Carbazole donor and carbazole or bithiophene bridged sensitizers for dye-sensitized solar cells

Article abstract of DOI:10.1016/j.jphotochem.2011.08.001

Three metal-free organic sensitizers consisting of carbazole as an electron donor, carbazole or bithiophene as the linker and cyanoacrylic acid as the electron acceptor and anchoring groups were designed and synthesized for use in dye-sensitized solar cel

Full text of DOI:10.1016/j.jphotochem.2011.08.001

Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Carbazole donor and carbazole or bithiophene bridged sensitizers for
dye-sensitized solar cells
Anthony C. Onicha, Krishna Panthi, Thomas H. Kinstle, Felix N. Castellano∗
Center for Photochemical Sciences and Department of Chemistry, Bowling Green State University, Bowling Green, OH 43403, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 April 2011
Received in revised form 15 June 2011
Accepted 1 August 2011
Available online 9 August 2011
Three metal-free organic sensitizers consisting of carbazole as an electron donor, carbazole or bithiophene
as the linker and cyanoacrylic acid as the electron acceptor and anchoring groups were designed and
synthesized for use in dye-sensitized solar cells (DSSCs). The sensitizers were characterized by 1H and
13C NMR, MALDI-TOF (or HRMS), UV–Vis, photoluminescence, and electrochemistry. The HOMO and
LUMO electron distributions of the sensitizers were calculated using density functional theory on a B3LYP
level for geometry optimization. The photovoltaic performance of the sensitizers in operational liquid
junction-based DSSCs under AM 1.5 G one-sun excitation (100 mW/cm2) indicate that the sensitizers
are promising candidates for use in DSSCs. Sensitizers 1 and 2 produce a power conversion efficiency of
2.70% with a maximum IPCE of 75% at 450 nm, while sensitizer 3 has a power conversion efficiency of
2.23% with a maximum IPCE of 66% at 440 nm. The sensitizers thus exhibit excellent photon-to-current
conversion efficiencies in the blue region of the spectrum and serve as candidates for further strategic
optimization in tandem cells.
Keywords:
Organic DSSC
Carbazole donor
Carbazole linker
Photochemically stable organic sensitizer
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
Novel organic sensitizers such as coumarin [12–14], tetrahy-
droquinoline [15,16], merocyanine [17,18], cyanine [19,20],
Dye-sensitized solar cells (DSSCs) have been attracting a great
deal of interest due to the potential of high energy conversion effi-
ciency at low cost [1,2]. Since the original breakthrough in DSSCs
[1,3], ruthenium polypyridyl complexes have been widely used as
sensitizers in DSSCs, and the highest reported NREL-certified power
conversion efficiency of 11.18% has been obtained for the N719
sensitizer, a derivative of cis-Ru(dcbpyH2)(NCS)2 (the N3 sensi-
tizer) [4]. In addition to ruthenium polypyridine sensitizers, other
organometallic sensitizers such as osmium(II) polypyridine sen-
sitizers [5–7], platinum(II) sensitizers [8], and zinc(II) sensitizers
[9,10], amongst others, have also been investigated by researchers.
Although the metal-complexed sensitizers exhibit high efficiency
and stability, they are expensive and difficult to purify compared
to the metal-free organic sensitizers. Metal-free organic sensitizers
also show promise due to advantages such as low cost, high molar
extinction coefficient, and facile molecular design. Recently, an
unprecedented power conversion efficiency of 10.3% was reported
for an organic sensitizer [11], thus indicating that with further
strategic optimization, metal-free organic sensitizers may overtake
their organometallic counterparts in photovoltaic performance.
