Simple sensitizers of low band gap based on 4-nitro-a-cyanostilbene prepared from a one-step reaction for efficient dye-sensitized solar cells

Article abstract of DOI:10.1016/j.orgel.2010.05.002

Two simple dyes D1 and D2 based on 4-nitro-a-cyanostilbene were conveniently synthesized from a one-step reaction. These dyes were prepared from the condensation of 4-carboxybenzaldehyde or 4-hydroxybenzaldehyde with 4-nitrobenzyl cyanide to afford 4-nitro-40-carboxy-a-cyanostilbene (D1) and 4-nitro-40-hydroxy-a-cyanostilbene (D2) respectively. Photophysical and electrochemical properties of the dyes were investigated by UV-vis spectroscopy and cyclic voltammetry. Their absorption spectra were broad with long-wavelength absorption maximum at 617-655 nm and optical band gap of 1.63 eV. These two dyes were used as sensitizers in quasi solid state dye-sensitized solar cells (DSSCs). Photovoltaic devices with D1 showed a maximum monochromatic incident photon to current efficiency (IPCE) of 70% and overall power conversion efficiency (PCE) of 4.8%, under illumination intensity 100 mW/cm². The photovoltaic performance of the cell with D2 was lower because of less dye loading on the TiO₂ surface, since it has OH anchoring group, and lower electron lifetime. Additionally, we have investigated the photovoltaic response of the DSSCs with nitrogen doped TiO2 photoanode and found that the PCE has been enhanced. The enhancement in PCE has been attributed to the increase in open circuit voltage and fill factor.

Full text of DOI:10.1016/j.orgel.2010.05.002

Organic Electronics 11 (2010) 1242–1249
Contents lists available at ScienceDirect
Organic Electronics
journal homepage: www.elsevier.com/locate/orgel
Simple sensitizers of low band gap based on 4-nitro-a-cyanostilbene
prepared from a one-step reaction for efficient dye-sensitized solar cells
a
J.A. Mikroyannidis a, , D.V. Tsagkournos , P. Balraju b,d, G.D. Sharma b,c,
*
**
a Chemical Technology Laboratory, Department of Chemistry, University of Patras, GR-26500 Patras, Greece
b Physics Department, Molecular Electronic and Optoelectronic Device Laboratory, JNV University, Jodhpur (Raj.) 342005, India
c Jaipur Engineering College, Kukas, Jaipur (Raj.), India
d Molecular Electronics Laboratory, JNCASR, Bangalore, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Two simple dyes D1 and D2 based on 4-nitro-a-cyanostilbene were conveniently syn-
thesized from a one-step reaction. These dyes were prepared from the condensation
of 4-carboxybenzaldehyde or 4-hydroxybenzaldehyde with 4-nitrobenzyl cyanide to
afford 4-nitro-40-carboxy- -cyanostilbene (D1) and 4-nitro-40-hydroxy-
a a-cyanostilbene
Received 25 March 2010
Received in revised form 27 April 2010
Accepted 1 May 2010
Available online 9 May 2010
(D2) respectively. Photophysical and electrochemical properties of the dyes were inves-
tigated by UV–vis spectroscopy and cyclic voltammetry. Their absorption spectra were
broad with long-wavelength absorption maximum at 617–655 nm and optical band
gap of 1.63 eV. These two dyes were used as sensitizers in quasi solid state dye-sensi-
tized solar cells (DSSCs). Photovoltaic devices with D1 showed a maximum monochro-
matic incident photon to current efficiency (IPCE) of 70% and overall power
conversion efficiency (PCE) of 4.8%, under illumination intensity 100 mW/cm2. The pho-
tovoltaic performance of the cell with D2 was lower because of less dye loading on the
TiO2 surface, since it has OH anchoring group, and lower electron lifetime. Additionally,
we have investigated the photovoltaic response of the DSSCs with nitrogen doped TiO2
photoanode and found that the PCE has been enhanced. The enhancement in PCE has
been attributed to the increase in open circuit voltage and fill factor.
Keywords:
Dye-sensitized solar cells
Low band gap
Cyanovinylene
Nitro
Organic sensitizers
Photovoltaic
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
conversion of solar energy into electrical energy and has
attracted the attention of material science community in
Silicon solar cells are presently playing a crucial role in
harvesting the solar energy efficiently. Nevertheless, their
high cost of production is one of the major stumbling
blocks towards their reach at common mass level as a
clean energy substitute. Dye-sensitized solar cells (DSSCs)
are a cheap alternatives to silicon solar cells for efficient
the recent past [1,2]. However, DSSCs based on organic
dyes give a respectable power conversion efficiency of
10–11%, which is still lower than the efficiency obtained
by silicon solar cells (20%) [3,4]. One of the most important
and challenging factor attributed to the lower efficiency of
the DSSCs, is the narrow optical absorption window
(400–800 nm) of organic dyes, which is narrower than
the optical absorption window of crystalline silicon
(500–1100 nm).
* Corresponding author. Tel.: +30 2610 997115; fax: +30 2610 997118
(J.A. Mikroyannidis).
** Corresponding author at: Physics Department, Molecular Electronic
and Optoelectronic Device Laboratory, JNV University, Jodhpur (Raj.)
342005, India. Tel.: +91 0291 2720857; fax: +91 0291 2720856 (G.D.
Sharma).
E-mail addresses: mikroyan@chemistry.upatras.gr (J.A. Mikroyannidis),
sharmagd_in@yahoo.com (G.D. Sharma).
The DSSC is typically composed of a photoanode, made
of a mesoporous dye-sensitized TiO2 film on a F-doped
SnO2 conductive substrate (FTO), a counter electrode,
made of a discontinuous platinum particulate layer on
FTO, and a Iꢀ3 /Iꢀ redox electrolyte or a solid state organic
1566-1199/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.orgel.2010.05.002

J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
1243
hole transporter [5]. Among them, the dye is essential in
2. Experimental
DSSC since it is the origin of the photocurrent generation.
