Electrochemical-assisted photodegradation of dye on TiO2thin films:
investigation on the effect of operational parameters
Zulkarnain Zainal∗, Chong Yong Lee, Mohd Zobir Hussein, Anuar Kassim, Nor Azah Yusof
Department of Chemistry, Faculty of Science, Universiti Putra Malaysia, 43400 UPM Serdang,
Selangor Darul Ehsan, Malaysia
Received 19 February 2004; received in revised form 12 September 2004; accepted 2 November 2004
Available online 28 December 2004
dip-coating method. Several operational parameters to achieve optimum efficiency of this electrochemical-assisted photodegradation system
have been tested. Photoelectrochemical degradation was studied using different light sources and light intensity. The light sources chosen
ranged from ultraviolet to visible light. The effect of agitation of the solution at different speeds has also been studied. Slight improvement of
can be reused up to 20 times with percentage of deficiency less than 15%. The study on the effect of solution temperature indicated that the
activation energy of the methyl orange degradation is 18.63kJmol−1.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Photoelectrochemical degradation; Thin films; Methyl orange; Sol–gel; TiO2
In the last two decades, numerous studies have been per-
formed on heterogeneous photocatalysis process, which has
researchers due to the prospect of complete mineralisation of
the pollutants into harmless compounds [1–3]. In addition,
the availability of abundance and inexpensive catalyst with
non-toxicity, high photoactivity and high stability (biologi-
cally and chemically) features have accelerated the research
activities in this field. Among the semiconductors, titanium
dioxide (TiO2) has been proven to be an excellent photo-
catalyst material, which many organic substances have been
shown to be oxidatively or reductively degraded [4–9].
The high degree of recombination between photogener-
ated electrons and holes in semiconductor particles is a ma-
∗Corresponding author. Tel.: +603 89466810; fax: +603 89435380.
E-mail address: email@example.com (Z. Zainal).
jor limiting factor for photodegradation process [10,11]. In
ducting substrates, on which electrical bias potential can be
applied. The application of the external potential can drive
could minimise the charge recombination process [12,13].
Another advantage of using the immobilising technique is
that the electrode can be easily recycled, while in suspen-
sion or slurry system costly post-filtration processes are
TiO2 (anatase) is a semiconductor that has bandgap
energy of 3.2eV, where only the light with wavelengths
below 380nm can be absorbed to generate the electron–hole
pairs. It is known that <5% of the solar energy reaching
the surface of the earth is ultraviolet light ; therefore,
significant photodegradation efficiency is difficult to be
achieved by natural sunlight in short irradiation period.
This factor places some limitations on the application
of this technology in the treatment of wastewater, since
artificial light sources need high electrical power and are
0304-3894/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
Fig. 1. Chemical structure of methyl orange.
expensive. Besides that, the efficiency of controlling the
parameters such as solution agitation and temperature in
aqueous pollutants in real wastewater treatment also seems
to determine the photoreactor efficiency.
In this paper, TiO2films were fabricated by sol–gel dip-
coating method. This study was undertaken to investigate the
effects of light source, light intensity, agitation, electrode re-
ical degradation process of a model pollutant, methyl orange
dye. Methyl orange (Fig. 1) was selected as model dyeing
pollutant because azo-dyes are among the largest group of
colorants used in a variety of industrials such as textile and
in designing a photoreactor with optimum efficiency.
2.1. Preparation of precursor solutions
system that containing titanium tetraisopropyl-orthotitanate,
polyethelene (glycol) (molecular weight, Mw=2000), di-
sequences. The molarity of alkoxide in the ethanol was
0.94mol/dm3. The molar ratio of diethanolamine to the
alkoxide was one. The concentration of polyethylene gly-
col and water to alkoxide was 6 and 0.8wt.%, respectively.
solution before adding other chemicals in the following
sequences; diethanolamine, titanium tetraisoprophyl ortho-
tianate and water. The mixture was stirred in a sealed con-
dition for several hours at room temperature. The resulting
sol–gel was clear and transparent.