phenothiazine [21], indoline [22,23], and hemicyanine [24,25] have
been used in DSSCs with good results. Still, a more strategic molec-
ular design of organic sensitizers is required to achieve higher
efficiency (ꢀ) values. The requirements [17,24,26] of efficient sen-
sitizers for use in DSSCs include (a) wide absorption range and
high absorption coefficient that give high light harvesting effi-
ciency, (b) excited-state redox potential should match the energy
of the conduction band, (c) light excitation associated with vecto-
rial electron flow from the light-harvesting moiety of the sensitizer
towards the surface of the semiconductor, (d) conjugation across
the donor and the anchoring group, (e) good anchoring group
and, (f) electronic coupling between the lowest unoccupied molec-
ular orbital (LUMO) of the sensitizer and the TiO2 conduction
band. The major factors that result in low conversion efficiency
of many organic sensitizers in the DSSCs are (a) the aggregation
of sensitizer on the semiconductor surface and (b) recombina-
tion of conduction-band electrons with the iodide/triiodide redox
mediator [27]. Many metal-free organic sensitizers usually exhibit
excellent incident photon-to-current efficiency in the blue region
of the spectrum with little or no photoaction in the lower energy
regions [28,29]. A tendency to ␲-stack exists in ␲-conjugated pla-
nar molecules due to strong intermolecular interactions between
␲-electrons and may result in a dissipative intermolecular energy
transfer which could have adverse effects on the photovoltaic per-
formance of DSSCs incorporating these types of molecules [30,31].
∗
Corresponding author. Tel.: +1 419 372 7513; fax: +1 419 372 9809.
E-mail addresses: castell@bgsu.edu, castell@bgnet.bgsu.edu (F.N. Castellano).
1010-6030/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotochem.2011.08.001

58
A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
as an electron acceptor and as an anchoring group and has been
t-Bu
shown to outperform the carboxylic acid group in these roles
[38].
N
NC
NC
N
COOH
COOH
2. Experimental
t-Bu
1
2.1. General
t-Bu
t-Bu
CF3
N
Triphenylphosphine, 5,5ꢀ-dibromo-2,2ꢀ-bithiophene, CuI, N,N-
diisopropylethylamine, Na2SO4, NaHCO3, CS2CO3, PdCl2(PPh3)2,
cyanoacetic acid, and ammonium acetate were purchased from
Sigma Aldrich. Lithium iodide, 4-tert-butylpyridine (4-tBupy),
iodine and 1-n-propyl-3-methylimidazolium iodide (PMII) were
available from our previous study [7]. 3,6-Di-tert-butylcarbazole,
and 3,6-di-tert-butyl-9-(4-ethynylphenyl)-9H-carbazole were syn-
thesized following the reported literature procedure [39]. The
synthesis of sensitizers 1, 3 and their precursors 7, 9 and 10 is
described here. The synthesis of sensitizer 2, and the precursors
4–6 is described elsewhere [37,40]. Relevant structural character-
ization spectra (1H and 13C NMR) for compounds 1, 3, 7, 9 and 10
are given in Supporting information.
N
2
t-Bu
t-Bu
CF3
S
S
N
NC
COOH
3
Fig. 1. Chemical structures of the sensitizers 1–3.
2.2. Characterization, physical measurements and
instrumentation
Minimization of the extent of aggregation of organic sensitizers
and charge recombination through appropriate structural modifi-
cation becomes necessary in order to achieve optimal performance.
On the basis of these strategies, we have designed and synthe-
sized organic dyes for use as sensitizers of DSSCs. The very few
organic compounds incorporating carbazole [32–34] which have
been used as sensitizers in DSSCs exhibit exciting results. Also, a
few reports on sensitizers incorporating thiophene-based spacers
and exhibiting good device efficiencies have appeared [35]. To date,
there are no reports related to sensitizers bearing the carbazole
moiety, in which it functions as both the donor and the linker.
Therefore, we are developing and evaluating the potential of com-
pounds bearing carbazole donor and carbazole or thiophene linker
as sensitizers in organic solar cells. Carbazole-based sensitizers
have been shown to exhibit excellent photovoltaic characteristics
in the blue region of the spectrum [33,34]. These sensitizers may
eventually be mixed with other sensitizers that exhibit excellent
photovoltaic responses in the red in a tandem cell. This sort of
tandem arrangement becomes necessary since extending the spec-
tral response of individual sensitizers to low energy regions of the
spectrum will ultimately result in ground and excited states that
are not thermodynamically well aligned for sensitizer regeneration
and electron injection. Also, we are interested in an excellent hole-
transporting material such as carbazole moiety, and solid-state
DSSCs [29,36].