In most cases, Ru-based organometallic compounds such
as N719 (red dye) [6] and N749 (green dye) [7] as well as
other black dyes have been used as sensitizers for the
DSSCs with PCE of more than 11%. However, due to the
high price of RuCl3 and the difficulties of synthesizing Ru
(II)–polypyridine complexes, metal-free organic dyes have
recently attracted a lot of attention because of higher mo-
lar extinction coefficient [8–17] and impressive efficiencies
in the range 8–9.7% have been achieved. Moreover, the
highest occupied molecular orbital (HOMO) and the lowest
unoccupied molecular orbital (LUMO) energy levels can be
rather easily tuned by controlling their molecular struc-
tures, which leads to various colors. Certain metal-free
photosensitizers have been very recently synthesized and
used for DSSCs [18].
Overall power conversion efficiency (PCE) of a DSSC is
basically decided by the short circuit current density (Jsc),
open circuit voltage (Voc) and fill factor (FF). There are re-
ports about a number of organic dyes giving very high Jsc
but the overall PCE of the cell was mainly attributed to
small observed Voc [19,20]. Voc plays an important role in
controlling the PCE and is basically decided by the differ-
ence in the conduction band level of TiO2 and potential
of Iꢀ3 /Iꢀ redox couple in DSSCs. In an interesting report
O’Regan and Grätzel [21] advocated the role of electron
recombination in governing the Voc based on their tran-
sient photovoltage and photocurrent measurements in
the DSSCs.
2.1. Reagents and solvents
4-Nitrobenzyl cyanide was synthesized from the nitra-
tion of benzyl cyanide with concentrated nitric and sulfuric
acid [26]. 4-Carboxybenzaldehyde and 4-hydroxybenzal-
dehyde were recrystallized from ethanol and water,
respectively. Tetrahydrofuran (THF) was dried by distilla-
tion over CaH2. All other reagents and solvents were com-
mercially purchased and were used as supplied.
2.2. Synthesis of dyes
2.2.1. 4-Nitro-40-carboxy-
a-cyanostilbene (D1)
A flask was charged with a solution of 4-carboxybenzal-
dehyde (0.5102 g, 3.33 mmol) and 4-nitrobenzyl cyanide
(0.5400 g, 3.33 mmol) in ethanol (30 mL). Sodium hydrox-
ide (0.60 g, 15.00 mmol) dissolved in ethanol (10 mL) was
added portionwise to the stirred solution. The mixture
was stirred for 1 h at room temperature under N2 and then
it was concentrated under reduced pressure to remove
ethanol. Water was added to the concentrate and then di-
lute hydrochloric acid was added to this solution to precip-
itate D1 as a dark-green solid. It was filtered and washed
thoroughly with water (0.70 g, 72%).
FT-IR (KBr, cmꢀ1): 3408 (O–H deformation of carboxyl);
2170 (cyano); 1684 (carboxylic C@O); 1592 (aromatic);
1534, 1384 (nitro).
1H NMR (CDCl3, ppm): 12.09 (broad, 1H, carboxyl);
8.19–8.12 (m, 4H, aromatic ortho to nitro and carboxyl);
7.70 (s, 1H, olefinic); 7.51–7.45 (m, 4H, aromatic meta to
nitro and carboxyl).
A series of low band gap small molecules and poly-
mers bearing 4-nitro-a-cyanostilbene moieties were syn-
thesized in our laboratory and used for bulk
heterojunction solar cells [22] and DSSCs [23] very re-
cently. In continuation of this research line, herein we
successfully synthesized two simple photosensitizers D1
Anal. Calc. for C16H10N2O4: C, 65.31; H, 3.43; N, 9.52.
Found: C, 64.94; H, 3.40; N, 9.58%.
and D2 based on 4-nitro-a-cyanostilbene which contain
2.2.2. 4-Nitro-40-hydroxy-
a-cyanostilbene (D2)
carboxy or hydroxy anchoring group, respectively. These
dyes were synthesized from a one-step reaction in high
yields utilizing widely available and inexpensive starting
materials. A literature survey revealed that dye D1 has
not been previously synthesized. Dye D2 has been syn-
thesized by condensing equimolar quantities of 4-nitro-
This dye was similarly prepared in 60% yield from the
reaction of 4-hydroxybenzaldehyde with 4-nitrobenzyl
cyanide (mol ratio 1:1) in ethanol in the presence of so-
dium hydroxide. The reaction mixture was stirred for 1 h
at room temperature under N2 and then it was concen-
trated under reduced pressure to remove ethanol. Water
was added to the concentrate and D2 precipitated as a
dark-green solid.
benzylcyanide with 4-hydroxybenzaldehyde [24].
A
photoresponsive effect has been observed with polyam-
ide/cyanostilbene systems [25]. Upon comparing the
present dyes D1 and D2 to those of reference 23 the fol-
lowing differences are observed. (i) The dyes D1 and D2
are very simple, while the previous dyes are complex. D1
and D2 are among the simplest photosensitizers which
have been reported in the literature. (ii) The most
remarkable difference is that D1 and D2 are prepared
from one step reaction only, while the previous from
multiple step reactions. (iii) D1 and D2 are unsymmetri-
cal molecules while the previous are symmetrical. The
dyes D1 and D2 were characterized in DSSCs using qua-
si-solid-state polymer electrolyte, FTO/TiO2 photoanode
and Pt/FTO counter electrode. The DSSC based on D1
showed higher PCE than that for D2, which is attributed
to the higher dye loading and electron lifetime for the
D1 based cell.
FT-IR (KBr, cmꢀ1): 3422 (O–H stretching of hydroxyl);
2168 (cyano); 1592 (aromatic); 1530, 1342 (nitro).
1H NMR (CDCl3, ppm): 8.18 (m, 2H, aromatic ortho to
nitro); 7.71 (s, 1H, olefinic); 7.48 (m, 2H, aromatic meta
to nitro); 7.24 (m, 2H, aromatic meta to hydroxyl); 6.84
(m, 2H, aromatic ortho to hydroxyl); 5.34 (s, 1H, hydroxyl).
