Enhancement of photovoltaic performance in dye-sensitized solar cells fabricated with dendritic photosensitizer containing site-isolated chromophores

Article abstract of DOI:10.1016/j.dyepig.2013.08.005

We designed and synthesized a series of phenothiazine-based donor-acceptor type molecules which consist of different numbers of chromophores in a molecule for use as photosensitizers for dye-sensitized solar cells (DSSCs). Intriguingly, a dendritic molecule containing three chromophores not only gave a new possibility to modify the three dimensional structure, but also reduced aggregation between chromophores inducing dipole-dipole interaction. The DSSCs made of a dendritic photosensitizer system exhibited much higher cell efficiencies than those with the single- or double-chromophoric photosensitizers due to efficient electron extraction pathways in the dendritic molecule which lead to a significantly reduce recombination rate of electrons from the TiO ₂ to the electrolyte when the same numbers of chromophores were loaded on the TiO₂ surface. In particular, the DSSC based on the dendritic molecule exhibited improved open-circuit voltage than that of the single- or double-chromophoric photosensitizers. This can be attributed to strong adsorption properties of the TiO₂ electrode and a screening effect to the electrolyte ions provided by the network structure of the dendritic photosensitizer. The different behavior of these DSSCs was explained by comparing the results of electrochemical impedance spectroscopy and measurement of open-circuit voltage decay.

Full text of DOI:10.1016/j.dyepig.2013.08.005

Dyes and Pigments 99 (2013) 986e994
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Enhancement of photovoltaic performance in dye-sensitized solar
cells fabricated with dendritic photosensitizer containing site-isolated
chromophores
,
,
Myeong Seok Kim a 1, Min Ju Cho b 1, Young Cheol Choi a, Kwang-Soon Ahn a,
Dong Hoon Choi b , Kyungkon Kim , Jae Hong Kim
,
c
a,
**
*
a Department of Chemical Engineering, Yeungnam University, Gyungsan, Gyeongbuk 712-749, Republic of Korea
b Department of Chemistry, Research Institute for Natural Sciences, Korea University, Sungbuk-gu, Seoul 136-701, Republic of Korea
c Department of Chemistry, Ewha Womans University, Seodaemun-gu, Seoul 120-750, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
We designed and synthesized a series of phenothiazine-based donoreacceptor type molecules which
consist of different numbers of chromophores in a molecule for use as photosensitizers for dye-sensitized
solar cells (DSSCs). Intriguingly, a dendritic molecule containing three chromophores not only gave a new
possibility to modify the three dimensional structure, but also reduced aggregation between chromo-
phores inducing dipoleedipole interaction. The DSSCs made of a dendritic photosensitizer system
exhibited much higher cell efficiencies than those with the single- or double-chromophoric photosen-
sitizers due to efficient electron extraction pathways in the dendritic molecule which lead to a signifi-
cantly reduce recombination rate of electrons from the TiO2 to the electrolyte when the same numbers of
chromophores were loaded on the TiO2 surface. In particular, the DSSC based on the dendritic molecule
exhibited improved open-circuit voltage than that of the single- or double-chromophoric photosensi-
tizers. This can be attributed to strong adsorption properties of the TiO2 electrode and a screening effect
to the electrolyte ions provided by the network structure of the dendritic photosensitizer. The different
behavior of these DSSCs was explained by comparing the results of electrochemical impedance spec-
troscopy and measurement of open-circuit voltage decay.
Received 18 July 2013
Received in revised form
13 August 2013
Accepted 13 August 2013
Available online 22 August 2013
Keywords:
Dye-sensitized solar cells
Chromophore
Dendritic molecule
Phenothiazine
Photosensitizer
Electron life-time
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
photon-to-current conversion mechanism of DSSCs is based on the
injection of electrons from the excited photosensitizers into the
After the historic paper on dye-sensitized solar cells (DSSCs) was
published in 1991 by Grätzel et al., DSSCs have been widely
investigated because of their special features such as low-cost
fabrication with fairly high solar energy-to-conversion efficiencies
relative to conventional pen junction solar cells [1,2]. The
improvement of solar energy-to-electricity conversion efficiency
has continued to be an important research area for DSSCs.
The standard structure of the DSSC is an electrochemical cell
composed of a Ru-complex photosensitizing dye-adsorbed wide
band-gap electrode such as TiO2 or ZnO, an electrolyte containing
Iꢀ/Iꢀ3 redox couples, and a Pt-coated counter electrode [3e5]. The
conduction band of nanocrystalline TiO2 or ZnO. The oxidized
photosensitizers are reduced by electron injection from the elec-
trolyte. Thus, the photosensitizing dye plays an important role in
capturing the photons and generating the electron/hole pair,
transferring the charges to the interface of the semiconductor and
the electrolyte. The sensitization of nanocrystalline TiO2 by Ru-
complex photosensitizers (e.g., N3, N719) have been investigated
intensively and power conversion efficiencies (PCEs) higher than
11% under AM 1.5 irradiation have been achieved [4,6,7], whereas
metal-free organic sensitizers have only showed PCEs between 6%
and 9% [8e12]. Nevertheless, organic dyes have many advantages
such as high molar extinction coefficients, ease of customized
molecular design for showing desired photophysical and photo-
chemical properties, low cost without the need for transition
metals, and environment friendliness.
* Corresponding author. Tel.: þ82 2 3290 3140; fax: þ82 2 925 4284.
** Corresponding author. Tel.: þ82 53 810 2521; fax: þ82 53 810 4631.
E-mail addresses: dhchoi8803@korea.ac.kr (D.H. Choi), jaehkim@ynu.ac.kr
(J.H. Kim).
Conversely, it is known that the thickness, composition, and
morphology of the dye layer at the interface on TiO2 plays an
1
These authors contributed equally to this work.
0143-7208/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.dyepig.2013.08.005

M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
987
important role in the performance of DSSCs [13]. If the electrolyte
2.3. Assembly and characterization of DSSCs
ions penetrate the dye layer easily, the recombination rate of
electrons generated at the dye/TiO2 interfaces with the electrolyte
The fabrication method of the DSSCs is as follows. The con-
ducting glass substrate (FTO; TEC8, Pilkington, 8
/cm2, thickness
will increase in the DSSC, reducing open-circuit voltage (Voc
)
U
significantly. If electrolyte ions do not penetrate into the dye layer,
the potential drop will extend over a larger distance on the pho-
toanode composed of TiO2 semiconductors, thus inducing a longer
electron lifetime in DSSCs [14]. Thus, it is obvious that the operation
and efficiency of DSSCs will be strongly influenced by the
morphology of the adsorbed dye molecules, layer thickness, and
chemical structure of photosensitizers.
of 2.3 mm) was cleaned in ethanol by ultrasonication. TiO2 pastes
(TiO2 particle size approximately 20e30 nm) were prepared using
ethyl cellulose (Aldrich), lauric acid (Fluka), and terpineol (Aldrich).
