH as well as the initial dye concentration. The
Show, G. Swam, Diamond Relat. Mater. 12 (2003) 1940.
key advantage of the ITO conducting glass anode is that the [29] G. Chen, Sep. Purif. Technol. 38 (2004) 11.
deposition of polymeric materials on the anode surface during [30] B. Marselli, J. Garcia-Gomez, P.A. Michaud, M.A. Rodrigo, Ch. Com-
electro-degradation of pollutant is absent and therefore the elec- meinellis, J. Electrochem. Soc. 150 (2003) 79.
trode fouling is not observed. Hence, cheap ITO anodes can be [31] A. Kraft, M. Stadelmann, M. Blaschke, J. Hazard. Mater. 103 (2003) 247.
[32] P.-C. Liu, C.-Y. Hou, Y.-S. Wu, Thin Solid Films 478 (2005) 280.
employed for an extended period without loss of activity for the
[33] I.-M. Chan, F.C.-N. Hong, Thin Solid Films 444 (2003) 254.
electrodegradation of textile dyes. [34] J. Zhang, J. Hu, Z.Q. Zhu, H. Gong, S.J. Oshea, Collod. Surf. A: Physic-
ochem. Eng. Aspects 236 (2004) 23.
Acknowledgement [35] V. Georgieva, M. Ristov, Sol. Energy Mater. Sol. Cells 73 (2002) 67.
[36] G. Hodes, L. Thompson, J. Dubow, K. Rajeshwar, J. Am. Chem. Soc. 105
Authors are thankful to Buddini Dissanayake for her help to (1983) 324.
[37] J. Pla, M. Tamasi, R. Rizzoli, M. Losurdo, E. Centurioni, O. Summonte, F.
conduct some of the experiments. Rubinelli, Thin Solid Films 425 (2003) 185.
[38] J. Lei, N.S. Trofimova, Chem. Lett. 32 (2003) 610.
References [39] J. Lei, H. Zu, O. Ikeda, J. Electroanal. Chem. 567 (2004) 331.
[40] N.S. Trofimova, A.Y. Safronov, O. Ikeda, Electchim. Acta 50 (2005) 4627.
[1] P. Chudoba, P.P. Pitter, Biodegradability of Organic Substances in the [41] N.V. Klassen, D. Marchington, H.C.E. McGowan, Anal. Chem. 66 (1994)
Aquatic Environment, CRC Press, Boca Raton, FL, 1990. 2921.