Anal. Calc. for C15H10N2O3: C, 67.67; H, 3.79; N, 10.52.
Found: C, 67.26; H, 3.85; N, 10.63%.
2.3. Fabrication of dye-sensitized TiO2 electrode and DSSC
assembly
TiO2 paste was prepared by mixing 1 g of TiO2 powder
(P25, Degussa), 0.2 mL of acetic acid, 1 mL of water. Then
60 mL of ethanol was slowly added while sonicating the

1244
J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
mixture for 3 h. Finally, Triton X-100 was added and a well
dispersed colloidal paste was obtained (TiO2). The whole
procedure is slow under vigorous stirring. The mixture
was stirred vigorously for 2–4 h at room temperature and
then stirred for 4 h at 80 °C to form a transparent colloidal
paste. The TiO2 paste was deposited on the F-doped tin
oxide (FTO) coated glass substrates by the doctor blade
technique. The TiO2 coated FTO films were sintered at
450 °C for 30 min. Before immersing into the dye solution,
these films were soaked in a 0.2 M aqueous TiCl4 solution
overnight in a closed chamber. After being washed with
deionized water and fully rinsed with ethanol, the films
were heated again at 450 °C for 30 min, followed by cool-
ing to 60 °C.
The nitrogen doped TiO2 paste was synthesized via sol
gel method using urea as nitrogen source. A solution of
urea was prepared by dissolving an appropriate amount
of urea in 90 mL of deionised water and it was added to
the paste of TiO2 under vigorous stirring at 80 °C. The
deposition of the nitrogen doped TiO2 film is the same as
described earlier. The crystallite size of nitrogen doped
TiO2 was determined using X-ray diffraction analysis with
The current–voltage (J–V) characteristics of the devices
in dark and under illumination were measured by semi-
conductor parameter analyzer (Keithley 4200-SCS). A xe-
non light source (Oriel, USA) was used to give an
irradiance of 100 mW/cm2 (the equivalent of one sun at
AM 1.5) at the surface of the device. The photoaction spec-
trum of the devices was measured using a monochromator
(Spex 500 M, USA) and the resulted photocurrent was mea-
sured with Keithley electrometer (model 6514), which is
interfaced to the computer by LABVIEW software. The elec-
trochemical impedance spectra (EIS) measurements were
carried out by applying bias of the open circuit voltage
(Voc) and recorded over a frequency range of 1 mHz to
105 Hz with ac amplitude of 10 mV. The above measure-
ments were recorded with an electrochemical impedance
analyzer equipped with FRA.
3. Results and discussion
3.1. Synthesis and characterization
Dyes D1 and D2 were conveniently synthesized by the
one-step reaction outlined in Scheme 1. In particular, 4-
carboxybenzaldehyde or 4-hydroxybenzaldehyde reacted
with an equivalent amount of 4-nitrobenzyl cyanide in
ethanol in the presence of sodium hydroxide to afford D1
and D2, respectively. D1 was obtained as sodium salt and
the acidification of the latter by hydrochloric acid yielded
the free dye. tert-Butoxide could be used instead of sodium
hydroxide for the condensation reaction. Both compounds
were obtained as dark-green solids which were soluble in
common organic solvents like THF, chloroform and dichlo-
romethane. Moreover, they were soluble in ethanol due to
the presence of the carboxy or hydroxy group. Even though
these dyes are small molecules, films of them could be ob-
tained from their solutions by spin-casting.
Structural characterization of dyes was accomplished
by FT-IR and 1H NMR spectroscopy. Since these dyes are
differentiated only from their anchoring groups, their spec-
tra were similar. They showed common absorption bands
around 1530, 1380 (NO2), 2170 (cyano) and 1592 cmꢀ1
(aromatic). Besides, D1 displayed absorptions at 3408
and 1684 cmꢀ1 (carboxyl), while D2 at 3422 cmꢀ1 (hydro-
xyl). The 1H NMR spectra showed common peaks approxi-
mately at 8.18 (aromatic ortho to nitro) and 7.70 ppm
(olefinic). Finally, D1 and D2 displayed characteristics sig-
nals at 12.09 and 5.34 ppm assigned to the carboxyl and
hydroxyl, respectively.
Cu K
a as radiant source and wavelength (k = 0.1541 nm).
The crystallite size
equation.
D was calculated by the Sherrer
The dyes were dissolved in THF solution (0.5 mM) and
the TiO2 electrode was immersed in the dye solution for
10 h and after sensitization the dye electrode was washed.
For the composite electrolytes preparation, 0.0383 g of
TiO2 powder (P25, Degussa) was dried for about 24 h at
250 °C. After drying a quasi-solid-state polymer electrolyte
containing 0.0383 g of P25 TiO2 powder, 0.1 g of LiI, 0.019 g
of I2, 0.264 g of PEO, and 44 lL of 4-tert-butylpyridine in
1:1 acetone/propylene carbonate was prepared. The mix-
ture was continuously stirred at 80 °C in water bath for
about 4–5 h. The polymer electrolyte was spread on the
sensitized TiO2 film by spin coating to form a hole-con-
ducting layer. The counter electrode was platinized by spin
coating H2PtCl6 solution (2 mg in 1 mL isopropanol) onto
the FTO coated glass substrate and then heated at 450 °C
for 30 min in air. The dye-sensitized photoelectrode con-
taining the quasi-solid-state polymer electrolyte and the
counter electrode were clamped together and separated
by 20 lm spacer.
2.4. Characterization methods
IR spectra were recorded on a Perkin–Elmer 16PC FT-IR
spectrometer with KBr pellets. 1H NMR (400 MHz) spectra
were obtained using a Brucker spectrometer. Chemical
shifts (d values) are given in parts per million with tetra-
methylsilane as an internal standard. UV–vis spectra were
recorded on a Beckman DU-640 spectrometer with spect-
rograde THF. Elemental analyses were carried out with a
Carlo Erba model EA1108 analyzer.