2.2. Preparation of TiO2thin films
The titanium plates (5cm×2cm) were used as the con-
by silicon carbide paper (Bioanalytical system PK-4 polish-
ing kit) and later cleaned with acetone in an ultrasonic bath
for 15min and then dip-coated with sol–gel solution and left
to dry at room temperature. The coated electrode was heated
at 100◦C for 5min in an oven followed by subsequent dip-
was annealed at 500◦C in a Thermolyne 21100 furnace for
2.3. Analytical measurements
in the two-compartment cell equipped with a quartz plane
were separated by Polytetrafluoroethylene (PTFE) 0.45?m
membrane. The working electrode was a TiO2/Ti plate and
the counter electrode was a platinum plate (1cm2). All the
potentials were specified to the Ag/AgCl reference elec-
trode, which was connected to the assembly via a salt bridge.
ing AMEL general-purpose potentiostat–galvanostat Model
2049. All the potentials were fixed at 1.0V during pho-
Princeton Applied Research (PAR) VersaStat driven by
model 270 Electrochemical Analysis System software with
PC control was used for linear sweep voltammetry (LSV)
measurement. The temperature of the reactor solutions was
maintained at 313K throughout the experiments by using a
the study on the effect of solution temperatures). The light
source was placed 8cm away from the sample. Photoreac-
tor cell (total volume of 140cm3) was filled with 120cm3
methyl orange solution containing 0.1M NaCl as a support-
ing electrolyte. The samples were withdrawn every 30min
thereafter for a period of 120min. The concentration of the
methyl orange in the solution was determined by measuring
the absorbance values using UV–vis Perkin-Elmer Lambda
20 Spectrophotometer at 464.5nm.
3. Results and discussion
3.1. Effect of different type of light source
In the studies reported elsewhere, various types of lamps
uate the operating cost and the efficiency of photoreactor.
Three types of lamps were chosen as light sources namely
halogen (Tungsten type, 300W) fluorescent (energy save
type, 15W equal to 75W) and near UV lamp (100W). Halo-
gen lamp was chosen because it closely resembles sunlight.
Meanwhile, fluorescent lamps are widely used for household
and industrial purpose provide a cheap and energy saving
Fig. 3 shows the effect of various types of light source
towards photoelectrochemical degradation of methyl orange
dye. The experimental results are reported as ratio of (C/C0)
versus illumination time (t), where C0is the initial concen-
tration of dye and C is the concentration at t minute. The dye
concentration was determined through standard calibration
curve of absorbance (recorded at λmax=464.5nm) versus
Fig. 2. The experimental set-up for the photoelectrochemical degradation process from cross-sectional view.
concentration. Control experiment was carried out by illu-
mination of methyl orange solution in the absence of TiO2,
which shows no changes of dye concentration for period of
120min irradiation times.
The photoelectrochemical degradation rate using UV
lamp with power of 100W is almost equivalent to a 300W
halogen lamp. This is due to UV radiation that possesses
higher intensity of photon with energy equal or higher than
TiO2bandgap (Eg). Thereby, more electrons were excited
from the valence band to the conduction band. In this study,
TiO2has bandgap energy about 3.2eV (∼380nm) as indi-
cated by UV–vis absorption spectrum in Fig. 4. In conjunc-
tion with this, only small amount of photon energy generated
Fig. 3. The methyl orange degradation dependence on different types of
light source. C0=10ppm, containing 0.1M NaCl; potential=1.0V.
by halogen lamp, which possesses longer wavelength, was
suitable for electron excitation. However, probably due to
high intensity of the halogen lamp its performance is compa-
rable to the near UV lamp.
Besides that, the mechanism of photodegradation using
light with energy less than Egwill be different from that tak-
ing place using light with energy more than Egof TiO2. The
former excite the dye compounds directly, while the latter
generate electron–hole pairs . In another study, the pho-
toassisted degradation of dyes preadsorbed on the surface of
TiO2particles with visible light was reported , where
the photoreaction system was almost water free and only the
molecules that were in direct contact with the TiO2surface
(the molecules at monolayer or submonolayer coverage) un-
component of the light plays an important role in generating
electron–hole pairs for the photodegradation reaction.
