A simple recipe for an efficient TiO2 nanofiber-based dye-sensitized solar cell.

Abstract

Development of highly efficient dye-sensitized solar cells (DSSCs) with good photovoltaic parameters is an active research area of current global interest. In this article, we provide a simple recipe for the fabrication of electrospun TiO(2) nanorod-based efficient dye-sensitized solar cell using a Pechini-type sol. The Pechini-type sol of TiO(2) nanofibers produces a highly porous and compact layer of TiO(2) upon doctor-blading and sintering without the need for an adhesion and scattering layers or TiCl(4) treatment. The best nanofiber DSSCs with an area of ~0.28 cm(2) shows an efficiency of ~4.2% under standard test conditions (100 mW/cm(2), 25°C and AM1.5 G) and an incident photon-to-electron conversion efficiency (IPCE) of ~50%. Impedance measurements show lower charge transfer resistance that improved the fill factor. We believe that simple approaches such as the present one to develop nanofiber DSSCs would open up enormous possibilities in effective harvesting of solar energy for commercial applications, considering the fact that electrospinning is a cost-effective method for the mass scale production of nanofibers and nanorods.

Full text

Page 1

A simple recipe for an efficient TiO2nanofiber-based dye-sensitized solar cell
A. Sreekumaran Naira,⇑, Rajan Joseb,⇑, Yang Shengyuanc, Seeram Ramakrishnaa,d,e,⇑
aHealthcare and Energy Materials Laboratory, Nanoscience and Nanotechnology Initiative, National University of Singapore, 117576 Singapore, Singapore
bFaculty of Industrial Science and Technology, Universiti Malaysia Pahang, 26300 Pahang, Malaysia
cNUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore 117456, Singapore
dInstitute of Materials Research and Engineering, Singapore 117602, Singapore
eKing Saud University, Riyadh 11451, Saudi Arabia
a r t i c l ei n f o
Article history:
Received 31 May 2010
Accepted 15 September 2010
Available online 22 September 2010
Keywords:
Electrospinning
TiO2nanofibers
Nanorods
Dye-sensitized solar cells
a b s t r a c t
Development of highly efficient dye-sensitized solar cells (DSSCs) with good photovoltaic parameters is
an active research area of current global interest. In this article, we provide a simple recipe for the fab-
rication of electrospun TiO2nanorod-based efficient dye-sensitized solar cell using a Pechini-type sol.
The Pechini-type sol of TiO2nanofibers produces a highly porous and compact layer of TiO2upon doc-
tor-blading and sintering without the need for an adhesion and scattering layers or TiCl4treatment.
The best nanofiber DSSCs with an area of ?0.28 cm2shows an efficiency of ?4.2% under standard test
conditions (100 mW/cm2, 25 ?C and AM1.5 G) and an incident photon-to-electron conversion efficiency
(IPCE) of ?50%. Impedance measurements show lower charge transfer resistance that improved the fill
factor. We believe that simple approaches such as the present one to develop nanofiber DSSCs would
open up enormous possibilities in effective harvesting of solar energy for commercial applications, con-
sidering the fact that electrospinning is a cost-effective method for the mass scale production of nanof-
ibers and nanorods.
? 2010 Elsevier Inc. All rights reserved.
1. Introduction
Development of efficient dye-sensitized solar cells (DSSCs) by
simple methodologies offer immense scope and potential in the
global strive towards harvesting solar energy [1–14]. Most DSSCs
employ TiO2as the porous material for anchoring light harvester
(dye) molecules. In a DSSC, the photoexcited electrons of the dye
are transferred to the conduction band of TiO2, which are taken
out through an external circuit using a working electrode (fluo-
rine-doped tin oxide, FTO) and a counter electrode (platinum sput-
tered FTO), respectively, in the presence of an electrolyte. For
realizing high efficiency in DSSCs, effective adhesion of the TiO2
layer to the FTO, a very high surface area for the porous TiO2films
for efficient absorption of the sensitizer, and good necking (connec-
tivity) between the particles are essential [1–6,12]. The TiO2nano-
particle (?10–20 nm size)-based DSSCs fabricated using multi-step
approaches (such as hole-blocking and scattering layers, TiCl4
treatment, etc.) have shown an efficiency as high as 11.1%
[12,13], which is roughly half of that of most efficient polycrystal-
line silicon (Si) solar cells. There exist several other approaches for
the fabrication of efficient solar cells using chemically synthesized
anisotropic TiO2 nanostructures [15–22]. However, mass scale
synthesis of such nanostructures is difficult from the perspective
of commercialization.
Of the anisotropic TiO2nanostructures, nanofibers (NFs), nano-
wires (NWs) and nanorods (NRs) are interesting for DSSCs from a
number of perspectives including semi-directed charge transport
(reduced grain boundaries in comparison to spherical nanoparti-
cles and hence lesser recombination of charge carriers) [20]. There
exist a number of studies on DSSCs using such TiO2nanostructures
[23–31].
Electrospinning is a versatile method for the mass production of
1-dimensional (1-D) nanostructures of oxides, non-oxides and
polymer nanofibers (NFs) [32–34]. We and others have used elec-
trospun TiO2NFs and NRs for fabricating efficient DSSCs [23–31]
and the present approach is a unique extension of our previous at-
tempts. It must be noted that the DSSCs fabricated using the above
approaches employed an adhesion layer and hot-pressing for effec-
tive contact of the electrode material to the FTO plate or a thin
dense hole-blocking film of TiO2by TiCl4treatment, etc. [23–31].
Without the adhesion layer or hot-pressing, the NF film has found
to undergo multiple cracks upon sintering. In the present approach,
we are proposing a simple single-step methodology for the fabrica-
tion of efficient TiO2NR-based DSSCs without the need for the
above mentioned steps. According to the methodology, a desired
amount of the electrospun TiO2NFs was ground to NRs and mixed
with a polymer (polyester). The resultant mixture upon sonication
formed a good paste of appropriate rheology (called Pechini-type
0021-9797/$ - see front matter ? 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2010.09.042
⇑Corresponding authors.
E-mail addresses: nniansn@nus.edu.sg (A. Sreekumaran
gmail.com (R. Jose), seeram@nus.edu.sg (S. Ramakrishna).
Nair), joserajan@
Journal of Colloid and Interface Science 353 (2011) 39–45
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science
www.elsevier.com/locate/jcis