3.2. Photophysical properties
The photophysical properties of the dyes were investi-
gated in both dilute (10ꢀ5 M) THF solution and thin film.
The electrochemical properties of both dyes were stud-
ied by using cyclic voltammetry (CV) (HCH instrument).
The dyes were coated on platinum disk and immersed in
0.1 M Bu4NPF6 acetonitrile solution. CV was recorded using
Platinum disk as working electrode and Ag/Ag+ as the ref-
erence electrode at a scan rate of 20 mV sꢀ1
.
Scheme 1. Synthesis of dyes D1 and D2.

J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
1245
Table 1
Fig. 1 depicts the UV–vis absorption spectra of the dyes
Optical and electrochemical properties of D1 and D2.
which have been normalized with respect the long-wave-
length absorption. The photophysical characteristics of
the dyes are summarized in Table 1.
Dye
D1
D2
a
ka,max in solution (nm)
617
652
761
1.63
639
655
761
1.63
a
Generally, the absorption curves of the dyes were broad
and covered a wide range of the spectrum from 300 to
ꢁ750 nm. This attractive feature was attributed to the
ka,max in thin film (nm)
Thin film absorption onset (nm)
Eogpt (eV)
b
HOMO (eV)
LUMO (eV)
Eegl (eV)
c
ꢀ5.2
ꢀ3.5
1.7
ꢀ5.4
ꢀ3.7
1.7
4-nitro-a-cyanostilbene segment as it has been well
established in our previous publications [22,23]. The
long-wavelength absorption maximum (ka,max) was lo-
cated at 617–655 nm. The thin film absorption curves of
the dyes absolutely coincided above 550 nm and exhibited
absorption onset at 761 nm corresponding to an optical
band gap (Eogpt) of 1.63 eV. This value of Eogpt conforms to
a
ka,max: the long-wavelength absorption maxima from the UV–vis
spectra in THF solution or in thin film.
Eogpt: optical band gap determined from the absorption onset in thin
b
film.
Eegl: electrochemical band gap determined from cyclic voltammetry.
c
those of other related materials which contain 4-nitro-
a-
cyanostilbene moieties [22,23].
The UV–vis spectra of the sensitizers absorbed on TiO2
film are shown in Fig. 2. The absorption spectrum of the
blank TiO2 film was subtracted from the curve. It can be
seen in this figure that the absorption spectra were broad-
ened, when anchored at the TiO2 surface, as compared to
pure dye spectra. Such broadening has been observed in
Fig. 2. UV–vis absorption spectra of the dyes adsorbed on TiO2 films.
other organic sensitizers on TiO2 electrodes [27]. Strong
interaction between the adsorbed dye molecules and oxide
molecules in the TiO2 surface leads to aggregate formation
and consequently broadening the absorption spectrum.
This might be correlated with the dye–TiO2 and dye–dye
interactions. The broadened absorption of dyes loaded on
the TiO2 film is advantageous for light harvesting of the so-
lar spectrum. The concentration of the D2 dye adsorbed on
the film is lower than that of D1. This is attributable to im-
paired surface packing due to OH anchoring group of
D2.The energetic alignment of the HOMO and LUMO en-
ergy levels is crucial for an efficient operation of the sensi-
tizer in a DSSC. To ensure efficient electron injection from
the excited dye into the conduction band of the TiO2 films,
the LUMO level must be higher in energy than the TiO2
conduction band edge. The HOMO level of the sensitizer
must be lower in energy than the redox potential of the
Iꢀ=Iꢀ3 redox couple for efficient regeneration of the sensi-
tizer cation after photoinduced electron injection into the
TiO2 film. The electrochemical behaviour of the sensitizers
was investigated by CV [28] and the data are listed in Table
1. The HOMO and LUMO levels of both sensitizers are suit-
able for DSSCs. The difference between the LUMO of dye
and the conduction band of TiO2 should be higher than
0.2 eV for efficient electron injection. The LUMO energy
levels (D1: ꢀ3.5 eV and D2: ꢀ3.7 eV) are sufficiently higher
than the conduction band of TiO2 (ꢀ4.0 eV). Hence, an effi-
cient electron transfer from the excited dye to the TiO2 is
Fig. 1. Normalized absorption spectra of dyes in THF solution (top) and
thin film (bottom).

1246
J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
oxidized species (tri-iodide) in the electrolyte at the
uncovered TiO2 surface.
The J–V characteristics of the DSSCs based on D1 and D2
are shown in Fig. 4. The photovoltaic parameters i.e. short
circuit photocurrent (Jsc), open circuit voltage (Voc), fill fac-
ensured. The HOMO energy levels of the dyes (D1: ꢀ5.2 eV
and D2: ꢀ5.4 eV) are lower than the standard potential of
the Iꢀ=Iꢀ3 redox couple (ꢀ4.83 eV vs. vacuum) [29]. Thus
sufficient driving force for the regeneration of the oxidized
sensitizer is available in DSSCs using these materials.
