Efficient electron injection due to a special adsorbing group's combination of carboxyl and hydroxyl: Dye-sensitized solar cells based on new hemicyanine dyes

Article abstract of DOI:10.1039/b418906j

A series of new benzothiazolium hemicyanine dyes (HC-1, HC-2, HC-3, HC-4, and HC-5 in Scheme 1) were designed and synthesized for sensitization of nanocrystalline TiO₂ electrodes by introducing carboxyl, hydroxyl, or sulfonate anchoring groups onto the dyes' skeletons. A naphthothiazolium hemicyanine with both sulfonate and hydroxyl (HC-6) was also prepared for comparison. The photophysical and photoelectrochemical studies revealed that three kinds of efficiencies, i.e. the fluorescence quenching efficiencies of the dyes by colloidal TiO₂, the monochromatic incident photon-to-current conversion efficiencies (IPCEs) for the dye-sensitized TiO₂ electrodes, and the overall photoelectric conversion efficiencies (η) for the dye-sensitized solar cells (DSSCs) based on these hemicyanines, all depended strongly on the anchoring group types and decreased in the order: carboxyl + hydroxyl > carboxyl > sulfonate + hydroxyl, indicating the importance of the dyes' adsorbing groups for their sensitization effects in DSSCs. The combination of carboxyl and hydroxyl as anchoring groups led to highly efficient IPCEs over a wide spectrum region with the maximum IPCE of 73.6% and a η of 5.2% under AM1.5 Global simulated light (80 mW cm^-2) for the HC-1 based DSSC, which may result from the complex formation between HC-1 and TiO₂ and the cathodic shift of the excited state oxidation potential. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b418906j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Efficient electron injection due to a special adsorbing group’s combination
of carboxyl and hydroxyl: dye-sensitized solar cells based on new
hemicyanine dyes{
ab
You-Sheng Chen, Chao Li, Zhang-Hua Zeng, Wei-Bo Wang, Xue-Song Wang* and
ab
ab
a
a
a
Bao-Wen Zhang*
Received 16th December 2004, Accepted 19th January 2005
First published as an Advance Article on the web 23rd February 2005
DOI: 10.1039/b418906j
A series of new benzothiazolium hemicyanine dyes (HC-1, HC-2, HC-3, HC-4, and HC-5 in
Scheme 1) were designed and synthesized for sensitization of nanocrystalline TiO electrodes by
2
introducing carboxyl, hydroxyl, or sulfonate anchoring groups onto the dyes’ skeletons. A
naphthothiazolium hemicyanine with both sulfonate and hydroxyl (HC-6) was also prepared for
comparison. The photophysical and photoelectrochemical studies revealed that three kinds of
2
efficiencies, i.e. the fluorescence quenching efficiencies of the dyes by colloidal TiO , the
monochromatic incident photon-to-current conversion efficiencies (IPCEs) for the dye-sensitized
TiO electrodes, and the overall photoelectric conversion efficiencies (g) for the dye-sensitized
2
solar cells (DSSCs) based on these hemicyanines, all depended strongly on the anchoring group
types and decreased in the order: carboxyl + hydroxyl . carboxyl . sulfonate + hydroxyl,
indicating the importance of the dyes’ adsorbing groups for their sensitization effects in DSSCs.
The combination of carboxyl and hydroxyl as anchoring groups led to highly efficient IPCEs over
a wide spectrum region with the maximum IPCE of 73.6% and a g of 5.2% under AM1.5 Global
2
2
simulated light (80 mW cm ) for the HC-1 based DSSC, which may result from the complex
formation between HC-1 and TiO and the cathodic shift of the excited state oxidation potential.