On the other hand, a 15W fluorescent lamp shows almost
equal photoelectrochemical degradation rate with a 50W
Fig. 4. UV spectra of TiO2thin film coated on ITO glass.
Apparent first-order kinetic rate constants, half times and correlation factors for the photoelectrochemical degradation of methyl orange illuminated with
different types of light with medium stirred speed
Apparent rate constant, kapp(×10−3min−1)
Tungsten halogen lamp 300W 21.9
Ultraviolet Lampl 100W19.9
Tungsten halogen lamp 50W0.6
Fluorescent lamp 15W (1 bulb)0.6
Fluorescent lamp 15W (2 bulbs)1.0
Type of lamp
Half time, t1/2(min)Correlation factor, R2
more photon with suitable energy for excitation. This result
may also be attributed to fluorescent lamp exhibits with en-
ergy saving feature. An increase in the photoelectrochemical
degradation rate was observed when two fluorescent bulbs
were used. The Langmuir–Hinshelwood first-order kinetic
rate constant as shown in Table 1 was increased by almost
two-fold when two fluorescent bulbs were used.
3.2. Effect of light intensity
The intensity of a 300W halogen lamp irradiation was
varied from without illumination to full light intensity using
a dimmer. A lux meter was used to determine the percentage
light intensity, the reaction pathway seems to follow strictly
the mechanism for visible light as discussed. Nevertheless,
at high intensity more photons with energy larger than Egof
TiO2are generated, which leads to more electron–hole pairs
on the TiO2surface. This produces more hydroxyl radical
Fig. 5. The methyl orange dye degradation dependence on percentage of
light intensity. C0=10ppm, containing 0.1M NaCl; potential=1.0V. Inset
on the surface or bulk solution to involve in the oxidation of
for dye molecules, which are in direct contact with the TiO2
As seen in inset in Fig. 5, the kinetic constant of photo-
electrochemical degradation shows exponential relationship
there is a possibility to drastically promote the degradation
rate by increasing the light intensity to a certain level where
significant concentrations of electron–hole pairs were gener-
3.3. Effect of agitation
The rate of photodegradation is affected not only by the
electrode reaction itself but also by the transport of species
to and from bulk solution. This transport can occur by dif-
fusion, convection or migration. Migration is an essential
electrostatic effect that arises due to the application of volt-
age which creates a charged interface on the electrodes. Any
charged species near the interface will either be attracted or
repelled from it by electrostatic forces.
In this study, the effect of migration can be neglected.
This is due to the presence of supporting electrolyte (0.1M
NaCl), which acts to transport almost all the current in the
cell and reduce the cell resistance. Consequently, the effect
of the electrical field is limited to a small distance from the
electrode. Therefore, mass transport due to the diffusion and
convection of the dye molecules to the interface may affect
the overall efficiency.
Diffusion is the natural movement of species in solution,
be charged or neutral. In normal circumstances, the rate of
diffusion depends on the concentration gradient as expresses
by Fick’s first law :
J = −D?c
in solution and D the proportionality constant known as the
There is no significant concentration gradient in the
bulk solution due to the homogenous distribution of dye
molecules in the solution. However, concentration gradient
exists at the interface between the electrode surface and
the solution during the photoelectrochemical degradation
experiment. Dye molecules near to the electrode surface
reacted with the reactive holes from the semiconductor
leading to decrease of dye concentration at that region.
On the other hand, hydroxyl radicals attacked the dye in
the bulk solution far from electrode surface also resulting
in the decrease of dye concentration . Therefore, the
variation of the concentration gradient is widely dependent
on whether the reaction is more dominant at the electrode
surface or to the bulk solution. If both the reaction occurred
at the same rate, no concentration gradient was established.
Another type of mass transport is convection, which re-
sults from the action of a force on the solution. This could be
due to a pump, flow of gas, stirrer and gravity. There are two
forms of convection; natural and force convection. The nat-
ural convection occurs in any solution is generated by small
thermal or density differences and acts to mix the solution in
the photo reactor cell was equipped with water jacket to keep
the temperature thermostated with the aim of avoiding big
Beside thermal effect, we try to draw out the natural con-
erately introducing convection into the cell. Magnetic stirrer
was placed in the bottom of the cell to agitate the reactants.