1H and 13C NMR spectra were measured on
a Bruker
500 MHz spectrometer internally referenced to TMS. UV–Vis
absorption spectra were measured on a Shimadzu UV-2401 spec-
trophotometer and photoluminescence spectra were recorded
on Jobin-Yvon-Fluorolog-3 spectrometer. Electrochemical data
were obtained using a BAS Epsilon electrochemistry worksta-
tion with a conventional three-electrode arrangement. Cyclic
voltammetry measurements were carried out in either anhy-
drous acetonitrile or dichloromethane solution containing 0.1 M
tetrabutylammonium hexafluorophosphate as the supporting elec-
trolyte, a gold microdisk (1.6 mm diameter) working electrode
(BAS model MF-2014), a platinum wire auxiliary electrode (BAS
model MW-4130) and a Ag/AgCl (3 M NaCl) reference elec-
trode (BAS model MF-2079), respectively. Measurements were
conducted in ca. 1 mM electroactive substrate in an argon gas
atmosphere with a scan rate of 300 mV/s and ferrocene was
used as internal standard. For all measurements, potentials were
recorded vs the ferrocenium/ferrocene (Fc+/Fc0) internal stan-
dard, and finally converted to E1/2 vs the normal hydrogen
electrode (NHE) using E1/2(Fc+/Fc0) = +0.69 V vs NHE [41]. Incident
photon-to-current conversion efficiency (IPCE) measurements
were carried out using an instrument from PV Measurements,
Inc., equipped with a xenon arc lamp and calibrated with a sil-
icon reference photodiode. Current–voltage characteristics were
measured on an I–V data acquisition system (PV Measure-
ments, Inc.) equipped with a small area solar simulator (AM
1.5 Global) and an NREL-certified silicon reference solar cell
(PVM 274, PV Measurements, Inc.) for calibrating the intensity
of the simulated sunlight to 100 mW/cm2, with the measured
photocurrent being within 2% of its calibration value. Photocur-
rent density (JSC) values directly measured using I–V curves
were typically comparable to those estimated from the inte-
grated EQE (IPCE) spectra. Estimation of JSC was performed by
the I–V software (PV Measurements Inc.) according to the ASTM
standard E1021. The sandwiched solar cells were illuminated
directly through the transparent conductive glass support con-
taining the TiO2 photoanode. The photovoltaic characteristics
reported herein are overall yields that were not corrected for losses
due to light absorption and reflection by the conductive glass
substrate.
In this article we report the design and synthesis of
(2-cyano-3-[4-((3ꢀ,6ꢀ-di-tert-butyl-9-[4-(trifluoromethyl)phenyl]-
9H-2,9ꢀ-bicarbazol-7yl)ethynyl)phenyl]acrylic
acid)
(1)
and
(2-cyano-3-(4-(2-(5-(2-(4-(3,6-
di-tert-butyl-9H-carbazolyl)phenyl)ethynyl)-2,2ꢀ-
bithiophene)ethynyl)phenyl)acrylic acid) (3) that exhibit
maximum photovoltaic performance in the blue region of the
spectrum. The design and synthesis of sensitizer 2 is reported
in our previous paper [37]. DSSCs using these carbazole-based
sensitizers (1–3) are discussed herein. The chemical structures
of these sensitizers are shown in Fig. 1. All the sensitizers have
the same acceptor cyanoacrylic acid, which incorporates the