tor (FF) and power conversion efficiency (g) are summa-
rized in Table 2. Depending on different anchoring
groups, these dyes show different PCE in spite of the same
band gap. The lower Jsc for DSSC based on D2 is consistent
with the lower IPCE, because the adsorbed amount of D2
on the surface of TiO2 is lower than D1. The lower value
of Voc can be explained by the increased charge recombina-
tion with the redox species which are present in the elec-
trolyte.In DSSCs, the charge separation occurs in the
adsorbed dyes by the injection of excited electrons into
the conduction band of TiO2 and reduction of the resulting
dye cations by Iꢀ in the electrolyte. Then, at short circuit
conditions, the electrons and Iꢀ3 diffuse towards the pho-
toanode and Pt counter electrodes, respectively. Conse-
quently, the charge collection efficiency is related to
3.3. Photovoltaic performance
Fig. 3 shows the incident photon to current efficiency
(IPCE) spectra for the DSSCs based on D1 and D2. The IPCE
data of D1 sensitizer exhibit efficiency higher than 55% in
the range 550–750 nm with a maximum value about 70%
at 650 nm. On the other hand, the IPCE value for the DSSC
based on D2 dye is lower than that for D1 in the same
wavelength region. The IPCE for a DSSC can be expressed as
IPCEðkÞ ¼ LHEðkÞ/injgc
where LHE(k) is the light harvesting efficiency of the pho-
toelectrode, /inj is the charge injection efficiency from the
excited state of dye into conduction band of the nano-crys-
electron diffusion coefficient (D) and electron lifetime (s)
in TiO2. The electron lifetime is defined as the average time
talline semiconductor, and gc is the collection efficiency of
spent in the TiO2 electrode. If the electron diffusion length,
the injected electron by the electrode. The parameter
LHE(k) is related to the dye loading on the surface of the
TiO2 photoanode. We have estimated the amount of dye
loading for both devices and found that it is slightly higher
for D1 than D2. In particular, the amount of the adsorbed
dyes (dye loading) on the TiO2 films were estimated by
desorbing the dyes with basic solution, and the surface
concentrations were determined to be 3.21 ꢂ 10ꢀ8 and
3.14 ꢂ 10ꢀ8 mol cmꢀ2 for D1 and D2, respectively. There-
fore, the LHE(k) does not much affect the IPCE. The
improvement in the IPCE for DSSCs based on the D1 may
be mainly due to the other parameters, such as /inj and
which is (Ds)
0.5, is shorter than the thickness of the TiO2
electrode, the electrons in TiO2 are recombined with the
dye cations and/or Iꢀ3 during the diffusion. The high IPCE
of the present DSSCs indicates that the recombination at
short circuit conditions is negligible. At open circuit condi-
tions, the injected electrons in TiO2 are accumulated until
the charge injection rate is balanced with the recombina-
tion rate. The recombination rate depends upon the elec-
tron lifetime in the TiO2 electrode. The electron lifetime
of the electron also depends on the reorganization energy
of dye used in the DSSC [30].
The electrochemical impedance spectra of the DSSCs
based on D1 and D2 were measured to investigate the ki-
netic process about the back reaction. The electrochemical
gc. The parameter /inj depends on the energy difference be-
tween the LUMO of the dye and conduction band of TiO2.
This energy difference is higher for the DSSCs based on
D1 than D2. Therefore, the electron injection efficiency is
higher for solar cells based on D1 than D2 leading to im-
proved IPCE. The loss in the IPCE is probably caused by
the reduced charge collection efficiency. The slightly lower
dye loading favors the increased charge recombination
between free electrons in the TiO2 conduction band and
Fig. 4. J–V characteristics under illumination of DSSCs based on D1 and
D2 dye.
Fig. 3. The IPCE spectra of DSSCs fabricated with D1 and D2.

J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
1247
Table 2
Photovoltaic performance of the quasi solid state DSSCs for the two dyes on
different measured conditions.
Dye Short-circuit
Open circuit
Fill
Power
current (Jsc
)
voltage (Voc
(V)
)
factor
(FF)
conversion
efficiency (
(%)
(mA/cm2)
g)
D1a 11.8
0.65
0.61
0.69
0.65
0.63
0.56
0.66
0.60
4.80
3.50
5.42
4.00
D2a 10.2
D1b 11.9
D2b 10.4
a
With undoped TiO2 photoanode.
With nitrogen doped TiO2 photoanode.
b
impedance spectra of the DSSCs based on D1 and D2 in the
form of Bode plots in dark at forward bias of 0.65 V are pre-
sented in Fig. 5. The peak at intermediate frequencies
(ꢁ100–1 Hz) is mainly attributed to the charge transfer
at the TiO2/electrolyte interface [31]. The electron lifetime
Fig. 6. Comparison of charge recombination rate in DSSCs based on the
D1 and D2.
(
s
) can be calculated by the reciprocal of peak frequency
and subsequent recombination with tri-iodide [33]. The
recombination rate for DSSC based on D2 is about two
times higher than that for DSSC with D1, which can be
attributed to an increased current (dark current) due to
the recombination of electrons in conduction band of
TiO2 with the tri-iodide ions in the electrolyte. The higher
recombination rate also leads to a decrease in the free
charge carriers in the conduction band of TiO2, which in-
duces a downward shift of conduction band of TiO2 to-
wards the iodide /tri-iodide potential. Both the higher
recombination rate and the downward shifted conduction
band in the DSSC based on D2 explain the lower Voc mea-
sured in the J–V characteristics. Additionally, the higher
charge recombination rate reduces the charge collection
efficiency and consequently the Jsc and PCE.
It has been reported that the addition of nanoparticles
such as TiO2 and SiO2 in the electrolyte can increase the io-
nic conductivity [34]. In these nanocomposite electrolytes,
the anion species are aligned around the nanoparticles,
facilitating the electron exchange reaction [33]. It is noted
that the TiO2 nanoparticles with particle size of 5–10 nm
are selected because they are sufficiently smaller that the
pore diameter of the typical nanoporous TiO2 layer (15–
25 nm). In our case the ionic conductivity of electrolyte
with and without TiO2 nanoparticle is 0.11 and
0.18 mS cmꢀ1, respectively. For comparison, we have also
measured the J–V characteristics of the DSSC of D1 dye
based on the quasi solid state electrolyte without TiO2
nanoparticles, under illumination intensity of 100 mW/
cm2 and found that the Jsc and Voc are about 9.6 mA/cm2
and 0.72 V, respectively. This is attributed to the lower
concentration of free tri-iodide ions near the dye attached
to TiO2 surface, because the surfaces of the nanoparticles
can immobilize the ions. This also results in slightly smal-
ler Voc owing to the high electron concentration at TiO2
surface of electrolyte. The charge transport in electrolyte
medium may be effectively facilitated by adding nanopar-
ticles to the polymer electrolyte. The enhancement in the
solar cell performance upon the introduction of the TiO2
nanoparticles in the polymer electrolyte is due to the im-
proved ion conductivity.
