Appl. Phys.

A (2015) 120:199–206
DOI 10.1007/s00339-015-9150-z

Efficient dye-sensitised solar cell based on uniform In-doped TiO2

spherical particles
A. M. Bakhshayesh1 • N. Farajisafiloo1

Received: 4 December 2014 / Accepted: 30 March 2015 / Published online: 7 April 2015
Ó Springer-Verlag Berlin Heidelberg 2015

Abstract A facile deposition of uniform photoanode generation of positive and negative charge carriers in the
electrodes by a novel anatase-stabilised gel for dye-sensi- electrolyte and TiO2 band gap, respectively. The electrons
tised solar cells (DSCs) applications is reported. Highly diffuse through TiO2 particles and eject from the inner
crystalline anatase–TiO2 phase is stabilised by indium ni- circuit. High electron injection, in conjunction with rapid
trate at 500 °C. The electrodes are composed of uniform electron transportation, contributes to a high-performance
spherical particles with diameter around 3 lm, containing DSC. In order to deposit a TiO2 film, conventional pastes
small nanoparticles with the average grain size of 40 nm, are usually prepared by mixing TiO2 powder, different
deposited by dip-coating method. X-ray photoelectron solvents, dispersing agents, and binders [2–5]. However,
spectroscopy reveals that 6 at.% In3? was incorporated into the quality of the resulting pastes is both difficult to predict
titania crystal lattice and stabilised anatase phase by lim- and difficult to control, which gives rise to non-uniform
iting the transformation from anatase to rutile phase. UV– films containing aggregates with irregular shapes [6, 7].
Visible spectra show that the stabilised film has lower band Various attempts were made to rectify this by recreating
gap energy than that of undoped TiO2, extending the ab- the processing history of the films using sol–gel techniques,
sorption of TiO2 into visible region. Electrochemical which enjoy excellent compositional control and high ho-
impedance spectroscopy demonstrates that the anatase- mogeneity. In our previous work [8], a novel simple syn-
stabilised DSC enjoys less recombination and internal re- thetic procedure for the fabrication of photoanode electrode
sistances, improving the photovoltaic performance of the with uniform particles for DSCs application was reported,
cell. The anatase-stabilised DSC has higher power con- so-called polymeric gel process. Photovoltaic measure-
version efficiency of 7.48 % than that of unstabilised cell ments showed that the TiO2 solar cells made of the poly-
(6.37 %). meric gel process had higher photovoltaic performance,
light scattering, and dye loading than those made of con-
ventional pastes as a consequence of the uniform distri-
1 Introduction bution of spherical particles with diameter around 2.5 lm,
containing small nanoparticles with average grain size of
During the last two decades, dye-sensitised solar cells 60 nm. The main drawback of the polymeric gel process is
(DSCs) have been regarded as new resources of supplying that the deposited films are composed of a mixture of
renewable energy efficiently and cost-effectively [1]. In a anatase and rutile phases, where rutile content is dominant.
DSC, similar to the role of chlorophyll in plants, dye To the best of our knowledge, anatase phase is more
molecules absorb the incident light, resulting in the favourable compared with rutile phases for DSC applica-
tions due to lower internal resistances and better dye
loading [1]. In order to improve both electron injection and
& A. M. Bakhshayesh electron transportation rate, a wide spectrum of research
[email protected]; [email protected]
focused on preparing photoelectrodes with pure anatase
1
Department of Research and Development, Royin Kar Shams crystal structure. Anatase-based photoanode has a larger
Company, 45371-59885 Zanjan, Iran dye-loading capability and delivered a higher electron