carboxylic acid anchoring group, and identical carbazole donor but
differ in the linker. Sensitizers 1 and 2 have 2-, 7-, 9-substituted
carbazole linker and the sensitizer 3 has bithiophene group as
a linker. Cyanoacrylic acid has evolved as the acceptor group of
choice in metal-free organic sensitizers since it functions both

A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
59
2.3. Synthesis and characterization
2.3.4. 2-Cyano-3-[4-((3ꢀ,6ꢀ-di-tert-butyl-9-[4-(trifluoromethyl)-
phenyl]-9H-2,9ꢀ-bicarbazol-7-yl)ethynyl)-phenyl]acrylic acid (1)
2.3.1. 4-((3ꢀ,6ꢀ-Di-tert-butyl-9-[4-(trifluoromethyl)phenyl]-
To a round bottomed flask containing a mixture of compound
7 (100 mg, 0.14 mmol), cyanoacetic acid (11 mg, 0.14 mmol), and
ammonium acetate (1 mg) was added acetic acid (5 ml). The mix-
ture was heated at 120 ◦C for 6 h and allowed to cool to room
temperature. The resulting solid was filtered and washed with dis-
tilled water, diethyl ether and methanol to give a bright orange
solid (82%). Mass spectrum (MALDI-TOF) m/z M+ = 783. 1H NMR
(500 MHz, DMSO): ı 1.40 (s, 18H), 7.38 (d, 2H), 7.47 (dd, 2H), 7.59
(dd, 1H), 7.61 (s, 1H), 7.64 (dd, 1H), 7.10 (s, 1H), 7.90 (d, 2H), 8.40 (s,
4H), 8.10 (d, 2H), 8.30 (d, 2H), 8.38 (s, 1H), 8.47 (d, 1H) and 8.59 (d,
1H). 13C NMR (DMSO): ı 32.3, 35.0, 89.7, 94.5, 108.3, 109.6, 113.5,
117.2, 119.9, 121.9, 122.1, 123.1, 123.3, 124.0, 124.2, 125.6, 127.2,
128.0, 128.1, 131.3, 132.5, 136.9, 139.3, 140.7, 141.8, 143.0, 153.5
and 163.6.
9H-2,9ꢀ-bicarbazol-7-yl)ethynyl)benzaldehyde (7)
To a solution of compound 6 (200 mg, 0.38 mmol) in dry toluene
(30 ml) was added 3,6-di-tert-butylcarbazole (160 mg, 0.42 mmol).
The Pd(OAc)2 (3%), P(t-Bu)3 7% and Cs2CO3 (495 mg, 1.5 mmol)
were also added and stirred under argon at 110 ◦C for about 24 h.
The reaction mixture was then cooled to room temperature and
toluene was removed completely under vacuum. The solid mix-
ture was dissolved in tetrahydrofuran and unreacted Cs2CO3 was
removed under gravity filtration. The organic residue was then
purified by column chromatography (silica gel, 8% ethyl acetate
in petroleum ether) to give the product (180 mg, 66%). 1H NMR
(500 MHz, CDCl3): ı 1.50 (s, 18H), 7.36 (d, 2H), 7.48 (dd, 2H), 7.55
(dd, 1H), 7.61 (dd, 2H), 7.68 (s, 1H), 7.30 (d, 2H), 7.78 (d, 2H), 7.90
(d, 2H), 7.91 (d, 2H), 8.18 (d, 2H), 8.22 (d, 1H), 8.34 (d, 1H) and
10.05 (s, 1H). 13C NMR (500 MHz, DMSO): ı 32.0, 34.8, 88.9, 94.4,
108.3, 109.0, 113.2, 116.4, 120.1, 120.3, 120.6, 121.9, 122.2, 123.4,
123.7, 124.8, 127.1, 127.2, 127.5, 127.6, 129.5, 129.6, 129.7, 132.0,
135.4, 139.5,140.8, 142.0, 143.0, and 191.4. HRMS EI+ calculated for
C48H39ON2F3 716.30146, measured 716.30024.
2.3.5. 2-Cyano-3-(4-(2-(5-(2-(4-(3,6-di-tert-butyl-9H-
carbazolyl)phenyl)ethynyl)-2,2ꢀ-bithiophene)-
ethynyl)phenyl)acrylic acid (3)
To a round bottomed flask containing a mixture of compound
10 (150 mg, 0.22 mmol), cyanoacetic acid (9 mg), and ammonium
acetate (1 mg) was added acetic acid (10 ml). The mixture was
heated at 120 ◦C for 8 h and allowed to cool to room temperature.
