Substituent effect in direct ring functionalized squaraine dyes on near infra-red sensitization of nanocrystalline TiO2 for molecular photovoltaics

Article abstract of DOI:10.1016/j.jphotochem.2010.07.010

We have synthesized a series of novel blue colored symmetrical squaraine sensitizers with variable alkyl chain length for their application towards the fabrication of dye-sensitized solar cells (DSSCs). It has been found that an increase in the alkyl chain length substituted at N-position of indole ring exhibits enhanced electron diffusion length and electron life-time resulting in better passivation of nanocrystalline TiO₂ surface leading to enhancement in the cell performance. Based on HOMO and LUMO energy measurement of squaraine dyes, it has been demonstrated that about 0.16 eV energy barrier is sufficient for electron injection from LUMO of dye to TiO₂ conduction band and dye regeneration after photo-excitation. Performances of DSSCs using model squaraine dyes indicate that dodecyl alkyl substituent is optimum giving highest Voc and use of chenodeoxycholic acid along with the dye SQ-4 shows a photoconversion efficiency of 3.5% under AM 1.5 irradiation.

Full text of DOI:10.1016/j.jphotochem.2010.07.010

Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Substituent effect in direct ring functionalized squaraine dyes on near infra-red
sensitization of nanocrystalline TiO2 for molecular photovoltaics
Shyam S. Pandeya,∗, Takafumi Inouea, Naotaka Fujikawaa, Yoshihiro Yamaguchib, Shuji Hayasea
a Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu, Kitakyushu 808-1096, Japan
b Nippon Steel Chemical Company Limited, Nakabaru, Tobata, Kitakyushu, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We have synthesized a series of novel blue colored symmetrical squaraine sensitizers with variable alkyl
chain length for their application towards the fabrication of dye-sensitized solar cells (DSSCs). It has been
found that an increase in the alkyl chain length substituted at N-position of indole ring exhibits enhanced
electron diffusion length and electron life-time resulting in better passivation of nanocrystalline TiO2 sur-
face leading to enhancement in the cell performance. Based on HOMO and LUMO energy measurement
of squaraine dyes, it has been demonstrated that about 0.16 eV energy barrier is sufficient for electron
injection from LUMO of dye to TiO2 conduction band and dye regeneration after photo-excitation. Perfor-
mances of DSSCs using model squaraine dyes indicate that dodecyl alkyl substituent is optimum giving
highest Voc and use of chenodeoxycholic acid along with the dye SQ-4 shows a photoconversion efficiency
of 3.5% under AM 1.5 irradiation.
Received 9 March 2010
Received in revised form 3 June 2010
Accepted 8 July 2010
Available online 16 July 2010
Keywords:
Dye-sensitized solar cells
Aggregation
Surface passivation
Squarine dyes
© 2010 Elsevier B.V. All rights reserved.
Nanocrystalline TiO2
1. Introduction
judicious selection of aromatic donor moieties with extended ␲-
conjugation surrounding the electron deficient squaric acid central
A new paradigm in harnessing the immense solar energy driven
by cast and quality has attracted the attention of material scien-
tists in the recent past. Report of conversion efficiency over 10%
by ruthenium based dye-sensitized solar cells (DSSCs) has brought
up new momentum in the energy research [1,2]. A perusal of
absorption spectrum of ruthenium sensitizers and solar spectrum
clearly indicates a need for novel sensitizers absorbing in near
infra-red (NIR) wavelength region for further enhancement in the
photoconversion efficiency. Recently, there are reports about the
extension of optical absorption window associated with photon
harvesting in the wide wavelength region by selective adsorption
of two dyes in dye double architecture leading to the enhance-
ment in the conversion efficiency [3,4]. Our approach is to develop
NIR sensitizers and fabricate the efficient DSSCs by combining
them with ruthenium based sensitizers in dye double layer archi-
tecture. In this regard, squaric acid based dyes are one of the
potential candidates owing to their intense sharp absorption in
NIR region, high molar extinction coefficient and their stability
along with application as sensitizers in the xerography and opti-
cal data storage [5,6]. At the same time, optical absorption of such
dyes are possible to tailor in extended wavelength region by the
core [7].
Yum et al. [8] have reported an unsymmetrical squaraine dye
bearing carboxylic anchoring group directly substituted in the aro-
matic ring giving photoconversion efficiency of 4.5% with photon
harvesting up to 700 nm. Such an efficiency given by squaraine sen-
sitizer broke the myth about the poor power conversion efficiency
of squaraine dyes reported by various research groups [9–11]. A
perusal of squaraine based sensitizers bearing carboxyl anchor-
ing group clearly corroborates that dyes bearing carboxylic group
directly substituted to aromatic ring are superior in performance
as compared to that of their alkyl side chain carboxy substituted
counterparts. This has been explained by the strong conjugation
across the chromophoric main body and anchoring group leading
to good electronic coupling between the lowest unoccupied molec-
ular orbital (LUMO) of the sensitizer and conduction band of TiO2.