The prepared TiO2 paste was coated on the pre-cleaned glass sub-
strate using a doctor blade and sintered at 450 ꢁC for 30 min. The
thickness of the sintered TiO2 layer was measured with an Alpha-
step IQ surface profiler (KLA Tencor). The other TiO2 paste was re-
coated over the sintered layer using TiO2 particles (250 nm) as a
scattering layer and sintered again at 450 ꢁC for 30 min. The pre-
pared TiO2 film was dipped in a TiCl4 aqueous solution (0.04 M) at
70 ꢁC for 30 min. For dye adsorption, the annealed TiO2 electrodes
were immersed in a dye solution (0.3 mM of dye in DMF) at 30 ꢁC
for 8 ꢀ 12 h. Pt counter electrodes were prepared by thermal
reduction of a thin film formed from 7 mM of H2PtCl6 in 2-propanol
at 400 ꢁC for 20 min. The dye-adsorbed TiO2 electrode and Pt
In this work, we demonstrate the synthesis of organic photo-
sensitizing dyes with varying numbers of phenothiazine (PTZ)-
based chromophores in a molecule. The anchoring property of dye
on the TiO2 surface is critical in achieving high photovoltaic effi-
ciency in DSSCs. The concept of multiple acceptors in a dye was
already known in Ru-complexes and organic dyes which revealed
that the multiple acceptors could enhance the Jsc, but reduce the
Voc in the DSSCs [15]. Thus, we designed and synthesized the new
dyes bearing multiple PTZ-based chromophores which are con-
nected through non-conjugated alkyl chains. PTZ-based chromo-
phores have already provided potential advantages for nonlinear
optical materials and photosensitizing dyes in DSSCs owing to
high electron donability of the PTZ moiety [16]. The photovoltaic
properties of the solar cells composed of new organic dyes were
measured and compared with those of ruthenium dye (N3).
Intriguingly, the tri-chromophoric dendritic dye-based cell
showed much higher PCE values than the other two cells bearing
mono- or di-chromophoric dyes, although the number of dye
molecules anchored onto the TiO2 was almost similar in the three
devices. To investigate these solar cell performances, we measured
decaying curve of Voc, impedance spectrum, and electron life-time,
respectively.
counter electrode were assembled using 60
m
m-thick SurlynÒ
(Dupont 1702) as a bonding agent. A liquid electrolyte was intro-
duced through a pre-punctured hole on the counter electrode. The
electrolyte was composed of 3-propyl-1-methyl-imidazolium io-
dide (PMII, 0.7 M), lithium iodide (LiI, 0.2 M), iodine (I2, 0.05 M),
and t-butylpyridine (TBP, 0.5 M) in acetonitrile/valeronitrile
(85:15). The active areas of the dye-adsorbed TiO2 films were
estimated using a digital microscope camera with image-analysis
software (Moticam 1000). The photovoltaic IeV characteristics of
the prepared DSSCs were measured under 1 sunlight intensity
(100 mW/cm2, AM 1.5), as verified with an AIST-calibrated Si-solar
cell (PEC-L11, Peccell Technologies, Inc.). The incident mono-
chromatic photon-to-current efficiencies (IPCEs) were plotted as a
function of light wavelength using an IPCE measurement instru-
ment (PEC-S20, Peccell Technologies, Inc.). The impedance spectra
(EIS) and open-circuit photovoltage decay (OCVD) curves were
acquired using an electrochemical impedance analyzer (Iviumstat
Tec.).
2. Experimental
2.1. Materials
2.4. Synthesis
All commercially available starting materials and solvents were
purchased from Aldrich, TCI, and ACROS Co. and were used without
further purification unless otherwise stated. HPLC grade toluene
and tetrahydrofuran (THF) were purchased from Samchung
Chemical and distilled from CaH2 immediately before use.
2.4.1. 10-Hexyl-10H-phenothiazine (1)
An oven dried, 250 mL round bottom flask (RBF) was charged
with dimethylformamide (100 mL), 1-bromohexane (10.7 g,
0.065 mol) and sodium hydride (1.8 g, 0.075 mol). Then, a DMF
solution of phenothiazine (10 g, 0.05 mol) was added to the mother
solution, and the reaction mixture was maintained for 10 h at room
temperature. After completion of the reaction, the solution was
neutralized with a dilute HCl aqueous solution. The mixture was
extracted with chloroform/water. The dried solution was concen-
trated. The resulting crude, oily product was purified by column
chromatography on silica gel with hexane. The product was ob-
tained as a colorless liquid. Yield: 11.2 g (79%). 1H NMR (300 MHz,
2.2. Instrumental analysis
All 1H NMR spectra were recorded on a Varian Mercury NMR
300 MHz spectrometer using CDCl3 and DMSO-d6 purchased from
Cambridge Isotope Laboratories, Inc. The redox properties of three
dyes were examined by using cyclic voltammetry (Model: CV-BAS-
Epsilon). The electrolyte solution used was 0.10 M tetrabuty-
lammonium hexafluorophosphate (TBAPF6) in freshly dried DMF.
The Ag/AgCl and Pt wire (0.5 mm in diameter) electrodes were
used as reference and counter electrodes, respectively. The scan
rate was 100 mV/s. The mass spectrum was acquired using a JEOL
HX-110 mass spectrometer. Elemental analysis was performed
with Flash F. A1112 (Automatic Elemental Analyzer, CE. In-
struments, Italy). Infrared spectra were obtained by the Perkin
Elmer Fourier Transformed Infrared (FT-IR) spectrophotometer
with a Miracle single bounce diamond ATR cell from PIKET Tech-
nologies. Dye desorption was conducted by detaching the dye from
TiO2 surface in ethanol/H2O (1:1 v/v) with NaOH (50 mM). The
solution bearing detached dyes was monitored with UVeVis
spectrophotometer.
DMSO-d6):
d
(ppm) 7.20 (d, J ¼ 5.7 Hz, 2H), 7.14 (d, J ¼ 6.9 Hz, 2H),
6.99 (d, J ¼ 8.1 Hz, 2H), 6.91 (t, J ¼ 7.2 Hz, 2H), 3.84 (t, J ¼ 6.9 Hz, 2H),
1.67e1.60 (m, 2H), 1.41e1.27 (m, 2H), 1.23e1.19 (m, 4H), 0.86e0.80
(m, 3H).