The resulting solid was filtered and washed with distilled water,
diethyl ether and methanol to give a bright red solid (142 mg, 86%).
Mass spectrum (MALDI-TOF) m/z M+ = 738. 1H NMR (500 MHz,
DMSO): ı 1.43 (s, 18H), 7.40 (d, 2H), 7.47–7.54 (m, 6H), 7.71 (d,
2H), 7.75 (d, 2H), 7.84 (d, 2H), 8.07 (d, 2H), 8.29 (s, 1H) and
8.1 (s, 2H). 13C NMR (500, DMSO): ı 31.2, 32.3, 83.4, 86.2, 94.8,
109.7, 117.3, 120.3, 121.2, 122.0, 123.6, 124.3, 126.1, 126.2, 126.6,
131.2, 132.2, 133.5, 134.8, 135.5, 137.7, 138.4, 138.5, 138.6, 143.5
and 163.6.
2.3.2. 4-(2-(5-Bromo-2,2ꢀ-bithiophene)ethynyl)benzaldehyde (9)
5,5ꢀ-Dibromo-2,2ꢀ-bithiophene
(8)
(0.45 g,
1.38 mmol),
PdCl2(PPh3)2 (16 mg), CuI (12 mg), N,N-diisopropylethylamine
(1 ml) and DMF (9 ml) were mixed together in 100 ml round bottom
flask. Then argon was bubbled for 20 min. Ethylene benzaldehyde
(0.1 g, 0.77 mmol) was added and stirred for 12–14 h at 80 ◦C under
argon atmosphere. Then the reaction mixture was poured into dis-
tilled water and extracted with DCM. The extract was washed with
0.5 M HCl, then it was washed using saturated NaHCO3 and distilled
water. The solution was dried using anhydrous Na2SO4, followed
by the addition of pentane (1/3 volume of DCM). The resultant
solution was filtered using a Buckner funnel with a layer of silica
pad. The filtrate was dried under vacuum. The crude product was
purified by flash chromatography (silica gel, 30% DCM in hexane) to
obtain pure compound (0.17 g, 65%). 1H NMR (500 MHz, CDCl3): ı
6.96 (d, 1H), 6.99 (d, 1H), 7.03 (d, 1H), 7.22 (d, 1H), 7.65 (d, 2H), 7.87
(d, 2H) and 10.03 (s, 1H). 13C (500 MHz, CDCl3): ı 86.5, 93.6, 120.1,
121.4, 123.0, 123.6, 129.0, 129.6, 130.0, 131.7, 133.7, 135.5, 137.9,
138.8 and 191.3. HRMS EI+ calculated for C17H9OS2Br 371.92783,
measured 371.92912.
2.4. Preparation of nanocrystalline TiO2 electrode and
transparent platinum cathode
The sol–gel synthesis of the colloidal TiO2 paste has been
described in detail elsewhere [7,42]. The prepared TiO2 paste was
doctor-bladed onto the conductive glass substrate (Hartford Glass,
TEC-15) to give the transparent layer of TiO2 film with a typical
thickness of 13 ␮m. The obtained nanoparticle film was then dried
at 125 ◦C for 6 min and a 5 ␮m thick scattering layer of mesoscopic
TiO2 (Solaronix, Ti-Nanoxide 300) was doctor-bladed on top of it.
The resulting TiO2 films were subsequently annealed for 30 min at
500 ◦C under oxygen flow in a tube furnace with ramped heating
control of 5 ◦C per minute. Upon cooling to 100 ◦C, TiO2 electrodes
were immersed in 0.5 mM sensitizer solution in acetonitrile/tert-
butanol (50:50, v/v, %) for 48 h at RT. Due to solubility limitations,
sensitization of sensitizer 3 was carried out in THF solution. Trans-
parent platinum-coated FTO cathodes were prepared as described
elsewhere [7].