Page 2

sol) necessary for doctor-blading. The doctor-bladed mixture upon
polymer evaporation produced a highly porous dense film of the
NRs on FTO, suited for excellent anchoring of the sensitizer dyes.
The virtues of the semi-directed electron transport are maintained
by retaining the rod/wire structure in DSSCs. The best NR-based
DSSCs fabricated by the methodology gave an efficiency of 4.2%
and an incident photon-to-electron conversion efficiency (IPCE)
of ?50%. Details of the charge transport analyzed through imped-
ance measurements showed a lower charge transport resistance
and hence a good fill factor. Application of the Pechini-type sol
for preparation of the nanorod paste, therefore, reduced the num-
ber of steps for DSSC fabrication while maintaining the properties
of the one-dimensional nanostructures.
2. Experimental
2.1. Materials and methods
2.1.1. Chemicals
Titanium (IV) isopropoxide (97%) and polyvinylpyrrolidone
(PVP, Mw = 1.30 ? 106, mp > 300? C) were from Aldrich (Steinheim,
Germany) and used as received. Acetic acid (99.7%) and citric acid
(99.5+%, A.C.S. reagent) were from LAB-SCAN Analytical Sciences,
Thailand and Sigma Aldrich, respectively. Methanol (CHROMASOLV,
Aldrich), ethanol (absolute, Fischer scientific, Leicestershire, UK) and
acetone (AR grade, Fisher Scientific, UK) were used as received. Fluo-
rine-doped tin oxide (FTO, 1.5 ? 1 cm2, sheet resistance of 25 X/
h) was from Asahi Glass, Japan. N3 dye ((cis-bis(4,40-dicarboxy-
2,20-bipyridine)dithiocyanato ruthenium(II)) was purchased from
Solaronix (Switzerland) and used as received. The FTO plates were
cleaned in ethanol, acetone and 2-propanol successively and dried
at 80 ?C in an oven before being subjected to cleaning under an air
flow.
2.1.2. Synthesis of TiO2NFs by electrospinning
PVP (0.6 g) was dissolved in a mixture of 7 mL of ethanol and
2 mL of glacial acetic acid. One milliliter of titanium (IV) isopropox-
ide was then added to the mixture and stirred for ?12 h for homo-
geneity. A clear pale-yellow solution thus obtained was then
subjected to electrospinning using a commercial instrument, NA-
NON (MECC, Japan) at 30 kV with a feed rate of 1 mL/h. The needle
specification was 27G1/2. The distance between the needle tip and
the collector was ?10 cm. The electrospun fibers were collected on
a grounded plate in the form of sheets. The sheets were annealed at
500 ?C for 1 h when the polymer matrix evaporated leaving the
TiO2in the form of continuous porous fibers.
2.1.3. Synthesis of the liquid polymer
The synthesis of the Pechini polymer for making a TiO2NR paste
of good rheology was done as per literature [35–37]. Ethylene gly-
col (8.15 g) taken in an RB flask was heated to 80 ?C in oil bath. At
80 ?C, 1.42 g of titanium (IV) isopropoxide was added and stirred
well. The mixture turned slightly turbid upon the addition of the
titanium precursor and became clear upon prolonged stirring.
Once the titanium precursor was completely dissolved in ethylene
glycol, 6.3 g of citric acid was added and the temperature was in-
creased to 100 ?C. The mixture was kept at the same temperature
under slow stirring for nearly ?5 h. The resulting clear polyester
polymer was cooled to room temperature and used in the fabrica-
tion of DSSCs as described below.
2.1.4. Fabrication of NR-based DSSCs
One hundred milligram of the TiO2NFs was ground to NRs in a
mortar. The polyester polymer synthesized as described above
(100 lL) was added to the NRs and mechanically mixed well (the
quantities were selected after several rounds of optimization tri-
als). The mixture was then subjected to sonication for ?12 h when
the TiO2NRs and the polymer formed a unique paste of appropriate
rheology (Pechini-type sol) necessary for doctor-blading. The paste
was doctor-bladed to a thickness of 15 lm, 25 lm and 30 lm,
respectively, on cleaned FTO plates on an area of ?0.28 cm2. After
relaxation of the doctor-bladed thin films in vacuum for several min-
utes, these were kept in an oven at 80 ?C for ?30 min for a slight
hardening of the composite. The FTOs were then annealed at
450 ?C for 1 h in air when the polymer evaporated leaving a highly
porous TiO2NR films on FTO plates (the TiO2electrodes fabricated
using other compositions of TiO2and the polymer were either not
good for doctor-blading or giving cracks when annealed). The pro-
cesses of the TiO2 electrode fabrication steps are illustrated in
Scheme 1. When the temperature became ?80 ?C (during the cool-
ing stage), the TiO2electrodes were taken out and soaked in the N3
dye solution (0.5 mM in 1:1 acetonitrile-tertiary butyl alcohol mix-
ture) for ?24 h for saturate-loading of the dye. The electrodes were
washed in absolute ethanol and dried in vacuum.
2.1.5. Characterization
The electrospun TiO2NFs were characterized BET surface area
measurements (NOVA 4200E Surface Area and Pore Size Analyzer,
Quantachrome, USA), powder X-ray diffraction (XRD, Bruker-AXS
D8 ADVANCE Powder X-ray diffractometer), field-emission scan-
ning electron microscopy (FE-SEM, Quanta 200 FEG System, FEI
Company, USA, operated at 10 kV) and high-resolution transmis-
sion electron microscopy (HR-TEM, JEOL 3010 operated at
300 kV). The TiO2fibers were dried under flowing N2at 150 ?C
overnight prior to BET measurements (under standard protocols
at 77 K). The flaky material obtained after sintering (?10 mg)
was powdered and spread on standard antireflection cavity
mounts for XRD measurements under 40 kV and 20 mA. The scan
speed and scan step were 2?/min and 0.01?, respectively. For SEM
measurements, the flakes of the NFs on glass substrates were
coated with a thin film of gold (Au) for good conductivity. A drop
of a dispersion of the NFs in methanol was casted on a holey car-
bon-coated copper grid, dried under ambience and vacuum and
used for TEM measurements. Photocurrent measurements were
carried out by sandwiching the TiO2cells (0.28 cm2) on fluorine-
doped tin oxide (FTO) against a platinum (Pt)-sputtered FTO plate
(in presence of parafilm spacers) using a XES-151 S solar simulator
(San Ei, Japan) under AM1.5 G condition. The level of standard irra-
diance (100 mW/cm2) was set with a calibrated c-Si reference solar
cell. Data acquisition was facilitated by an Autolab PGSTAT30 (Eco
Scheme 1. Steps illustrating the TiO2electrode fabrication processes.
40
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45