123
200 A. M. Bakhshayesh, N. Farajisafiloo

diffusion coefficient than rutile-based electrodes [9]. (Aldrich, UK) at 70 °C for 15 min [15]. Two different
Therefore, the preparation of a polymeric gel containing films were deposited on FTO substrate by dip-coating
anatase-TiO2 nanoparticles may improve the photovoltaic method using the TiO2 gel (unstabilised gel) and the gel
performance of DSCs. Some studies have focused on the modified with indium nitrate (stabilised gel). In order to
stabilisation of anatase crystal structure by the introduction control the thickness of the photoelectrodes, coating cycle
of foreign ions into TiO2 lattice [10–12] in general and was repeated four times, followed by drying at 50 °C for
indium in particular. Tahir et al. [13] synthesised In-doped 5 min after each coating cycle. The films were annealed at
TiO2 particles using a controlled sol–gel method and re- 500 °C for 2 h in air atmosphere. The films were again
ported that In ions inhibited crystal growth and produced TiCl4-treated and sintered at 450 °C for 30 min to improve
anatase phase. Sun et al. [14] prepared In-doped TiO2 thin interfacial contact with the nanocrystalline TiO2. Dye
film at the interface of fluorine-doped tin oxide substrate sensitisation was performed by immersing the films in
and mesoporous TiO2 film by spin-coating method. They 0.5 mM N-719 dye (Dyesol, Australia) solution (mixture
showed that the overall energy conversion efficiency of the of acetonitrile and methanol, volume ratio 1:1) at 80 °C for
DSC with In-doped TiO2 compact layer was enhanced by 20 h in a dark room. The sensitised electrodes were washed
11.9 and 6.9 % compared with the DSC without compact with acetonitrile and dried in air. The cell size was
layer and with pure TiO2 compact layer, respectively. In 0.25 cm2, which corresponds to the working electrode. The
introduction enhances TiO2 photoactivity and selectivity, counter electrode was platinised by applying a drop of
producing a large number of electrons due to vacant 5 mM dry H2PtCl66H2O solution (8 mM in isopropanol)
d-orbits and hindering photo-generated charges over TiO2. onto a conducting glass substrate and annealing in air at
In the present research work, we undertook some studies 400 °C for 10 min. The DSCs were assembled by sand-
to prepare TiO2 DSCs based on a novel gel, modified with wiching the working and counter electrodes using a 60-lm-
indium nitrate to stabilise anatase phase. Furthermore, the thick thermoplastic frame (Surlyn 1601, Dupont). The
effect of In introduction on the optical properties, internal electrolyte consists of 0.6 M butylmethylimidazolium
resistances, and photovoltaic performance of the DSCs was iodide, 0.05 M I2, 0.1 M LiI, and 0.5 M tert-butylpyridine
studied. in 1:1 acetonitrile/valeronitrile.

2.3 Characterisation and measurements

2 Experimental
The photoanode electrodes were characterised by X-ray
2.1 Preparation of TiO2 gels diffraction diffractometer (XRD) using a Philips X’pert
PW3020, Cu-Ka, and field emission scanning electron
In a typical experiment, TiO2 gels were prepared according microscope (FE-SEM) using a JEOL 6340. The stabilised
to the method reported elsewhere [8] except that the co- TiO2 gel was characterised in thermal behaviour using
agulation and flocculation of colloidal suspensions were Thermogravimetric analysis Linseis STA PT1600, with a
controlled to improve the morphology and photovoltaic heating rate of 5 °C/min in air up to 600 °C. X-ray pho-
properties of spherical particles. At first, a TiO2 sol was toelectron spectroscopy (XPS) was performed under ul-
prepared using titanium (IV) isopropoxide (TTIP) with a trahigh vacuum at a pass energy of 93.90 eV on a
normal purity of 97 % (Merck, Germany) as a titanium PerkinElmer PHI 5000 CESCA system equipped with a
precursor and nitric acid 70 % (Merck, Germany) as a dual X-ray source by using a Mg KR (1253.6 eV) anode
catalyst for the peptisation. The mixture of water and acid and a hemispherical energy analyser. All binding
(pH = 1.5) was heated at 80 °C under reflux system. energies (Eb) were calibrated with contaminant carbon (C
Afterwards, indium nitrate with a normal purity of 99.9 % 1s = 285 eV) as a reference. The absorption spectrum was
(Sigma, UK) was dissolved in the mixture. TTIP was analysed by ultraviolet–visible (UV–Vis) spectroscopy
gradually added to form a TiO2 sol after 5 h. The TiO2 sol using a 6705 JENWAY spectrometer. The photovoltaic
was prepared with TTIP:HNO3:H2O molar ratio of measurements of the fabricated solar cells were carried out
1:0.3:245. Finally, 1.5 mol hydroxypropyl cellulose (HPC, using a solar simulator under white light irradiance of
Aldrich, UK) and 2.0 mol acetic acid (Merck, Germany) 100 mW/cm2 (AM 1.5). The electrochemical impedance
were added into 0.5 mol sol to form the stabilised gel. spectroscopy (EIS) measurements of the DSCs were
recorded with a potentiostat/galvanostat (PGSTAT 302 N,
2.2 Preparation of DSCs Autolab, Eco-Chemie, the Netherlands) under 100 mW/cm2.
The frequency range was explored from 0.01 to 106 Hz. The
A thin TiO2 blocking layer was deposited on FTO (14 X/sq) thickness of the films was measured with a Tensor Alpha-
substrates using a 40 mM aqueous solution of TiCl4 step Profiler.