2.4.2. 10-Hexyl-10H- phenothiazine-3-carbaldehyde (2)
An oven dried, 250 mL RBF was charged with a solution of DMF
(13.2 g, 0.18 mol) and 1,2-dichloroethane (20 mL). Then, 10-hexyl-
10H-phenothiazine, 1(5.0 g, 0.017 mol) in 1,2-dichloroethane
(20 mL) was slowly added to the mixture for 30 min. Next, POCl3
(10.7 g, 0.07 mol) was added dropwise over 30 min. The mixture
was stirred for 10 h at 90 ꢁC. It was poured into ice water (300 mL)
and neutralized with an aqueous solution of NaOH. The solution

988
M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
was extracted with chloroform. The resulting product was purified
by column chromatography on silica gel with ethyl acetate/hexane
(1:5 v/v). The product was obtained as a yellow solid. Yield: 2.7 g
(0.94 g, 11.1 mmol), and ammonium acetate (0.86 g, 11.1 mmol)
were dissolved in acetic acid (20 mL) under argon. After heating the
solution at 110 ꢁC, a trace amount of acetic acid was added drop-
wise. After 12 h of reaction, the solvent was evaporated to obtain a
dark solid. The resulting product was purified by recrystallization
using methanol. The product was obtained as a black solid. Yield:
1.9 g (42%). Mp: 165e167 ꢁC. FT-IR (KBr, cmꢀ1): 3446 (OeH); 2223
(-CN); 1695 (carboxylic C]O). 1H NMR (300 MHz, DMSO-d6):
(51%). 1H NMR (300 MHz, CDCl3):
d (ppm) 9.79 (s, 1H), 7.63 (d,
J ¼ 8.4 Hz, 1H), 7.58 (s, 1H), 7.17 (d, J ¼ 7.5 Hz, 1H), 7.11 (d, J ¼ 7.2 Hz,
1H), 6.97 (d, J ¼ 7.2 Hz, 1H), 6.92 (d, J ¼ 8.1 Hz, 1H), 6.88 (d,
J ¼ 7.5 Hz, 1H), 3.88 (t, J ¼ 7.2 Hz, 2H), 1.85e1.76 (m, 2H), 1.43e1.30
(m, 6H), 0.87 (m, 3H).
d
(ppm) 8.14 (s, 2H), 7.89 (d, J ¼ 6.6 Hz, 2H), 7.78 (s, 2H), 7.21 (d,
2.4.3. 2-Cyano-3-(10-hexyl-10H-phenothiazine-3-yl)acrylic acid
(1(PTZCA)) (3)
J ¼ 7.2 Hz, 2H), 7.16 (d, J ¼ 8.4 Hz, 2H), 7.13 (d, J ¼ 6.3 Hz, 2H), 7.02 (d,
J ¼ 8.4 Hz, 2H), 6.98 (d, J ¼ 7.8 Hz, 2H), 3.88 (t, J ¼ 7.8 Hz, 4H), 1.66
(m, 4H), 1.41 (m, 4H). HRFAB(m/z): Calculated C38H30N4O4S2,
670.17; Found: 670.41. Anal.: C38H30N4O4S2 (670.17); Calcd: C 68.04,
H 4.51, N 8.35, S 9.56; Found: C 67.68, H 4.51, N 8.12, S 9.87.
In
a dried, 100 mL RBF, 10-hexyl-10H-phenothiazine-3-
carbaldehyde, 2(2.0 g, 6.4 mmol), cyano acetic acid (0.71 g,
8.3 mmol), and ammonium acetate (0.64 g, 8.3 mmol) were dis-
solved in acetic acid (10 mL) under argon. After heating the solution
at 110 ꢁC, a trace amount of piperidine was added dropwise. After
12 h of reaction, the solvent was evaporated to obtain a dark solid.
The resulting product was purified by recrystallization in acetoni-
trile. The product was obtained as a black solid. Yield: 1.9 g (42%).
Melting point (Mp): 155e156 ꢁC. FT-IR (KBr, cmꢀ1): 3425 (OeH);
2221 (-CN); 1685 (carboxylic C]O). 1H NMR (300 MHz, CDCl3):
2.4.7. 10-(6-Bromohexyl)-10H-phenothiazine (7)
An oven dried, 250 mL RBF was charged with DMF (100 mL), 1,6-
dibromohexane (10 g, 0.041 mol) and sodium hydride (0.98 g,
0.041 mol) under argon at 0 ꢁC. Then, a DMF solution of pheno-
thiazine (5.0 g, 0.025 mol) was added to the mother solution, and
the reaction mixture was maintained for 10 h at room temperature.
After completion of the reaction, the solution was neutralized with
a dilute HCl aqueous solution. The mixture was extracted with
chloroform/water, and the organic layer was dried under MgSO4.
After the organic layer was concentrated, the resulting crude, oily
product was purified by column chromatography on silica gel with
chloroform/hexane (1:5 v/v). The product was obtained as a
colorless oil. Yield: 4.5 g (49.7%). 1H NMR (300 MHz, CDCl3):
d
(ppm) 8.09 (s, 1H), 7.91 (d, J ¼ 8.4 Hz, 1H), 7.68 (s, 1H), 7.17 (d,
J ¼ 6.9 Hz,1H), 7.09 (d, J ¼ 7.5 Hz,1H), 6.96 (d, J ¼ 7.5 Hz,1H), 6.87 (d,
J ¼ 7.8 Hz, 2H), 3.88 (t, J ¼ 6.9 Hz, 2H), 1.80 (d, J ¼ 6.9 Hz, 2H), 1.44e
1.31 (m, 6H), 0.88 (m, 3H). HRFAB (m/z): Calculated for C22H22N2O2S,
378.14; Found: 378.33. Anal.: C22H22N2O2S (378.14); Calcd: C 69.81,
H 5.86, N 7.40, S 8.47; Found: C 69.64, H 5.90, N 7.57, S 8.46.
2.4.4. 1,6-Di(10H-phenothiazine-10-yl)hexane (4)
d
(ppm) 7.13 (d, J ¼ 5.1 Hz, 2H), 7.09 (d, J ¼ 7.5 Hz, 2H), 6.98 (d,
An oven dried, 250 mL RBF was charged with DMF (150 mL), 1,6-
dibromohexane (5.0 g, 20.5 mmol) and sodium hydride (1.48 g,
61.5 mmol). Then, a DMF solution of phenothiazine (8.99 g,
45.1 mmol) was added to the mother solution, and the reaction
mixture was maintained for 12 h at room temperature. After
completion of the reaction, the solution was neutralized with a
dilute HCl aqueous solution. The mixture was extracted with
chloroform/water. The dried solution was concentrated. The
resulting crude, oily product was purified by recrystallization using
methanol. The product was obtained as a white powder. Yield: 8.3 g
J ¼ 7.2 Hz, 2H), 6.82 (d, J ¼ 8.1 Hz, 2H), 3.84 (t, J ¼ 6.9 Hz, 2H), 3.33 (t,
J ¼ 6.6 Hz, 2H), 1.87e1.76 (m, 4H), 1.47e1.40 (m, 4H).
