Semiconductor-catalyzed solar photooxidation of iodide ion

Article abstract of DOI:10.1016/j.molcata.2006.10.016

With natural sunlight TiO₂ (anatase), ZrO₂, MoO₃, Fe₂O₃, ZnO, CeO₂ and Al₂O₃ microparticles photocatalyze the oxidation of iodide ion but CuO, ZnS, CdO, CdS, HgO, PbO,

Full text of DOI:10.1016/j.molcata.2006.10.016

Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
Semiconductor-catalyzed solar photooxidation
of iodide ion
C. Karunakaran∗, P. Anilkumar
Department of Chemistry, Annamalai University, Annamalainagar 608002, Tamilnadu, India
Received 19 May 2006; received in revised form 4 October 2006; accepted 11 October 2006
Available online 17 October 2006
Abstract
With natural sunlight TiO2 (anatase), ZrO2, MoO3, Fe2O3, ZnO, CeO2 and Al2O3 microparticles photocatalyze the oxidation of iodide ion but
CuO, ZnS, CdO, CdS, HgO, PbO, Sb2O3 and Bi2O3 microparticles do not. The photocatalysis is not slowed down at least up to 120 min with ZrO2,
MoO3, Fe2O3, CeO2 and Al2O3 whereas it is 45 min with TiO2 and 30 min with ZnO. The iodine generation depends on [I−], surface area and pH
and is enhanced by the addition of ethanol. The catalysts show sustainable photocatalysis. Nanoparticles exhibit higher photocatalytic activities
than microparticles. The catalytic efficiencies are of the order: Fe2O3 > MoO3 > TiO2 > CeO2 > ZnO > ZrO2 > Al2O3.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Photocatalysis; Semiconductor; Sunlight; Iodide ion; TiO2; ZrO2; MoO3; Fe2O3; ZnO; CeO2; Al2O3
1. Introduction
2. Experimental
Harnessing solar radiation is a means of sustainable, eco-
friendly energy generation. Production of energy bearing chem-
icalsthroughthermodynamicallyuphillreactionsistheobjective
of solar energy conversion and storage and one such reaction is
iodide ion oxidation (ꢀG◦ = +51.6 kJ mol−1). Further, light-to-
electrical energy conversion is possible through the oxidation
of iodide ion at TiO2 electrodes in photodriven electrochemical
cells [1]. Although photocatalyzed oxidation of iodide ion on
TiO2 [2–6], colloidal TiO2 [7], Pt-loaded TiO2 [8], TiO2 sen-
sitized with phthalocyanines [9] and flower pigment cyanidin
[10], ZrO2, V2O5, Fe2O3, ZnO and Al2O3 [6] using artifi-
cial UV light have been made there is no report with natural
sunlight and hence this work. The problem of fluctuation of
sunlight intensity even under clear sky is overcome by carrying
out set of experiments simultaneously, side-by-side, thus mak-
ing the quantum of sunlight absorbed in the set of experiments
identical.
2.1. Materials
2.1.1. Microparticles
TiO2 (Merck), ZrO2 (Chemco, India), V2O5 (Johnson
Matthey), MoO3 (SD Fine, India), Fe2O3 (Fischer, India), CuO
(SD Fine, India), ZnO (Merck), ZnS (SD Fine, India), CdO
(Chemco, India), CdS (Chemco, India), HgO yellow (Chemco,
India), HgO red (SD Fine, India), CeO2 (SD Fine, India),
Al2O3 (Merck), SnO2 (BDH), PbO (Fischer, India), Pb2O3,
PbO2 (BDH), Sb2O3 (Qualigens, India), Bi2O3 (SD Fine, India),
KI (Qualigens, India), KBr (SD Fine, India) and KCl (Merck)
were used as received. Deionized distilled water was employed
throughout the study. The TiO2 used is of anatase form (99%+);
the XRD pattern of the sample totally matches with the standard
pattern of anatase (JCPDS) and the rutile lines are insignifi-
˚
cant (Siemens D-5000 XRD, Cu K␣ X-ray, λ = 1.54 A, scan:
5–60◦, scan speed: 0.2◦ s−1). The BET surface areas were deter-
mined as: TiO2 14.68, ZrO2 15.12, V2O5 16.14, Fe2O3 17.84,
CuO 1.51, ZnO 12.16, ZnS 7.67, CdO 14.45, CdS 15.47, HgO
6.39, CeO2 11.0, Al2O3 10.63, SnO2 114.7, PbO 0.28, PbO2
1.95, Bi2O3 2.75 m2 g−1; the corresponding values for MoO3
and Pb2O3 are too small to be determined by BET method.