This form of convection is termed forced convection. Fig. 6
depicts a slight increase of photocurrent, when higher ag-
itation rates were applied. The sinusoidal curves could be
observed for stirred solution compared to a smooth line for
unagitated solution for both in dark and under illumination.
slightly lower C/C0value indicating better degradation when
agitated. By agitating the solution, more dye molecules will
On the other hand, hydroxyl radical produced at the surface
can react more randomly with the dye in bulk solution.
Fig. 6. The effect of agitation speed on photocurrent under illumination of
300W halogen lamp. C0=10ppm, containing 0.1M NaCl. ((—) Without
stirred, (···) slow, (–––) medium, (-·-·-·-) high).
3.4. Effect of electrode repeated use
The electrode performance upon repeated usage has great
significance in justifying the effectiveness of the electrode. If
the electrode shows good ability to be repeatedly used, there
is a good possibility for it to be applied in actual wastewater
In this study, the electrode was tested for total 20 runs for
treatment of methyl orange solution with an anodic bias po-
tential of 1.0V as shown in Fig. 8. The result indicates that
The decrease in the efficiency may be due to the factor such
as dissolution of the photocatalyst, development of insula-
tion layer and poisoning of the catalyst surface by impurities
Fig. 8. The electrode repeated usage performance for total 20 times.
Fig. 9. Graph kinetic constant vs. electrode repeated usage time.
. However, high degradation percentage for 4, 8, 10, 13,
15, and 17 repetitions may be attributed to the increase in
the surface area, chemical composition and the presence of
bound water in the reused catalyst.
better performance than the reused one and vice versa. Fig. 9
shows the graph for first-order kinetic rate constants versus
the electrode used times. The general trend is shown by the
straight line, which indicates that the electrode suffered only
3.5. Effect of photoreactor temperature
cess on the temperature of methyl orange solution ranges
from 298 to 318K was studied. An increase of photoelectro-
chemical degradation process with increasing temperature
was observed. Table 2 shows increasing of pseudo first-order
kinetic rate constants with temperature.
Arrhenius suggested that the rate of most reactions varies
with temperature in such a way that :
k = Ae−Ea/RT
where k is the rate constant, A is the frequency factor or pre-
exponential factor, and T is the temperature (K).
Taking natural logarithm of both sides of Eq. (3) resulted
in the following equation.
lnk = −Ea
In this experiment, the rate constants were determined
from the first-order kinetic graphs. The values of the rate
constants are listed in Table 2. A linear plot of lnk versus
1/T was obtained as shown in Fig. 10 with the slope value
of 2240.5K−1. The overall activation energy obtained was
18.63kJmol−1. This value is quite similar to the activation
tion energy on immobilised TiO2compared to suspension is
However, only a slight difference in the activation energy for
methyl orange is observed compared to the one in suspen-
photodegradation system applied in this study successfully
reduce the activation energy for photodegradation reaction.
Apparent first-order kinetic rate constants, half times and correlation factors for the photoelectrochemical degradation of methyl orange at different solution
temperatures using 300W halogen lamp with medium stirred speed
Apparent rate constant, kapp(×10−3min−1)
Temperature T (K)
Half time, t1/2(min) Correlation factor, R2
Comparison of activation energy obtained in this study with the findings in literatures
Chen and Chou 
Mills and Morris 
Hofstadler et al. 
The activation energy obtained is small, which indicates
that the photocatalytic reaction is less temperature depen-
dent. This is due to irradiation is the primary source of
electron–hole pair generation as the band gap energy is too
high to be overcome by thermal activation. Therefore, the ef-
fect of temperature is most likely due to the increase in colli-
tion to compete efficiently with electron–hole recombination
The results indicate the importance of the operational
parameters towards obtaining high photoelectrochemical
degradation rate. Types of light sources as well the intensity
significantly influenced the photoelectrochemical degra-
dation. Agitation acts to promote the degradation rate by
bringing more dye molecules closer to the electrode surface
and hydroxyl radical produced at the interface can also react
more randomly with the dye in the bulk solution. The film
showed only slight decrease in the performance after re-
peated usage. The activation energy of photoelectrochemical
degradation of methyl orange in aqueous solution was found
to be 18.63kJmol−1, which is comparable to the value
obtained in suspension system.