123
Efficient dye-sensitised solar cell based on uniform In-doped TiO2 spherical particles 201

3 Results and discussion 3.2 Crystal structure

3.1 Thermogravimetric analysis (TGA) Figure 2 shows the XRD patterns of the films prepared by
the gels. It can be seen that the stabilised film showed pure
Figure 1 shows the thermal behaviour of the stabilised gel anatase phase by dominant peak at 2 h = 25.31° (101),
dried at 50 °C for 15 min. The weight loss of 2.5 % at- whereas the unstabilised film was composed of a mixture
tributes to the evaporation of water below 106 °C. From of anatase and rutile phases by the strong peaks at
106 to 271 °C, the organic compound (i.e. HPC) decom- 2h = 25.31° and 2h = 27.51°, respectively. The reflection
poses and the weight loss of 3.4 % occurs. Further de- pattern of anatase nanoparticles was matched with JCPDS
composition of the organic compound (14.7 %) takes place card number 01-071-1166, with lattice constants of
between 271 and 500 °C. Since the weight loss nearly a(Å) = b(Å) = 3.784 and c(Å) = 9.515. The reflection
stabilised after 500 °C, the minimum annealing tem- pattern of rutile nanoparticles was matched with JCPDS
perature to obtain organic-free film can be determined at card number 01-078-1510, with lattice constants of
500 °C. a(Å) = b(Å) = 4.582 and c(Å) = 2.953. The XRD results
gave no information about the indium species such as
indium oxide and indium titanates solid solutions. It is
evident that the XRD patterns shifted to the small angle
direction with the introduction of indium nitrate, which
proves that In3? has been well doped into the TiO2 crystal
structure. The content of anatase (fa) and rutile (fr) in the
crystal structure can be calculated based on the relationship
between the integrated intensities of anatase (1 0 1) and
rutile (1 1 0) peaks by the following formulas [16]:
1
fa ¼   ð1Þ
Ia
1 þ 1:26 Ir

f r ¼ ð1  f a Þ ð2Þ
Fig. 1 TGA curve of the stabilised gel dried at 50 °C

Fig. 2 XRD patterns of the films annealed at 500 °C: a the unstabilised film and b the stabilised film

123
202 A. M. Bakhshayesh, N. Farajisafiloo

Table 1 Crystallite size and Electrode Crystallite size (nm) Phase composition (%) Average crystallite size (nm)
phase composition of the
deposited films Anatase Rutile Anatase Rutile