This work ignited the more intense research for this class of dyes
leading towards the further improvement in efficiency by fabrica-
tion of hybrid DSSCs using squaraine dyes with other sensitizers
[12–14].
Amongst the squaric acid based dyes bearing ring substituted
carboxyindoles, there are scattered reports about the selection and
suitability of the alkyl substituents at N-position of the indole ring.
Although, Yum et al. [8] have used octyl indole while Chen et al.
[15] have used the butyl indole. There is no discussion about the
reason for the selection of a particular alkyl group substituted at
∗
Corresponding author. Tel.: +81 93 695 6044; fax: +81 93 695 6005.
E-mail address: shyam@life.kyutech.ac.jp (S.S. Pandey).
1010-6030/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotochem.2010.07.010

270
S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
N-position of the indole ring in their investigation. In this regard,
Sayama et al. [16] have although advocated the importance of long
alkyl substituents towards the enhancement of photoconversion
efficiency but their report was mainly focused on merocyanine dyes
absorbing effectively in lower wavelength region. In the present
investigation, we would like to report our systematic study of role
of substituents on the sensitization behavior of model squaraine
dyes bearing variable alkyl chain length as well as fluoroalkyl sub-
stituent. Results of the cell performance have been elucidated using
dark current–voltage (I–V) characteristics, extent of dye adsorp-
tion, dye surface potential, electron life-time and electron diffusion
length measurements. At the same time novelty also lies in the fact
that role of substituent on controlling the HOMO and LUMO of the
squaraine dyes which is helpful for the design and development of
novel NIR sensitizers.
2.2. Synthesis of SQ dyes and dye intermediates
Aromatic ring carboxy functionalized indole derivative 2,3,3-
trimethyl-3H-indole-5-carboxylic acid was synthesized following
the methodology reported by Pham et al. [20]. Symmetrical
SQ dyes and corresponding dye intermediates of 5-carboxy-
2,3,3-trimethyl-indole have been synthesized following the
methodology as shown in Scheme 1.
2.2.1. Synthesis of 2,3,3-trimethyl-3H-indole-5-carboxylic acid
[1]
In a round bottom flask fitted with condenser and N2 purg-
ing, 4-hydrazinobenzoic acid (5.0 g; 32.85 mmol), glacial acetic acid
(80 ml) + sodium acetate (5.5 g; 67 mmol) + 3-methyl-2-butanone
(4.45 g; 51.5 mmol) were added. Reaction mixture was refluxed at
120 ◦C for 8 h leading to brown suspension. Acetic acid was evap-
orated followed by addition of 9:1 water methanol mixture on
ice-bath leading to precipitation. Residue was filtered and dried giv-
ing 3.7 g of titled compound as off white powder in 56% yield. HPLC
analysis of product suggests that compound was 100% pure. FAB-
mass (measured 203; calculated 203.09) and 1H NMR (500 MHz,
CDCl3): d/ppm = 7.99 (s, H-4), 7.93 (d, J = 8.0 Hz, H-6), 7.59 (d,
J = 8.0 Hz, H-7), 2.26 (s, 3H, H-10), 1.28 (s, 6H, H11 + 12) verifies the
successful synthesis of the compound.
2. Experimental
2.1. Materials, instruments and methods
1-Iodoethane (2), 1-iodobutane (3), 1-iodooctane (4), 1-
iodododecane (5), 1-iodooctadecane (6) and 1,1,1-trifluoro-4-
iodobutane (7) used in the present synthesis were purchased
from Tokyo Kasei Co. Ltd. Solvents (reagent grade, Wako
Chemical Company) and squaric acid were purchased from
Alfa Aesar and used as received. Synthesized squaraine (SQ)
dyes and dye intermediates were analyzed by high perfor-
mance liquid chromatography (JASCO) equipped with ODS
2.2.2. Synthesis of
5-carboxy-2,3,3-trimethyl-1-alkyl-3H-indolium iodide [8–12]
2,3,3-trimethyl-3H-indole-5-carboxylic acid (1, 1 equiv.) and
1-iodoalkane (2–7, 3 equiv.) were dissolved in dehydrated ace-
tonitrile and reaction mixture was refluxed for (2: 24 h, 3: 48 h;
4: 72 h; 5: 96 h and 6: 144 h) under nitrogen atmosphere to give
corresponding 5-carboxy-N-alkyl-indoium iodides (8–12). In the
case of 7, reaction was carried at 130 ◦C for 24 h using propionitrile
solvent. After completion of the reaction as monitored by HPLC,
solvent was evaporated and the crude product was washed with
ample diethyl ether giving the titled compound. The physical and
spectral data of N-alkyl-indolium iodides (8–12) are as follows.
analytical column (CD-C18,
␾
4.6 mm × 150 mm) and UV as
well as multi channel photodiode array detector for mon-
itoring the reaction progress and final purity of the com-
pound.