Page 3

Chemie B.V., The Netherlands) integrated with a potentiostat. IPCE
measurements were done using the standard kit from Oriel Instru-
ments (model QE-PV-SI, Silicon detector, Newport Corporation,
US).
3. Results and discussion
3.1. Characterization of the sintered nanofibers and the DSSC electrode
Fig. 1 shows the SEM images of the sintered TiO2NFs (Image A)
showing the fiber morphology. The fibers were continuous, 50–
200 nm in diameters and are randomly oriented. A magnified im-
age of the same is shown in Fig. 1B. The fibers were porous and
grainy due to evaporation of the polymer and the crystallization
of TiO2(BET surface area was 44 ± 2 m2/g and the average pore size
was estimated to be 3.2 nm). Energy-dispersive X-ray spectroscopy
(EDS) confirmed the elemental composition of the NRs. An XRD
spectrum of the sintered (500 ?C for 1 h) NFs showed the charac-
teristic anatase pattern (Supporting information 1), similar to our
previous report [38]. Fig. 1C shows a large area TEM image of the
NFs. A magnified TEM image of a single NF is shown in Fig. 1D,
revealing the presence of TiO2grains of ?12–20 nm size on the sur-
faces. Insets of C and D show a selected area diffraction pattern
(SAED) assigned to the anatase pattern and a lattice resolved image
of the periphery of a NF, respectively. A complete structural char-
acterization of electrospun TiO2NFs was reported previously [38].
Fig. 2 shows the SEM image of the TiO2electrodes fabricated
for DSSCs using the Pechini-type sol of the NRs (see also the
experimental Section 2.1.3). The large area image A shows a
dense assembly of the NRs in the TiO2electrode. A high-magnifi-
cation image of the surface is shown in Fig. 2B, which reveals the
presence of randomly oriented interconnecting NRs. A cross-sec-
tional SEM image of the electrode is shown in Fig. 2C, revealing
the thickness of the electrode (12 lm) and the compact packing
of the NRs. The thickness of the electrode in DSSCs was slightly
lower than that of the as-fabricated (doctor-bladed) electrode
(?15 lm) because of the evaporation of the polymer during the
sintering process. The thicknesses of the electrodes after sintering
in the other two cases were 20 lm and 25 lm, respectively
(thicknesses of the as-doctor-bladed electrodes were ?25 lm
and 30 lm, respectively). We believe that the compact packing
of the NRs in the present case was because of the chemical bind-
ing of the side-carboxylic acid groups of the polyester with the
TiO2surfaces in the doctor-bladed mixture. The TiO2photoelec-
trodes were also characterized by TEM analysis (Supporting infor-
mation 2). We wish to point out that the TiO2 NR-electrodes
Fig. 1. Figures A and B show a large area and a magnified SEM image, respectively, of the sintered porous TiO2NFs showing the lengthy fiber morphology. Figure C shows a
large area TEM image of the NFs. A magnified image showing the grains in a single NF is given in Figure D. Insets of C and D show a selected area electron diffraction (SAED)
pattern and a lattice resolved image, respectively.
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45
41