2
2
20
sources of the injected electrons. Huang et al. synthesized
three hemicyanine dyes with b-naphthothiazolium propyl-
sulfonate as electron acceptor, N,N-dialkylaniline as electron
donor and a double bond as p bridge, studied their photo-
sensitization abilities in DSSCs, and found the introduction of
a hydroxyl group onto the aniline improved the electron
injection efficiency. It has been well recognized that the
electron injection efficiencies of dyes upon nanocrystalline
wide band-gap semiconductors depend not only on their
respective intrinsic properties such as energy levels and excited
state lifetimes, but also on the manner in which they are
connected, such as physically or chemically adsorbed, the
nature of anchoring groups, and the distance of the dye
skeleton from the nanocrystalline surface. In relation to these
factors, it is expected that the electron injection efficiencies of
1
Introduction
Dye-sensitized solar cells (DSSCs) based on Ru(II) polypyridyl
complexes have been actively investigated since Gr a¨ tzel and
coworkers reported highly efficient photoelectric conversion
1
–6
efficiencies of up to 10%. Compared with Ru(II) polypyridyl
complexes, organic dyes not only exhibit higher molar
extinction coefficients, but also can be prepared and purified
in easier procedures at lower cost, and therefore are attracting
more and more interest. Recently great progress has been
made in this field and the highest overall photoelectric
conversion efficiency of solar cells sensitized by organic dyes
7
has reached 8%, comparable with the performances of Ru(II)
polypyridyl complex-based DSSCs. Lots of organic dyes such
12–14
8
,9
as cyanine dyes, perylene dyes,
10,11
merocyanine dyes,
18,19
1
5–17
coumarin dyes
and hemicyanine dyes
have been
such hemicyanine dyes may be improved further by rational
2
structure design, because (1) sulfonate (–SO
investigated in such applications. Among them hemicyanine
dyes, a group of interesting photosensitizers, with electron
donor and electron acceptor linked by a p-conjugation bridge
3
) is a weaker
21
adsorbing group on TiO than carboxyl (–COOH) and (2)
2
the tethering group of the carbon chain (–CH
be shortened further to ensure closer contact between dye
and TiO . Therefore, to better understand the effects of the
2 2 2
CH CH –) can
(
D–p–A) can effectively produce a charge-separation excited
state upon photoinduced intramolecular charge transfer, from
which a high quantum yield of electron injection into the
conduction band of wide band-gap semiconductor is expected
as in the case of Ru(II) polypyridyl complexes where metal-to-
ligand charge transfer states (MLCT states) serve as the
2
anchoring groups on the photoelectrochemical performance of
hemicyanine dyes in DSSCs, five new benzothiazolium based
hemicyanine dyes (HC-1, HC-2, HC-3, HC-4 and HC-5, see
Scheme 1) were designed and synthesized and detailed studies
of their photophysical and photoelectrochemical properties
have been undertaken compared with a reference compound
1
Electronic supplementary information (ESI) available: H NMR and
{
MALDI-TOF data for the hemicyanine dyes. See http://www.rsc.org/
suppdata/jm/b4/b418906j/
*g203@mail.ipc.ac.cn (Xue-Song Wang)
(
HC-6) which was the best photosensitizer among those
2
reported by Huang et al. As expected, the substitution of
0
1
654 | J. Mater. Chem., 2005, 15, 1654–1661
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 1 Molecular structures of the hemicyanine dyes.
carboxyl for sulfonate enhanced the photosensitization cap-
abilities of these hemicyanine dyes remarkably. Moreover,
we found the introduction of a hydroxyl group on the C2
position of the aniline moiety (HC-1 and HC-3) led to the
formation of a complex between the corresponding hemi-
When cooled to room temperature, the resulting precipitate
was filtered and recrystallized from ethanol to afford
pure hemicyanine dyes in the yield of 70–90%: 2-(2-hydroxyl-
4-N,N-diethylaminophenylethenyl)-benzothiazolium-1-acetate
(HC-1),
2-(4-N,N-diethyl
aminophenylethenyl)-benzo-
cyanine dye and nanocrystalline TiO
2
, which may play a
thiazolium-1-acetate (HC-2), 2-(2-hydroxyl-4-N,N-diethyl-
aminophenylethenyl)-benzothiazolium-1-propionate (HC-3),
2-(4-N,N-diethylaminophenylethenyl)-benzothiazolium-1-pro-
pionate (HC-4), 2-(2-hydroxyl-4-N,N-diethylaminophenyl-
key role in the further improvement of the electron injection
efficiency.
ethenyl)-benzothiazolium-1-propylsulfonate
(HC-5)
and
2
Experimental
2
-(2-hydroxyl-4-N,N-diethylaminophenylethenyl)-b-naphtho-
2
.1 General procedures for the synthesis of hemicyanine dyes
thiazolium-1-propylsulfonate (HC-6). The identity of the
1
hemicyanine dyes was confirmed by H NMR and MALDI-
Hemicyanine dyes (HC-1, HC-2, HC-3, HC-4, HC-5 and
HC-6) were synthesized according to the procedures illustrated
in Scheme 2.
TOF (see ESI{).
2
.2 Preparation of dye-sensitized TiO electrodes
2
2
.1.1 Preparation of the benzothiazolium and naphthothiazo-
0,22
The TiO paste (Nanoxide-T from Solaronix, Switzerland) was
2
2
lium intermediates (3)
. 10 mmol thiazole compound (1) and
coated by means of the doctor blade technique on fluorine-
1
0 mmol compound (2) were refluxed overnight in 50 ml
doped tin oxide (FTO) conducting glass (transmission .80%,
2
2
benzene. The solvent was removed on a rotary evaporator.
Column chromatography on silica gel with chloroform as
eluent yielded 60–80% of pure thiazolium salts (3a), (3b), (3c)
and (3d).
sheet resistance 20 V cm ), followed by sintering at 450 uC in
air atmosphere for 30 min. The TiO electrode was soaked in
2
0.02 M TiCl stock solution overnight at room temperature
4
2
and then washed with distilled water. Finally, it was fired
again for 30 min at 450 uC. The thickness of the TiO film was
9 mm determined by a Dektak 3 Profilometer. The obtained
2
2
0,23
2.1.2 Preparation of the target hemicyanine dyes
. 1 mmol
of (3) and 1 mmol of benzaldehyde compound (4) were
added into 25 ml anhydrous ethanol and refluxed for a
whole day in the presence of a catalytic amount of piperidine.