The particle sizes, measured using particle sizer Horiba LA-910
∗
Corresponding author. Tel.: +91 4144221820.
E-mail address: karunakaranc@rediffmail.com (C. Karunakaran).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.10.016

154
C. Karunakaran, P. Anilkumar / Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
or Malvern 3600E (focal length 100 mm, beam length 2.0 mm,
wet (methanol) presentation), are as below: TiO2: 2.6–27.6,
ZrO2: 3.5–27.6, V2O5: 8.5–57.7, MoO3: 4.19–19.3, Fe2O3:
2.6–27.6, CuO: 5.69–30.5, ZnO: 3.5–27.6, ZnS: 0.115–2.60,
CdO: 2.6–11.4, CdS: 3.0–9.8, HgO (yellow): 0.17–5.69, CeO2:
0.23–10.5, Al2O3: 2.6–57.7, SnO2: 1.15–17.4, PbO: 4.47–13.3,
Pb2O3: 0.20–7.72, PbO2: 1.95–10.5, Bi2O3: 0.17–0.49 ␮m; the
detailed distributions are given as supplementary materials.
2.1.2. Nanoparticles
TiO2 P25 Degussa (ca. 80% anatase, 20% rutile) of mean par-
ticle size 30 nm and BET surface area ∼50 m2 g−1, TiO2 anatase
Hombikat (Fluka) of surface area ≥300 m2 g−1, TiO2 anatase
nanopowder (99.7%, Sigma–Aldrich) of average particle size
15 nm and BET surface area 190–290 m2 g−1, WO3 nanoparti-
cles (Sigma–Aldrich) of average diameter 49.6 nm and of sur-
face area 16.9 m2 g−1, ZnO nanoparticles (Sigma–Aldrich) of
mean particle size 50–70 nm and of surface area 15–25 m2 g−1
were used as supplied.
Fig. 1. Fe2O3-catalyzed solar photooxidation of I−. The UV–vis spectra of AM
1 sunlight illuminated KI solution diluted five-times and recorded at 10 min
intervals (↑); [I−] = 0.05 M, Fe2O3 bed: 15.1 cm2, Fe2O3 loading = 2 g, volume
of air saturated KI solution = 50 mL.
dation (85 ␮M iodine liberated in 30 min from 50 mL of 0.05 M
KI solution using AM 1 sunlight with a catalyst bed of 15.1 cm2
and a catalyst loading of 2.0 g) it liberates iodine in dark as well
(8 ␮M under identical conditions but in dark). The results with
insulator Al2O3 (Eg 9.0 eV [13]) under identical conditions pro-
vide a comparison. Fig. 1, the UV–vis spectra of the air saturated
KI solution illuminated with AM 1 sunlight and recorded at 10-
min intervals, shows formation of iodine; the spectrum is similar
to that of the authentic iodine–iodide solution. Chemical tests
also confirm the formation of iodine; the solution turns purple
with starch and discharged by thiosulfate.
