Tuning of spacer groups in organic dyes for efficient inhibition of charge recombination in dye-sensitized solar cells

Article abstract of DOI:10.1016/j.dyepig.2012.04.002

Most of the studied organic dyes for dye-sensitized solar cell have exhibited low photovoltage due to fast charge recombination. Here, we report on synthetic strategy of organic dye to get high photovoltage. Phenothiazine (PTZ) moieties are incorporated between electron donating triphenylamine (D) and electron accepting cyanoacrylic acid (A) to enhance non-planarity. Compared to one PTZ incorporation (D-PTZ-A), hypsochromic shift in absorption is observed by incorporation of two PTZs (D-PTZ1-PTZ2-A) due to transition from HOMO-1 to LUMO as confirmed by density functional theory (DFT) calculation. D-PTZ1-PTZ2-A shows higher photovoltage (0.804 V) than D-PTZ-A (0.781 V), those which are even higher than that of the ruthenium-based N719-snesitized one (0.775 V). Little difference in electron transport rate is observed for PTZ derivatives and N719, while electron life time is increased in the order of D-PTZ1-PTZ2-A > D-PTZ-A > N719. Retardation of electron recombination is responsible for the high photovoltage, which is associated with bent conformation induced by PTZ spacer group.

Full text of DOI:10.1016/j.dyepig.2012.04.002

Dyes and Pigments 95 (2012) 134e141
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Tuning of spacer groups in organic dyes for efficient inhibition of charge
recombination in dye-sensitized solar cells
,
Mi-Jeong Kim a, Yong-Jae Yu b, Jong-Hyung Kim b, Young-Sam Jung b, Kwang-Yol Kay b
,
**
,
Soo-Byung Ko a, Chang-Ryul Lee a, In-Hyuk Jang a, Young-Uk Kwon c, Nam-Gyu Park a
*
a School of Chemical Engineering and Department of Energy Science, Sungkyunkwan University, Suwon 440-746, Republic of Korea
b SolarSys Co., Ltd., BI Center, Ajou University, Suwon 443-749, Republic of Korea
c Department of Chemistry, Sungkyunkwan University, Suwon 440-746, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Most of the studied organic dyes for dye-sensitized solar cell have exhibited low photovoltage due to fast
charge recombination. Here, we report on synthetic strategy of organic dye to get high photovoltage.
Phenothiazine (PTZ) moieties are incorporated between electron donating triphenylamine (D) and
electron accepting cyanoacrylic acid (A) to enhance non-planarity. Compared to one PTZ incorporation
(D-PTZ-A), hypsochromic shift in absorption is observed by incorporation of two PTZs (D-PTZ1ePTZ2-A)
due to transition from HOMO-1 to LUMO as confirmed by density functional theory (DFT) calculation.
D-PTZ1ePTZ2-A shows higher photovoltage (0.804 V) than D-PTZ-A (0.781 V), those which are even
higher than that of the ruthenium-based N719-snesitized one (0.775 V). Little difference in electron
transport rate is observed for PTZ derivatives and N719, while electron life time is increased in the order
of D-PTZ1ePTZ2-A > D-PTZ-A > N719. Retardation of electron recombination is responsible for the high
photovoltage, which is associated with bent conformation induced by PTZ spacer group.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 28 February 2012
Received in revised form
3 April 2012
Accepted 4 April 2012
Available online 11 April 2012
Keywords:
Dye-sensitized solar cell
Open-circuit voltage
Organic dye
Phenothiazine
Non-planarity
Recombination
1. Introduction
potential. The conduction-band energy can be modified by inter-
facial environment. Thus, photovoltage can be tuned by different
Dye-sensitized solar cell (DSSC) has been considered as
a promising next generation solar cell because of cost-effective
technology and fascinating appearance thanks to colorful dyes
and semi-transparency [1]. Nanocrystalline TiO2 film, redox elec-
trolyte and dye are inevitable constituent in preparing DSSC.
Among these constituents, dye is key component since it absorbs
light and generates electrons. Band gap energy and molar extinc-
tion coefficient of dye molecule are most important physical
quantities because photocurrent density is mostly affected by such
optoelectronic parameters. It is obvious that photocurrent will
increase as band gap energy decreases and molar extinction coef-
ficient increases. However, when considering DSSC as a photo-
electrochemical solar cell, that is, solid-liquid junction structure,
modification of dye can also have influence on photovoltage. Pho-
tovoltage is determined by the difference between conduction-
band edge (more precisely, Fermi energy) of TiO2 and redox
dye attached on TiO2 surface because of the interfacial modifica-
tion. Dye is generally classified into organometallic dye and organic
dye. Ruthenium is commonly used as a central metal in organo-
metallic dyes and N719-coded dye is a typical example [2]. In case
of organic dye, the molecular structure is constructed based on
electron donor, electron acceptor and bridging unit to connect
them. Electron donor plays a role in pushing electrons in order for
the photo-excited electrons to be injected into TiO2. Various elec-
tron donors have been investigated to get high photovoltaic
performance, including an electron-hole transporting triphenyl-
amine unit [3], a coumarin unit with improved electron injection
rate [4], a synthetically facile carbazole unit [5], a cost-effective
indoline unit [6] and a fluorene unit less degradable than aryl-
amine unit [7]. Although organic dyes have advantages over Ru
complex dyes because of relatively low cost and high molar
extinction coefficient, most of the studied organic dyes showed
lower photovoltage than N719. For instance, compared with the
open-circuit voltage (VOC) of more than 0.8 V for N719 dye, organic
dyes showed relatively lower VOC ranging from 0.55 V to 0.75 V for
triphenylamine dyes [8e10] coumarin dyes [11,12] carbazole dyes
[5] indoline dyes [13,14] and fluorene dyes [7,15]. It has been argued
* Corresponding author. Tel.: þ82 312907241; fax: þ82 312907272.
** Corresponding author. Tel.: þ82 312192602; fax: þ82 312191592.
E-mail addresses: kykay@ajou.ac.kr (K.-Y. Kay), npark@skku.edu (N.-G. Park).
0143-7208/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dyepig.2012.04.002

M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
135
that the low voltages observed generally from organic dyes might
be due to fast recombination kinetics [16,17]. A strong interaction
between organic dye and iodine in electrolyte was proposed to be
responsible for accelerating recombination [18,19] and the lack of
electron donating moiety in the oxidized organic dye was also
proposed to have influence on fast recombination, compared to
N719 with electron donating NCS ligand [12,20]. In addition, dye
aggregation due to planar or sterically less-hindered structure was
also proposed as one of major factors for lowering voltage [21,22]. It
is therefore required to design molecular structure of organic dye to
reduce recombination and/or prevent dye aggregation.