(fp) [31]
s
¼ 1=xp ¼ 1=2
pfp
The electron lifetime for the DSSC based on D1 is 25 ms
which is about 1.6 times longer than the DSSC based on D2
with electron lifetime of 15 ms. The open circuit voltage
decay analysis also shows that DSSC based on D1 has a
longer electron lifetime than that of DSSC with D2. The
longer electron lifetime of the DSSC with D1 contributes
to its superior photovoltaic performance.
We measured photovoltage decay of the devices at var-
ious white light intensities in order to study the charge
recombination rates between electrons in TiO2 conduction
band and tri-iodide ions in the electrolyte [32]. The recom-
bination rates are plotted in Fig. 6 versus the open circuit
voltage of the DSSC, adjusted by varying the white light
intensities. The Voc in the cell is given by the difference be-
tween the redox level of the electrolyte and the quasi Fer-
mi level of TiO2. It can be determined by the concentration
of free charge carriers. This plot allows us to compare the
recombination rate at equal charge density concentration
in the TiO2 films. The recombination rate is known to in-
crease exponentially with increasing light intensity due
to a filling of intra-band trap states in the TiO2. This allows
a faster de-trapping of electrons to the conduction band
Fig. 5. The electrochemical impedance spectra of the DSSCs based on D1
and D2 in the form of Bode plots in dark at forward bias of 0.65 V.

1248
J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
3.4. Effect of nitrogen doped TiO2 photoanode
One crucial component in DSSCs is that the nanostruc-
ture TiO2 electrode has some oxygen deficiency in the crys-
tal structure [35]. The oxygen deficiency could create
electron–hole pairs. The oxidizing holes can react with dye
or be scavenged by iodide ions [36]. These processes result
in shortening the electron life time. Consequently doping
with nitrogen into the TiO2 crystal structure can improve
the oxygen deficiency and decrease the back reaction. Asahi
et al. reported at 2001 that photocatalytic activity can be
achieved by nitrogen doping. Considerable research on
nitrogen doping in TiO2 film has been extensively conducted
[37]. Recently, doping of nitrogen into TiO2 for the improve-
ment of DSSCs performance has been reported [38]. We
have attempted to improve the photovoltaic performance
of the DSSCs using nitrogen doped TiO2 photoelectrode.
The nitrogen doped TiO2 materials prepared by modi-
fied sol gel method using urea as nitrogen carrier display
a yellow color and are characterized by X-ray diffraction
(XRD) analysis. XRD spectrum of the nitrogen doped TiO2
powder shows anatase phase. The average crystallite size
calculated from the broadening of the (1 0 1) XRD peak of
the anatase phase was 20.3 and 17.5 nm for doped and un-
doped TiO2, respectively. We have observed a little shift in
the (1 0 1) peak position for nitrogen doped TiO2 which is
consistent with the work reported earlier [39].
Fig. 7a shows the J–V characteristics of the DSSCs based
on the nitrogen doped photoelectrode using D1 and D2.
The photovoltaic performance of the DSSCs is shown in Ta-
ble 2. It is apparent from this table that Voc and FF shift to
higher values with the nitrogen doped TiO2 electrode,
whereas Jsc remains almost unchanged. The increase of Voc
in the presence of nitrogen compared with that in the ab-
sence of nitrogen may be explained by the change in the
quasi Fermi level (Ef) of TiO2 photoelectrode. Using Fermi le-
vel pinning, these two parameters are described by the rela-
tionship [40]:
Fig. 7b. Dark J–V characteristics under illumination of DSSCs based on D1
and D2 using nitrogen doped and undoped TiO2 photoelectrode.
where Ered is the standard reduction potential of redox
couple, assuming that the Ered does not vary with the addi-
tion of nitrogen. Therefore, the increase in Voc of the DSSCs
with the addition of nitrogen in the TiO2 photoelectrode,
may be attributed to the retarding of the dark current, as
shown in Fig. 7b, or to the decrease of the charge recombi-
nation. The addition of nitrogen in the TiO2 can suppress
the production of the dark current. By comparing the dark
current of DSSCs based on nitrogen doped TiO2 and bare
TiO2 photoanode, it is observed that the nitrogen doped
DSSC exhibit smaller dark current. This indicates that the
nitrogen doping could successfully retard the charge
recombination at the TiO2/dye/electrolyte interface. Thus,
the nitrogen doped TiO2 introduced formation of O–Ti–N
structure which can suppress the reduction of Iꢀ3 on the
TiO2 electrode. The onset voltage in the dark current for
DSSC also shifts to higher voltage with nitrogen doped
TiO2 electrode as compared to undoped TiO2 electrode.
This means that Efb of TiO2 shifts towards the vacuum level
with doping, leading to higher value of Voc
.
In the Bode plots of electrochemical impedance spectral
data, it can be seen that adding nitrogen into TiO2 shifts the
middle frequency peak to lower frequency, as shown in
Fig. 8. This corresponds to an increase in the electron life-
time and suppressing the back transfer of photoexcited
electron from nitrogen doped TiO2 nanoparticles to the
electrolyte at TiO2/dye/electrolyte interface. This confirms
that the nitrogen doping into TiO2 photoanode can im-
prove the performance of the DSSC.
Voc ¼ jEf ꢀ Ered
j
Fig. 8. The electrochemical impedance spectra of the DSSCs based on D1
and D2 using nitrogen doped TiO2 electrode in the form of Bode plots in
dark at forward bias of 0.65 V.
Fig. 7a. J–V characteristics under illumination of DSSCs based on D1 and
D2 dye using nitrogen doped TiO2 photoelectrode.

J.A. Mikroyannidis et al. / Organic Electronics 11 (2010) 1242–1249
1249
[12] R. Chen, X. Yang, H. Tian, X. Wang, A. Hagfeldt, L. Sun, Chem. Mater.
19 (2007) 4007.