Mass of the intermediates as well as final SQ dyes was con-
firmed by MALDI-TOF-mass (Applied Biosystems) or fast ion
bombardment (FAB) mass in positive ion monitoring mode. For
final SQ dyes, high resolution FAB-mass (HR-MS) in positive ion
monitoring mode was also measured. Nuclear magnetic reso-
nance (NMR) spectra were recorded on a JEOL JNM A500 MHz
spectrometer in CDCl3 or d6-DMSO with reference to TMS for
structural elucidation. Electronic absorption spectroscopic inves-
tigations in solution and thin film adsorbed on TiO2 surface were
conducted using UV–vis spectrophotometer (JASCO V 550). Sur-
face potential measurement of SQ dyes was performed using
scanning Kelvin probe microscope (SKPM model JSPM 5200,
JEOL Datum). Surface potential of bare TiO2 surface was first
measured and its value was taken as reference followed by
measurement of surface potential of dye adsorbed on TiO2
film.
2.2.2.1. 5-Carboxy-2,3,3-trimethyl-1-ethyl-3H-indolium iodide (8).
Yield 79% having 98% purity as confirmed by HPLC. FAB-mass (mea-
sured 232.0; calculated 232.13) and 1H NMR (500 MHz, d6 DMSO):
d/ppm = 8.40 (s, H-4), 8.19 (d, J = 8.0 Hz, H-6), 8.08 (d, H = 8.0 Hz, H-
7), 4.52 (q, 2H, H-13), 2.87 (s, 3H, H-10), 1.28 (s, 6H, H11 + 12), 1.09
(t, 3H, H-14) confirms the identity of the compound.
2.2.2.2. 5-Carboxy-2,3,3-trimethyl-1-butyl-3H-indolium iodide (9).
Yield 77% having 97% purity as confirmed by HPLC. FAB-mass (mea-
sured 260.0; calculated 260.16) confirms successful synthesis of the
compound.
Electron diffusion length (Ln) and electron life-time (tn) of
SQ dyes adsorbed on TiO2 layer were estimated with intensity
modulated photocurrent/photovoltage spectroscopic (IMPS/IMVS)
measurements [17–19]. The amount of dye molecules adsorbed
on TiO2 layers was measured spectrophotometrically after des-
orption of adsorbed dye molecules using NaOH aqueous solution
and standard calibration curve of respective SQ dye. Optical
absorption at ꢀmax was used for the standard calibration curve
and calculation of the number of dye molecules quantitatively.
Highest occupied molecular orbital (HOMO) energy level of the
squaraine dyes used in the present investigation has been deter-
mined by photoelectron spectroscopy in air (PESA model AC3)
from Riken Keiki Co. Ltd. Japan while their lowest unoccupied
molecular orbital (LUMO) energy level was determined from
the edge of optical absorption using electronic absorption spec-
troscopy.
2.2.2.3. 5-Carboxy-2,3,3-trimethyl-1-octyl-3H-indolium iodide (10).
Yield 72% having 96% purity as confirmed by HPLC. FAB-mass
(measured m/z: 316.0; calculated m/z: 316.23) confirms successful
synthesis of the compound.
2.2.2.4. 5-Carboxy-2,3,3-trimethyl-1-dodecyl-3H-indolium
iodide
(11). Yield 66% having 98% purity as confirmed by HPLC. FAB-mass
(measured 372.0; calculated 372.29) confirms the identity of the
compound.
2.2.2.5. 5-Carboxy-2,3,3-trimethyl-1-octadecyl-3H-indolium iodide
(12). Yield 56% having 98% purity as confirmed by HPLC. MALDI-
TOF-mass (measured 456.69; calculated 456.38) confirms the
successful synthesis of the compound.

S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
271
Scheme 1. Synthesis of symmetrical squaraine dyes.
2.2.2.6. 5-Carboxy-2,3,3-trimethyl-1-Trifluorobutyl-3H-indolium
iodide (13). Yield 46% having 99% purity as confirmed by HPLC.
FAB-mass (measured 441.0; calculated 441.04) confirms the
successful synthesis of the compound.
H-19), 0.83 (t, 3H, H-20) confirms the successful synthesis of the
dye SQ-3.
2.2.3.4. N-dodecyl substituted squarine dye SQ-4 (17). Yield 55%
and HPLC purity 98%. MALDI-TOF-mass (calculated 820.54 and
observed 821.84 [M+H]+). HR-MS (calculated 820.539 and observed
820.537 [M]+).
2.2.3. Synthesis of symmetrical squarine dyes [14–19]
Symmetrical SQ Dyes (SQ-1–6) were synthesized using corre-
sponding carboxy functionalized trimethyl-indolium iodide salt
8–13 ((2 equiv.) and squaric acid (1 equiv.) in 1-butanol:toluene
mixture (1:1, v/v). Reaction mixture was refluxed for 18 h using
Dean–Stark trap for azeotropic removal of water. After com-
pletion of reaction, reaction mixture was cooled, solvent was
evaporated and product was purified by silica gel column chro-
matography using chloroform: methanol as eluting solvent. The
physical and spectroscopic data of symmetrical SQ dyes are as fol-
lows;
2.2.3.5. N-octadecyl substituted squarine dye SQ-5 (18). Yield 64%
and HPLC purity 98%. MALDI-TOF-mass (calculated 988.73 and
observed 989.40 [M+H]+). HR-MS (calculated 988.727 and observed
989.0000 [M]+).