D-PTZ1ePTZ2-A are also compared to that of the conventional
ruthenium-based N719 dye.
2. Experimental
2.1. Materials and reagents
3-Bromo-10-hexyl-10H-3-phenothiazine-carboxaldehyde (1)
[29], N-(4-vinylphenyl)-N0,N0-diphenylamine (2) [30] and 10-hexyl-
3-vinyl-10H-phenothiazine (3) [31] were synthesized according to
the reported methods. Reagents and solvents were purchased as
reagent grade and used without further purification. All reactions
were performed using dry glassware under nitrogen atmosphere.
Analytical thin layer chromatography (TLC) was carried out on
Merck 60 F254 silica gel plate and column chromatography was
performed on Merck 60 silica gel (230e400 mesh).
Recently, heterocyclic phenothiazine (PTZ), originally used in
drug application [23,24], has been adopted as a novel electron
donor in organic dye because PTZ contains electron-rich sulfur-
nitrogen heteroatoms and its ring has sterically hindered non-
planar butterfly conformation, incorporation of PTZ derivatives in
organic dye backbone is expected to inhibit molecular aggregation
[25e28]. Yang and co-workers synthesized PTZ based organic dye
series and compared their solar energy-to-electricity conversion
efficiency [25]. Among them, 3-(10-butyl-10H-phenothiazin-3-yl)-
2-cyanoacrylic acid (T2-1) based DSSC showed an efficiency of 5.5%
with short-circuit current density (JSC) of 10.9 mA/cm2, VOC of
0.712 V and fill factor (FF) of 0.71. Another attempt to modify PTZ
units for DSSCs was made by Hua and co-workers, where PTZ-
thiophene dyes with hole transporting group such as triphenyl-
amine and 1,1,2-triphenylethene were synthesized [27]. But these
dyes showed only 2.48e4.41% conversion efficiency due to low VOC
ranging 0.505e0.592 V. Despite a variety of molecular engineering
of the PTZ species, VOC of PTZ-contained dye has not been
dramatically improved as expected over those of other organic
dyes. For overcoming these limitations, we design and synthesize
new organic dyes for high voltage DSSC based on PTZ moiety,
where two different structural modifications are attempted to
induce bent conformation and steric hindrance as well. Fig. 1 shows
the synthesized organic dyes, where one PTZ is placed as a space-
r group between electron donating triphenylamine (D) and elec-
tron accepting carboxylic acid (A) (D-PTZ-A) and two PTZs
are introduced between electron donor and electron acceptor
(D-PTZ1ePTZ2-A). Photocurrent-voltage characteristics, electron
transport and electron life time of D-PTZ-A and D-PTZ1ePTZ2-A
dyes are compared. Photovoltaic properties of D-PTZ-A and
2.2. Analytical instruments and measurements
Melting points were determined on an Electrothemal IA 9000
series melting point apparatus. NMR spectra were recorded on
a Varian Mercury-400 (400 MHz) spectrometer with TMS peak as
reference. MALDI-TOF MS spectra were recorded with an Applied
Biosystems Voyager-DE-STR. Elemental analyses were performed
with a Perkin-Elmer 2400 analyzer.
2.3. Synthesis of dyes
7-[{4-(N,N-Diphenylamino)phenyl}ethen-3-yl]-10-hexyl-
10H-phenothiazine-3-carboxaldehyde (4). To
a solution of
compound 2 (398 mg, 1.46 mmol) and compound 1 (569 mg,
1.46 mmol) in DMF (20 ml) were added Pd(OAc)2 (65 mg
0.29 mmol), Bu4NBr (941 mg, 2.91 mmol), K2CO3 (703 mg,
5.83 mmol) and the mixture was stirred at 105 ꢀC for 16 h. After
cooling, the solution was poured into water (50 ml) and the crude
product was extracted with dichloromethane (3 ꢁ 50 ml). The
combined organic layer was dried over Na2SO4 and the solvent was
removed under reduced pressure. The residue was chromato-
graphed on silica gel with dichloromethane/n-hexane (1:2 v/v) to
give compound 4 (452 mg, 53.1%) in a yellowish solid. Melting
point was 54 ꢀC. 1H NMR (400 MHz, CDCl3):
d
¼ 9.78 (s, 1H), 7.61
(dd, J ¼ 2 Hz, 2 Hz, 1H), 7.48 (s, 1H), 7.35e7.21 (m, 8H), 7.13e7.00
(m, 8H), 6.92e6.81 (m, 4H), 3.88 (t, 2H), 1.80 (m, 2H), 1.43
(m, 2H), 1.31 (m, 4H), 0.88 (t, 3H). Anal. calcd (%) for C39H36N2OS: C
80.65, H 6.25, N 4.82. Found: C 80.38, H 6.61, N 4.61.
n-C6H13
N
CN
S
COOH
3-[7-[{4-(N,N-diphenylamino)phenyl}ethen-3-yl]-10-hexyl-
10H-phenothiazinyl]-2-cyanoacrylic acid (D-PTZ-A). A mixture of
compound 4 (290 mg, 0.5 mmol) and cyanoacetic acid (425 mg,
5 mmol) in dry chloroform (30 ml) was refluxed in the presence of
piperidine (1 ml) for 16 h. After cooling, the solutionwas poured into
water (50 ml) and the crude product was extracted with chloroform
(3 ꢁ 50 ml). The combined organic layer was dried over Na2SO4 and
the solvent was removed under reduced pressure. The crude product
was purified by column chromatography with dichloromethane/
MeOH (10:1 v/v) to give compound D-PTZ-A (226 mg, 69.7%) in a red
solid. Melting point was 123 ꢀC. 1H NMR (400 MHz, DMSO-d6):
N
D-PTZ-A
n-C6H13
N
S
N
COOH
CN
d
¼ 7.97 (s,1H), 7.80 (d, J ¼ 8.8 Hz,1H), 7.73 (s,1H), 7.44 (d, J ¼ 8.4 Hz,
S
2H), 7.37 (m, 2H), 7.30 (m, 4H), 7.12e7.01 (m,11H), 6.92 (m, 2H), 3.91
(t, 2H), 1.68(m, 2H), 1.38 (m, 2H), 1.24 (m, 4H), 0.81 (t, 3H). MS
(MALDI-TOF): m/z 647.8 [Mþ]. Anal. Calcd (%) for C42H37N3O2S: C
77.87, H 5.76, N 6.49. Found: C 77.53, H 5.99, N 6.37.