TiO2 electrodes were dipped in a 0.3 mM hemicyanine dye
solutions in CHCl for 8 h at room temperature and rinsed
3
3
with CHCl and dried under a stream of hot air.
Scheme 2 The synthetic procedures of the hemicyanine dyes.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 1654–1661 | 1655

View Article Online
2
.3 Photophysical measurements
Electronic absorption spectra of the hemicyanine dyes in
MeOH solutions, colloidal TiO solutions and adsorbed on
Table 1 Photophysical properties of the hemicyanine dyes in MeOH
solutions and adsorbed on nanocrystalline TiO films
2
HC-1 HC-2 HC-3 HC-4 HC-5 HC-6
2
TiO2 films were measured using a Shimadzu UV-1601PC
spectrophotometer. Fluorescence emission spectra of the
hemicyanine dyes in MeOH solutions and TiO2 colloidal
solutions were recorded on a Hitachi F-4500 fluorescence
ab
a
l
e (MeOH)/10 M cm
max (MeOH)/nm
535
7.25 7.67 7.92 7.08 6.27 5.41
532.5 544
535.5 547
554
4
21
21b
fl
c
l
max (MeOH)/nm
Stokes shift/nm
584
49
596 586
63.5 42
600 588
64.5 41
602
48
rel
d
W
l
fl (MeOH)
1.00 0.67 0.78 0.62 0.99 1.38
494.5 501 485 495 485 521
7.03 7.58 9.78 11.70 6.53 5.74
spectrophotometer. The stock colloidal TiO
2
2
solution (5 6
M) in acetonitrile was prepared by the hydrolysis of
ab
e
max (on TiO
2
)/nm
N/10 molecules cm
3
16
22f
1
0
2
4
a
b
Absorption maximum in MeOH. Molar extinction coefficient at
titanium(IV) isopropoxide following the reported method,
and diluted with acetonitrile to the desired concentrations.
ab
max
c
d
Fluorescence maximum in MeOH. Relative fluorescence
l
quantum yield. Absorption maximum adsorbed on TiO
.
e
2
film.
f
2
Adsorption amount per unit area of TiO film.
2
.4 Electrochemical and photoelectrochemical measurements
Cyclic voltammetry measurements were carried out on a
Potentiostat/Galvanostat model 283A (EG&G Princeton
Applied Research) to determine the oxidation potentials of
the hemicyanine dyes in DMSO solutions. A three-electrode
cell was composed of a Pt working electrode, a Pt counter
electrode and a saturated calomel reference electrode (SCE).
Tetrabutyl ammonium perchlorate (0.01 M) served as support-
introduction of a hydroxyl group at the C2 position. When
comparisons are made between HC-1, HC-3 and HC-5, or
HC-2 and HC-4, it appears that the absorption maxima move
to longer wavelengths with increasing carbon number of the
alkyl chain between the terminal carboxyl or sulfonate group
and the benzothiazolium part. The absorption bathochromic
shift of HC-6 relative to HC-5 implies naphthothiazolium is a
stronger electron acceptor than benzothiazolium, partly due to
its larger p system which can accommodate more charge
density. All six hemicyanine dyes exhibit large molar extinction
coefficients at their absorption maxima, ranging from 5.41 6
2
1
ing electrolyte. The scan rate was 200 mV s
A sandwich-type solar cell was constructed by introducing
the redox electrolyte containing 0.1 M LiI, 0.05 M I , and
.3 M 1,2-dimethyl-3-propylimidazolium iodide (DMPImI) in
.
2
0
2
methoxypropionitrile between a dye-sensitized TiO electrode
4
4
21
21
1
0
to 7.92 6 10
M
cm , and being about four-to-six
and a counter electrode of Pt-coated FTO conducting glass.
The dye-sensitized TiO film was illuminated through the
conducting glass support and the effective irradiated area was
times larger than those of Ru(II) polypyridyl complexes (1.40 6
2
4
21
21 2
1
0 M cm ), which favors light harvesting in DSSCs.
While hydroxyl red shifts the absorption maxima, its
2
0
.175 cm . The photovoltaic performance of the solar cell
was taken from a computer-controlled digital sourcemeter
Keithley 2400) under illumination by AM1.5 Global simu-
introduction gave rise to blue shifts of the fluorescence
emission maxima of the hemicyanines (HC-1 vs. HC-2 and
HC-3 vs. HC-4). As an electron donating group, hydroxyl can
promote ground state charge transfer from the aniline to the
benzothiazolium moiety, as a result making the ground state
more similar in charge population to the charge transfer
excited state, which will lead to a diminished Stokes shift, and
therefore a blue shift is reflected in the fluorescence emission
maximum. Besides, the introduction of hydroxyl favors the
radiative decay of the corresponding hemicyanine excited state
(
2
2
lated light (80 mW cm ) from an Oriel 91192 Solar Simulator.