2.2. Method
The semiconductor-catalyzed solar photooxidations were
carried out with AM 1 sunlight under clear sky in summer
(March–July). The intensity of sunlight (W m−2) was mea-
sured using Global pyranometer, MCPT, supplied by Industrial
Meters, Bombay; the solar radiation was beyond the range of
measurement of lux meter Lx-101A, Lutron, Taiwan, far larger
than the upper limit of 50,000 lx. As near UV and visible light
up to 560 nm is effective in photoexcitation of the active semi-
conductors employed in this study, the intensity of solar radi-
ation (einstein L−1 s−1) was also determined using ferrioxalate
actinometer, which covers a wavelength range of 250–577 nm;
under the experimental conditions, 440 W m−2 corresponds to
22 ␮einstein L−1 s−1. Fresh solutions of KI of required concen-
trations were taken in wide cylindrical glass vessels of uniform
diameter and appropriate height; the catalyst powder covered the
entire bottom of the vessel. 50 mL of the solution was air sat-
urated and used for each experiment. The iodine formed was
determined from the absorbance of the illuminated solution,
measured at 350 nm using Hitachi U-2001 spectrometer; cali-
bration curve was constructed for each KI concentration.
3.2. Obtaining solar results
The measurement of solar radiation (W m−2) shows fluctu-
ation of sunlight intensity during the course of the experiment
even under clear sky. Also, the sunlight intensity differs from day
to day. Now, the sunlight intensity for a set of solar experiments
of required reaction conditions was kept identical by carrying
out the experiments simultaneously, side-by-side, thus making
possible the comparison of the solar results.
The solar photogeneration of iodine from air saturated KI
solution is not slowed down at least up to 120 min with Fe2O3,
MoO3, ZrO2, CeO2 and Al2O3 as catalysts while it is 45 minutes
with TiO2 and 30 min with ZnO as catalysts. Fig. 2 presents the
formation of iodine from air saturated KI solution with TiO2,
ZrO2, MoO3, Fe2O3, ZnO, CeO2 and Al2O3 as catalysts using
AM 1 sunlight under identical sunlight intensities; the exper-
iments were carried out simultaneously, side-by-side, under
clear sky. The fluctuation of sunlight intensity during this study
was insignificant. As the generation of iodine is uniform at least
up to 30 min of illumination the iodine-formation rates were
obtained by estimating iodine after 30 min irradiation in all the
cases. The slowdown of the formation of iodine on prolonged
illumination is not unknown; the UV photooxidation of iodide
ion with platinum-loaded TiO2 [8] and with TiO2 but sensitized
with phthalocyanines [9] show similar results. The results of
TiO2, ZrO2, MoO3, Fe2O3, ZnO, CeO2 and Al2O3-catalyzed
solar photooxidations are consistent. For each catalyst, a couple
3. Results and discussion
3.1. Effective semiconductors
TiO2 anatase (bandgap energy, Eg 3.2 eV [11]), ZrO2 (Eg
5.0 eV [11]), MoO3 (Eg 2.9 eV [12]), Fe2O3 (Eg 2.2 eV [11]),
ZnO (Eg 3.2 eV [11]) and CeO2 (Eg 3.4 eV [13]) catalyze the
solar photooxidation of iodide ion whereas CuO (Eg 1.7 eV
[11]), ZnS (Eg 3.6 eV [11]), CdO (Eg 2.2 eV [11]), CdS (Eg
2.4 eV [11]), HgO yellow (Eg 1.9 eV [11]), HgO red (Eg 1.9 eV
[11]), PbO (Eg 2.8 eV [11]), Sb2O3 (Eg 3.0 eV [11]), and Bi2O3
(Eg 2.8 eV [11]) fail; V2O5 (Eg 2.8 eV [11]) dissolves, Pb2O3
oxidizes iodide ion in dark and PbO2 (Eg 1.4 eV [14]) reacts with
iodide. Although SnO2 (Eg 3.5 eV [11]) photocatalyzes the oxi-

C. Karunakaran, P. Anilkumar / Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
155
Fig. 2. Photooxidation of I− on TiO2, Fe2O3, MoO3, ZrO2, ZnO, CeO2 and
Al2O3 with AM 1 sunlight under identical intensities (Inset shows linearity
up to 45 min). [I−] = 0.05 M, catalyst bed = 15.1 cm2, catalyst loading = 2.0 g,
volume of air saturated KI solution = 50 mL.