N
n-C6H13
7-{(10-Hexyl-10H-phenothiazin-3-yl)ethen-3-yl}-10-hexyl-
D-PTZ1-PTZ2-A
10H-phenothiazine-3-carboxaldehyde (5). To
a solution of
compound 3 (7.92 g, 25.6 mmol) and compound 2 (10 g, 25.6 mmol)
in DMF (100 ml) were added Pd(OAc)2 (1.14 g, 5 mmol), Bu4NBr
Fig. 1. Molecular structures of 10-hexyl-10H-phenothiazine incorporated D-PTZ-A and
D-PTZ1ePTZ2-A dyes.

136
M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
(16.5 g, 51.1 mmol), K2CO3 (12.3 g, 102 mmol) and the mixture was
stirred at 105 ꢀC for 16 h. After cooling, the solution was poured into
water (150 ml) and the crude product was extracted with
dichloromethane (3 ꢁ 150 ml). The combined organic layer was
dried over Na2SO4 and the solvent was removed under reduced
pressure. The residue was chromatographed on silica gel with
dichloromethane/n-hexane (1:4 v/v) to give compound 5 (11.3 g,
depositing TiO2 layer on fluorine-doped tin oxide (FTO) substrate,
FTO glasses (Pilkington, TEC-8, 8 U/sq) were washed in ethanol
using an ultrasonic bath for 10 min. The FTO layer was first covered
with 0.1 M Ti(IV) bis(ethyl acetoacetato)-diisopropoxide (Aldrich)
in 1-butanol (Aldrich) solution by spin-coating method, on which
the nanocrystalline TiO2 paste was deposited using a doctor-blade
technique. After the TiO2 films were heated at 550 ꢀC for 1 h, the
sides of TiO2 films was trimmed to 9 mm wide and 5 mm long. The
thickness of TiO2 films was measured by Alpha step profiler (KLA
Tencor). The annealed TiO2 electrodes were immersed in ethanol
71.5%) in red oil.1H NMR (400 MHz, CDCl3):
d
¼ 9.80 (s,1H), 7.62 (dd,
J ¼ 2 Hz, 2 Hz,1H), 7.57 (d, J ¼ 2 Hz,1H), 7.26e7.19 (m, 4H), 7.14e7.11
(m, 2H), 6.92e6.78 (m, 7H), 3.85 (m, 4H), 1.81(m, 4H), 1.43 (m, 4H),
1.30 (m, 8H), 0.78 (m, 6H). Anal. Calcd (%) for C39H42N2OS: C 75.69, H
6.84, N 4.53. Found: C 75.35, H 7.09, N 4.69.
containing
0.5 mM
(E)-2-cyano-3-(7-((E)-4-(diphenylamino)
styryl)-10-hexyl-10H-phenothiazin-3-yl)acrylic acid (D-PTZ-A)
and 0.5 mM (E)-2-cyano-3-(7-((E)-2-(7-((E)-4-(diphenylamino)
styryl)-10-hexyl-10H-phenothiazin-3-yl)vinyl)-10-hexyl-10H-phe-
nothiazin-3-yl)acrylic acid (D-PTZ1ePTZ2-A) for 5 h at 40 ꢀC. For
comparison, ethanolic solution of 0.5 mM N719 dye (Esolar,
TBA2[RuL2(NCS)2], where L and TBA represent 4-carboxylic acid-40-
carboxylate-2,20-bipyridine and tetrabutylammonium, respec-
tively) was used for N719 sensitization. Pt counter electrode was
prepared by spreading a droplet of 7.0 mM H2PtCl6 in 2-propanol
on top of a FTO substrate and heated at 400 ꢀC for 20 min. The
7-{(7-Bromo-10-hexyl-10H-phenothiazin-3-yl)ethen-3-yl}-
10-hexyl-10H-phenothiazine-3-carboxaldehyde (6). To a solution
of compound 5 (11.3 g, 18.2 mmol) in DMF (50 ml), was added
N-bromosuccinimide (3.41 g, 19.1 mmol). The reaction mixture was
stirred at room temperature for 5 h, and then water (150 ml) was
added to the solution. The reaction solution was extracted with
dichloromethane (3 ꢁ150 ml). The combined organic layer dried
over MgSO4 and evaporated under reduced pressure. The crude
product was purified by column chromatography dichloromethane/
n-hexane (1:3 v/v) to give compound 6 (8.62 g, 67.8%) in red oil. 1H
two electrodes were sealed with 25-mm thick Surlyn (Meltonix
NMR (400 MHz, CDCl3):
d
¼ 9.78 (s,1H), 7.64e7.56 (m, 2H), 7.25e7.11
1170-25, Solaronix). Redox electrolyte was introduced through
a small hole drilled in the counter electrode. The electrolyte was
composed of 0.7 M 1-methyl-3-propylimidazolium iodide (MPII),
0.05 M I2 (Aldrich, 99.8%), 0.2 M LiI, 0.5 M 4-tert-butylpyridine
(TBP) (Aldrich, 96%) in acetonitrile (Fluka, 99.9%) and valeronitrile
(Aldrich, 99.5%) (85:15 v/v).
(m, 5H), 6.91e6.78 (m, 7H), 3.85 (m, 4H), 1.80(m, 4H), 1.41 (m, 4H),
1.29 (m, 8H), 0.89 (m, 6H). Anal. Calcd. (%) for C39H41N2OS2Br: C
67.13, H 5.92, N 4.01. Found: C 67.26, H 6.27, N 3.88.