2.3.3. 4-(2-(5-(2-(4-(3,6-Di-tert-butyl-9H-carbazolyl)phenyl)-
ethynyl)-2,2ꢀ-bithiophene)ethynyl)benzaldehyde (10)
To
a solution of compound 9 (100 mg, 0.269 mmol) in
diisopropylethylamine (0.5 ml) and DMF (7 ml) was added 3,6-di-
tert-butyl-9-(4-ethynylphenyl)-9H-carbazole (344 mg, 0.9 mmol).
After the solution was degassed with argon for 30 min while stir-
ring, PdCl2(PPh3)2 (6.7 mg), CuI (5.1 g), and triphenylphosphine
(8 mg) were added. The reaction was then stirred for 14 h at 80 ◦C
under argon. The solvent was evaporated using vacuum and crude
product was obtained. The product obtained was recrystallized
from DCM and hexane to obtain the product (150 mg, 82%). 1H NMR
(500 MHz, CDCl3): ı 1.5 (s, 18H), 7.17 (d, 1H), 7.18 (d, 1H), 7.27 (d,
1H), 7.28 (d, 1H), 7.42 (d, 2H), 7.50 (ds, 2H), 7.61 (d, 2H), 7.69 (d,
2H), 7.75 (d, 2H), 7.91 (d, 2H), 8.16 (d, 2H) and 10.08 (s, 1H). 13C
(500 MHz, CDCl3): ı 31.0, 32.0, 76.6, 76.8, 77.0, 77.3, 83.1, 86.7, 93.9,
94.3, 109.2, 116.3, 120.9, 121.6, 122.6, 123.6, 123.8, 124.2, 124.4,
126.4, 129.0, 129.7, 131.8, 132.6, 133.1, 133.9, 135.5, 138.0, 138.5,
138.8, 139.2, 143.3 and 191.4. HRMS EI+ calculated for C45H37ONS2
671.23167, measured 671.22981.
2.5. Sandwiched solar cell assembly
The active device area of the sensitized TiO2 photoanode was
adjusted to 0.25 cm2. Stretched Parafilm-M was used as a spacer
between the photoanode and the platinum counter electrode. The
typical thickness of the spacer was 20–30 ␮m. A few drops of the
redox electrolyte were placed on top of the active electrode area
and a platinized FTO-glass counter electrode was placed on top.
The electrodes were then sealed together using binder clips. For
these studies, the redox electrolyte solution consisted of 0.2 M LiI,
0.05 M I2, 0.7 M PMII and 0.5 M 4-tBupy in anhydrous acetonitrile
[43].

60
A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
in Table 1. The minimum energy between the ground and excited
states (E0-0) was estimated as half the sum of the absorption and
emission maxima, and the obtained value was subsequently con-
1
2
3
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
verted to electron volts (eV). The excited state potential (ELUMO
)
was determined as described elsewhere [7].
These sensitizers are composed of three main parts: a carbazole
donor, a carbazole or bithiophene as a linker and cyanoacrylic
acid as an acceptor and anchoring group. The donor groups may
efficiently inhibit I3− from approaching the surface of the nanocrys-
talline TiO2. The phenylethynyl spacer is a semi-rigid conjugated
linker [44] which should favor the electron injection and the car-
bazole linker in sensitizers 1 and 2 can suppress the aggregation
of dye molecules by the steric hindrance as trifluoromethyl benzyl
group is attached at the 9H position of it.
300
400
500
600
700
800
Wavelength (nm)
3.3. Electrochemistry
Fig. 2. Absorption–emission spectra of sensitizers 1–3 in dichloromethane.