2.2.3.6. N-Trifluorobutyl substituted squarine dye SQ-6 (19). Yield
71% and HPLC purity 99%. FAB-mass (calculated 704.23 and
observed 705.0 [M+H]+) confirms the identity of the compound.
2.2.3.1. N-ethyl substituted squarine dye SQ-1 (14). Yield 58%
and HPLC purity 98%. MALDI-TOF-mass (calculated 540.22 and
observed 541.49 [M+H]+). HR-MS (calculated 540.226 and observed
540.223 [M]+). 1H NMR (500 MHz, d6-DMSO): d/ppm = 8.04 (dd, H-
6), 7.98 (dd, H-4), 7.42 (dd, H-7), 5.89 (s, H-10), 4.17 (q, 2H, H-13),
1.71 (s, 6H, H11 + 12), 1.30 (t, 3H, H-14) confirms the successful
synthesis of the dye SQ-1.
2.3. DSSC fabrication and measurement of solar cell performance
DSSC were fabricated using Ti-Nanoxide D paste (Solaronix
SA) which was coated on a Low E glass (Nippon Sheet Glass
Co., Ltd.) by
a doctor-blade. The substrate was then baked
at 450 ◦C to fabricate TiO2 layers of about 12 ␮m thickness.
The substrate was dipped in the solution containing dye in
the presence and/or absence of CDCA. The dye concentra-
tion was fixed to be 0.25 mM while CDCA concentration was
2.5 mM. A Pt sputtered SnO2/F glass substrate was employed
as the counter electrode. Electrolyte (WWS30) containing LiI
(500 mM), iodine (50 mM), t-butylpyridine (580 mM), MeEtIm-
DCA (ethylmethylimidazolium dicyanoimide) (4:6 w/w) (600 mM)
in acetonitrile, was used to fabricate the DSSC. A Himilan film
(Mitsui-DuPont Polychemical Co., Ltd.) of 25 ␮m thickness was
used as a spacer. The cell area was 0.25 cm2. Solar cells were
evaluated by using a photo-mask on the solar cell in order to
remove the effect of the optical reflection under the irradia-
tion of 100 mW/cm2 at AM 1.5. No special efforts were made
towards cell optimization such as, electrode (TiCl4 treatment, use
of scattering layer) and at electrolyte level in order to attain
the maximum possible efficiency. Photoconversion efficiencies
were calculated without the correction incorporated pertaining
2.2.3.2. N-butyl substituted squarine dye SQ-2 (15). Yield 64%
and HPLC purity 98%. MALDI-TOF-mass (calculated 596.29 and
observed 597.25 [M+H]+). HR-MS (calculated 596.288 and observed
596.296 [M]+). 1H NMR (500 MHz, d6-DMSO): d/ppm = 12.85 (b,
–COOH), 8.04 (dd, H-6), 7.96 (dd, H-4), 7.43 (dd, H-7), 5.90 (s, H-10),
4.13 (q, 2H, H-13), 1.76 (m, 2H, H-14), 1.70 (s, 6H, H11 + 12), 1.40
(m, 2H, H-15), 0.95 (t, 3H, H-16) confirms the successful synthesis
of the dye SQ-2.
2.2.3.3. N-octyl substituted squarine dye SQ-3 (16). Yield 46%
and HPLC purity 97%. MALDI-TOF-mass (calculated 708.41 and
observed 708.48 [M]+). HR-MS (calculated 708.413 and observed
708.412 [M]+). 1H NMR (500 MHz, d6-DMSO): d/ppm = 12.84 (b,
–COOH), 8.04 (dd, H-6), 7.96 (dd, H-4), 7.43 (dd, H-7), 5.90 (s, H-10),
4.1 (q, 2H, H-13), 1.70 (s, 6H, H11 + 12), 1.40–1.20 (m, 12H, H-14 to

272
S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
Fig. 2. Photocurrent–voltage characteristics for the DSSC based on squaraine dyes
bearing variable alkyl chains (C2–C18).
porting information) clearly corroborates that efficiency increases
with the increasing alkyl chain length. This increase in the effi-
ciency was associated with increase in both of the Voc as well as Jsc
as a function of alkyl chain length. Although variation of alkyl chain
length was lacking, Koumura et al. have also advocated the impor-
tance of the presence of long alkyl chain in their MK dyes towards
efficiency enhancement [22]. They advocated that long alkyl chain
not only prevent the approach of the acceptor molecules to the TiO2
surface but suppress the dye aggregation also, making favorable
condition for electron transport.
Fig. 1. Absorption spectra of symmetrical squaraines in DMF solution and thin film
adsorbed on nanoporous TiO2 (4 ␮m).
to the light absorption and reflection caused by conducting glass
substrate.