7-[[7-[{4-(N,N-Diphenylamino)phenyl}ethen-3-yl]-10-hexyl-
10H-phenothiazinyl]-ethen-3-yl]-10-hexyl-10H-phenothiazine-
3-carboxaldehyde (7). To a solution of compound 2 (94 mg,
0.35 mmol) and compound 6 (244 mg, 0.35 mmol) in DMF (10 ml)
were added Pd(OAc)2 (15 mg, 0.067 mmol), Bu4NBr (225 mg,
0.69 mmol), K2CO3 (168 mg, 1.39 mmol) and the mixture was stir-
red at 105 ꢀC for 16 h. After cooling, the solution was poured into
water (50 ml) and the crude product was extracted with
dichloromethane (3 ꢁ 50 ml). The combined organic layer was
dried over Na2SO4 and the solvent was removed under reduced
pressure. The residue was chromatographed on silica gel with
dichloromethane/n-hexane (1:2 v/v) to give compound 7 (198 mg,
2.5. Characterization of photoelectrochemical and photovoltaic
properties
Photocurrent and voltage were measured from a solar simulator
(Oriel Sol 3A class AAA) equipped with 450 W Xenon lamp (New-
port 6279NS) and a Keithley 2400 source meter. Light intensity was
adjusted with the NREL-calibrated Si solar cell having KG-2 filter for
approximating one sun illumination (100 mW/cm2). The cell was
covered with an aperture mask to measure photocurrent and
voltage accurately. The incident photon-to-current efficiency (IPCE)
spectra were measured under DC mode using an IPCE system
(PV Measurements, Inc.), where 75 W Xe light source was used for
monochromatic light. The electrochemical impedance spectra were
obtained at open-circuit voltage under aperture mask and one sun
illumination with a potentiostat (Metrohm Autolab B.V.). Cyclic
voltammetry was carried out with a potentiostat (Metrohm Auto-
lab B.V.) using a three electrode cell in 0.1 M tetrabutylammonium
perchlorate ((C4H9)4NClO4) in methylene chloride and scan rate
was 100 mV/s. Pt sheet, Au electrode and Ag/AgCl electrode were
used as a counter electrode, a working electrode and a reference
electrode, respectively. The redox potentials are reported with
a ferrocenium/ferrocene (Fcþ/Fc) redox couple as an internal
reference [33]. The absorbance was recorded by a UVeVis spec-
trophotometer (Agilent 8453). The emission spectra were recorded
using a photoemission yield spectrometer (RIKEN KEIKI). The
concentration of dyes solution for absorbance and emission was
0.01 mM in methylene chloride. The time constants for photo-
injected electron transport and recombination were measured
according to published procedures [32].
63.8%) in red oil. 1H NMR (400 MHz, CDCl3):
d
¼ 9.78 (s, 1H),
7.64e7.57 (m, 3H), 7.33e7.05 (m, 17H), 6.91e6.79 (m, 10H), 3.85
(m, 4H), 1.81(m, 4H), 1.43 (m, 4H), 1.31 (m, 8H), 0.87 (m, 6H). Anal.
Calcd. (%) for C59H57N3OS2: C 79.78, H 6.47, N 4.73. Found: C 79.46,
H 6.78, N 4.51.
3-[7-[[7-[{4-(N,N-Diphenylamino)phenyl}ethen-3-yl]-10-
hexyl-10H-phenothiazinyl]-ethen-3-yl]-10-hexyl-10H-pheno-
thiazin-3-yl]-2-cyanoacrylic acid (D-PTZ1ePTZ2-A). A mixture of
compound 7 (198 mg, 0.22 mmol) and cyanoacetic acid (187 mg,
2.2 mmol) in dry chloroform (30 ml) was refluxed in the presence of
piperidine (1 ml) for 16 h. After cooling, the solution was poured
into water (50 ml) and the crude product was extracted with
chloroform (3 ꢁ 50 ml). The combined organic layer was dried over
Na2SO4 and the solvent was removed under reduced pressure. The
crude product was purified by column chromatography dichloro-
methane/MeOH (10:1 v/v) to give compound D-PTZ1ePTZ2-A
(122 mg, 58.3%) in a red solid. Melting point was 114 ꢀC. 1H NMR
(400 MHz, DMSO-d6):
d
¼ 7.89 (s, 2H), 7.77 (d, J ¼ 8.4 Hz, 2H), 7.68
(s, 1H), 7.57 (m, 2H), 7.34e7.28 (m, 6H), 7.20e6.90 (m, 17H),
6.81e6.73 (m, 4H), 3.89 (m, 4H), 1.67(m, 4H), 1.38 (m, 4H), 1.24
(m, 8H), 0.82 (m, 6H). MS: m/z 955.0 [Mþ]. Anal. Calcd. (%) for
C62H58N4O2S2: C 77.95, H 6.12, N 5.86. Found: C 77.62, H 6.46, N 5.61.
3. Results and discussion
2.4. Fabrication of dye-sensitized solar cells
3.1. Synthesis
Anatase TiO2 nanoparticle and its non-aqueous paste were
prepared according to the method reported elsewhere [32]. For
Synthesis of the organic dyes D-PTZ-A and D-PTZ1ePTZ2-A is
performed by following the steps as depicted in Scheme 1. Each

M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
137
step of the reaction sequence proceeds smoothly and efficiently to
give a good or moderate yield of the product (see Section 2 for the
synthetic details). The Heck reaction [34] of brominated pheno-
thiazine aldehyde 1 with vinyl group of compound 2 produces
triphenylamine substituted phenothiazine aldehyde 4 in 53.1%
yield. The aldehyde 4 is reacted with cyanoacetic acid in the pres-
ence of piperidine to produce the dye D-PTZ-A in 69.7% yield. The
dye D-PTZ1ePTZ2-A is synthesized starting from compound 1,
which is reacted with the vinylphenothiazine 3 under Heck reac-
tion condition to give compound 5 in 71.5% yield. Subsequent
bromination with NBS (N-bromosuccinimide) produces compound
6 in 67.8% yield. The aldehyde 7 is synthesized by the Heck type
cross coupling reaction of compound 6 with compound 2 in 63.8%
yield. Finally, Knoevenagel condensation of aldehyde 7 and cya-
noacetic acid give the dye D-PTZ1ePTZ2-A in 58.3% yield.
The sensitizers D-PTZ-A and D-PTZ1ePTZ2-A and their
precursor compounds 4e7 are highly soluble in aromatic solvents
(i.e., toluene, o-dichlorobenzene and benzonitrile) and other
common organic solvents (i.e., acetone, ethanol, CH2Cl2, CHCl3,
DMF, DMSO and THF). The structure and purity of the new
compounds are confirmed by 1H NMR and elemental analysis. 1H
NMR spectra of these compounds are consistent with the proposed
structures, showing the expected features with the correct inte-
gration ratios. The MALDI-TOF mass spectra provide a direct
evidence for the structures of D-PTZ-A and D-PTZ1ePTZ2-A
showing a singly charged molecular ion peak at m/z ¼ 647.8 and
955.0, respectively, that matches the calculated value for the
molecular weight of each compound.
(
3
¼ 15,930 Mꢂ1 cmꢂ1). Compared to D-PTZ-A, D-PTZ1ePTZ2-A
shows hypsochromic shift and decreases its extinction coefficient,
which is related to structure, associated with electron delocaliza-
tion [35,36], and photo-excitation (see details from time
dependent-density functional theory (TD-DFT) calculation below).