4. Conclusions
[13] S. Hwang, J.H. Lee, C. Park, H. Lee, C. Kim, C. Park, M.-H. Lee, W. Lee, J.
Park, K. Kim, N.-G. Park, C. Kim, Chem. Commun. 46 (2007) 4887.
[14] K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo, S. Suga, K.
Sayama, H. Sugihara, H. Arakawa, J. Phys. Chem. B 107 (2003) 597.
[15] H. Choi, C. Baik, S.O. Kang, J. Ko, M.S. Kang, Md.K. Nazeeruddin, M.
Grätzel, Angew. Chem., Int. Ed. 47 (2008) 327.
Two new photosensitizers D1 and D2 based on 4-nitro-
-cyanostilbene with carboxy or hydroxyl anchoring group
a
were synthesized in high yields from a one-step reaction
only. The long-wavelength absorption maximum was lo-
cated at 617–655 nm with thin film absorption onset at
761 nm which corresponds to an optical band gap of
1.63 eV. Quasi solid state dye-sensitized solar cells were
fabricated using polymer gel electrolyte and dyes D1 and
D2. In DSSC based on D1, IPCE about 70% at the absorption
maxima was observed with overall PCE of 4.8%. The lower
PCE in the case of D2 is ascribed to the reduction in the
photocurrent due to less dye loading on the TiO2 surface,
since D2 has the OH anchoring group. The lower PCE of
the DSSC based on the D2 has also been ascribed in terms
of increased recombination rate and lower electron life-
time. It was found that the PCE of the DSSCs has been en-
hanced when a nitrogen doped TiO2 photoanode is used.
The increase in PCE is attributed to the retardation of
charge recombination and the introduction of nitrogen,
which replaced the oxygen deficiency in the titania crystal
lattice. The enhanced electron lifetime for doped TiO2
DSSCs could be attributed to the formation of O–Ti–N in
the TiO2 electrode to retard the recombination reaction at
the TiO2/electrolyte interface, as compared to DSSC based
on undoped TiO2 electrode. Additionally, the Efb of TiO2 is
also affected slightly by the presence of nitrogen which is
responsible for the improvement in Voc and PCE.
[16] S. Kim, H. Choi, D. Kim, K. Song, S.O. Kang, J. Ko, Tetrahedron 63
(2007) 9206.
[17] W.-H. Liu, I.-C. Wu, C.-H. Lai, C.-H. Lai, P.-T. Chou, Y.-T. Li, C.-L. Chen,
Y.-Y. Hsub, Y. Chi, Chem. Commun. 41 (2008) 5152.
[18] (a) C.P. Hsieh, H.P. Lu, C.L. Chiu, C.W. Lee, S.H. Chuang, C.L. Mai, W.N.
Yen, S.J. Hsu, E.W.G. Diau, C.Y. Yeh, J. Mater. Chem. 20 (2010) 1134;
(b) J.A. Mikroyannidis, M.M. Stylianakis, P. Suresh, M.S. Roy, G.D.
Sharma, Energy Environ. Sci. 2 (2009) 1301;
(c) J.A. Mikroyannidis, M.M. Stylianakis, M.S. Roy, P. Suresh, G.D.
Sharma, J. Power Sources 194 (2009) 1179;
(d) W. Zhang, Z. Fang, M.J. Su, M. Saeys, B. Liu, Macromol. Rapid
Commun. 30 (2009) 1537;
(e) C.W. Lee, H.P. Lu, C.M. Lan, Y.L. Huang, Y.R. Liang, W.N. Yen, Y.C.
Liu, Y.S. Lin, E.W.G. Diau, C.Y. Yeh, Chem. Eur. J. 15 (2009) 1412;
(f) N. Koumura, Z.-S. Wang, M. Miyashita, Y. Uemura, H. Sekiguchi,
Y. Cui, A. Mori, S. Mori, K. Hara, J. Mater. Chem. 19 (2009) 4829.
[19] (a) Q.H. Yao, F.S. Meng, F.Y. Li, H. Tian, C.H. Huang, J. Mater. Chem.
13 (2003) 048;
(b) M. Xu, S. Wenger, H. Bala, D. Shi, R. Li, T. Zhou, S.M. Zakeeruddin,
M. Grätzel, P. Wang, J. Phys. Chem. C 113 (2009) 2966.
[20] Z.S. Wang, F.Y. Li, C.H. Huang, Chem. Commun. (2000) 2063.
[21] B.C. O’Regan, S. Scully, A.C. Mayer, E. Palomares, J. Durrant, J. Phys.
Chem. B 109 (2005) 4616.
[22] (a) J.A. Mikroyannidis, M.M. Stylianakis, P. Balraju, P. Suresh, G.D.
Sharma, ACS Appl. Mater. Interfaces 1 (8) (2009) 1711;
(b) J.A. Mikroyannidis, M.M. Stylianakis, P. Suresh, P. Balraju, G.D.
Sharma, Org. Electron. 10 (7) (2009) 1320;
(c) J.A. Mikroyannidis, S.S. Sharma, Y.K. Vijay, G.D. Sharma, ACS
Appl. Mater. Interfaces 2 (1) (2010) 270.
[23] J.A. Mikroyannidis, A. Kabanakis, P. Balraju, G.D. Sharma, J. Phys.
Chem. C, submitted for publication.
[24] R. Merck, Bull. Soc. Chim. Belg. 58 (1949) 460.
[25] H.S. Blair, T.-K. Law, Polymer 21 (1980) 1475.
[26] G.R. Robertson, Org. Synth. Coll. 1 (1941) 396, 2 (1922) 57.
[27] (a) N. Cho, H. Choi, D. Kim, K. Song, M.S. Kang, S.O. Kang, J. Ko,
Tetrahedron 65 (2009) 6236;
(b) K. Sayama, K. Hara, N. Mori, M. Satsuki, S. Suga, S. Tsukagoshi, Y.
Abe, H. Sugihara, H. Arakawa, Chem. Commun. (2000) 1173.