2.4.8. 10-(6-Chlorohexyl)-10H-phenothiazine-3-carbaldehyde (8)
An oven dried, 250 mL RBF was charged with a solution of DMF
(10.88 g, 0.15 mol) and 1,2-dichloroethane (20 mL). Then, 10-(6-
bromohexyl)-10H-phenothiazine, 7 (4.5 g, 12.4 mmol) in 1,2-
dichloroethane (20 mL) was slowly added to the mixture for
30 min. Next, POCl3 (7.6 g, 50 mmol) was added dropwise over
30 min. The mixture was stirred for 10 h at 90 ꢁC. It was poured into
ice water (300 mL) and neutralized with an aqueous solution of
NaOH. The solution was extracted with chloroform. The resulting
product was purified by column chromatography on silica gel with
chloroform/hexane (1:3 v/v). The product was obtained as a yellow
(84.3%). 1H NMR (300 MHz, CDCl3):
d
(ppm) 7.15 (d, J ¼ 8.4 Hz, 4H),
7.09 (d, J ¼ 7.5 Hz, 4H), 6.90 (d, J ¼ 8.4 Hz, 4H), 6.81 (d, J ¼ 8.4 Hz,
4H), 3.81 (t, J ¼ 8.4 Hz, 4H), 1.79e1.75 (m, 4H), 1.44 (m, 4H).
2.4.5. 10,100-(Hexane-1,6-diyl)bis(10H-phenothiazine-3-
solid. Yield: 2.5 g (51.6%). 1H NMR (300 MHz, CDCl3):
d (ppm) 9.78(s,
carbaldehyde) (5)
1H), 7.63 (d, J ¼ 8.7 Hz, 1H), 7.59 (s, 1H), 7.17 (d, J ¼ 7.2 Hz, 1H), 7.12
(d, J ¼ 8.7 Hz, 1H), 6.97 (d, J ¼ 7.5 Hz, 1H), 6.92 (d, J ¼ 9.9 Hz, 1H),
6.87 (d, J ¼ 7.5 Hz,1H), 3.90 (t, J ¼ 7.2 Hz, 2H), 3.49 (t, J ¼ 6.3 Hz, 2H),
1.82e1.79 (m, 4H), 1.45 (m, 4H).
An oven dried, 250 mL RBF was charged with a solution of DMF
(13.7 g, 0.187 mol) and 1,2-dichloroethane (20 mL). Then, 1,6-
di(10H-phenothiazine-10-yl)hexane, 4 (10.0 g, 0.024 mol) in 1,2-
dichloroethane (20 mL) was slowly added to the mixture for
30 min. Next, POCl3 (9.56 g, 62.4 mmol) was added dropwise over
30 min. The mixture was stirred for 10 h at 90 ꢁC. It was poured into
ice water (300 mL) and neutralized with an aqueous solution of
NaOH. The solution was extracted with chloroform. The resulting
product was purified by column chromatography on silica gel with
ethyl acetate/hexane (1:3 v/v). The product was obtained as a yel-
2.4.9. 10,100,1000-(6,60,600-(4,40,400-(Ethane-1,1,1-triyl)tris(benzene-
4,1-diyl))tris(oxy)tris (hexane- 6,1-diyl))tris(10H-phenothiazine-3-
carbaldehyde) (9)
10-(6-Chlorohexyl)-10H-phenothiazine-3-carbaldehyde, 8 (2.5
g, 6.4 mmol), 1,1,1-tris(4-hydroxyphenyl)ethane (0.64 g, 2.1 mmol),
K2CO3 (4.15 g, 12.8 mmol), and a catalytic amount of 18-crown-6
were dissolved in DMF (20 mL) under argon. The reaction mixture
was stirred at 110 ꢁC for 24 h. The resulting mixture was filtered,
and the organic layer was diluted with 100 mL of chloroform and
washed with water (200 mL) several times. The organic layer was
dried over MgSO4, and the solvent was removed in vacuo after
filtration. The resulting product was purified by column chroma-
tography on silica gel with ethylacetate/chloroform (1:10 v/v). The
product was obtained as a yellow solid. Yield: 1.3 g (50.1%). 1H NMR
low solid. Yield: 2.4 g (43%). 1H NMR (300 MHz, DMSO-d6):
d (ppm)
9.77 (s, 2H), 7.68 (d, J ¼ 8.4 Hz, 2H), 7.55 (s, 2H), 7.20 (d, J ¼ 8.1 Hz,
2H), 7.15 (d, J ¼ 6.9 Hz, 2H), 7.10 (d, J ¼ 8.7 Hz, 2H), 7.02 (d, J ¼ 8.7 Hz,
2H), 6.97 (d, J ¼ 7.2 Hz, 2H), 3.89 (t, J ¼ 6.9 Hz, 4H), 1.65 (m, 4H), 1.38
(m, 4H).
2.4.6. 3,30-(10,100-(Hexane-1,6-diyl)bis(10H-phenothiazine-10,3-
diyl))bis(2-cyanoacrylic acid) (2(PTZCA)) (6)
In a dried 100 mL RBF, 10,100-(hexane-1,6-diyl)bis(10H-pheno-
thiazine-3-carbaldehyde), 5 (2.0 g, 3.7 mmol), cyano acetic acid
(300 MHz, CDCl3):
(s, 3H), 7.16 (d, J ¼ 7.8 Hz, 3H), 7.10 (d, J ¼ 7.8 Hz, 3H), 6.96 (d,
d
(ppm) 9.77 (s, 3H), 7.62 (d, J ¼ 8.4 Hz, 3H), 7.56

M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
989
J ¼ 8.1 Hz, 9H), 6.88 (d, J ¼ 8.1 Hz, 6H), 6.73 (d, J ¼ 7.2 Hz, 6H), 3.89
(m, 12H), 2.09 (s, 3H), 1.84e1.75 (m, 12H), 1.49 (m, 12H).
3(PTZCA) was synthesized by using aldehyde-functionalized PTZs,
which were attached to 1,1,1-tris(4-hydroxyphenyl)ethane as a core
unit. Each chromophore in 2(PTZCA) and 3(PTZCA) was connected
through a nonconjugated alkylene spacer which acted as a blocking
moiety to prevent the recombination of electrons on the TiO2 sur-
face into electrolytes.
2.4.10. (2E,20E,200E)-3,30,300-(10,100,1000-(6,60,600-(4,40,400-(Ethane-
1,1,1-triyl)tris(benzene-4,1-diyl))tris(oxy)tris(hexane-6,1-diyl))
tris(10H-Phenothiazine-10,3-diyl))tris(2-cyanoacrylic acid)
(3(PTZCA)) (10)
In a dried, 100 mL RBF, 9 (1.3 g, 1.05 mmol), cyano acetic acid
(0.45 g, 5.25 mmol) and ammonium acetate (0.40 g, 5.25 mmol)
were mixture under argon. After heating the solution at 110 ꢁC, a
trace amount of acetic acid was added dropwise. After 12 h of re-
action, the solvent was evaporated to obtain a red solid. The
resulting product was purified by recrystallization in acetonitrile.