It can be seen from the emission spectra that there is relatively
large Stocks shift for the dye D-PTZ1ePTZ2-A as compared to
that of D-PTZ-A, which underlines that the emitted photon has
much less energy than the absorbed photon for the case of
D-PTZ1ePTZ2-A. Such a large shift in the absorption and emission
bands for the D-PTZ1ePTZ2-A dye may be attributed to large
change in the geometrical distortion of the molecule in the excited
state [37], associated with double PTZ spacer molecules. From both
absorption and emission spectra, band gap of D-PTZ-A is deter-
mined to be 2.55 eV from intersection wavelength of 487 nm and
that of D-PTZ1ePTZ2-A is 2.48 eV from 500 nm.
To determine highest occupied molecular orbital (HOMO)
energy levels, cyclic voltammetric measurements are performed.
Fig. 3 shows cyclic voltammogram for D-PTZ-A and D-PTZ1ePTZ2-
A, where 1.0 mM of each dye is prepared in methylene chloride
and 0.1 M tetrabutylammonium perchlorate is used as
supporting electrolyte. The HOMO energy levels of D-PTZ-A and
D-PTZ1ePTZ2-A are figured out from the onset oxidation potential
of cyclic voltammogram. The HOMO energy level of D-PTZ-A is
determined to be 1.02 V versus saturated calomel electrode (SCE)
(ꢂ5.76 eV versus vacuum), and that of D-PTZ1ePTZ2-A lies at
0.84 V versus SCE (ꢂ5.58 eV versus vacuum). Based on the band gap
data observed from absorption-emission spectra and HOMO ener-
gies, lowest unoccupied molecular orbital (LUMO) energy levels are
determined to be ꢂ1.53 V and ꢂ1.62 V (versus SCE) for D-PTZ-A
and D-PTZ1ePTZ2-A, respectively. Energetic data for D-PTZ-A and
D-PTZ1ePTZ2-A are listed in Table 1.
3.2. Optical and electrochemical properties
Fig. 2 shows absorption and emission spectra of D-PTZ-A and D-
PTZ1ePTZ2-A. Absorption peaks at 317 nm (molar extinction
coefficient
and 476 nm (
3
¼ 40,360 Mꢂ1 cmꢂ1), 384 nm
(
3
¼ 31,180 Mꢂ1 cmꢂ1
)
3.3. Photovoltaic performance
3
¼ 18,680 Mꢂ1 cmꢂ1) appear from D-PTZ-A dye
solution and D-PTZ1ePTZ2-A shows three peaks at 260 nm
Fig. 4 shows photocurrent-voltage curves and IPCE spectra for
(
3
¼ 38,500 Mꢂ1 cmꢂ1), 319 nm (
3
¼ 25,350 Mꢂ1 cmꢂ1) and 468 nm
the (D-PTZ-A)-sensitized and (D-PTZ1ePTZ2-A)-sensitized DSSCs,
Scheme 1. Synthesis of D-PTZ-A and D-PTZ1ePTZ2-A; (a) compound 2, Pd(OAc)2, Bu4NBr, K2CO3, DMF, 105 ꢀC, 16 h, 53.1% for 3, 63.8% for 7. (b) Cyanoacetic acid, piperidine,
chloroform, reflux, 16 h, 69.7% for D-PTZ-A, for 53.8% for D-PTZ1ePTZ2-A. (c) compound 3, Pd(OAc)2, Bu4NBr, K2CO3, DMF, 105 ꢀC, 16 h, 71.5%. (d) NBS, DMF, rt, 5 h, 67.8%.

138
M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
Table 1
Electrochemical properties of D-PTZ-A and D-PTZ1ePTZ2-A dyes.
E0e0 (eV)b
ELUMOc (V)
a
Dye
Epa (V)
Epc (V)
EHOMO (V)
versus SCE
versus SCE
D-PTZ-A
D-PTZ1ePTZ2-A
1.30
0.94
0.75
0.75
1.02
0.84
2.55
2.46
ꢂ1.53
ꢂ1.62
The conduction-band level of the TiO2 electrode and the redox potential of Iꢂ/I3ꢂ
were estimated to be ꢂ0.7 V versus SCE and 0.2 V versus SCE, respectively [38,39].
a
HOMO level was measured by EHOMO ¼ (Epa ꢂ Epc)/2.
b
E0e0 was estimated from intersection between absorbance and emission.
c
LUMO level was estimated by subtracting E0e0 from HOMO.
sensitized one. It is noticed that VOC of 0.781 V for the (D-PTZ-A)-
sensitized DSSC and 0.804 V for the (D-PTZ1ePTZ2-A)-sensitized
DSSC are higher than that for the N719-sensitized one (0.775 V)
at the given 4.8-
even at the increased film thickness of about 14
line under layer: 10 m, light scattering overlayer: 4
m
m thick TiO2 film. Such the tendency is preserved
m (nanocrystal-
m), where
m
m
m
D-PTZ-A and D-PTZ1ePTZ2-A show VOC of 0.746 V and 0.759 V,
respectively, whereas VOC of N719 is 0.713 V (Table 2). The higher
VOCs for the organic dyes with PTZ spacer group are probably
related to the recombination kinetics at TiO2/electrolyte interface,
which will be discussed in detail with transient photovoltage and
impedance spectra.
To gain a further insight into the effect of the introduction of PTZ
on photovoltaic performance, DFT is used to calculate the electronic
structures and TD-DFT is adopted to investigate electronic transi-
tion from the ground state to the excited state, where the compu-
tational calculations are performed using Gaussian09 software
package [40]. B3LYP exchange-correlation functional [41] and
6e31G(d) basis set [42] in DFT calculation are applied for the
geometry optimization. The optimized geometries of both D-PTZ-A
and D-PTZ1ePTZ2-A are shown in Fig. 5. The optimized geomet-
rical parameters are quite reasonable for the X-ray structure of PTZ
species [43], indicating that the geometry is bent along the
Fig. 2. Absorption (black line) and emission (blue line) spectra of D-PTZ-A and
D-PTZ1ePTZ2-A dyes (0.01 mM solution in methylene chloride). (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version
of this article.)
along with those of N719-sensitized one. Photovoltaic parameters
are summarized in Table 2. Photocurrent of (D-PTZ-A)-sensitized
DSSC is higher than that of (D-PTZ1ePTZ2-A)-sensitized one,
which is due to higher molar extinction coefficient and more
delocalized electron distribution in HOMO state as well (see iso-
density surface plot in Fig. 6). VOC is 23 mV higher for the
(D-PTZ1ePTZ2-A)-sensitized DSSC than for the (D-PTZ-A)-
Fig. 4. (a) Photocurrent densityevoltage curves and (b) incident photon-to-current
conversion efficiency (IPCE) spectra for (D-PTZ-A)-, (D-PTZ1ePTZ2-A)- and N719-
Fig. 3. Cyclic voltammogram of D-PTZ-A and D-PTZ1ePTZ2-A (1 mM solution in
methylene chloride).
sensitized DSSCs. TiO2 film thickness was 4.8 mm.