Unstabilised 19 37 35 65 28
Stabilised 08 – 100 0 08

Fig. 3 High-resolution XPS spectrum of: a Ti 2p peak, b O 1s peak, and c In 3d for the unstabilised and stabilised films

where Ia and Ir are the integrated peak intensities of the stabilising anatase phase. Additionally, the FWHM of the
anatase and rutile peaks, respectively. stabilised film was broadened, resulting in reducing the av-
In addition, the average crystallite size of crystalline erage crystallite size of the stabilised film. This phenomenon
anatase and rutile phases was calculated by the Scherrer has been previously observed in literatures [18]. The re-
equation [17]: duction of rutile content as well as anatase crystallite size in
the stabilised film gives rise to the improvement of electron
kk
d¼ ð3Þ injection since the lower surface area of rutile nanoparticles
B cos h
than that of anatase nanoparticles constrains dye sensitisa-
where d is the crystallite size, k is a dimensionless shape tion and correspondingly the short-circuit current.
factor, which has a typical value of about 0.9, k is the X-ray
wavelength of Cu, which is 1.5406 Å, h is the Bragg angle in 3.3 X-ray photoelectron spectroscopy (XPS)
degree, and B is the line broadening at half the maximum
intensity (FWHM). Table 1 shows the average crystallite X-ray photoelectron spectroscopy was carried out to mea-
size as well as the phase composition of anatase and rutile sure the elemental composition of the deposited films. The
phases. It is evident that indium introduction into TiO2 lat- Shirley method [19] of background removal was used in
tice retarded the anatase to rutile phase transformation, the least squares fitting. Figure 3 shows that Ti, O, and In

123
Efficient dye-sensitised solar cell based on uniform In-doped TiO2 spherical particles 203

Fig. 4 FE-SEM images of the surface morphology of the stabilised electrode at different magnifications: a 9500, b 96 K, and c 960 K

elements exist on the surface of the stabilised film. The Ti into the film as a result of the deposition method, in-
2p XPS spectrum of the stabilised film shows changes creasing dye loading, light scattering, and power conver-
with a major contribution from TiO2 bonds since the core sion efficiency of a DSC. The spherical particles were
levels of Ti4? 2p3/2 (Eb = 458.7 eV) and Ti4? 2p1/2 made of nanoparticles with the average grain size of
(Eb = 464.5 eV) shifted to higher binding energies as 40 nm. Coagulation is the most likely process for the for-
compared to that of undoped TiO2 (458.3 and 464.0 eV, mation of such morphology so that nanoparticles are
respectively), as can be seen in Fig. 3a. Figure 3b shows combined into larger agglomerates by double-layer com-
the XPS spectrum of O 1s for the unstabilised and sta- pression mechanism. The process is shown schematically
bilised electrodes. The peak at Eb = 532.0 eV for unsta- in Fig. 5. On the one hand, acetic acid adds additional ionic
bilised film is due to oxygen ions into the TiO2 lattice (Ti– species, which surround the nanoparticles, decreasing the
O–Ti), while the peak at Eb = 532.8 eV can be attributed thickness of double layer, and therefore, aggregates are
to the surface hydroxyl groups (Ti–OH), formed during gel formed. On the other hand, the addition of large polymeric
synthesis. The peaks at Eb = 532.5 eV and Eb = 533.3 eV molecules leads to the attachment of the polymer chains to
are ascribed to the stabilised lattice oxygen and surface- the aggregates, which hinders the further aggregate growth
bound hydroxyl groups, respectively. The XPS spectrum of [8]. The process parameters such as the type and concen-
the stabilised film in Fig. 3c illustrates two picks at Eb tration of the coagulants were controlled to prevent desta-
(In3? 3d5/2) = 445.3 and Eb (In3? 3d3/2) = 452.7 eV. The bilisation. The stabilised gel is an effective method for the
XPS measurements showed that 6 at.% indium was doped preparation of uniform sponge-like TiO2 photoelectrode
into the TiO2 lattice during the stabilisation process and with high surface area for DSC applications.
formed In0.006Ti0.994O2.
3.5 UV–Vis analysis
3.4 Microstructure
Figure 6 shows the UV–Visible spectra of the stabilised
Figure 4 shows the surface morphology of the stabilised and unstabilised TiO2 nanoparticles. A strong absorption in
electrode. It can be observed that the spherical particles, the visible light range was observed for the doped
with average diameter of 3 lm, were uniformly distributed nanoparticles, which can be ascribed to the newly formed