3. Results and discussion
3.1. Electronic absorption of squaraine dyes
UV–vis absorption spectra of squarine dyes in DMF solution as
shown in Fig. 1 exhibit a sharp absorption peak at 640–643 nm
associated with ␲–␲* electronic transition which undergo spectral
broadening and red shift upon adsorption onto thin TiO2 film and
could be attributed to the interaction between carboxyl functional-
ity of squaraine dyes with the titania surface. It is interesting to note
that spectral broadening and red shift increases with the increas-
ing alkyl chain length. This can probably be explained by facile
J-aggregate formation by squaraine dyes bearing longer alkyl chains
due to improved self-aggregation. Similar kind of spectral broaden-
ing and red shift in the edge of IPCE as a function of increase in the
alkyl chain length have also been observed by Sayama et al. [16].
It is noteworthy to see that SQ-1 having ethyl substation shows
the pronounced blue shift in the absorption spectra along with rel-
atively enhanced lower wavelength shoulder peak and could be
attributed to enhanced H-aggregate formation. Kim et al. [21] have
also observed the formation of blue shifted H-aggregates formed by
squarine dyes on SnO2 surface absorbing in the lower wavelength
region. This lower wavelength absorption is much clearly evident
in the IPCE spectrum of SQ-1 bearing ethyl substituent and is neg-
ligible for squaraine dyes having butyl or higher alkyl substituent.
This clearly indicates that at least butyl substitution of the carboxy
functionalized indole ring at N-position is necessary to suppress
the dye aggregation. DFT calculations performed on squaraine dye
bearing carboxyl group directly attached to chromophore indicates
that HOMO is mainly centralized over squaric acid core while there
is sufficient diversion of LUMO over carboxyl moiety along with
the indole ring (see Fig. S1) suggesting possibility of facile electron
injection from photoexcited dye to TiO2 conduction band.
3.2.1. Elucidation of Voc increase as a function of alkyl chain
length
Enhancement of the Voc as a function of alkyl chain length
in these squaraine dyes was elucidated by dark current–voltage
(I–V) characteristics, electron life-time and surface potential mea-
surements (see Figs. S2–S4 in supporting information). Shift of the
onset of dark currents towards higher voltage indicates the sup-
pression of recombination leading to improved Voc for squaraine
dyes bearing longer alkyl chains. This was further confirmed by
the electron life-time measurement, which increases with increas-
ing alkyl chain length. Using different kind of organic sensitizers
Miyashita et al. [23] have also emphasized that organic dye hav-
ing longer alkyl chain exhibits the increased electron life-time.
Recently, Sakaguchi et al. [24] have reported the increase in the
Voc with the increase of surface potential for different sensitizer,
which was reasoned with the upward shift of the conduction band
edge of TiO2. Surface potential of squaraine dyes was also found
to increase with the increasing alkyl chain length (see Fig. S4) and
could be, therefore, attributed to the increase in Voc. Fig. 3 exhibits a
correlation between electron life-time and Voc for model squaraine
dyes and clearly indicates that squaraine dye bearing longer alkyl
chain exhibits higher Voc due to longer electron life-time. It also
suggests that dodecyl alkyl chain length is optimum for attaining
the highest observed Voc.
3.2.2. Elucidation of Jsc improvement as a function of alkyl chain
length
Enhancement of Jsc was elucidated with incident photon to
electron conversion efficiency (IPCE), electron diffusion length
(Fig. S5) and extent of dye adsorption on the TiO2 surface (Fig. S6).
With the increasing alkyl chain length, spectral broadening of
IPCE along with its increase can be attributed to the enhanced
Jsc as a function of increasing alkyl chain length as shown
in Fig. 4. Increase in the Jsc was also found to be in accor-
dance with the increased electron diffusion length in the case
3.2. Effect of alkyl chain length of squaraine dyes on DSSC
performance
A perusal of the photovoltaic performance of DSSC based on
squaraine dyes as shown in Fig. 2 and Table 1 (for table see the sup-

S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
273
Fig. 3. Relationship between measured electron life time and observed Voc for DSSC
based on squaraine dyes having variable alkyl chain length.
Fig. 6. Effect of CDCA addition on the photovoltaic performance of SQ-4. Thin solid
line (action spectra) while line with symbol represents the I–V curve.
length which ultimately results in to enhanced conversion effi-
ciency.