In the measurements of photocurrent action spectra, the
solar cell was irradiated by monochromatic light, obtained
from a 500 W Xe lamp in combination with a SpectraPro-150
monochromator (Acton Research Co.), in the range of 400–
8
00 nm at 10 nm intervals, and the generated short-circuit
photocurrents were recorded on a model 283A Potentiostat/
Galvanostat. The incident monochromatic light intensity was
measured with a S370 single-channel optometer equipped with
a Model 262 flat response sensor head (Graseby Optronics).
2
0
(
Table 1); this phenomenon was also observed by Huang.
3
.2 Photophysical properties of the hemicyanine dyes adsorbed
films
on nanocrystalline TiO
2
3
Results and discussion
Fig. 1 shows the absorption spectra of the hemicyanine dyes in
methanol solutions and adsorbed on nanocrystalline TiO
3
.1 Photophysical properties of the hemicyanine dyes in
2
methanol solutions
films of 3 mm thickness (the absorption spectra of HC-3 and
HC-4 are very similar to those of HC-1 and HC-2, respectively,
and are not shown in Fig. 1). Compared with their absorption
in methanol solutions, it is clear that the absorption bands for
these dyes are all broadened significantly on both wings upon
Electronic absorption and fluorescence emission properties of
the hemicyanine dyes in methanol are listed in Table 1. These
hemicyanine dyes are composed of an electron donor aniline
moiety and an electron acceptor benzothiazolium or naphtho-
thiazolium moiety and are typical D–p–A chromophores,
therefore their absorption bands in the visible region are of
charge-transfer character and the absorption maxima will
undergo a red shift upon increasing the electron donating or
withdrawing abilities of the D or A moieties. So the red shift of
the absorption maxima for HC-1 and HC-3 with respect to
HC-2 and HC-4 respectively can be attributed to the enhanced
electron donating property of the aniline moiety upon
adsorption on TiO
remarkably (30–60 nm, Table 1), suggesting the formation
of H-aggregates on the TiO surface. For TiO films of 9 mm
, and the absorption maxima blue shift
2
2
2
thickness, more dye molecules were adsorbed, and wide
absorption bands with high absorbance over the region of
450–650 nm were observed, implying both H and J aggregates
may have formed and the aggregation extent for each kind
1
4
of aggregate increased in such cases. The formation of
1
656 | J. Mater. Chem., 2005, 15, 1654–1661
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Fig. 1 Absorption spectra of the hemicyanine dyes in MeOH solutions (a) and adsorbed on nanocrystalline TiO
2
films of 3 mm thickness (b).
aggregates broadens the absorption spectra and favors light
harvesting.
the result of the participation of the hydroxyl in adsorption.
Once the hydroxyl attaches on to the TiO2 surface as an
ancillary anchor, the contact area of HC-1 or HC-3 with the
surface will increase, and some adsorbing sites will be occupied
by hydroxyl or be shielded by the dye’s skeleton, accordingly,
the adsorbed amount decreases. The interaction of the
Besides the absorbance and the overlap of the absorption
with the solar spectrum, the light harvesting efficiency of a
photosensitizer in DSSCs depends also on its adsorbed amount
on the nanocrystalline semiconductor electrode. To determine
the exact amounts of the hemicyanine dyes adsorbed on TiO
2
hydroxyl with TiO was confirmed further by the absorption
2
films, dye-coated films (9 mm) were soaked in methanol to
2
desorb the dye. It was found that HC-5 and HC-6 were prone
and emission spectra of HC-1 and HC-3 in colloidal TiO2
solutions (see below). Moreover, the increased flexibility of the
alkyl spacer between the terminal carboxyl and the benzothia-
zolium part seems to facilitate dye adsorption on the TiO2
surface (HC-1 vs. HC-3 and HC-2 vs. HC-4), probably because
the dye molecules may adjust their mutual orientation more
easily in such cases.
to be desorbed completely in neat methanol during several
minutes, while the desorption processes of the other hemi-
cyanine dyes took about 1 h in the presence of alkali,
indicating carboxyl is much superior to sulfonate in anchoring
the dyes on the TiO
spectra of the desorption solutions, the numbers of adsorbed
dye molecules per square centimetre of TiO film were
2
surface. By measuring the absorption
2
3
.3 Photophysical properties of the hemicyanine dyes in colloidal
calculated and collected in Table 1. Due to the weaker
adsorbing ability of sulfonate than carboxyl and the larger
molecular size of the naphthothiazolium moiety than the
TiO solutions
2
To disclose the interaction nature of the hemicyanine dyes with
benzothiazolium moiety, HC-6 and HC-5 adsorbed on TiO in
2
nanocrystalline TiO
dyes in acetonitrile were measured in the presence of increased
concentrations of colloidal TiO and are shown in Fig. 2 (the
2
electrodes, the absorption spectra of these
the first and second smallest amounts, respectively. The
adsorbed amounts of HC-1 and HC-3 decreased by 7% and
2
1
6% with respect to HC-2 and HC-4, respectively, owing to the
spectra of HC-3, HC-4 and HC-5 were omitted due to their
similarities to those of HC-1, HC-2 and HC-6, respectively).