Page 4

fabricated by the process were not contaminated with bulk TiO2
(from the titanium precursor in the polymer, see the experimen-
tal Sections 2.1.3 and 2.1.4 above) as the amount of titanium pre-
cursor in the sol used for doctor-blading was ?8 lg/100 mg of
TiO2NRs. The SEM and TEM images of the electrodes further sub-
stantiate this.
The sintered TiO2NR-electrodes were coated with N3 dyes as
explained in section d of the experimental section. A comparison
of the absorption spectra of pure N3 dye and that anchored on
the TiO2NRs in the DSSC (the dye-chemisorbed TiO2is scratched
from the electrode, dispersed in methanol and a dilute solution
was used for measurements) is shown in Fig. 3. Pure N3 dye in ace-
tonitrile–tertiary butyl alcohol solvent mixture showed two
absorption bands centered on ?397 nm and ?538 nm, respec-
tively, due to the metal to ligand charge transfer. The absorption
spectrum of N3 anchored on TiO2(dispersed in methanol with son-
ication) showed broadened and red-shifted absorption peaks due
to its chemical interaction with the nano TiO2surfaces (H-type
aggregation) [39] through bidentate and ester linkages. The dye-
loading on TiO2 was found to be excellent to the tune of
?1.3 ? 10?7mol/cm2(from (dye) desorption measurements).
3.2. Photocurrent measurements
Fig. 4 shows the results of photocurrent measurements of the
DSSCs. The TiO2 photoelectrode with 12 lm thickness (trace a)
gave good photovoltaic parameters than the 20 lm and 25 lm
thicknesses (traces b and c, respectively). The PV measurements
of the ?12 lm electrode under 1 Sun conditions showed a current
density (Jsc) of 8 mA/cm2, open-circuit voltage (Voc) of 0.81 V, an
efficiency (g) of 4.2% and a fill factor (FF) of 62%. The inset shows
the IPCE trace the cell. We believe that the high efficiency of the
DSSC was the result of the combined effect of the following two
factors: (a) highly porous nature of the NR film (high inner
Fig. 2. Figures A and B show a large area and a magnified SEM image, respectively, of the NRs in the TiO2electrode fabricated for DSSCs. Figure C is a cross-sectional view of
the electrode revealing the compact packing of the TiO2layer. Figure D is a magnified view of C.
Fig. 3. A comparison of the absorption spectrum of pure N3 dye (trace a) and that
anchored on TiO2NFs (trace b).
42
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45