3.3. Effect of chenodeoxycholic acid as co-adsorbent on DSSC
performance
Chenodeoxycholic acid (CDCA) has been widely employed as
co-adsorbent along with the sensitizers for the enhancement of
efficiency of DSSC owing to its ability to suppress the electron
recombination by disrupting the ␲-aggregates usually formed by
squaraine and other organic dyes[24,26]. We employed the CDCA
with one of our squaraine dye SQ-4 to fabricate DSSC which results
into pronounced improvement in both of the IPCE and efficiency
as shown in Fig. 6. In the presence of CDCA, IPCE of SQ-4 shows
an enhancement from 27% (without CDCA) to 51% having onset of
IPCE edge at about 745 nm. In an attempt towards the search for
efficient NIR dye, Cid et al. [14] have reported novel unsymmetri-
cal phthalocyanine dye having onset of IPCE edge of about 725 nm
giving power conversion efficiency of 3.5%. SQ-4 in the presence
of CDCA gave a Jsc of 8.12 mA/cm2, Voc of 0.63 and a fill factor
(ff) of 0.69 corresponding to overall power conversion efficiency of
3.53% under global AM 1.5 condition. We would like to mention
here that using our model squarine sensitizers, DSSC performance
have not yet been optimized in terms TiO2 surface treatment using
TiCl4, most suitable electrolyte and use of scattering layer on to the
nanoporous TiO2 layer. Such optimizations are certainly expected
to lead in to further enhancement in power conversion efficiency.
Fig. 4. Photocurrent action spectra of monochromatic incident photon-to-current
conversion efficiency for DSSC based on squaraine dyes.
3.4. Effect of substituent on HOMO and LUMO: implication on
new dye design
Fig. 5. Correlation between measured electron diffusion length and Jsc for DSSC
based on alkyl substituted model squaraine dyes.
To design a novel sensitizer for DSSC a judicious control
of energy level of HOMO and LUMO of dye with respect to
the energy level of I3−/I− redox couple and conduction band
of TiO2, respectively, is highly desired. Suitability of a partic-
ular dye for NIR sensitization depends on the minimization of
energy gap for electron injection to TiO2 and energy required
for dye regeneration. We have also synthesized a new sym-
metrical squaraine dye SQ-6 having trifluoro-butyl substitution
(Fluoro substituted analogue of SQ-2). DSSC fabricated with
this dye gave a Jsc of 6.63 mA/cm2, Voc of 0.57 and a fill
factor (ff) of 0.70 corresponding to the overall power conver-
sion efficiency of 2.65% under global AM 1.5 condition (see
of squaraines bearing longer alkyl chains. Recently, Ogomi et al.
[25] have demonstrated that increase in the extent of adsorbed
dyes on TiO2 surface leads to enhanced surface trap passivation
resulting in to increased Jsc. It is interesting to note that the
extent of the dye adsorbed on TiO2 surface increases with the
increasing alkyl chain length which supports the increased Jsc for
squaraine dyes having longer alkyl chain length. A perusal of Fig. 5
clearly corroborates that squaraine dyes bearing longer alkyl chain
length exhibits increased Jsc owing to longer electron diffusion

274
S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
and I3−/I− redox system, respectively. Taking this into consider-
ation it is, therefore, possible to design a dye sensitizer having
an energy gap of about 1.3 eV [0.9 eV (TiO2-redox couple energy
gap) + 0.4 eV (energy barrier for electron injection and dye regen-
eration)] i.e. photon harvesting up to 950 nm. Thus it possible to
design a novel NIR sensitizer by judicious selection of aromatic moi-
eties surrounding the squaraine core while control of HOMO and
LUMO energy level can be made by suitable alkyl or fluoro alkyl
substituents.
4. Conclusion
In conclusion, role of alkyl chain length towards the enhance-
ment of photoconversion efficiency of DSSC based on NIR
wavelength absorbing squaraine dyes has been demonstrated. Our
results indicate that alkyl chain length more than butyl is necessary
to prevent the dye aggregation and dodecyl alkyl substituent was
found to be optimum giving highest Voc. We would like to empha-
size that model squaraine dyes having longer alkyl chain exhibit
the enhanced adsorption of dyes on TiO2 surface leading to bet-
ter passivation of surface traps. Enhanced surface passivation leads
to increase in the electron life-time and electron diffusion length
resulting in to increase in the observed Voc and Jsc, respectively.
Measurement of HOMO and LUMO of model squaraine dyes used
in the present investigation reveal that it possible to design novel
squaraine dyes having photon harvesting up to 950 nm by judicious
selection of alkyl and fluoroalkyl substituents along with aromatic
moieties around squaric acid core.
Fig. 7. HOMO and LUMO energy level of squaraine dyes used in the present inves-
tigation. Values in parentheses represent the maximum IPCE for the DSSC based on
dyes under investigation.
Fig. S6 for I–V characteristics and IPCE in the supporting informa-
tion). Redox potential of I3−/I− redox couple has been reported to
be 0.44 V vs. NHE [27] which can be translated into −4.9 eV with
respected to the vacuum level as shown in Fig. 7 by dotted line.
Conduction band (CB) of TiO2 was considered as –4.0 eV consid-
ering the most negative quasi Fermi level corresponds to the flat
band potential of TiO2 (0.7 V vs. SCE) as reported by Ogomi et al.