presence of a hydroxyl in the aniline moiety, which might be
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 1654–1661 | 1657

View Article Online
Upon increasing the concentrations of colloidal TiO
2
the
600 nm can be attributed to the scattering effect induced by
absorption spectra of HC-2, HC-4, HC-5, and HC-6 were all
broadened, and the broadening extents of HC-2 and HC-4
are much greater than HC-5 and HC-6, suggesting again that
TiO2 particles. However, along with the broadening of the
absorption band upon addition of TiO , the spectra of HC-1
2
and HC-3 underwent a remarkable red shift and an isosbestic
point was observed. This suggests that a new species formed
and the hydroxyl must be involved in its formation. This new
species is not likely to be a dye aggregate because the presence
of the hydroxyl decreases the dye adsorption amount on TiO2
(see Table 1) and therefore disfavors aggregate formation
between dye molecules. A possible explanation is that the
combination of the carboxyl and the hydroxyl leads to a
complexation reaction of the corresponding dye molecule with
the carboxyl favors the interaction of these dyes with TiO
2
particles due to its stronger adsorbing capability than the
sulfonate. The absorption maxima changed marginally for
HC-2, HC-4, HC-5, and HC-6, indicative of no formation of
any new species such as dye aggregates or complexes between
the dye and TiO2 under the experimental conditions. The
significant absorbance increase below 450 nm and above
4+
Ti ion, which induced the observation of the red shift and the
isosbestic point. It was reported that a triphenylmethane dye
with hydroxyl–carboxylic (salicylate type) groups adsorbed on
a nanocrystalline TiO
2
film by a complexation reaction with
4+
Ti , and an ultrafast and efficient electron injection was
2
5
observed from this complex. So, it is expected HC-1 and
HC-3 may exhibit high electron injection efficiency due to
an enhanced electronic coupling within the complexation
2
6
interaction, which was supported by fluorescence emission
measurements in TiO2 colloid solutions and photoelectro-
chemical studies in DSSCs.
The addition of TiO
of the hemicyanine dyes resulted in the fluorescence quenching
see Fig. 3b as an example). Fig. 3a shows the fluorescence
quenching efficiencies of the hemicyanine dyes by TiO colloid.
2
colloid into the acetonitrile solutions
(
2
It is obvious that the quenching efficiencies depended greatly
on the anchoring types of the dye molecules, and decreased
in the order: carboxyl + hydroxyl . carboxyl . sulfonate +
hydroxyl. The fluorescence quenching can be attributed to
electron injection from the excited hemicyanine dyes to the
TiO2 conduction band. The complexation interaction men-
tioned above for HC-1 and HC-3 may contribute to their
highly efficient fluorescence quenching behavior. It is interest-
ing that the fluorescence quenching of HC-1 and HC-2 is less
efficient than that of HC-3 and HC-4, respectively, implying
the substituents on the N atom of benzothiazolium for HC-3
and HC-4 may adopt a bending rather than an extended
conformation so that their chromophore skeletons approach
the TiO surface closer than HC-1 and HC-2, leading to more
2
efficient electron injection.
3
.4 Energetic consideration for electron injection of the
hemicyanine dyes into TiO electrodes
2
The excited state oxidation potentials (Eox*) of the hemi-
cyanine dyes can be estimated by subtracting the 0–0 transition
energies (E0–0) from the oxidation potentials (Eox), and E0–0
can be deduced from the wavelength where the absorption and
2
fluorescence emission spectra cross each other. As shown in
Table 2, the excited state energy levels for the hemicyanine
dyes are in the range of 21.24 to 21.61 V (vs. SCE), far more
negative than the band edge energy of the nanocrystalline TiO2
electrode (20.8 V vs. SCE), indicating that the electron
injection processes from the excited dyes to TiO conduction
2
bands are energetically permitted. Though the dyes’ oxidation
Fig. 2 Absorption spectra of the hemicyanine dyes in acetonitrile
26
(
5 6 10 M) at varied TiO
2
colloid concentrations: a. 0, b. 0.5 6
potentials may change upon adsorption to the TiO
2
surface,
2
5
25
25
25
25
10
, c. 1.5 6 10 , d. 2.5 6 10 , e. 3.5 6 10 , f. 4.5 6 10 M.
the fluorescence emission quenching by TiO colloid confirms
2
1
658 | J. Mater. Chem., 2005, 15, 1654–1661
This journal is ß The Royal Society of Chemistry 2005

View Article Online
circuit divided by the number of monochromatic incident
2
photons, is represented by the following equation:
À
Á
{
2
1
240 Isc mAcm
IPCEð%Þ~
À
chromatic light and l and P are the wavelength and the intensity
Á
{
lðnmÞPin Wm
where Isc is the short-circuit photocurrent generated by mono-
2
in
of the monochromatic light, respectively.