Page 5

porosity), which allowed for a high dye-loading, and (b) compact
packing and hence good necking of the NRs in the electrode. The
high porosity of the TiO2electrode resulted from the evaporation
of the polymer from the doctor-bladed mixture. The compactness
and good necking in the TiO2layer was the result of the formation
of mono- and bidentate bonds between the TiO2and the polyester.
Song et al. [24] reported photovoltaic parameters of Jsc of
14.77 mA/cm2, Voc of 0.7 V, FF of 60% and an g of 6.2% using elec-
trospun TiO2NRs, however the cell area was 0.16 cm2which is
nearly half of that in the present case (0.28 cm2). An g of ?4.2%
was recently reported using TiO2 NFs and NWs also, however,
the electrode fabrication methodologies used an adhesion layer
of the electrospun solution and hot-pressing of the electrospun
membranes for firm contact of the electrode material to the FTO
[26,27]. The present methodology aimed to avoid the above men-
tioned steps in the fabrication of efficient NR-based DSSCs.
The efficiency decreased with increase in the thickness of the
TiO2electrode layer. When the thickness was increased to 20 lm,
the Jsc, Voc, FF and g were changed to 7.4 mA/cm2, 0.75 V, 62.8%
and 3.4%, respectively. With further increase of thickness (to
25 lm), the parameters became 5.7 mA/cm2, 0.72 V, 63% and
2.49%, respectively. The decrease in Voc with increase in the TiO2
electrode thickness could be attributed to increase in the electron
diffusion pathways and the electron back-transfer reaction, respec-
tively. The FF slightly increased with increase in thickness, as ex-
pected. The lower Jsc resulted from the increase in the diffusion
pathways of I?/I3?moieties through the NR network. Thus, we
have also confirmed that the optimum thickness of the TiO2photo-
electrode for attaining best efficiencies in DSSCs was 12 lm in
accordance with previous observations with nanoparticle DSSCs
[2].
3.3. Impedance measurement
The charge transport though the NRs in the TiO2photoelectrode
was studied by electrochemical impedance spectroscopy (EIS)
[12,40,41]. The EIS measurements were carried out using the
NRs-based DSSCs in the dark using a 10 mV ac voltage superim-
posed on a forward-biased DSSC (0.4–0.7 V) with frequency rang-
ing from 30 kHz to 0.05 Hz. Fig. 5 shows the electrochemical
equivalent circuit of the DSSCs which has been successfully applied
to determine its charge transport properties [12,40,41]. The TiO2
NR/electrolyte interface is modeled as a transmission line that
exhibits complex interfacial processes in this circuit. The imped-
ance of the I3?in the electrolyte has been modeled using a finite
Warburg element and the Pt counter electrode can be described
by an RC element RPtand CPt. Similarly RFTOand CFTOare the inter-
face between the exposed FTO and the electrolyte. For a TiO2layer
of thickness L, the electron transport resistance in the nanofiber is
given by Rt= rtL. The charge transfer resistance Rct= rct/L and bar-
rier layer capacitance Qk= qkL are indicative of recombination with
the electrolyte. The rt, rct, and qkare components of the transmis-
sion line as in Fig. 5. Conventionally, a capacitor in the transmis-
sion line is replaced with a constant phase element (CPE) due to
the fact that the electrospun nanorods could exhibit frequency
dependant interfacial capacitance due to the formation of a
space-charge-region [42]. The electron diffusion coefficient as a
function of the bias voltage (Dn) through the nanorods in the pres-
ence of the present electrolyte was extracted from fitted parame-
ters of the transmission line using the expressions Dn= L2/sd,
where sd= sn(Rw/Rk)1/b, is the electron transit time. The snis the
electron lifetime given by sn= 1/(rkqk)1/band b is the CPE exponent
[42].
Fig. 6 shows typical measured impedance spectra of the NRs-
based DSSCs duly fitted to the above described model using the
Z-View program (Scribner Associates). The high frequency offset
in the spectra is the series resistance of the cell, primarily due
to the FTO resistance. The impedance at the platinum/electrolyte
interface can be seen in the spectra as a small arc at high frequen-
cies. The charge transport and charge transfer resistance in the