The product was obtained as a red solid. Yield: 0.8 g (53.3%). Mp:
182e185 ꢁC. FT-IR (KBr, cmꢀ1): 3448 (OeH); 2220 (-CN); 1685
3.2. Optical and electrochemical properties
Fig. 1A and B display the UVeVis absorption spectra of the
dyes in DMF solution and on the TiO2 layer, respectively. The
detailed optical properties of the dyes are summarized in
Table 1. The maximum absorption wavelength (lmax) corre-
sponding to the
pep* transition of each of the three dyes
appeared at around 420 nm; however, the molar extinction
(carboxylic C]O). 1H NMR (300 MHz, DMSO-d6):
d (ppm) 8.09 (s,
coefficients ( max) of 1(PTZCA), 2(PTZCA), and 3(PTZCA) were
3
3H), 7.87 (d, J ¼ 8.1 Hz, 3H), 7.77 (s, 3H), 7.22 (d, J ¼ 7.8 Hz, 3H), 7.17
(d, J ¼ 5.7 Hz, 3H), 7.13 (d, J ¼ 6.0 Hz, 3H), 7.07 (d, J ¼ 8.1 Hz, 3H),
6.98 (d, J ¼ 7.5 Hz, 3H), 6.88 (d, J ¼ 8.7 Hz, 6H), 6.76 (d, J ¼ 8.7 Hz,
6H), 3.94e3.88 (m, 12H), 2.00 (s, 3H), 1.69e1.66 (m, 12H), 1.44 (m,
12H). HRFAB (m/z): Calculated for C86H78N6O9S3 (1434.50); Found:
1434.80. Anal.: C86H78N6O9S3 (1434.50); Calcd.: C 71.94, H 5.48, N
5.85, S 6.70; Found: C 71.25, H 5.23, N 6.12, S 6.93.
determined to be 1.2 ꢂ 104, 2.1 ꢂ 104, and 3.0 ꢂ 104 Mꢀ1 cmꢀ1
,
respectively. For a larger number of chromophores induced in a
molecule, a higher molar extinction coefficient was observed,
following a linear proportionality. The formation of aggregates
of dyes on the TiO2 surface induced the bathochromic shift of
absorption spectra, in comparison with that of the solution state
(Table 1) [17].
The highest occupied molecular orbital (HOMO) and lowest
unoccupied molecular orbital (LUMO) energy levels are typically
used to estimate the efficiency of electron injection from the dye in
the excited state to the TiO2, as well as the efficiency of dye
regeneration in DSSCs [18]. The HOMO energy levels of the dyes
were determined by measuring the oxidation potentials in cyclic
voltammograms (Table 1). The optical band gaps (Eogpt) were
employed to determine the LUMO energy levels. The results
showed that all the dyes had more positive HOMO levels than the
redox potential (0.4 V) of Iꢀ/I3ꢀ in the electrolyte: þ0.827, þ0.836,
and þ0.822 V for 1(PTZCA), 2(PTZCA), and 3(PTZCA), respectively.
This implies that the oxidized dyes could be regenerated easily by
the Iꢀ/Iꢀ3 redox couple.
3. Results and discussion
3.1. Synthesis and characterization
Three different organic photosensitizers containing cyanoacrylic
acid as an electron acceptor and PTZ as an electron donor were
prepared (Scheme 1). The synthetic procedures used to access
1(PTZCA), 2(PTZCA), and 3(PTZCA) photosensitizers are illustrated
in Schemes 2 and 3. The compounds 1, 4, and 7 were synthesized
from phenothiazine and bromoalkane in the presence of sodium
hydride. The formylation of each compound was conducted to yield
2, 5, and
8 through VilsmeiereHaack reaction. Knoevenagel
condensation with cyanoacetic acid in the presence of piperidine
provided the final desired products. In particular, 2(PTZCA) was
synthesized by using dimerized PTZ with a hexylene bridge and
The dyes containing different numbers of chromophores were
used to fabricate DSSCs in order to study their photovoltaic per-
formance. To measure the dye adsorption, the TiO2 electrodes were
Scheme 1. Structure of photosensitizers containing a PTZ-based chromophore, and Ru-complex dyes (N3 and N719).

990
M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
Scheme 2. Synthesis of organic photosensitizers (3) 1(PTZCA) and (6) 2(PTZCA): i) POCl3, DMF, 1,2-dichloroethane; ii) cyanoacetic acid, ammonium acetate, acetic acid.
adsorption process. The carbonyl stretching band in eCOOH
observed at w1690 cmꢀ1 in the neat sample disappeared in dyes
adsorbed on the TiO2 surface and shifted to a higher wave number
region at 1735 cmꢀ1, which suggests the presence of a monodentate
binding mode of dye molecules on the TiO2 surface during the dye
adsorption [19b,19c].
3.3. Photovoltaic performance of DSSCs
The synthesized PTZ-based photosensitizers were used for
fabricating DSSCs in order to study their current densityevoltage
characteristics under almost identical numbers of chromophores
adsorbed on the TiO2 electrode. The photovoltaic performances of
the DSSCs including three different dyes (i.e., 1(PTZCA), 2(PTZCA),
and 3(PTZCA)) were compared using the variation of flow current
with the bias voltage, incident photon-to-current (IPCE) effi-
ciencies, impedance, and electron life-time analysis.
The photovoltaic properties of the DSSCs were measured and
compared with those of a DSSC containing N3 dye, as shown in
Fig. 3A, and the measured parameters are summarized in Table 2.
Fig. 3B shows the IPCE spectra of the DSSCs containing different
organic dyes under simulated solar light (AM 1.5G, 100 mW/cm2).
More importantly, we loaded the number of dye molecules in order
to provide nearly the same number of chromophores on the TiO2
surface, which made it possible to compare the photovoltaic per-
formance of three devices with different dye structures. To calcu-
late the amount of dye adsorbed on the TiO2 film, we obtained
molar desorptivities for 1(PTZCA), 2(PTZCA), and 3(PTZCA) of 9.48,
5.50, and 3.80 (ꢂ10ꢀ8) mol cmꢀ2, respectively, as listed in Table 2.
The molar desorptivity was measured at the maximum absorbance
wavelength of all dyes in 50 mM NaOH [20]. Thus, it is interesting
that the numbers of chromophores on the TiO2 surfaces containing
Scheme 3. Synthesis of organic photosensitizer 3(PTZCA): i) POCl3, DMF, 1,2-
dichloroethane; ii) 1,1,1-tris(4-hydroxyphenyl)ethane, K2CO3, 18-crown-6, DMF; iii)
cyanoacetic acid, ammonium acetate, acetic acid.
immersed in each dye solution in DMF (0.3 mM) at 30 ꢁC for about
8e12 h. It is known that the interaction and adsorption mechanism
of anchoring groups in dyes on the TiO2 surface directly affects the
electron transfer and performance in DSSCs [19a]. Thus, we
observed the attenuated total reflection Fourier transform infrared
(ATR-FT-IR) spectra to investigate the adsorption mechanism of
dyes on the TiO2 surface, and the results are shown in Fig. 2. The
wave number assigned to eC^N stretching in organic dyes
(w2220 cmꢀ1) did not change in the spectra before or after the dye
adsorbed on the TiO2 surface, which indicates that no interaction
occurred between the cyano moieties and the TiO2 during the
(B)
(A)
Fig. 1. Absorption spectra of synthesized dyes in the DMF solution (A) and adsorbed dyes on TiO2 films (B).