The cyclic voltammogram of 1 in acetonitrile solution shows a
quasi-reversible oxidation process centered at 1178 mV vs Ag/AgCl
with the Fc+/Fc0 wave centered at 443 mV vs Ag/AgCl. The cyclic
voltammogram of 2 in dichloromethane solution shows a quasi-
reversible oxidation process centered at 1191 mV vs Ag/AgCl with
the Fc+/Fc0 wave centered at 428 mV vs Ag/AgCl. Due to solubility
limitations, the electrochemical properties of 3 could not be mea-
sured. Although sensitizer 3 has high solubility in tetrahydrofuran
(THF), no oxidation waves were observed in THF solution within
the solvent’s potential window [45]. The oxidation potentials were
further converted to E1/2 vs NHE using E1/2(Fc+/Fc0) = +0.69 vs
NHE [41] yielding E1/2(1) = 1.43 V vs NHE and E1/2(2) = 1.45 V vs
NHE. The sensitizers have oxidation potentials that are more pos-
3. Results and discussion
3.1. Synthesis and characterization
The preparation of 2,7-functionalized carbazoles is not straight-
forward, since both the 2- and 7-positions are in the meta position
of the amino group of the carbazole unit and cannot be directly
functionalized by standard electrophilic aromatic substitution.
Successful synthetic strategies usually require functionalized 4,4ꢀ-
biphenyl precursors with an additional reactive group at the 2
position used for a subsequent ring closure reaction. In this way
4,4ꢀ-biphenyl compound was converted into 2,7-dibromocarbazole
(4) following a literature procedure [40]. N-arylation of 4 was fol-
lowed by the Sonogashira coupling with 4-ethynylbenzaldehyde
to obtain 6 (Scheme 1). Compound 6 reacts with 3,6-di-tert-
butylcarbazole in presence of Pd(OAc)2, P(t-Bu)3 and Cs2CO3 to
obtain 7 which further reacts with cyanoacetic acid, ammonium
acetate and acetic acid to give the sensitizer 1. On the other
hand 5,5ꢀ-dibromo-2,2ꢀ-bithiophene (8) was converted into com-
pound 9 by the Sonogashira coupling with 4-ethynylbenzaldehyde
which further undergoes Sonogashira coupling with 3,6-di-tert-
butyl-9-(4-ethynylphenyl)-9H-carbazole to obtain compound 10.
Compound 10 reacts with cyanoacetic, ammonium acetate and
acetic acid to give the sensitizer 3.
itive than that of the I−/I3 redox mediator (0.4 V vs NHE) [46]
−
which provide a thermodynamic driving force of 1.03–1.05 V for
regeneration of the oxidized sensitizers by the iodide/triiodide
redox mediator. The excited state oxidation potentials of the
sensitizers (Table 1) are sufficiently more negative than the con-
duction band edge of TiO2, which is −0.5 V vs NHE [47], providing
a great amount of thermodynamic driving force for electron
injection.
3.4. Photovoltaic measurements
The photoaction spectra and I–V curves were measured under
identical conditions for all the sensitizers investigated herein. Four
independent DSSCs were assembled and measured in parallel and
the results reported herein are the average of the four cells, rep-
resented as overall yields that were not corrected for any kind of
losses. All devices have an active area of 0.25 cm2. The photoac-
tion spectra of dye-sensitized solar cells incorporating sensitizers
1–3, along with that of the N3 sensitizer, are presented in Fig. 3.
The carbazole-based sensitizers convert visible light to photocur-
rent efficiently across the higher energy region over the wavelength
range of 350–550 nm. A maximum photon-to-current conversion
efficiency of about 75.1% at 450 nm was realized for sensitizers 1
and 2, while sensitizer 3 has a maximum IPCE of 66.5 0.64% at 440
(Fig. 3 and Table 2). The current–voltage properties of the DSSCs
based on sensitizers 1–3, as well as the N3 sensitizer, measured
under standard one-sun conditions (AM 1.5 G, 100 mW/cm2) are
shown in Fig. 4. Devices based on the carbazole-linked sensitizers
1 and 2 displayed the highest power conversion efficiency of about
2.7%, those based on the thiophene-linked sensitizer 2 displayed
a power conversion efficiency of 2.23 0.04%. Importantly, DSSCs
based on these metal-free organic sensitizers displayed excellent
fill factors which suggests that there is reduced contribution to the
internal resistance within these devices, especially from the charge
transport at the TiO2/dye/electrolyte interface [48,49]. Compared
with sensitizers 1 and 2, sensitizer 3 possesses a higher short-circuit
Reagents
and
conditions:
(a)
1-bromo-4-
trifluoromethylbenzene, K2CO3, DMF, reflux 12 h; (b) Pd(PPh3)2Cl2,
CuI, PPh3, N,N-diisopropylethylamine, reflux 10 h; (c) Pd(OAc)2
P(t-Bu)3, Cs2CO3, toluene, reflux 22 h; (d) cyanoacetic acid, ammo-
nium acetate, CH3COOH, reflux 5 h; (e) Pd(PPh3)2Cl2, CuI, PPh3,
N,N-diisopropylethylamine/DMF, reflux 14 h; (f) Pd(PPh3)2Cl2,
CuI, PPh3, N,N-diisopropylethylamine, reflux 10 h; (g) cyanoacetic
acid, ammonium acetate, CH3COOH, reflux 5 h.