TiO2can be seen as the big semicircle at the intermediate frequen-
cies, whose radius increases with decrease in the bias voltage.
Diffusion in the electrolyte is usually expressed in the spectra as
a small arc at low frequencies. However, this component is
Fig. 4. Traces a, b and c show the Jsc vs. V plots of the DSSCs having different
thicknesses of the TiO2electrode. The thicknesses of the electrodes in a, b and c
were ?12 lm, 20 lm and 25 lm, respectively. Inset shows the IPCE spectrum of the
?12 lm sample.
Fig. 5. The equivalent circuit of a DSSC represented as a transmission line.
Fig. 6. Experimental EIS spectra fitted the transmission line model shown in Fig. 5
using Z-View program for three bias conditions. Inset: the EIS spectrum of the
measurement for 0.7 V for clarity.
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45
43

Page 6

overshadowed in the present spectra, even for the lower bias volt-
ages, by the large charge transport resistances observed in the
TiO2.
Fig. 7 shows the transmission line parameters determined from
the EIS spectra. Some data from the electrospun continuous NFs-
based DSSCs [27] and a best performing nanoparticulate based
DSSCs [12] have also been included in the graphs for a comparison.
The charge transport appears severely affected upon breaking
down of the continuous NFs into short NRs of aspect ratio ?3:1.
This is because the grain-boundary density and hence the grain-
boundary scattering increased upon breaking the continuous NFs
into the NRs. The Rctis comparable to that of the nanoparticulate
DSSCs which is consistent with the above observation. Owing to
the poor crystallinity (particle size ?12–20 nm) and inferior Rct,
the Dn of the nanorods has further decreased than that of the
nanofibers. It should be noted that the TiO2NFs reported in an ear-
lier study [27] and the present NRs were thermally processed for
nucleation and crystal growth under similar conditions and that
their crystallinities are similar. Thus a lowering of Dncould nor-
mally be expected. The capacitance Qkof the nanorods are higher
than that of the nanoparticles indicating a high electron concentra-
tion and a thin layer of space-charge-region which could help in
reducing the charge recombination. Occurance of this space-
charge-region is reflected in the fill factor and Rct even though
the Vocwas slightly higher for the NFs-based cells.
4. Conclusions
In conclusion, we have found that a Pechini-type paste formula-
tion works well for the TiO2NR-based DSSCs for realizing high
efficiency. The TiO2 electrodes fabricated by the methodology
without the use of an adhesion layer, hot-pressing, TiCl4and the
associated multiple heat treatments gave compact packing with
high porosity and necking. The photovoltaic parameters of the best
NR-based DSSCs were a Jsc of 8 mA/cm2, Voc of 0.81 V, FF of 62%
and g of 4.2%. The efficiency of the DSSCs decreased with increase
in the thickness of the TiO2electrode, the ideal thickness being
?12 lm. The impedance spectroscopic measurement and analyses
showed that the NR-based DSCs are characterized by a lower
charge transfer resistance and higher space charge capacitance that
improved the fill factor. We believe that simple energy-efficient
approaches such as the present one to fabricate TiO2NR-based
DSSCs would help in their commercialization at low-cost.
Acknowledgments
Clean Energy Program Office (CEPO) of National Research Foun-
dation (NRF), Singapore is thanked for financial assistance (Grant
No: NRF2007EWT-CERP01-0531). We thank Dr. Qing Wang and
Dr. J.R. Jennings of the Department of Materials Science and Engi-
neering, National University of Singapore for help with IPCE mea-
surements. We also thank Prof. B.V.R. Chowdari and Dr. M.V.V.
Reddy of the Dept. of Physics, National University of Singapore
for surface area and powder XRD measurements.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcis.2010.09.042.
Fig. 7. The fitted transmission line parameters of the present electrospun nanorods (squared), nanoparticles (circles) in Ref. [41], and electrospun nanofiber-based DSCs
(triangles).
44
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45