M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
991
Table 1
Absorption and electrochemical parameters of the organic dyes.
Dyes
lmaxa/nm (
3
ꢂ 10ꢀ4 Mꢀ1 cmꢀ1
)
lmaxb/nm
E0-0c/eV
Eox/V
Eox ꢀ E0-0/eV
HOMO/eV
LUMO/eV
1(PTZCA)
2(PTZCA)
3(PTZCA)
424 (1.2)
417 (2.1)
413 (3.0)
426
423
427
2.505
2.535
2.574
0.827
0.836
0.822
ꢀ1.678
ꢀ1.699
ꢀ1.752
ꢀ5.153
ꢀ5.148
ꢀ5.141
ꢀ2.648
ꢀ2.613
ꢀ2.567
a
Measured in the DMF solution.
b
c
Absorption maximum on TiO2 film.
Determined from the intersection of the absorption and emission spectra in the DMF solution.
different organic dyes were similar, although the performance of
the DSSCs was observed to be significantly different.
The molar extinction coefficient of 3(PTZCA) was three times
higher than that of 1(PTZCA) (Fig. 1B) but the amount of the
adsorbed 3(PTZCA) on the TiO2 was 2/5-fold of that of the adsorbed
1(PTZCA), indicating that the similar light harvesting efficiency
(LHE) values should be expected for all the dyes in the DSSCs.
However, a higher photon-to-current efficiency was achieved in the
DSSC based on the 3(PTZCA) photosensitizer compared to either
1(PTZCA) or 2(PTZCA), primarily because of the higher Jsc and Voc
(A)
(B)
Fig. 2. ATR-FT-IR spectra of 1(PTZCA) (a), 2(PTZCA) (b), and 3(PTZCA) (c). A: neat dye samples, B: dyes adsorbed on the TiO2.
(B)
(A)
(D)
(C)
Fig. 3. (A) Current densityevoltage characteristics for DSSCs containing organic dyes under simulated solar light (AM 1.5, 100 mW/cm2). (B) IPCE curves for DSSCs based on organic
dyes: (a) N3, (b) 1(PTZCA), (c) 2(PTZCA), and (d) 3(PTZCA). *Sample conditions: Blocking layer, TiO2 paste:B32, organic dye 0.5 mM e PMII (0.7 M) þ LiI (0.2 M) þ I2 (0.05 M) þ TBP
(0.5 M) in ACN/VN ¼ 85:15 with a Pt electrode solution (7 mM), 60
m
m SurlynÒ, and a doctor blade (2T). (C) Measured impedance spectra of DSSC at the light illumination. (D) The
Bode-phase plots for the DSSC cells. Square: 1(PTZCA), Circle: 2(PTZCA), Triangle: 3(PTZCA).

992
M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
Table 2
Photovoltaic performance and results of impedance analysis of DSSCs. (Scattering layer: 250 nm, 4
m
m þ TiCl4, cell areas: 0.24 cm2).
Dyes
Jsc/mA cmꢀ1
Voc/V
FF/%
h/%
Amount of dye/mol cmꢀ2
R1a/
U
R2b/
U
R3c/
U
sd/ms
N3
17.02
9.03
10.32
11.96
0.678
0.631
0.628
0.681
60.07
62.79
60.22
60.15
6.93
3.57
3.91
4.90
e
e
e
e
e
1(PTZCA)
2(PTZCA)
3(PTZCA)
9.48 ꢂ 10ꢀ8
5.50 ꢂ 10ꢀ8
3.80 ꢂ 10ꢀ8
16.38
14.58
13.61
22.88
15.82
14.30
34.77
24.93
22.84
11.2
15.9
17.4
a
R1 is FTO interface resistance.
b
c
R2 is due to the resistance between the counter electrode and the electrolyte.
R3 is possibly originated from the backward charge transfer from TiO2 to the electrolyte and the electron conduction porous TiO2 film.
s
d
is life-time of an electron in DSSC.
values (Table 2). The DSSC based on 3(PTZCA) produced
a
variations in DSSCs composed of various organic photosensitizers
allowed us to compare the electron life-times in three different
DSSCs [21].
maximum IPCE of approximately 70% at 460 nm, while DSSCs based
on the other dyes produced a maximum IPCE of approximately 60%
for both 1(PTZCA) and 2(PTZCA).
We observed and analyzed the EIS under light illumination
conditions, as shown in Fig. 3C, and the results are summarized in
Table 2. There are two well-defined semicircles in the high fre-
quency (f w1000 Hz) and medium frequency (f w10 Hz) regions. The
ohmic serial resistance (R1) corresponds to the resistance between
the electrolyte and the fluorine doped-tin-oxide (FTO) substrate,
and R2 and R3 are the resistances retarding the charge transfer
process occurring at the Pt counter electrode and at the interfaces of
the TiO2/dye/electrolyte, respectively. Thus, the R3 value possibly
originates from the charge transfer process from the TiO2 to the
electrolyte and from the electron conduction in porous TiO2 films.
The R3 values significantly decreased for the 3(PTZCA) dye, indi-
cating much faster electron transport. It also implies that the alkyl
chains present in 3(PTZCA) retarded the backward electron transfer
from the TiO2 to the electrolyte and facilitated forward electron
transport from the TiO2 to the FTO substrate. The recombination rate
caused by the backward electron transfer was estimated by the
Bode-phase spectra for the DSSCs, as shown in Fig. 3D. The electron
life-times were calculated and are illustrated in Table 2. The peak
shift from high frequency to low frequency reveals longer electron
life-time because the frequency (fmed) can be related tothe inverse of
Table 1 shows that the LUMO level slightly increased as the
number of chromophores in the dye molecule increased. This led to
a higher driving force for the electron injection from the dyes to the
conduction band of the TiO2, indicating that the 3(PTZCA) dye
could provide more efficient electron extraction pathways and
superior electron injection efficiency (hinj). The charge collection
efficiency (hcc) was also influenced by the number of chromophores
in the dye molecular structure, which will be discussed later.
It is noteworthy that the Voc for the DSSC based on the 3(PTZCA)
photosensitizer, which contains three isolated chromophores in a
molecule, was higher than that of either 1(PTZCA) or 2(PTZCA).
3.4. Electrochemical impedance spectroscopy
Electrochemical impedance spectra (EIS) analysis was per-
formed to study the electrode kinetics and interfacial charge
transfer process in the DSSCs. The precise analysis of the impedance
(A)
(B)
electron life-time (
s
e) inTiO2 films as (se ¼ 1/2p med
f
) [22,23]. A much
longer electron life-time (17.4 ms) was obtained for the 3(PTZCA)-
based device, compared with 11.2 ms and 15.9 ms for the 1(PTZCA)-
and 2(PTZCA)-based devices, respectively.