The IPCEs are strongly dependant on the anchoring types of
the dyes and decreased in the order: carboxyl + hydroxyl .
carboxyl . sulfonate + hydroxyl. The lowest IPCEs for HC-5
and HC-6 indicate sulfonate, as attaching group, is much
inferior to carboxyl in dye sensitization upon nanocrystalline
2
TiO . More importantly, the introduction of the hydroxyl
improved the IPCEs greatly (HC-1 vs. HC-2 and HC-3 vs.
HC-4). The IPCE at any wavelength is the production of three
efficiencies, i.e. the light harvesting efficiency, the electron
2
injection efficiency, and the electron collection efficiency. Due
to the high absorbance (.1.0) over the region of 450–650 nm
2
for the hemicyanine dye-sensitized TiO electrodes, the light
harvesting efficiencies for them all exceeded 90% and could not
account for the significant IPCE enhancement by introduction
of the hydroxyl. The electron collection efficiency is associated
with (1) the electron transport within the TiO network, (2)
2
the dark current arising from the injected electrons to the
2
electrolyte (I3 ), and (3) the recombination between the
injected electrons with the dye cations. The former two terms
should be the same for these DSSCs because the same
electrodes and electrolytes were used. The introduction of the
hydroxyl shifts the oxidation potentials of the dyes cathodi-
cally (Table 2) and should promote electron recombination
2
and restrict dye cation reduction by electrolyte (I ) (because
the electron recombination in DSSCs generally fell in the
Marcus inverted region while the dye cation reduction by
electrolyte fell in the Marcus normal region), and therefore
should have decreased rather than increased the IPCEs if the
Fig. 3 (a) Fluorescence quenching efficiencies of the hemicyanine
6
2
dyes (5 6 10 M) in acetonitrile by TiO
2
colloid. (b) Fluorescence
2
6
emission spectra of HC-1 (5 6 10 M) in acetonitrile in the presence
2
5
25
of TiO
2
with concentrations of: a. 0, b. 0.5 6 10 , c. 1.5 6 10 , d.
25 25 25
2.5 6 10 , e. 3.5 6 10 , f. 4.5 6 10 M.
that the electron injection occurred from the excited states of
the hemicyanine dyes.
3
.5 Photoelectrochemical performances of the hemicyanine dyes
in DSSCs
Photocurrent action spectra for the dye-coated TiO electrodes
2
are shown in Fig. 4, where the monochromatic incident
photon-to-current conversion efficiencies (IPCEs) are plotted
as a function of wavelength. The IPCE, defined as the number
of photogenerated electrons flowing through the external
Table 2 Electrochemical properties of the hemicyanine dyes in
DMSO
HC-1 HC-2 HC-3 HC-4 HC-5 HC-6
Fig. 4 Photocurrent action spectra for DSSCs based on the hemi-
E
E
E
ox/V (vs. SCE)
0–0/eV
0.78
2.20
0.89
2.19
0.58
2.19
0.85
2.18
0.94
2.18
0.88
2.12
2
cyanine dyes. The electrolyte was a solution of 0.1 M LiI, 0.05 M I ,
and 0.3 M 1,2-dimethyl-3-propylimidazolium iodide (DMPImI) in
methoxypropionitrile.
ox*/V (vs. SCE) 21.42 21.30 21.61 21.33 21.24 21.24
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 1654–1661 | 1659

View Article Online
2
2
22
electron collection efficiency was the decisive factor. So, the
improved IPCEs upon the introduction of the hydroxyl can
only be ascribed to the improved electron injection efficiency,
in good agreement with the fluorescence quenching behaviors
80 mW cm ), Jsc of 12.8 and 11.7 mA cm can be estimated, in
agreement with the experimental observations. Interestingly, HC-1
and HC-3 also showed enlarged Voc (520 mV for HC-1 and
499 mV for HC-3) than the other dyes, suggesting the cooperation
of the carboxyl and the hydroxyl may restrict the dark current. In
combination with the similar f (0.57 for HC-1 and 0.56 for HC-3)
to the other dyes’ f, HC-1 and HC-3 based solar cells yielded
enhanced g (5.2% for HC-1 and 4.4% for HC-3) over the DSSCs
sensitized by the other hemicyanine dyes. However, the obtained
open-circuit photovoltages are lower than Ru(II) polypyridyl-
based DSSCs, where open-circuit voltages higher than 0.6 V were
2
of these dyes by TiO colloid.