[28] H. Tian, X. Yang, J. Cong, R. Chen, C. Teng, J. Liu, Y. Hao, L. Wang, L.
Sun, Dyes Pigm. 84 (2010) 62.
[29] G. Zhang, H. Bala, Y. Cheng, D. Shi, X. Lv, Q. Yu, P. Wang, Chem.
Commun. (2009) 2198.
[30] (a) N. Koumura, Z.S. Wang, S. Mori, M. Miyashita, E. Suzuki, K. Hara,
J. Am. Chem. Soc. 128 (2006) 14256;
References
[1] B. O’Regan, M. Grätzel, Nature 353 (1991) 737.
[2] A. Hagfeldt, M. Grätzel, Acc. Chem. Res. 33 (2000) 269.
[3] M.K. Nazeeruddin, P. Pechy, M. Grätzel, Chem. Commun. (1997)
1705.
[4] A. Hagfeldt, M. Grätzel, Chem. Rev. 95 (1995) 49.
[5] H.J. Snaith, L. Schmidt-Mende, Adv. Mater. 19 (2007) 3187.
[6] (a) Md.K. Nazeeruddin, S.M. Zakeeruddin, R. Humphry-Baker, M.
Jirousek, P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer, M.
Grätzel, Inorg. Chem. 38 (1999) 6298;
(b) M. Miyashita, K. Sunahara, T. Nishikawa, Y. Uemura, N. Koumura,
K. Hara, A. Mori, T. Abe, E. Suzuki, S. Mori, J. Am. Chem. Soc. 130
(2008) 17874.
(b) 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;
(c) P. Wang, C. Klein, R. Humphry-Baker, S.M. Zakeeruddin, M.
Grätzel, J. Am. Chem. Soc. 127 (2005) 808;
(d) L. Schmidt-Mende, J.E. Kroeze, J.R. Durrant, M.K. Nazeeruddin, M.
Grätzel, Nano Lett. 5 (2005) 1315;
(e) C.S. Karthikeyan, H. Wietasch, M. Thelakkat, Adv. Mater. 19
(2007) 1091;
(f) A. Staniszewski, S. Ardo, Y. Sun, F.N. Castellano, G.J. Meyer, J. Am.
Chem. Soc. 130 (2008) 11586;
(g) Y. Cao, Y. Bai, Q. Yu, Y. Cheng, S. Liu, D. Shi, F. Gao, P. Wang, J.
Phys. Chem. C 113 (2009) 6290.
[31] M. Adachi, M. Sakamoto, J. Jin, Y. Ogata, S. Isoda, J. Phys. Chem. B 110
(2006) 13872.
[32] N.W. Duffy, L.M. Peter, K.G.U. Wijayantha, Electrochem. Commun. 2
(2000) 262.
[33] J. Bisquert, V.S. Vikhrenko, J. Phys. Chem. B 108 (2004) 2313.
[34] (a) M. Berginc, M. Hocevar, U.O. Krasovec, A. Hinsch, R. Santrawan,
M. Topic, Thin Solid Films 516 (2008) 4645;
(b) M.S. Kang, K.S. Ahn, J.W. Lee, J. Power Sources 180 (2008) 896.
[35] (a) I. Inakamura, N. Negishi, S. Kutsuma, T. Ihara, S. Sugihara, K.
Takeuchi, J. Mol. Catal. A: Chem. 161 (2000) 205;
(b) H. Irie, Y. Watanabe, K. Hashimoto, J. Phys. Chem. B 107 (2003)
5483.
[36] M. Mrowetz, W. Balcerski, A.J. Colussi, M.R. Hoffmann, J. Phys. Chem.
B 108 (2004) 17269.
[37] (a) R. Asahi, T. Morkawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293
(2001) 269;
[7] M.K. Nazeeruddin, P. Péchy, T. Renouard, S.M. Zakeeruddin, R.
Humphry- Baker, P. Comte, P. Liska, Le Cevey, E. Costa, V. Shklover,
L. Spiccia, G.B. Deacon, C.A. Bignozzi, M. Grätzel, J. Am. Chem. Soc.
123 (2001) 1613.
[8] S. Ito, S.M. Zakeeruddin, R. Humphry-Baker, P. Liska, R. Charvet, P.
Comte, Md.K. Nazeeruddin, P. Péchy, M. Takata, H. Miura, S. Uchida,
M. Grätzel, Adv. Mater. 18 (2006) 1202.
[9] K. Hara, Z.-S. Wang, T. Sato, A. Furube, R. Katoh, H. Sugihara, Y. Dan-
oh, C. Kasada, A. Shinpo, S. Suga, J. Phys. Chem. B 109 (2005) 15476.
[10] Z.-S. Wang, N. Koumura, Y. Cui, M. Takahashi, H. Sekiguchi, A. Mori,
T. Kubo, A. Furube, K. Hara, Chem. Mater. 20 (2008) 3993.
[11] S. Kim, J.K. Lee, S.O. Kang, J. Ko, J.H. Yum, S. Fantacci, F. De Angelis, D.
Di Censo, Md.K. Nazeeruddin, M. Grätzel, J. Am. Chem. Soc. 128
(2006) 16701.
(b) T. Lindgren, J.M. Mwabora, E.D. Avendano Soto, J. Jonsson, A.
Hoel, C.G. Granqvist, S.E. Lindquist, J. Phys. Chem. B 107 (2003) 5709;
(c) S. Sakthivel, H. Kisch, Chem. Phys. Chem. 4 (2003) 487.
[38] T. Ma, M. Akiyama, E. Abe, I. Imai, Nano Lett. 5 (2005) 2543.
[39] T.C. Jagadake, S.P. Takale, R.S. Sonawane, H.M. Joshi, S.I. Patil, B.B.
S.B. Ogale, J. Phys. Chem. C 112 (2008) 14595.
Kale,
[40] T.S. Kang, K.H. Chun, J.S. Hong, S.H. Moon, K. Kim, J. Electrochem. Soc.
147 (2000) 3049.