Page 7

References
[1] M. Grätzel, Acc. Chem. Res. 42 (2009) 1788.
[2] S. Ito, T.N. Murakami, P. Comte, P. Liska, C. Grätzel, M.K. Nazeeruddin, M.
Grätzel, Thin Solid Films 516 (2008) 4613.
[3] E.L. Tae, S.H. Lee, J.K. Lee, S.S. Yoo, E.J. Kang, K.B. Yoon, J. Phys. Chem. B 109
(2005) 22513.
[4] M. Grätzel, J. Photochem. Photobiol. C. Photochem. Rev. 4 (2003) 145.
[5] M. Grätzel, Prog. Photovolt. Res. Appl. 8 (2000) 171.
[6] A. Hagfeldt, M. Grätzel, Acc. Chem. Res. 33 (2000) 269.
[7] V. Thavasi, V. Renugopalakrishnan, R. Jose, S. Ramakrishna, Mater. Sci. Eng. R
63 (2009) 81.
[8] V. Thavasi, G. Sing, S. Ramakrishna, Energy Environ. Sci. 1 (2008) 205.
[9] R. Jose, V. Thavasi, S. Ramakrishna, J. Am. Ceram. Soc. 92 (2009) 289.
[10] M. Zukalová, A. Zukal, L. Kavan, M.K. Nazeeruddin, P. Liska, M. Grätzel, Nano
Lett. 5 (2005) 1789.
[11] P. Wang, B. Wenger, R. Humphry-Baker, J.-E. Moser, J. Teuscher, W. Kantlehner,
J. Mezger, E.V. Stoyanov, S.M. Zakeeruddin, M. Grätzel, J. Am. Chem. Soc. 127
(2005) 6850.
[12] Q. Wang, S. Ito, M. Grätzel, F. Fabregat-Santiago, I. Mora-Seró, J. Bisquert, T.
Bessho, H. Imai, J. Phys. Chem. B 110 (1) (2006) 25210.
[13] S. Chuangchote, T. Sagawa, S. Yoshikawa, Appl. Phys. Lett. 93 (1) (2008)
033310.
[14] R.Y. Ogura, S. Nakane, M. Morooka, M. Orihashi, Y. Suzuki, K. Noda, Appl. Phys.
Lett. 94 (1) (2009) 073308.
[15] D. Kuang, J. Brillet, P. Chen, M. Takata, S. Uchida, H. Miura, K. Sumioka, S.
Zakeeruddin, M. Grätzel, ACS Nano 2 (2008) 1113.
[16] S.H. Kang, S.-H. Choi, M.-S. Kang, J.-Y. Kim, H.-S. Kim, T. Hyeon, Y.-E. Sung, Adv.
Mater. 20 (2008) 54.
[17] B. Liu, E.S. Aydil, Am. Chem. Soc. 131 (2009) 3985.
[18] G.K. Mor, K. Shankar, M. Paulose, O.K. Varghese, C.A. Grimes, Nano Lett. 2
(2006) 215.
[19] K. Zhu, N.R. Neale, A. Miedaner, A.J. Frank, Nano Lett. 7 (2007) 69.
[20] K. Yu, J. Chen, Nanoscale Res. Lett. 4 (2009) 1.
[21] Q. Yao, J. Liu, Q. Peng, X. Wang, Y. Li, Chem. Asian J. 1 (2006) 737.
[22] J. Liu, Y-.T. Kuo, K.J. Klabunde, C. Rochford, J. Wu, J. Li, ACS Appl. Mater.
Interfaces 1 (2009) 1645.
[23] M.Y. Song, D.K. Kim, K.J. Ihn, S.M. Jo, D.Y. Kim, Nanotechnology 15 (2004) 1861.