[28]. For the normal DSSC operation, there should be finite energy
gap between the LUMO of dye and CB of TiO2 for electron injection
as well as HOMO of the dye and redox potential of I3−/I− for dye
regeneration after photo-excitation of the dye. Information about
this energy barrier is expected to help for the design of novel sen-
sitizers having absorption in longer wavelength region. The HOMO
and LUMO energy levels of SQ dyes used in the present investigation
have been shown in Fig. 7 which clearly indicates that these energy
level for symmetrical squaraine dyes increase with the increasing
alkyl chain length up to octyl substituent while fluorine substitu-
tion leads to a drastic decrease in the energy of both of the energy
level. The HOMO energy level of SQ-3 having octyl chain substituent
has been estimated to be −5.06 eV using AC3 measurement which
is about 0.16 eV below the energy level of the redox couple. IPCE
for DSSC based on this dye at its absorption maximum (54%) indi-
cates that even at this energy barrier dye is getting successfully
regenerated after photo-excitation.
LUMO level of the SQ dyes as shown in Fig. 7 was estimated
by the relation ELUMO = EHOMO − Eg, where, ELUMO, EHOMO and Eg
are LUMO energy level, HOMO energy level and band gap of the
dye molecule under investigation, respectively. Eg was estimated
from the optical absorption edge using the electronic absorption
spectroscopy. As indicated from the figure that lowest LUMO level
(−3.84 eV) was observed for SQ-6 having the Trifluorobutyl sub-
stituent. Observation of the LUMO energy level of SQ-6 above
0.16 eV from the CB of TiO2 and observation of 41% IPCE at absorp-
tion maximum of the dye suggests that even at such an energy
barrier electron injection from the photoexcited dye to the CB of
TiO2 is possible. Therefore, maximum HOMO and minimum LUMO
energy level of SQ-3 and SQ-6 amongst the dyes used in the present
work, clearly corroborates that an energy barrier of about 0.16 eV
seems to be sufficient for dye regeneration after photo excitation
and electron injection to TiO2 CB in DSSC based on TiO2 anode
Acknowledgement
Authors would like to acknowledge the financial support from
the New Energy Industrial Technology Development Organization
(NEDO), Japan
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jphotochem.2010.07.010.
References
[1] M.K. Nazeeruddin, F. De Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito,
B. Takeru, M. Gratzel, Combined experimental and DFT-TDDFT computational
study of photoelectrochemical cell ruthenium sensitizers, J. Am. Chem. Soc. 127
(2005) 16835–16847.
[2] M. Gratzel, Solar energy conversion by dye-sensitized photovoltaic cells, Inorg.
Chem. 44 (2005) 6841–6851.
[3] Y. Noma, K. Iizuka, Y. Ogomi, S.S. Pandey, S. Hayase, Preparation of double
dye-layer structure of dye-sensitized solar cells from cocktail solutions for
harvesting light in wide range of wavelengths, Jpn. J. Appl. Phys. 48 (2009)
020213.
[4] F. Inakazu, Y. Noma, Y. Ogomi, S. Hayase, Dye-sensitized solar cells consisting of
dye-bilayer structure stained with two dyes for harvesting light of wide range
of wavelength, Appl. Phys. Lett. 93 (2008) 93304.
[5] K.Y. Law, F.C. Bailey, Squarine chemistry-synthesis, characterization and optical
properties of a class of novel unsymmetrical squarines—4-(dimethylamino)-
phenyl (4ꢀ-methoxyphenyl)-squarine and its derivatives, J. Org. Chem. 57
(1992) 3278–3286.
[6] K.Y. Law, Organic photoconductive materials—recent trends and develop-
ments, Chem. Rev. 93 (1993) 449–486.
[7] S. Sreejith, P. Carol, P. Chithra, A. Ajayaghosh, Squaraine dyes: a mine of molec-
ular materials, J. Mater. Chem. 18 (2008) 264–274.
[8] J.H. Yum, P. Walter, S. Huber, D. Rentsch, T. Geiger, F. Nuesch, F. De Angelis,
M. Gratzel, M.K. Nazeeruddin, Efficient far red sensitization of nanocrystalline
TiO2 films by an unsymmetrical squaraine dye, J. Am. Chem. Soc. 129 (2007)
10320–10321.
[9] N. Takeda, B.A. Parkinson, Adsorption morphology, light absorption, and sensi-
tization yields for squaraine dyes on SnS2 surfaces, J. Am. Chem. Soc. 125 (2003)
5559–5571.
[10] S. Alex, U. Santhosh, S. Das, Dye sensitization of nanocrystalline TiO2: enhanced
efficiency of unsymmetrical versus symmetrical squaraine dyes, J. Photochem.
Photobiol. A 172 (2005) 63–71.

S.S. Pandey et al. / Journal of Photochemistry and Photobiology A: Chemistry 214 (2010) 269–275
275
[11] X. Chen, J. Guo, X. Peng, M. Guo, Y. Xu, L. Shi, S.C. Liang, L. Wang, Y. Guo, S.
Sun, S. Cai, Novel cyanine dyes with different methine chains as sensitizers for
nanocrystalline solar cell, J. Photochem. Photobiol. A 171 (2005) 231–236.