The improved electron injection efficiencies probably result
from the two changes brought about by the introduced
hydroxyl group. The combination of carboxyl and hydroxyl
led to the complexation of the corresponding dyes with TiO
2
as
mentioned above, and the enhanced electronic coupling within
2
6
the formed complex promoted the electron injection. Also,
the introduced hydroxyl led to the cathodic shift of the excited
state oxidation potential (Table 2), and the enlarged driving
force promoted the electron injection. Though HC-5 and HC-6
possess the hydroxyl group also, the lack of complexation
1–6
generally achieved. The relatively low Voc was also observed in
other organic dye sensitized solar cells, and probably originates
from the dark current due to the reduction of both the oxidized
2
27
dyes and I3 by the injected electrons. For Ru(II) polypyridyl
complexes, electron injection occurs from polypyridyl ligands, but
recombination occurs at an Ru(III) center, which is located
2
between them and TiO and their less negative excited state
oxidation potentials may be the reasons for their low IPCEs,
supporting the above explanations on how the hydroxyl
improves the IPCEs.
somewhat further away from the TiO surface than polypyridyl
2
ligands, partly accounting for the fast electron injection and slow
28
Though the fluorescence of HC-3 was quenched more
recombination kinetics. Organic photosensitizers having photo-
induced intramolecular electron transfer properties such as the
hemicyanine dyes shown in this work may mimic the electron
injection and recombination kinetics provided the dye molecules
2
efficiently by colloidal TiO than that of HC-1, HC-1 exhibited
higher IPCEs than HC-3 (similar behavior for HC-2 and
HC-4), which probably resulted from the dye aggregation
upon adsorption onto the TiO electrodes.
2
are assembled on the TiO surface in such a way that the electron
2
The cooperation of carboxyl and hydroxyl led to high IPCEs
accepting moiety attaches directly onto TiO , leaving the electron
2
(.60%) over a wide spectrum region with maximum IPCEs of
2
donating moiety away from the TiO surface. The validity of this
7
3.6% and 64.2% for HC-1 and HC-3, respectively. The func-
idea is worthy of experimental investigation.
tions of carboxyl + hydroxyl in improving the photoelectro-
chemical properties of HC-1 and HC-3 were further confirmed
by the photovoltaic data of the DSSCs based on them.
Table 3 shows the photovoltaic properties of DSSCs based
on the hemicyanine dyes. The overall photoelectrical conver-
sion efficiency, g, is calculated by the following equation:
4
Conclusion
The benzothiazolium hemicyanine dyes with carboxyl as
anchoring group (HC-1, HC-2, HC-3, and HC-4) exhibited
improved IPCEs and overall photoelectric conversion efficien-
cies in the DSSCs than the benzo- or naphthothiazole
hemicyanine dyes with sulfonate and hydroxyl as anchoring
groups (HC-5 and HC-6). More importantly, the combination
of the carboxyl and the hydroxyl in HC-1 and HC-3 improved
the IPCEs further than HC-2 and HC-4, probably resulting
from (1) the complex formation between HC-1 or HC-3
g 5 (JscVocf)/Pin
where Pin is the intensity of the incident light, Jsc the short-circuit
photocurrent density, Voc the open-circuit photovoltage, and f the
fill factor which is the ratio of the maximum output power of the
2
cell with Pin. The g values depended on the anchoring types
2
and TiO , and (2) the cathodic shift of the excited state
strongly and decreased in the order: carboxyl + hydroxyl .
carboxyl . sulfonate + hydroxyl. HC-1 and HC-3 with both
carboxyl and hydroxyl as anchoring groups exhibited much larger
oxidation potentials for HC-1 and HC-3. Accordingly, a
significantly enhanced overall photoelectrical efficiency of
5
.2% was achieved in a DSSC based on HC-1. These results
22
22
Jsc (13.9 mA cm for HC-1 and 12.7 mA cm for HC-3) than
the other dyes, obviously benefited from their highly efficient
IPCEs. From the overlap integral of the IPCE curves for HC-1
and HC-3 with AM1.5 Global solar spectrum (normalized to
illustrate how significantly the anchoring manner of a dye
molecule influences its sensitization behavior on a nano-
crystalline semiconductor, and the optimization on adsorbing
groups may result in more efficient dyestuffs for DSSC
applications.
Table 3 Photovoltaic performances of the hemicyanine dye-sensitized
solar cells
a
Acknowledgements
HC-1 HC-2 HC-3 HC-4 HC-5 HC-6
This work was financially supported by the Ministry of Science
and Technology of China (G2000028204).
2
2
J
V
f
g (%)
a
sc/mA cm
oc/mV
13.9
520
0.57
5.2
10.5
448
0.58
3.5
12.7
499
0.56
4.4
8.9
6.1
7.6
474
0.59
3.1
465
0.55
2.0
452
0.59
2.6
ab
You-Sheng Chen, Chao Li, Zhang-Hua Zeng, Wei-Bo Wang,
ab
ab
a
a
Xue-Song Wang* and Bao-Wen Zhang*
a
Conditions: irradiated light, AM1.5 Global simulated light
a
2
2
2
Technical Institute of Physics and Chemistry, Chinese Academy of
Sciences, Beijing 100101, P. R. China. E-mail: g203@mail.ipc.ac.cn
(
80 mW cm ); photoelectrode, TiO
electrolyte, 0.1 M LiI, 0.05 M I
propylimidazolium iodide (DMPImI) in methoxypropionitrile.