Fig. 3. Iodine formation as a function of [I−]. Catalyst bed: 15.1 cm2, catalyst
loading = 2.0 g, volume of air saturated KI solution = 50 mL.
0.12, 0.14, 0.84, 0.85, 0.09, 0.22, 0.01 nM cm−2 s−1 for TiO2,
ZrO2, MoO3, Fe2O3, ZnO, CeO2, Al2O3, respectively; figures
not given) and Fig. 4 is the corresponding plot of relative rates.
Fig. 5 presents the variation of iodine generation with pH; the
pH was adjusted by the addition of HCl or NaOH and measured
using a digital pH meter, after allowing the system to attain
equilibrium. Except TiO2 all the photocatalysts studied slow
down the iodine-generation with increase of pH. TiO2 enhances
photoformation of iodine with increase of pH. Plots of actual
rates provide similar profiles (slope: +1.4, −0.84, −0.86, −5.8,
−0.77, −1.1, −0.16 nM s-1 for TiO2, ZrO2, MoO3, Fe2O3, ZnO,
CeO2, Al2O3, respectively; figures not given). Measurement of
pH before and after solar irradiation reveals a small drop in pH
for all the photocatalysis examined. The metal oxides studied
show sustainable photocatalytic activity; reuse of the photocat-
alysts yield identical results. Addition of ethanol enhances the
photogeneration of iodine significantly in ZnO and ZrO2 pho-
of solar experiments of identical reaction conditions carried
out simultaneously, side-by-side, yield results with in 6% and
this is so on different days. This consistency is not surprising
as the quantum of light absorbed per unit area is the same in
control and text experiments. The photoactive metal oxides do
not oxidize iodide ion in dark, confirmed by experiments under
identical conditions but in dark.
3.3. Factors influencing photocatalysis
The influence of operational parameters such as [I−], area
of the catalyst and pH on the photoformation of iodine in each
catalysis was studied separately; the sunlight intensity in each
study was kept identical by performing a set of experiments
simultaneously, side-by-side. For example, the photogeneration
of iodine at different iodide ion-concentrations for a particular
catalysis was determined simultaneously, side-by-side. Under
identical sunlight intensity, the iodine-formation increases lin-
early with [I−]; plot of iodine-formation rate versus [I−] is linear
and this is so in all the stated seven active catalysts (slope:
1.2 × 10−7, 5.4 × 10−8, 3.1 × 10−7, 5.4 × 10−7, 9.5 × 10−8
,
5.5 × 10−8, 2.9 × 10−9 s−1 for TiO2, ZrO2, MoO3, Fe2O3, ZnO,
CeO2, Al2O3, respectively; figures not given). As the photogen-
eration depends on sunlight intensity and the sunlight intensity
is not identical in all the photocatalysis, as they were carried out
on different days, comparison of the rates is incorrect. But, rela-
tive rates, rates relative to that at a specified condition, deduced
by dividing the actual rates by that at a specified condition, all
determined simultaneously, side-by-side, yield rates indepen-
dent of sunlight intensity. These rates also vary linearly with
[I−] (Fig. 3). Also, the photoformation of iodine is a linear
function of the apparent surface area of the catalyst. And, this
is in accordance with the Langmuir-Hinshelwood model [15].
Plot of iodine-formation rate versus surface area of the cat-
alyst bed is linear for all the active catalysts studied (slope:
Fig. 4. Iodine formation as a function of area of catalyst bed. [I−] = 0.05 M,
catalyst loading = 2.0 g, volume of air saturated KI solution = 50 mL.

156
C. Karunakaran, P. Anilkumar / Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
recombination of photogenerated electron–hole pairs in semi-
conductors are so rapid, occurring in a picosecond time scale,
for an effective photocatalysis the reactants are to be adsorbed on
the photocatalysts [3]. The hole reacts with the adsorbed iodide
ion to form an iodine atom that further reacts with iodide ion to
−
produce I2−. The I2 disproportionates to form tri-iodide and
iodide ions [3,7].