123
204 A. M. Bakhshayesh, N. Farajisafiloo

energy levels between the valence and the conduction

bands in the TiO2 band structure. The optical absorption, as
a result of electron excitation from the valence band to the
conduction band, can be used to determine the value of the
optical band gap. To quantify the optical band gap of
the nanoparticles, Tauc model [20] was employed in the
high-absorbance region of the transmittance spectra ac-
cording to the following equation:

r
ahm ¼ A hm  Eg ð4Þ
where hm is the photon energy, Eg is the optical band gap,
A is a constant, r is equal to 2 for allowed indirect optical
transitions, and a is the absorption coefficient. The indirect Fig. 7 Photocurrent density–voltage curve of the fabricated DSCs

band gap of 3.16 and 2.73 eV was determined for the un-
stabilised and stabilised films, respectively. The smaller
band gap of the stabilised film suggests that In dopant
modifies the conductive band of TiO2 as a result of the
mixing of In and Ti states. Thanks to the lower band gap
energy of doped nanoparticles, it is expected that their light
absorption is greater than those of undoped particles. The
red shift of the absorption curve results in a reduction of the
band gap energy and the recombination rate, resulting in
improving photocatalytic activities.

3.6 Photovoltaic performance of DSCs

The photovoltaic performance of the fabricated DSCs is

described by their density–voltage (J–V) characteristics,
illustrated in Fig. 7. The key parameters determined from
the photocurrent J–V curve are VOC (open-circuit voltage),
JSC (short-circuit current density), and FF (fill factor). The
corresponding photovoltaic parameters are summarised in
Table 2. As can be observed, the stabilised DSC had lower
open-circuit voltage. VOC is given by the difference of the
quasi-Fermi level of the electrons in TiO2 and the potential
Fig. 5 Schematic for the formation mechanism of anatase-stabilised
gels: a In-doped TiO2 sol, b addition of polymeric agent (i.e. HPC),
of the counter electrode. The quasi-Fermi level depends on
c formation of aggregates due to acetic acid addition, and d a the accumulated charge in titania and will approach the
spherical particle composed of small nanoparticles conduction band edge when the concentration of

Fig. 6 a Optical absorbance spectra of the electrodes and the lot of (ahm)0.5 versus photon energy of: b the unstabilised electrode and c the
stabilised electrode

123
Efficient dye-sensitised solar cell based on uniform In-doped TiO2 spherical particles 205

Table 2 Photovoltaic parameters of the fabricated DSCs

DSC VOC (mV) JSC (mA/cm2) FF (%) PCE (%) Adsorbed dye (108 mol/cm2) Thickness (lm)