The observed electron life-time of the 3(PTZCA)-based device
showed a significant difference compared with that of either
1(PTZCA)- or 2(PTZCA)-based devices, as shown in Table 2. The
Fig. 4. (A) Voc decay curves of DSSCs with an anchoring process of PTZ-based organic
photosensitizers, depending on the number of chromophores recorded during relax-
ation from an illuminated quasi-equilibrium state to the dark equilibrium. (B) The
electron life-time derived from Equation (1) as a function of Voc. (a) 1(PTZCA), (b)
2(PTZCA), (c) 3(PTZCA).
Fig. 5. Schematic representation of the anchoring process of newly synthesized pho-
tosensitizers as a function of the number of chromophores: (a) 1(PTZCA), (b) 2(PTZCA),
and (c) 3(PTZCA).

M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
993
results suggest that the 3(PTZCA) photosensitizer containing three
“site-isolated” chromophores in a molecule hampered the contact of
electrolyte ions with charges generated from the photosensitizers.
Thus, the recombination rate of electrons on the TiO2 surface
decreases significantly, which extends the electron life-time on the
TiO2 surface, thus supporting the higher Voc value in 3(PTZCA)-
based DSSCs compared to that of either of the other two photo-
sensitizers [24]. To confirm the electron life-times in the DSSCs, we
also measured open-circuit photovoltage decay (OCVD) curves
[25,26]. The OCVD technique is a method of monitoring the sub-
sequent decay of photovoltage Voc after turning off the illumination
in a steady state. Fig. 4A shows that Voc decay curves of DSSCs with
organic PTZ-based photosensitizers were recorded during relaxa-
tion from an illuminated quasi-equilibrium state to the dark state.
Fig. 4B shows electron life-times calculated from the OCVD curves,
according to the following equation:
(KETEP) and Dongjin Semichem Co., Ltd. grant funded by the Korea
government Ministry of Knowledge Economy (No. 200930
1001002D). DH Choi particularly thanks to the support from Basic
Science Research Program through the National Research Founda-
tion of Korea (NRF) funded by the Ministry of Education
(NRF20100020209).
References
[1] O’Regan B, Grätzel M. A low-cost, high-efficiency solar cell based on dye-
sensitized colloidal TiO2 films. Nature 1991;353:737e40.
[2] Grätzel M. Photoelectrochemical cells. Nature 2001;414:338e44.
[3] Nazeeruddin MK, Kay A, Rodicio I, Humphry-Baker R, Mueller E, Liska P, et al.
Conversion of light to electricity by cis-X2bis( 2,20-bipyridyl-4,40-dicarbox-
ylate)ruthenium(II) charge-transfer sensitizers (X ¼ C1ꢀ, Brꢀ, Iꢀ, CNꢀ, and
SCNꢀ) on nanocrystalline TiO2 electrodes. J Am Chem Soc 1993;115:6382e90.
[4] Nazeeruddin MK, Zakeeruddin SM, Humphry-Baker R, Jirousek M, Liska P,
Vlachopoulos N, et al. Acid-base equilibria of (2,20-Bipyridyl-4,40-dicarboxylic
acid)ruthenium(II) complexes and the effect of protonation on charge-transfer
sensitization of nanocrystalline titania. Inorg Chem 1999;38:6298e305.
[5] Grätzel M. Solar energy conversion by dye-sensitized photovoltaic cells. Inorg
Chem 2005;44:6841e51.
[6] Nazeeruddin MK, De Angelis F, Fantacci S, Selloni A, Viscardi G, Liska P, et al.
Combined experimental and DFT-TDDFT computational study of photo-
electrochemical cell ruthenium sensitizers. J Am Chem Soc 2005;127:16835e47.
[7] Grätzel M. Dye-sensitized solar cells. J Photochem Photobiol C 2003;4:145e53.
[8] Horiuchi T, Miura H, Sumioka K, Uchida S. High efficiency of dye-sensitized
solar cells based on metal-free indoline dyes. J Am Chem Soc 2004;126:
12218e9.
ꢀ
ꢁꢀ1
kB
se ¼ ꢀ
T
e
dVoc
dt
(1)
where kBT is the thermal energy, e is the positive elementary charge,
and dVoc/dt is the first derivative of the open-circuit voltage. The
photovoltaic decay rate is directly related to electron life-time (se
)
because as the illumination of the DSSC at the open circuit is inter-
rupted, excess electrons are diminished through recombination
[27]. The 3(PTZCA) dye exhibited much longer electron life-times
[9] Wang ZS, Cui Y, Hara K, Dan-Oh Y, Kasada C, Shinpo A. A high-light-har-
vesting-efficiency coumarin dye for stable dye-sensitized solar cells. Adv
Mater 2007;19:1138e41.
(ses) compared to either 1(PTZCA) or 2(PTZCA), which is in good
[10] Hara K, Sato T, Katoh R, Furube A, Ohga Y, Shinpo A, et al. Molecular design of
agreement with the results obtained from Fig. 4B. These results can
suggest that the alkyl chains connecting the three PTZ-based chro-
mophores in one dye molecule significantly suppressed the
recombination rate, resulting in the much longer electron life-time.
Therefore, 3(PTZCA) provided much faster electron transport and
significantly prolonged electron life-time, giving rise to the signifi-
cantly improved electron collection efficiency (hcc) because the
value of hcc is determined by using two factors such as electron
transport and recombination rates. Consequently, the improved Jsc
and Voc of the DSSC with the 3(PTZCA) can be mainly attributed to
the improved hinj and hcc values.
Fig. 5 is a schematic representation of the possible anchoring
process of the three photosensitizers on the TiO2 surface, manifesting
the dependence of “site-isolation” of the chromophore on the
adsorbed geometry of the TiO2 surface. The dye 3(PTZCA) covered
the TiO2 surface effectively, thereby possibly acting as a blocking
layer for the recombination of electrons into the electrolyte.
coumarin dyes for efficient dye-sensitized solar cells.
2003;107:597e606.
J Phys Chem B
[11] Hwang S, Lee JH, Park C, Lee H, Kim C, Park C, et al. A highly efficient organic
sensitizer for dye-sensitized solar cells. Chem Commun 2007:4887e9.
[12] Park SS, Won YS, Choi YC, Kim JH. Molecular design of organic dyes with
double electron acceptor for dye-sensitized solar cell. Energy & Fuels 2009;23:
3732e6.
[13] Gregg BA. Interfacial processes in the dye-sensitized solar cell. Coor Chem Rev
2004;248:1215e24.
[14] Marquet P, Andersson G, Snedden A, Kloo L, Atkin R. Molecular scale char-
acterization of the titania-dye-solvent interface in dye-sensitized solar cells.
Langmuir 2010;26:9612e6.