SC + hν → h(vb)+ + e(cb)
−
I(ads)− + h(vb)+ → I
−
I + I− → I2
2I2− → I3− + I
In the presence of oxygen, transfer to the adsorbed oxygen
molecule resulting in highly active superoxide radical-anion,
O2•−, effectively removes the electron [17].
Fig. 5. Iodine formation as
a
function of pH. [I−] = 0.05 M, catalyst
bed = 15.1 cm2, catalyst loading = 2.0 g, volume of air saturated KI solu-
tion = 50 mL.
−
O2(ads) + e(cb)− → O2
•
O2 − + e− + 2H+ → H2O2
•
tocatalysis, moderately in MoO3 and TiO2 catalysis and not so
with CeO2 and Al2O3 (Fig. 6).
The H2O2 produced may oxidize iodide ion to iodine.
H2O2 + 2I− → I2 + 2HO−
3.4. Mechanism
Generally, adsorption of water on catalytic surfaces results
in surface hydroxyl groups. Hole trapping by either the surface
hydroxyl groups or adsorbed water molecules generates OH
Of the seven photocatalysts studied Al2O3 is an insulator
providing a non-reactive surface while others are semiconduc-
tors with finite bandgap energies. Although ZrO2 presents an
absorption maximum around 250 nm, some samples show a non-
negligible absorption in the near UV range (290–390 nm) and
marginal photocatalysis occurs under irradiation in this range
[16]. Illumination of the semiconductors with light of energy
greater than the bandgap leads to bandgap excitation of semicon-
ductors resulting in creation of electron–hole pairs; holes in the
valence band and electrons in the conductance band. Since the
•
radicals, which are the primary oxidizing agents [17]. However,
the clean first order dependence of iodine formation on [I−] dis-
counts the possibility of •OH radicals as the primary oxidizing
species of I−. If the photooxidation of I− were primarily due to
the •OH radicals generated, the iodine formation should not lin-
early depend on [I−]; the •OH radicals are short lived and react
almost instantaneously demanding non-dependence of the pho-
tooxidation rate on [I−]. Also, this is in agreement with the study
using terephthalic acid as a fluorescence probe for the quanti-
tative measurement of •OH production in TiO2 photocatalysis;
the oxidative reactions on TiO2 occur mainly via photogenerated
holes, not via •OH [18].
−OH(ads) + h(vb)
→
→
•OH
+
•OH + H+
+
H2O(ads) + h(vb)
Semiconductor-catalyzed photoreactions are generally gov-
erned by Langmuir–Hinshelwood kinetics [6,15] and the
observed first order dependence on [I−] points to insignificant
adsorption of iodide ion on the catalysts. The adsorption of
anionic species on the metal oxides depends also on the sur-
face excess charge on the semiconductor particles. At pH higher
than the point of zero charge (PZC), the catalyst surface is nega-
tively charged leading to electrostatic repulsion between iodide
ion and the semiconductor particles. Hence, the concentration
of iodide ion at the surface, in the double layer, is likely to
be smaller than that in the bulk of the solution; the adsorp-
tion isotherm becomes linear leading to a first order kinetics
of the photocatalysis. However, examination of Fig. 5 reveals,
Fig. 6. Iodine formation as a function of ethanol content. [I−] = 0.05 M, cat-
alyst bed = 15.1 cm2, catalyst loading = 2.0 g, volume of air saturated KI solu-
tion = 50 mL.