[24] M.Y. Song, Y.R. Ahn, S.M. Jo, D.Y. Kim, J.-P. Kim, J.-P. Ahn, Appl. Phys. Lett. 87
(2005) 113113.
[25] B.H. Lee, M.Y. Song, S.-Y. Jang, S.M. Jo, S.-Y. Kwak, D.Y. Kim, J. Phys. Chem. C 113
(2009) 21453.
[26] P.S. Archana, R. Jose, C. Vijila, S. Ramakrishna, J. Phys. Chem. C 13 (2009)
21538.
[27] K. Mukherjee, T.-H. Teng, R. Jose, S. Ramakrishna, Appl. Phys. Lett. 95 (1) (2009)
01210.
[28] R. Jose, A. Kumar, V. Thavasi, Nanotechnology 19 (1) (2008) 424004.
[29] K. Fujihara, A. Kumar, R. Jose, Nanotechnology 18 (1) (2007) 365709.
[30] R. Zhu, C.-Y. Jiang, X.-Z. Liu, B. Liu, A. Kumar, S. Ramakrishna, Appl. Phys. Lett.
93 (1) (2008) 013102.
[31] R. Jose, A. Kumar, V. Thavasi, K. Fujihara, S. Uchida, S. Ramakrishna, Appl. Phys.
Lett. 93 (1) (2008) 023125.
[32] W.E. Teo, S. Ramakrishna, Nanotechnology 17 (2006) R89.
[33] D. Li, Y. Xia, Adv. Mater. 16 (2004) 1151.
[34] D. Li, Y. Wang, Y. Xia, Nano Lett. 3 (2003) 1167.
[35] W. Guo, A.K. Datye, T.L. Ward, Mater. Chem. 15 (2005) 470.
[36] M. Pechini, US Patent no. 3 330 697 (1967). Pechini-type process generally
refers to the synthesis of polymerizable complexes involving the steps of
metal complexformation and
Multifunctional organic acids such as citric acid are used to form metal
complexes and polyesterification with ethylene glycol produces a polymeric
resin. A mixture of TiO2 NRs and the Pechini polymer upon sonication
produces a sol of appropriate rheology ideal for doctor blading.
[37] U.O. Krašovec, M. Berginc, M. Hoc ˇevar, M. Topic ˇ, Sol. Energy Mater. Sol. Cells 93
(2009) 379.
[38] A. Kumar, R. Jose, K. Fujihara, J. Wang, S. Ramakrishna, Chem. Mater. 19 (2007)
6536.
[39] F. Nüesch, J.E. Moser, V. Shklover, M. Grätzel, J. Am. Chem. Soc. 118 (1996)
5420.
[40] A. Zaban, M. Greenshtein, I. Mora-Sero, Chem. Phys. Chem. 4 (2003) 859.
[41] J. Bisquert, A. Zaban, M. Greenshtein, I. Mora-Sero, J. Am. Chem. Soc. 126
(2004) 13550.
[42] J. Bisquert, G. Garcia-Belmonte, F. Fabregat-Santiago, N.S. Ferriols, P.
Bogdanoff, E.C. Pereira, J. Phys. Chem. B 104 (2000) 2287.
in-situ
polymerizationoforganics.
A. Sreekumaran Nair et al./Journal of Colloid and Interface Science 353 (2011) 39–45
45