[15] (a) Shang H, Luo Y, Guo X, Huang X, Zhan X, Jiang K, et al. The effect of
anchoring group number on the performance of dye-sensitized solar cells.
Dyes Pigment 2010;87:249e56;
(b) Hara K, Sugihara H, Tachibana Y, Islam A, Yanagida M, Sayama K, et al. Dye-
sensitized nanocrystalline TiO2 solar cells based on ruthenium(II) phenan-
throline complex photosensitizers. Langmuir 2001;17:5992e9;
(c) Hara K, Horiuchi H, Katoh R, Singh LP, Sugihara H, Sayama K, et al. Effect of
the ligand structure on the efficiency of electron injection from excited Ru-
phenanthroline complexes to nanocrystalline TiO2 films.
2002;106:374e9;
J Phys Chem B
(d) Park H, Bae E, Lee J, Park J, Choi W. Effect of the anchoring group in Ru-
bipyridyl sensitizers on the photoelectrochemical behavior of dye-sensitized
TiO2 electrodes: carboxylate versus phosphonate linkages. J Phys Chem B
2006;110:8740e9;
4. Conclusion
(e) Yang C, Chen H, Chuang Y, Wu C, Chen C, Liao S, et al. Characteristics of
triphenylamine-based dyes with multiple acceptors in application of dye-
sensitized solar cells. J Power Sources 2009;188:627e34.
We demonstrate a series of phenothiazine based dyes contain-
ing different numbers of chromophores in a molecule were
employed as photosensitizers for a DSSC. Although these dyes
showed similar optical and electrochemical properties, the DSSC
device based on 3(PTZCA) exhibited much higher PCE value than
the other two devices bearing either 1(PTZCA) or 2(PTZCA) dyes.
These DSSC results were supported by studying the decay curve of
Voc, impedance spectrum, and electron life-time.
It is noteworthy that the performance of DSSCs can be improved
only by sheaving the ends of the dye molecules together to form the
dendritic photosensitizer, even though the number of adsorbed dye
molecules is almost identical on the semiconductor electrode.
[16] (a) Cho MJ, Lee SK, Jin JI, Choi DH, Dalton LR. Electro-optic property of
chromophore-terminated trifunctional dendrimer in a guestehost system.
Thin Solid Films 2006;515:2303e9;
(b) Iqbal Z, Wu WQ, Kuang DB, Wang L, Meier H, Cao D. Phenothiazine-based
dyes with bilateral extension of
p-conjugation for efficient dye-sensitized
solar cells. Dyes Pigment 2013;96:722e31;
(c) Kim SH, Sakong C, Chang JB, Kim B, Ko MJ, Kim DH, et al. The effect of N-
substitution and ethylthio substitution on the performance of phenothiazine
donors in dye-sensitized solar cells. Dyes Pigment 2013;97:262e71.
[17] Mooney III WF, Brown PE, Russell JC, Costa SB, Pedersen LG, Whitten DG. Photo-
chemistryandphotophysicsofsurfactanttrans-stilbenesinsupportedmultilayers
and films at the air-water interface. J Am Chem Soc 1984;106:5659e67.
[18] (a) Hara K, Wang ZS, Sato T, Furube A, Katoh R, Sugihara H, et al. Oligothiophene-
containing coumarin dyes for efficient dye-sensitized solar cells. J Phys Chem B
2005;109:15476e82;
(b) Kim DH, Lee JK, Kang SO, Ko JJ. Molecular engineering of organic dyes con-
taining N-aryl carbazole moiety for solar cell. Tetrahedron 2007;63:1913e22;
(c) Cho MJ, Park SS, Yang YS, Kim JH, Choi DH. Molecular design of donore
acceptor-type cruciform dyes for efficient dyes-sensitized solar cells. Synth Met
2010;160:1754e60.
Acknowledgments
This work was supported by the New & Renewable Energy of the
Korea Institute of Energy Technology Evaluation and Planning

994
M.S. Kim et al. / Dyes and Pigments 99 (2013) 986e994
[19] (a) Lee KE, Gomez MA, Elouatik S, Demopoulos GP. Further understanding of
the adsorption mechanism of N719 sensitizer on anatase TiO2 films for DSSC
applications using vibrational spectroscopy and confocal Raman imaging.
Langmuir 2010;26:9575e83;
[22] Sun S, Gao L, Liu Y. Enhanced dye-sensitized solar cell using graphene-TiO2 pho-
toanode prepared by heterogeneous coagulation. Appl Phys Lett 2010;96:083113.
[23] Kern R, Sastrawan R, Ferber J, Stangl R, Luther J. Modeling and interpretation
of electrical impedance spectra of dye solar cells operated under open-circuit
conditions. Electrochim Acta 2002;47:4213e25.
(b) Qin P, Yang X, Chen R, Sun L. Influence of
p-conjugation units in organic
dyes for dye-sensitized solar cells. J Phys Chem C 2007;111:1853e60;
(c) Leon CP, Kador L, Peng B, Thelakkat M. Characterization of the adsorption
of Ru-bpy dyes on mesoporous TiO2 films with UVꢀVis, Raman, and FTIR
spectroscopies. J Phys Chem B 2006;110:8723e30.
[24] Wang Q, Ito S, Grätzel M, Fabregat-Santiago F, Mora-Sero I, Bisquert J, et al.
Characteristics of high efficiency dye-sensitized solar cells. J Phys Chem B
2006;110:25210e21.
[25] Bisquert J, Zaban A, Greenshtein M, Mora-Sero I. Determination of rate con-
stants for charge transfer and the distribution of semiconductor and elec-
trolyte electronic energy levels in dye-sensitized solar cells by open-circuit
photovoltage decay method. J Am Chem Soc 2004;126:13550e9.
[26] Zaban A, Greenshtein M, Bisquert J. Determination of the electron lifetime in
nanocrystalline dye solar cells by open-circuit voltage decay measurements.
ChemPhysChem 2003;4:859e64.
[20] (a) Heo JH, Oh JW, Ahn HI, Lee SB, Cho SE, Kim MR, et al. Synthesis and
characterization of triphenylamine-based organic dyes for dye-sensitized
solar cells. Synth Met 2010;160:2143e50;
(b) Uam HS, Jung YS, Jun Y, Kim KJ. Relation of Ru(II) dye desorption from TiO2
film during illumination with photocurrent decrease of dye-sensitized solar
cells. J Photochem Photobiol A Chem 2010;212:122e8.
[21] Lee KM, Suryanarayanan V, Ho KC. Influences of different TiO2 morphologies
and solvents on the photovoltaic performance of dye-sensitized solar cells.
J Power Sources 2009;188:635e41.
[27] Fisher AC, Peter LM, Ponomarev EA, Walker AB, Wijayantha KGU. Intensity
dependence of the back reaction and transport of electrons in dye-sensitized
nanocrystalline TiO2 solar cells. J Phys Chem B 2000;104:949e58.