We gratefully acknowledge the financial support from the
04-0255-EA001 and 09-02-04-0369-EA001.
 M.A. Fox, M.T. Dulay, Chem. Rev. 93 (1993) 341.
 M.R. Hoffman, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem.
Rev. 95 (1995) 69.
 R.L. Pozzo, M.A. Baltanas, A.E. Cassano, Catal. Today 39 (1997)
 J.A. Bryne, A. Davidson, P.S.M. Dunlop, B.R. Eggins, J. Photochem.
Photobiol. A 148 (2002) 365.
 A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol. A 1
 K. Pirkanniemi, M. Sillanpaa, Chemosphere 48 (2002) 1047.
 K. Shaw, P. Christensen, A. Hamnett, Electrochem. Acta 41 (1996)
 N.N. Rao, A.K. Dubey, S. Mohanty, P. Khare, R. Jain, S.N. Kaul,
J. Hazard. Mater. B 101 (2003) 301.
 B. Neppolian, H.C. Choi, S. Sakthivel, B. Arabindoo, V. Murugesan,
J. Hazard. Mater. B 89 (2002) 303.
 N.S. Lewis, M.L. Rosenbluth, in: N. Serpone, E. Pelizzetti (Eds.),
Photocatalysis: Fundamentals and Applications, Wiley Interscience,
New York, 1989, pp. 79–90.
 A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem Rev. 95 (1995) 735.
 K. Vinodgopal, S. Hotchandani, P.V. Kamat, J. Phys. Chem. 97
 K. Vinodgopal, P.V. Kamat, Sol. Energy Mater. Sol. Cells 38 (1995)
 F. Zhang, J. Zhao, L. Zang, T. Shen, H. Hidaka, E. Pelizzetti, N.
Serpone, J. Mol. Catal. A 120 (1997) 173.
 K. Vidnogopal, J. Peller, O. Makogon, P.V. Kamat, Water Res. 32
(12) (1998) 3646.
 K. Kato, A. Tsuzuki, Y. Torii, H. Taoda, J. Mater. Sci. 30 (1995)
 K. Kato, A. Tsuge, K. Niihara, J. Am. Ceram. Soc. 79 (6) (1996)
 K. Kato, K. Niihara, Thin Solid Films 298 (1997) 76.
 A. Blazkova, I. Csolleova, V. Brezova, J. Photochem. Photobiol. A
113 (1998) 251.
 A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49.
 P.V. Kamat, K. Vinodgopal, in: D.F. Ollis, H. Al-Ekabi (Eds.), Pho-
tocatalytic Purification and Treatment of Water and Air, Elsevier
Science Publishers, Amsterdam, 1993, p. 83.
 C.M.A. Brett, A.M.O. Brett, Electrochemistry Principles Methods
Applications, Oxford University Press Inc., New York, 1993, p. 25.
 C.S. Turchi, D.F. Ollis, J. Catal. 122 (1990) 178.
 M.A. Fox, in: N. Serphone, E. Pelizzetti (Eds.), Photocatalysis Fun-
damentals and Applications, Wiley Interscience, New York, 1989, p.
 J.F. Bunnett, in: C.F. Bernasconi (Ed.), Investigation of Rates and
Mechanisms of Reactions Part 1, fourth ed., Wiley Interscience, New
York, 1986, p. 285.
 L.C. Chen, T.C. Chou, Ind. Eng. Chem. Res. 32 (1993) 1520.
 A. Mills, S. Morris, J. Photochem. Photobiol. A Chem. 71 (1993)
 K. Hofstadler, R. Bauer, S. Novalic, G. Heisler, Environ. Sci. Tech-
nol. 28 (1994) 670.
 D. Chen, A.K. Ray, Water Res. 32 (11) (1998) 3223.
 H. Al-Ekabi, P. Mayo, J. Phys. Chem. 89 (1985) 5815.