[12] J.H. Yum, S.R. Jang, P. Walter, T. Geiger, F. Neusch, S. Kim, J. Ko, M. Gratzel, M.K.
Nazeeruddin, Efficient co-sensitization of nanocrystalline TiO2 films by organic
sensitizers, Chem. Commun. (2007) 4680–4682.
[13] H. Choi, S. Kim, S.O. Kang, J. Ko, M.S. Kang, J.N. Clifford, A. Forneli, E. Palomares,
M.K. Nazeeruddin, M. Gratzel, Stepwise cosensitization of nanocrystalline TiO2
films utilizing Al2O3 layers in dye-sensitized solar cells, Angew. Chem. 120
(2008) 8383–8387.
[14] J.J. Cid, J.H. Yum, S.R. Jang, M.K. Nazeeruddin, E.M. Ferrero, E. Palomares, J. Ko,
M. Gratzel, T. Torres, Molecular cosensitization for efficient panchromatic dye-
sensitized solar cells, Angew. Chem. Int. Ed. 46 (2007) 8358–8362.
[15] Y.S. Chen, Z.H. Zeng, C. Li, W.B. Wang, X.S. Wang, B.W. Zhang, Highly efficient
co-sensitization of nanocrystalline TiO2 electrodes with plural organic dyes,
New J. Chem. 29 (2005) 773–776.
[16] K. Sayama, S. Tsugakoshi, K. Hara, Y. Ohga, A. Shinpou, Y. Abe, S. Suga, H.
Arakawa, Photoelectrochemical properties of j aggregates of benzothiazole
merocyanine dyes on a nanostructured TiO2 film, J. Phys. Chem. B 106 (2002)
1363–1371.
zinc nitrate bath and their photoelectrochemical characterizations, Electro-
chemistry 70 (2002) 470.
[20] W. Pham, W.F. Lai, R. Weissleder, C.H. Tung, High efficiency synthesis of
a bioconjugatable near-infrared fluorochrome, Bioconjug. Chem. 14 (2003)
1048–1051.
[21] Y.S. Kim, K.N. Liang, K.Y. Law, D.G. Whitten, An investigation of photocurrent
generation by squarine aggregates in monolayer modifies SnO2 electrodes, J.
Phys. Chem. 98 (1994) 984–988.
[22] N. Koumura, Z.S. Wang, S. Mori, M. Miyashita, E. Suzuki, K. Hara, Alkyl-
functionalized organic dyes for efficient molecular photovoltaics, J. Am. Chem.
Soc. 128 (2006) 14256–14257.
[23] M. Miyashita, K. Sunahara, T. Nishikawa, Y. Uemura, N. Koumura, K. Hara, A.
Mori, T. Abe, E. Suzuki, S. Mori, Interfacial electron-transfer kinetics in metal-
free organic dye-sensitized solar cells: combined effects of molecular structure
of dyes and electrolytes, J. Am. Chem. Soc. 130 (2008) 17874–17881.
[24] S. Sakaguchi, S.S. Pandey, K. Okada, Y. Yamaguchi, S. Hayase, Probing TiO2/Dye
interface in dye sensitized solar cells using surface potential measurement,
Appl. Phys. Exp. 1 (2008) 105001.
[25] Y. Ogomi, Y. Kashiwa, Y. Noma, Y. Fujita, S. Kojima, M. Kono, Y. Yamaguchi,
S. Hayase, photovoltaic performance of dye-sensitized solar cells stained with
black dye under pressurized condition and mechanism for high efficiency, Sol.
Energy Mater. Sol. Cells vol.93 (2009) 1009–1012.
[26] K. Hara, Y. Dan-oh, C. Kasada, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H. Arakawa,
Effect of additives on the photovoltaic performance of coumarin-dye-sensitized
nanocrystalline TiO2 solar cells, Langmuir 20 (2004) 4205–4210.
[27] A. Hagfeldt, M. Gratzel, Light induced redox reactions in nanocrystalline sys-
tems, Chem. Rev. 95 (1995) 49–68.
[17] G. Schlichthörl, S.Y. Huang, J. Sprague, A.J. Frank, Band edge movement and
recombination kinetics in dye-sensitized nanocrystalline TiO2 solar cells: a
study by intensity modulated photovoltage spectroscopy, J. Phys. Chem. B 101
(1997) 8141–8155.
[18] L.M. Peter, K.G.U. Wijayantha, Electron transport and back reaction in dye
sensitised nanocrystalline photovoltaic cells, Electrochim. Acta 45 (2000)
4543–4551.
[19] T. Yoshida, T. Oekermann, K. Okabe, D. Schlettwein, K. Funabiki, H. Minoura,
Cathodic electrodeposition of ZnO/EosinY hybrid thin films from dye added
[28] Y. Ogomi, T. Kato, S. Hayase, Dye sensitized solar cells consisting of ionic liquids
and solidification, J. Photopylm. Sci. Technol. 19 (2006) 403–408.