Unstabilised 740 13.83 62.1 06.37 09.46 12.1

Stabilised 716 16.97 61.4 07.48 10.35 12.9

conduction band electrons is high. The potential of the films were placed in a 0.1 M NaOH solution, and then
conduction band edge (VCB) depends on the surface charge Beer–Lambert law was used to determine the concentration
of the metal oxide. Hagfeldt et al. [21] showed that the of the desorbed dye [22]. The absorbance of N719 dye on
following relation is valid: the stabilised film was found to be greater than the other
  electrode at the dye spectral maximum (308 nm).
kT nc
VOC ¼ VCB  ln ð5Þ Another possible reason is that doped In favours elec-
e Nc
tron transport in the material because the In-doped TiO2 is
where kT is the thermal energy, e is the elementary charge, an n-type semiconductor. Since In3? has a higher number
nc is the concentration of electrons in the conduction band, of valence electrons than Ti4?, one Ti vacancy is created
and Nc is the effective density of states in the conduction per four In introduced, and the stoichiometric reduction of
band. Indium introduction into TiO2 lattice produces an Ti4? to Ti3? occurred. Consequently, the electron carrier
n-type semiconductor, altering the position of VCB. In density in the material is increased, resulting in the im-
Sects. 3, 4, we showed that the optical band gap energy of provement of photo-response by introducing additional
the stabilised films was decreased with the introduction of energy levels within the band gap of TiO2, which increases
In, which shifted VCB to lower values, resulting in de- the current density.
creasing the VOC.
The stabilised DSC showed higher PCE of 7.48 % and 3.7 Electrochemical impedance spectroscopy (EIS)
JSC of 16.97 than that of unstabilised cell. The most likely
reason for the enhanced JSC was attributed to the increase The effect of anatase stabilisation on the electron injection,
in dye adsorption on TiO2 surface, as a result of the lower electron transfer, and recombination process of DSCs is
crystallite size of anatase nanoparticles than rutile phase. discussed by EIS. Figure 8 shows the Nyquist plots for the
To determine if that is the case here, a spectroscopic DSCs. All spectra can be described by three typical
method was used to measure the amount of the dye ad- semicircles observed in the measured frequency at both
sorbed onto the photoanode electrodes. The dye-sensitised nanocrystalline TiO2 photoelectrodes, including electro-
chemical reaction at the Pt counter electrode, charge
transfer at the TiO2/dye/electrolyte interface, and Warburg
diffusion process of I-/I3- [23]. Semicircle at mid-fre-
quencies is associated with the impedance of the charge
transfer from the excited dye LUMO level to the conduc-
tion band of TiO2 and the recombination of injected elec-
tron with the electrolyte. The diameter of the arc radius on
the EIS Nyquist plot of the stabilised electrode is smaller
than that of the other, resulting in higher charge separation
and electron transformation for the anatase-stabilised
electrode [24]. The electron life time and the electron
transfer time were calculated by srec = 2p/xrec and
sd = 2p/xd, respectively, where xrec and xd are relaxation
constants [25, 26]. The derived parameters, after fitting by
a proper equivalent circuit, are presented in Table 3. It
Fig. 8 Electrochemical impedance spectroscopy Nyquist plots of the could be observed that for both cells, the charge recom-
DSCs bination resistance (Rct) was much larger than the transport

Table 3 Recombination and Parameter Rtr (X cm2) Rct (X cm2) xrec (Hz) srec (ms) xd (Hz) sd (ms) sd/srec
electronic transfer properties of
the DSCs Unstabilised DSC 01.96 10.75 13.5 465.1 30.1 208.5 0.447
Stabilised DSC 02.11 09.25 08.7 721.8 18.5 339.4 0.470