C. Karunakaran, P. Anilkumar / Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
157
for some catalysts at least (TiO2 and ZrO2), uniform trend in the
photocatalysis at pH higher as well as lower than PZC; the PZC
for TiO2, ZrO2, Fe2O3 and ZnO are 5.80, 6.70, 8.60 and 8.80,
respectively [11]. A possible explanation is the modification of
PZC values in presence of ions in solution [19,20]; the PZC for
TiO2 lowers from 6.4 to 4.5 [19]. Hence it is possible that the
PZC values for the photocatalysts in presence of ions are lower
than the experimental pH. Significant enhancement of iodine-
formation on the addition of ethanol, in some semiconductors
at least, indicates hole trapping by alcohol and the generated
radicals oxidize iodide ions.
Non-reactive surfaces such as Al2O3 provide an ordered
two-dimensional environment for effective electron transfer
from the donor to the acceptor. Either the donor iodide ion
or the acceptor oxygen molecule, both adsorbed on the pho-
tocatalyst, may undergo photoexcitation followed by electron
transfer. The donor excitation results in the transfer of excited
electron whereas the acceptor excitation leads to an electron
jump from the donor level to the vacant acceptor level [3].
The lack of enhancement of iodine-generation by ethanol in
Al2O3-photocatalysis is on expected lines; due to the absence
of generation of holes on Al2O3 there could be no hole trapping
by adsorbed alcohol.
4.2) > TiO2 anatase microparticles, Merck (1.0, 1.1); the relative
efficiencies with air saturated and continuous air purging are
given in parentheses. Unlike TiO2, ZnO and WO3 nanopowders
do not form suspension but settle at the bottom even on con-
tinuous air purging. As the catalyst loading is very small, they
do not cover the entire bottom of the reaction vessel and hence
there is no effective catalyst bed.
3.7. Lack of photooxidation of chloride and bromide ions
Experiments under identical conditions with all the micropar-
ticles and the nanoparticles stated show absence of photooxida-
tion of chloride and bromide ions. Irradiation of air saturated
0.2 M KCl or KBr solutions (25 mL) over the microparticles
(catalyst bed = 15.1 cm2, catalyst loading = 1.0 g) with natural
sunlight for 5 h fails to show positive results; only ZnO feebly
photooxidizes chloride ion in traces. Similar experiments with
TiO2 nano- and micro-particles (0.02 g) under suspension in air
saturated 0.2 M KCl or KBr solutions with an illumination area
of 15.1 cm2 do not show oxidation of chloride or bromide ions
even on irradiation with sunlight for 5 h.
4. Conclusions
3.5. Comparison of photocatalytic efficiencies
TiO2 (anatase), ZrO2, MoO3, Fe2O3, ZnO, CeO2 and Al2O3
microparticles photocatalyze the oxidation of iodide ion with
natural sunlight and the influence of [I−] and also surface area
on the generation of iodine is similar with all the photocatalysts
reported. While the increase of pH favors TiO2-photocatalysis
it is the other way with other catalysts. The catalytic efficiencies
are in line with the bandgap energies. Nanoparticles show better
photocatalytic efficiencies than microparticles.
Comparison of the photocatalytic efficiencies reveal
the order: Fe2O3 (4.5) > MoO3 (1.5) > TiO2 (1.0) > CeO2
(0.90) > ZnO (0.82) > ZrO2 (0.38) > Al2O3 (0.08); the solar
photooxidations were carried out under identical AM 1 sun-
light intensity and reaction conditions (0.05 M KI, catalyst
bed: 15.1 cm2, catalyst loading = 2.0 g, volume of KI solu-
tion = 50 mL, illumination = 30 min); the relative efficiencies are
given in parentheses. The observed photocatalytic efficiencies
are in line with the bandgap energies. The highly active Fe2O3
(optical absorption edge 620 nm [21]) can be photoexcited
with visible light whereas the least active semiconductor ZrO2
requires UV light for photoactivation. The bandgap energies of
TiO2, ZnO and CeO2 are very close (optical absorption edges
380, 396, 440 nm, respectively [21]) and hence the catalytic effi-
ciencies. MoO3 (optical absorption edge 443 nm [21]) shows a
better catalytic activity as it is susceptible to photoexcitation
with visible light.