2
(9 mm thickness and 0.175 cm );
2
, and 0.3 M 1,2-dimethyl-3-
b
Graduate School of Chinese Academy Sciences, Beijing 100039,
P. R. China
1
660 | J. Mater. Chem., 2005, 15, 1654–1661
This journal is ß The Royal Society of Chemistry 2005

View Article Online
1
4 K. Sayama, S. Tsukagoshi, T. Mori, K. Hara, Y. Ohga,
A. Shinpou, Y. Abe, S. Suga and H. Arakawa, Sol. Energy
Mater. Sol. Cells, 2003, 80, 47.
References
1
2
B. O’Regan and M. Gr a¨ tzel, Nature, 1991, 353, 737.
M. K. Nazeeruddin, A. Kay, I. Rodicio, R. H. Baker, E. Muller,
15 K. Hara, K. Sayama, Y. Ohga, A. Shinpo, S. Suga and
H. Arakawa, Chem. Commum., 2001, 569.
16 K. Hara, Y. Tachibana, Y. Ohga, A. Shinpo, S. Suga, K. Sayama,
H. Sugihara and H. Arakawa, Sol. Energy Mater. Sol. Cells, 2003,
77, 89.
17 K. Hara, M. Kurashige, Y. Danoh, C. Kasada, A. Shinpo, S. Suga,
K. Sayama and H. Arakawa, New J. Chem., 2003, 27, 783.
18 E. Stathatos and P. Lianos, Chem. Mater., 2001, 13, 3888.
19 Z. S. Wang, F. Y. Li and C. H. Huang, Chem. Commun., 2000,
2063.
P. Liska, N. Vlachopoulos and M. Gr a¨ tzel, J. Am. Chem. Soc.,
1993, 115, 6382.
3
4
M. K. Nazeeruddin, P. Pechy, T. Renouard, S. M. Zakeeruddin,
R. H. Baker, P. Comte, P. Liska, L. Cevey, E. Costa, V. Shklover,
L. Spiccia, G. B. Deacon, C. A. Bignozzi and M. Gr a¨ tzel, J. Am.
Chem. Soc., 2001, 123, 1613.
T. Renouard, R. A. Fallahpour, M. K. Nazeeruddin, R. Humphry,
S. I. Gorelsky, A. B. P. Lever and M. Gr a¨ tzel, Inorg. Chem., 2002,
41, 367.
5
6
7
8
9
M. Yanagida, T. Yamaguchi, M. Kurashije, K. Hara, R. Katoh,
H. Sugihara and H. Arakawa, Inorg. Chem., 2003, 42, 7921.
C. Klein, M. K. Nazeeruddin, D. D. Censo, P. Liska and
M. Gr a¨ tzel, Inorg. Chem., 2004, 43, 4216.
T. Horiuchi, H. Miura, K. Sumioka and S. Uchida, J. Am. Chem.
Soc., 2004, 126, 12218.
K. Sayama, K. Hara, Y. Ohga, A. Shinpou, S. Suga and
H. Arakawa, New J. Chem., 2001, 25, 200.
A. Ehret, L. Stuhi and M. T. Spitler, J. Phys. Chem. B, 2001, 105,
20 Q. H. Yao, L. Shan, F. Y. Li, D. D. Yin and C. H. Huang, New J.
Chem., 2003, 27, 1277.
21 O. Schwarz, D. V. Loyen, S. Jockusch, N. J. Turro and H. Durr,
J. Photochem. Photobiol. A : Chem., 2000, 132, 91.
22 W. H. Mills, J. Chem. Soc., 1922, 121, 455.
23 A. Hassner, D. Birnbaum and L. M. Loew, J. Org. Chem., 1984,
49, 2546.
24 P. V. Kamat and M. A. Fox, Chem. Phys. Lett., 1983, 102, 379.
25 G. Ramakrishna, H. N. Ghosh, A. K. Singh, D. K. Palit and
J. P. Mittal, J. Phys. Chem. B, 2001, 105, 12786.
26 J. M. Rehm, G. L. Mclendon, Y. Nagasawa, K. Yoshihara,
J. Moser and M. Gr a¨ tzel, J. Phys. Chem., 1996, 100, 9577.
27 K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo,
S. Suga, K. Sayama, H. Sugihara and H. Arakawa, J. Phys. Chem.
B, 2003, 107, 597.
9
960.
0 S. Ferrere, A. Zaben and B. A. Gregg, J. Phys. Chem. B, 1997, 101,
490.
1 S. Ferrer and B. A. Gregg, New J. Chem., 2002, 26, 1155.
2 A. C. Khazraji, S. Hotchandani, S. Das and P. V. Kamat, J. Phys.
Chem. B, 1997, 103, 4693.
1
4
1
1
13 K. Sayama, K. Hara, N. Mori, M. Satsuki, S. Suga, S. Tsukagoshi,
Y. Abe, H. Sugihara and H. Arakawa, Chem. Commun., 2000, 1173.
28 K. Kalyanasundaram and M. Gr a¨ tzel, Coord. Chem. Rev., 1998,
77, 347.
This journal is ß The Royal Society of Chemistry 2005
J. Mater. Chem., 2005, 15, 1654–1661 | 1661