123
206 A. M. Bakhshayesh, N. Farajisafiloo

resistances (Rtr), which is a generally necessary feature for 2. B. Tan, Y. Wu, J. Phys. Chem. B 110, 15932 (2006)
good performance of DSCs. The stabilised film showed 3. S. Ito, N. Murakami, P. Comte, P. Liska, C. Grätzel, M.K.
Nazeeruddin, M. Grätzel, Thin Solid Films 516, 4613 (2008)
less recombination as a result of higher Rtr and lower Rct in 4. T.H. Tsai, S.C. Chiou, S.M. Chen, Int. J. Electrochem. Sci. 63,
comparison with the unstabilised film. The electrons in the 333 (2011)
stabilised film were well transmitted from electrolyte to 5. Y. Chena, E. Stathatosb, D.D. Dionysioua, J. Photochem. Pho-
dye molecules and from dye molecules to the TiO2 film tobiol. A 15, 192 (2009)
6. Y.J. Kim, M.H. Lee, H.J. Kim, G. Lim, Y.S. Choi, N.G. Park, K.
than that of unstabilised film. Moreover, higher srec and sd Kim, W.I. Lee, Adv. Mater. 21, 3668 (2009)
were observed for the stabilised film thanks to the slower 7. J. Xi, Q. Zhang, S. Xie, S. Yodyingyong, K. Park, Y. Sun, J. Li,
recapture of conduction band electrons by I3-. The cell G. Cao, Nanosci. Nanotechnol. Lett. 3, 690 (2011)
with more power conversion efficiency (i.e. stabilised cell) 8. A.M. Bakhshayesh, M.R. Mohammadi, Electrochim. Acta 89, 90
(2013)
had higher sd/srec ratio, which confirms slower recombi- 9. N.G. Park, J. van de Lagemaat, A.J. Frank, J. Phys. Chem. B
nation process. 104(38), 8989 (2000)
10. S.C. Xua, S.S. Pan, Y. Xu, Y.Y. Luo, Y.X. Zhang, G.H. Lia, J.
Hazard. Mater. 283, 7 (2015)
11. S. Muduli, O. Game, V. Dhas, K. Vijayamohanan, K.A. Bogle, N.
4 Conclusions Valanoorb, S.B. Ogale, Sol. Energy 86, 1428 (2012)
12. G.R. Deng, X.H. Xia, M.L. Guo, Y. Gao, G. Shao, Mater. Lett.
This study presents a low-cost straightforward InxTi1-xO2 65, 2051 (2011)
gel, developed for the preparation of uniform films with pure 13. M. Tahir, N.S. Amin, Appl. Catal. B 162, 98 (2015)
14. X. Sun, Q. Zhang, Y. Liu, N. Huang, P. Suna, T. Peng, T. Peng,
anatase crystal structure for DSCs application. The deposited X.Z. Zhao, Electrochim. Acta 129, 276 (2014)
films showed porous structure composed of 3-lm spherical 15. S. Ito, P. Liska, P. Pechy, U. Bach, M.K. Nazeeruddin, A. Kay,
particles made of 40-nm nanoparticles. XRD analysis re- S.M. Zekeeruddin, M. Grätzel, Chem. Commun. 34, 4351 (2005)
vealed that indium nitrate stabilises anatase phase by hin- 16. R.A. Spurr, H. Myers, Anal. Chem. 29, 760 (1957)
17. B.D. Cullity, S.R. Stock, Elements of X-ray diffraction, 3rd edn.
dering anatase to rutile phase transformation. The peak (Prentice Hall, Lawrence, 2001)
location derived from XPS spectrum, in conjunction with 18. E. Wang, W. Yang, Y. Cao, J. Phys. Chem. C 113, 20912 (2009)
lack of In species derived from XRD pattern, provides a clear 19. D.A. Shirley, Rev. B 5, 4709 (1972)
proof of incorporating In ions into the TiO2 lattice through 20. J. Tauc, Amorphous and Liquid Semiconductors, 1st edn. (Plenum
Press, London and New York, 1974)
the formation of the gels. UV–Vis absorption verified that the 21. H. Alarcon, G. Boschloo, P. Mendoza, J.L. Solis, A. Hagfeldt, J.
band gap energy of nanoparticles decreased with the intro- Phys. Chem. B 109, 18483 (2005)
duction of In. The anatase-stabilised photoanode electrode 22. C. Feigenbrugel, S.L. Loew, P. Calvé, J. Mirabel, J. Photochem.
showed the maximum power conversion efficiency of Photobiol. A 174(76), 76 (2005)
23. C. Longo, J. Freitas, M.A. De Paoli, J. Photochem. Photobiol. A
7.48 % thanks to decreasing in the internal electrical resis- 159, 33 (2003)
tances and recombination, demonstrated by EIS. 24. L.W. Zhang, H.B. Fu, Y.F. Zhu, Adv. Funct. Mater. 18, 2180
(2008)
Acknowledgments Iran Nanotechnology Initiative Council is 25. A.A.B.F. Martinson, M.S. Goes, F. Fabregat-Santiago, J. Bis-
gratefully acknowledged for partially supporting this research. quert, M.J. Pellin, J.T. Hupp, J. Phys. Chem. A 113, 4015 (2009)
26. F. Fabregat-Santiago, J. Bisquert, E. Palomares, L. Otero, D.
Kuang, S.M. Zakeeruddin, M. Gratzel, J. Phys. Chem. C 111,
6550 (2007)
References

1. B. Oregano, M. Grätzel, Nature 353, 737 (1991)

123