Acknowledgements
The authors thank the University Grants Commission, New
Delhi, for the financial support through major research grant no
F.12-64/2003 (SR) and Degussa for gifting TiO2 P25 sample.
PA is grateful to UGC for PF.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.molcata.2006.10.016.
3.6. Photocatalysis by nanoparticles
References
Semiconductor nanoparticles are effective photocatalysts.
Experiments with nanoparticles and microparticles, under AM
1 sunlight of identical intensity and under identical conditions
(0.05 M KI, catalyst bed: 11.1 cm2, catalyst loading = 0.02 g,
volume of KI solution = 25 mL, illumination = 30 min), reveal
the higher photocatalytic efficiencies of nanoparticles com-
[1] B. O’Regan, N. Gratzel, Nature 353 (1991) 737.
[2] K. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, J. Photochem.
Photobiol. A 134 (2000) 139.
[3] A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem. Rev. 95 (1995) 735.
[4] P.V. Kamat, Chem. Rev. 93 (1993) 267.
[5] G.P. Lepore, C.H. Langford, J. Vichova, A. Vicek Jr., J. Photochem. Pho-
tobiol. A 75 (1993) 67.
[6] C. Karunakaran, S. Senthilvelan, S. Karuthapandian, K. Balaraman, Catal.
Commum. 5 (2004) 283.
pared to the microparticles. Also, purging of air (7.8 mL s−1
)
marginally improves the efficiency of photocatalysis by TiO2
Hombikat and TiO2 P25 Degussa. The photocatalytic efficien-
cies are of the order: TiO2 nanopowder, Sigma–Aldrich (6.1,
6.6) > TiO2 Hombikat, Fluka (3.6, 5.5) > TiO2 P25 Degussa (3.0,
[7] D.J. Fitzmaurice, M. Eschle, H. Frei, J. Phys. Chem. 97 (1993) 3806.
[8] T. Ohno, K. Fujihara, S. Saito, M. Matsumura, Sol. Energy Mater. Sol.
Cells 45 (1997) 169.

158
C. Karunakaran, P. Anilkumar / Journal of Molecular Catalysis A: Chemical 265 (2007) 153–158
[9] J. Hodak, C. Quinteros, M.I. Litter, E.S. Roman, J. Chem. Soc., Faraday
[16] S.G. Botta, J.A. Navio, M.C. Hidalgo, G.M. Restrepo, M.I. Litter, J. Pho-
tochem. Photobiol. A 129 (1999) 89, and the references cited therein.
[17] D. Chatterjee, S. Dasgupta, J. Photochem. Photobiol. C 6 (2005) 186.
[18] K-i. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, J. Photochem.
Photobiol. A 134 (2000) 139.
Trans. 92 (1996) 5081.
[10] K. Tennakone, A.R. Kumarasinghe, G.R.R.A. Kumara, K.G.U. Wijayan-
tha, P.M. Sirimanne, J. Photochem. Photobiol. A 108 (1997) 193.
[11] Y. Xu, M.A.A. Schoonen, Am. Mineral. 85 (2000) 543.
[12] Y. Zhao, J. Liu, Y. Zhou, Z. Zhang, Y. Xu, H. Naramoto, S. Yamamoto, J.
Phys.: Condens. Matter 15 (2003) L547.
[19] J. Gimenez, M.A. Aguado, S. Cervera-March, J. Mol. Catal. A 105 (1996)
67.
[13] C. Sol, R.J.D. Tilley, J. Mater. Chem. 11 (2001) 815.
[14] A. Kumar, A. Henglein, H. Weller, J. Phys. Chem. 93 (1989) 2262.
[15] C. Karunakaran, S. Senthilvelan, J. Mol. Catal. A 233 (2005) 1.
[20] J. Domenech, J. Munoz, Electrochim. Acta 32 (1987) 1383.
[21] M. Miyauchi, A. Nakajima, T. Watanabe, K. Nashimoto, Chem. Mater. 14
(2002) 2812.