M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
139
Table 2
Comparison of short-circuit photocurrent density (JSC), open-circuit voltage (VOC),
fill factor (FF) and conversion efficiency (s) for (D-PTZ-A)-, (D-PTZ1ePTZ2-A)- and
N719-sensitized solar cells.a
Dye
JSC
VOC (V) FF
h
(%) Area
(cm2)
Film thickness
m)
4.8
(mA/cm2)
(m
D-PTZ-A
D-PTZ1ePTZ2-A
N719
D-PTZ-A
D-PTZ1ePTZ2-A 12.2
10.7
9.1
9.8
0.781
0.804
0.775
0.746
0.759
0.713
0.55 4.6
0.56 4.1
0.73 5.5
0.58 6.1
0.58 5.3
0.69 7.9
0.442
0.461
0.401
0.457 14.5
0.400
14.1
N719
16.1
0.455
a
An aperture black mask was used during measurement at AM 1.5 G 1 sun light
illumination (100 mA/cm2). Electrolyte used was composed of 0.7 M 1-methyl-3-
propylimidazolium iodide (MPII), 0.05 M I2 (Aldrich, 99.8 %), 0.2 M LiI, 0.5 M 4-
tert-butylpyridine (TBP) (Aldrich, 96 %) in acetonitrile (Fluka, 99.9 %) and valeroni-
trile (Aldrich, 99.5 %) (85:15 v/v).
nitrogenesulfur (NeS) axis for PTZ species. The dihedral angle
between the two planes of the phenyl group is about 161ꢀ, leading
to a crooked geometry. At these results, we can expect that the
sterically more hindered D-PTZ1ePTZ2-A dye will be favorable for
counteracting the dye aggregation, which can be effective for
higher VOC than D-PTZ-A [44e46].
Fig. 5. Optimized molecular structures of D-PTZ-A and D-PTZ1ePTZ2-A at the B3LYP/
6-31G(d) level theory.
The calculated frontier molecular orbitals and electronic struc-
tures of D-PTZ-A and D-PTZ1ePTZ2-A are shown in Fig. 6. The first
three single excited transition parameters of the dyes are summa-
rized in Table 3. The electron density distribution in HOMO level of
D-PTZ-A is spread over triphenylamine through PTZ. The HOMO-1
(the first lower orbital than HOMO) and HOMO-2 (the second lower
orbital than HOMO) states are delocalized from PTZ to ethylene
linker. At LUMO level, the electron density distribution is concen-
trated on the cyanoacrylic acid unit. For the D-PTZ-A dye, the
TD-DFT calculation leads to a lowest transition at 2.21 eV, which
corresponds to an excitation from HOMO to LUMO. As compared to
experimental absorption maximum in Fig. 2, the absorption peak at
476 nm (2.61 eV) is attributed to the pep* transition from HOMO to
LUMO. The calculated transition is red-shifted by about 0.4 eV
because it is not properly captured by TD-DFT calculation
employing current exchange-correlation functional [7]. The reason
Fig. 6. Isodensity surface plot (isodensity contour ¼ 0.02) of D-PTZ-A and D-PTZ1ePTZ2-A.

140
M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
Table 3
TD-DFT calculated electronic transition configurations, excitation energy and oscil-
lator strengths (>0.01) for the single
D-PTZ1ePTZ2-A.
/ single transitions of D-PTZ-A and
Dye
Transition configurations
Excitation
energy (nm, eV)
Oscillator
strength
D-PTZ-A
HOMO / LUMO (97%)
HOMO-1 / LUMO (2%)
HOMO-1 / LUMO (92%)
HOMO / LUMO (3%)
559.8 (2.21)
0.4700
472.0 (2.63)
0.1648
HOMO / LUMO þ 1 (2%)
HOMO / LUMO þ 2 (93%)
HOMO-1 / LUMO (2%)
HOMO-3 / LUMO (3%)
HOMO / LUMO (97%)
HOMO-1 / LUMO (3%)
HOMO-1 / LUMO (85%)
HOMO-2 / LUMO (9%)
HOMO / LUMO (3%)
402.1 (3.08)
1.0433
Fig. 8. (a) Nyquist plots from the impedance measurement. umax represents the
frequency at the maximum imaginary resistance of the Nyquist plot. (b) VOC as
D-PTZ1ePTZ2-A
603.9 (2.05)
518.0 (2.39)
0.2742
0.3894
a function of the logarithm of the inverse of umax
.
N719 < D-PTZ-A < D-PTZ1ePTZ2-A, which indicates that electron
life time increases as going from N719 to D-PTZ1ePTZ2-A. Electron
life time is an order of magnitude higher for D-PTZ1ePTZ2-A than
for N719. Thus, compared to N719, higher VOCs for the organic dyes
with PTZ space group are due to slow charge recombination.
Electrochemical impedance measurements are performed to
confirm the relation between VOC and the charge recombination
rate. The frequency at the maximum imaginary resistance of the
second semicircle (umax) in the Nyquist plot (Fig. 8a) is related to
back reaction constant and VOC was reported to br proportional to
ln(1/umax) [47]. As can be seen in Fig. 8b, the observed VOCs are
linearly proportional to ln(1/umax), which indicates that the open-
circuit voltages for D-PTZ-A, D-PTZ1ePTZ2-A and N719 are related
to the interfacial charge transfer rate, associated with electron
recombination rate. The synthesized D-PTZ1ePTZ2-A dye retards
back reaction more effectively than D-PTZ-A and N719, which is
well consistent with the result observed from transient photo-
voltage measurement.
HOMO-2 / LUMO (79%)
HOMO-1 / LUMO (12%)
HOMO / LUMO þ 1 (6%)
472.6 (2.62)
0.3266
for the difference between calculation ob observation results is that
experimental results are for the dye molecules in the solution
where intermolecular interactions between the dye molecules
themselves and solvent-dye molecular interactions exist while
TD-DFTcalculation is generally performed on single molecule in gas
phase. For the case of D-PTZ1ePTZ2-A, the LUMO state is quite
similar to that of D-PTZ-A but electrons in the HOMO level are
merely delocalized from triphenylamine to the first PTZ moiety
(PTZ1).
As shown in Table 3, these disconnection between HOMO and
LUMO causes to decrease the intensity for the lowest transition at
2.05 eV. Compared to lowest transition, the second transition
occurs from HOMO-1 to LUMO at 2.39 eV and the intensity is much
higher. This is related to the electron distribution in HOMO-1 level,
where electrons spread over from triphenylamine to the second
PTZ moiety (PTZ2). Such the differences in electron distribution
and excited transition property can explain well the hypsochromic
shift in absorption spectra for D-PTZ1ePTZ2-A in Fig. 2 and the
larger photocurrent for D-PTZ-A.
In order to discuss in detail about high VOCs observed in D-PTZ-
A and D-PTZ1ePTZ2-A dyes, time constants for electron transport
and recombination are compared. No significant difference is seen
in the electron transport rate (Fig. 7a), indicating that bent
conformation of organic dye has little influence on electron trans-
port, while substantial change is observed in time constant
for electron recombination as can be seen in Fig. 7b. Time constant
for electron recombination increases in the following order
4. Conclusions
We introduced one or two phenothiazine (PTZ) spacer group
between electron donor and electron acceptor to induce bent
conformation of organic dye. As a result, compared to the reported
organic dyes, high voltages were obtained and those were even
higher than that of the ruthenium-based N719 dye. Incorporation
of two PTZs led to slower recombination than one PTZ, which
might be related to enhancement of non-planarity and thereby
enhancement of steric hindrance against electron back reaction.
Non-planar geometry in organic dye was proved to be one of
important factors affecting photovoltage.
Acknowledgements
This work was supported by the National Research Foundation
of Korea (NRF) grant funded by the Ministry of Education, Science
and Technology (MEST) of Korea under contracts no. 2011-0016441,
2011-0030359 and R31-2008-10029 (WCU program) and the Korea
Institute of Energy Technology Evaluation and planning (KETEP)
grant funded by the Ministry of Knowledge Economy under
contract no. 20103020010010.
References
[1] O’Regan B, Gräzel M. A low-cost, high-efficiency solar cell based on dye-
sensitized colloidal TiO2 films. Nature 1991;353:737e40.
[2] Nazeeruddin MK, Humphry-Baker R, Liska P, Gräzel M. Investigation of
sensitizer adsorption and the influence of protons on current and voltage of
a dye-sensitized nanocrystalline TiO2 solar cell. Journal of Physical Chemistry
B 2003;107:8981e7.
Fig. 7. Time constants for (a) electron transport and (b) recombination as a function of
light intensity, represented by photocurrent density. TiO2 film thickness was 4.8 mm.

M.-J. Kim et al. / Dyes and Pigments 95 (2012) 134e141
141
[3] Zhang GL, Bala H, Cheng YM, Shi D, Lv XJ, Yu QJ, et al. High efficiency and
[24] Levy L, Tozer TN, Tuck LD, Loveland DB. Stability of some phenothiazine free
radicals. Journal of Medicinal Chemistry 1972;15:898e905.
stable dye-sensitized solar cells with an organic chromophore featuring
a binary
p
-conjugated spacer. Chemical Communications; 2009:2198e200.
[25] Tian HN, Yang XC, Chen RK, Pan YZ, Li L, Hagfeldt A, et al. Phenothiazine
derivatives for efficient organic dye-sensitized solar cells. Chemical Commu-
nications; 2007:3741e3.
[26] Cho M-J, Park S-S, Yang Y-S, Kim J-H, Choi D-H. Molecular design of donor-
acceptor-type cruciform dyes for efficient dyes-sensitized solar cells.
Synthetic Metals 2010;160:1754e60.
[27] Wu WJ, Yang JB, Hua JL, Tang J, Zhang L, Long YT, et al. Efficient and stable
dye-sensitized solar cells based on phenothiazine sensitizers with thiophene
units. Journal of Materials Chemistry 2010;20:1772e9.
[28] Kim SH, Kim HW, Sakong C, Namgoong J, Park SW, Ko MJ, et al. Effect of five-
membered heteroaromatic linkers to the performance of phenothiazine-
based dye-sensitized solar cells. Organic Letters 2011;13:5784e7.
[29] Cho MJ, Kim JY, Kim JH, Lee SH, Dalton LR, Choi DH. Heterocyclic nonlinear
optical chromophores composed of phenothiazine or carbazole donor
and 2-cyanomethylene-3-cyano-4,5,5-trimethyl-2,5-dihydrofuran acceptor.
Bulletin of the Korean Chemical Society 2005;26:77e84.
[4] Hara K, Sayama K, Ohga Y, Shinpo A, Suga S, Arakawa H. A coumarin-deriv-
ative dye sensitized nanocrystalline TiO2 solar cell having a high solar-energy
conversion efficiency up to 5.6%. Chemical Communications; 2001:569e70.
[5] Wang Z-S, Koumura N, Cui Y, Takahashi M, Sekiguchi H, Mori A, et al. Hex-
ylthiophene-functionalized carbazole dyes for efficient molecular photovol-
taics: tuning of solar-cell performance by structural modification. Chemistry
of Materials 2008;20:3993e4003.
[6] Horiuchi T, Miura H, Uchida S. Highly-efficient metal-free organic dyes for
dye-sensitized solar cells. Chemical Communications; 2003:3036e7.
[7] Kim S, Lee JK, Kang SO, Ko J, Yum J-H, Fantacci S, et al. Molecular engineering
of organic sensitizers for solar cell applications. Journal of American Chemical
Society 2006;128:16,701e16,707.
[8] Kitamura T, Ikeda M, Shigaki K, Inoue T, Anderson NA, Ai X, et al. Phenyl-
conjugated oligoene sensitizers for TiO2 solar cells. Chemistry of Materials
2004;16:1806e12.
[9] Hagberg DP, Yum J-H, Lee H, De Angelis F, Marinado T, Karlsson KM, et al.
Molecular engineering of organic sensitizers for dye-sensitized solar cell
applications. Journal of American Chemical Society 2008;130:6259e66.
[10] (a) Zhang F, Luo Y-H, Song J-S, Cuo X-Z, Liu W-L, Ma C-P, et al. Triphenyla-
minebased dyes for dye-sensitized solar cells. Dyes and Pigments 2009;81:
224e30;
[30] Hagberg DP, Edvinsson T, Marinado T, Boschloo G, Hagfeldt A, Sun LC. A novel
organic chromophore for dye-sensitized nanostructured solar cells. Chemical
Communications; 2006:2245e7.
[31] Qiu XP, Lu R, Zhou HP, Zhang XF, Xu TH, Liu XL, et al. Synthesis of linear
monodisperse vinylene-linked phenothiazine oligomers. Tetrahedron Letters
2007;48:7582e5.
(b) Ko S-B, Cho A-N, Kim M-J, Lee C-R, Park N-G. Alkyloxy substituted organic
dyes for high voltage dye-sensitized solar cell: effect of alkyloxy chain length
on open-circuit voltage. Dyes and Pigments 2012;94:88e98.
[32] Kim M-J, Lee C-R, Jeong W-S, Im J-H, Ryu TI, Park N-G. Unusual enhancement
of photocurrent by incorporation of Brӧnsted base thiourea into electrolyte of
dye-sensitized solar cell. Journal of Physical Chemistry
C 2010;114:
[11] Hara K, Sato T, Katoh R, Furube A, Ohga Y, Shinpo A, et al. Molecular design of
coumarin dyes for efficient dye-sensitized solar cells. Journal of Physical
Chemistry B 2003;107:597e606.
19,849e19,852.
[33] Connelly NG, Geiger WE. Chemical redox agents for organometallic chemistry.
Chemical Reviews 1996;96:877e910.
[12] Hara K, Wang Z-S, Sato T, Furube A, Katoh R, Sugihara H, et al. Oligothiophene-
containing coumarin dyes for efficient dye-sensitized solar cells. Journal of
Physical Chemistry B 2005;109:15,476e15,482.
[13] Liu B, Zhu WH, Zhang Q, Wu WJ, Xu M, Ning ZJ, et al. Conveniently synthe-
sized isophorone dyes for high efficiency dye-sensitized solar cells: tuning
photovoltaic performance by structural modification of donor group in donor-
[34] Heck RF. Palladium reagents in organic synthesis. London: Academic Press;
1985.
[35] Nasipuri D. Stereochemistry of organic compounds: principles and applica-
tions. India: New Age International; 1994.
[36] Choi H, Lee JK, Song K, Kang SO, Ko J. Novel organic dyes containing bis-
dimethylfluorenyl amino benzo[b]thiophene for highly efficient dye-
sensitized solar cell. Tetrahedron 2007;63:3115e21.
p-acceptor system. Chemical Communications; 2009:1766e8.
[14] (a) Kim D, Song K, Kang M-S, Lee J-W, Kang SO, Ko J. Efficient organic sensi-
tizers containing benzo[cd]indole: effect of molecular isomerization for
photovoltaic properties. Journal of Photochemistry and Photobiology A 2009;
201:102e10;
[37] Dutta AK, Kamada K, Ohta K. Spectroscopic studies of nile red in organic
solvents and polymers. Journal of Photochemistry and Photobiology A 1996;
93:57e64.
[38] Hagfeldt A, Gräzel M. Light-induced redox reactions in nanocrystalline
systems. Chemical Reviews 1995;95:49e68.
[39] Kalyanasundaram K, Gräzel M. Applications of functionalized transition metal
complexes in photonic and optoelectronic devices. Coordination Chemistry
Reviews 1998;177:347e414.
[40] Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al.
Gaussian 09. Wallingford, CT: Gaussian, Inc.; 2009.
(b) Matsui M, Ito A, Kotani M, Kubota Y, Funabiki K, Jin J, et al. The use of
indoline dyes in a zinc oxide dye-sensitized solar cell. Dyes and Pigments
2009;80:233e8.
[15] Kim C, Choi H, Kim S, Baik C, Song K, Kang M-S, et al. Molecular engineering of
organic sensitizers containing p-phenylene vinylene unit for dye-sensitized
solar cells. Journal of Organic Chemistry 2008;73:7072e9.
[16] Peter LM. Dye-sensitized nanocrystalline solar cells. Physical Chemistry
Chemical Physics 2007;9:2630e42.
[41] (a) Lee C, Yang W, Parr RG. Development of the Colle-Salvetti correlation-
energy formula into a functional of the electron density. Physical Review B
1988;37:785e9;
(b) Becke AD. Density-functional thermochemistry. III. The role of exact
exchange. Journal of Chemical Physics 1993;98:5648e52.
[42] Krishnan R, Binkley JS, Seeger R, Pople JA. Self-consistent molecular orbital
methods. XX. A basis set for correlated wave functions. Journal of Chemical
Physics 1980;72:650e4.
[17] Koops SE, Barnes PRF, O’Regan BC, Durrant JR. Kinetic competition in
a coumarin dye-sensitized solar cell: injection and recombination limitations
upon device performance. Journal of Physical Chemistry C 2010;114:8054e61.
[18] O’Regan BC, Lopez-Duarte I, Martinez-Diaz MV, Forneli A, Albero J,
Morandeira A, et al. Catalysis of recombination and its limitation on open
circuit voltage for dye sensitized photovoltaic cells using phthalocyanine dyes.
Journal of American Chemical Society 2008;130:2906e7.
[43] McDowell JJ. The crystal and molecular structure of thiethylperazine,
[19] Hara K, Miyamoto K, Abe Y, Yanagida M. Electron transport in coumarin-dye-
sensitized nanocrystalline TiO2 electrodes. Journal of Physical Chemistry B
2005;109:23,776e23,778.
[20] Robertson N. Optimizing dyes for dye-sensitized solar cells. Angewandte
Chemie International Edition 2006;45:2338e45.
a
derivative of phenothiazine. Acta Crystallographica Section
B 1970;26:
954e64.
[44] Nishida J, Masuko T, Cui Y, Hara K, Shibuya H, Ihara M, et al. Molecular design
of organic dye toward retardation of charge recombination at semiconductor/
dye/electrolyte interface: introduction of twisted
p-linker. Journal of Physical
[21] Mishra A, Fischer MKR, Bauerle P. Metal-free organic dyes for dye-sensitized
solar cells: from structure: property relationships to design rules. Ange-
wandte Chemie International Edition 2009;48:2474e99.
[22] Li G, Jiang K-J, Li Y-F, Li S-L, Yang L-M. Efficient structural modification of
triphenylamine-based organic dyes for dye-sensitized solar cells. Journal of
Physical Chemistry C 2008;112:11,591e11,599.
[23] Vanderkooi JM. Photoreduction of membrane bound cytochrome c by excited-
state phenothiazine. Biochemical and Biophysical Research Communications
1976;69:1043e9.
Chemistry C 2010;114:17,920e17,925.
[45] Ning ZJ, Fu Y, Tian H. Improvement of dye-sensitized solar cells: what we
know and what we need to know. Energy & Environmental Science 2010;3:
1170e81.
[46] He JX, Wu WJ, Hua JL, Jiang YH, Qu SY, Li J, et al. Bithiazole-bridged dyes for
dye-sensitized solar cells with high open circuit voltage performance. Journal
of Materials Chemistry 2011;21:6054e62.
[47] Lee K, Park SW, Ko MJ, Kim K, Park N-G. Selective positioning of organic dyes
in a mesoporous inorganic oxide film. Nature Materials 2009;8:665e71.