3.2. Photophysical properties
These sensitizers (1–3) are soluble in common organic solvents,
such as dichloromethane (DCM), tetrahydrofuran (THF), dimethyl-
formamide (DMF), and dimethylsulfoxide (DMSO). The absorption
spectra of the sensitizers in DCM are shown in Fig. 2. All sensi-
tizers have a strong absorption in the ultraviolet and blue regions
with a maxima at 290 nm and around 350–430 nm, respectively.
The absorption maxima at around 290 (not shown) are due to car-
bazole moiety and assigned to a ␲–␲* transition while the bands
at around 350–430 nm are due to intramolecular charge trans-
fer (ICT) between the donor carbazole and the acceptor moiety.
A summary of photophysical properties of sensitizers 1–3 is given

A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
61
t-Bu
t-Bu
Br
Br
CHO
Br
H
Br
Br
N
N
N
H
OHC
N
H
a
c
b
5
4
CF3
6
CF3
t-Bu
t-Bu
OHC
N
d
1
N
7
t-Bu
CF3
CHO
Br
N
f
S
S
Br
CHO
Br
S
S
e
t-Bu
8
9
t-Bu
S
g
N
3
OHC
S
t-Bu
10
Scheme 1. Synthesis of sensitizers 1 and 3.
Table 1
Summary of photophysical properties of sensitizers 1–3.
Sensitizer
ꢁabs (nm) (ε × 10−3 M−1 cm−1
)
EOX (V vs NHE)
Estimated E0-0 (eV)
ELUMO (V vs NHE)
ꢁem (nm)
1
2
3
350 (35.76), 395 (34.12)
350 (48.43), 385 (54.74)
350 (11.09), 420 (16.42)
1.43
1.45
–
2.71
2.63
2.58
−1.28
524
545
540
–1.18
–
current despite having a lower value for the maximum IPCE. This
is probably due to the fact that the sensitizer has a much broader
absorption spectrum whose contributions are expected to enhance
the value of the photogenerated current. Sensitizers 1 and 2 are
structurally similar, except for the inclusion of an extra pheny-
lacetylene group in the motif of sensitizer 2, and as such should
display similar properties which are reflected in the photophysi-
cal, electrochemical and photovoltaic properties of the sensitizers.
The photogenerated short-circuit current and the values of the IPCE
maxima are identical, within experimental error, for both sensitiz-
ers and the observed differences are revealed in the open-circuit
voltage. There is a ∼25 mV increase in the value of the VOC of
sensitizer 1 relative to sensitizer 2 which can be attributed to the
effects of an extended ␲-conjugated backbone which enhances
charge separation between the carbazole donor and cyanoacrylic
acceptor moieties, enhances electron injection and occupancy of
the conduction band which raises the Fermi level and ultimately
results in a greater open-circuit voltage.
Due to concerns about the photostability of organic sensitiz-
ers in DSSCs, light-soaking tests were performed over 1 h with
sandwich DSSCs based on sensitizer 2 and the results (Fig. 5) indi-
cate a stable photovoltaic performance of the solar cells within the
period under review. We were limited to such a small irradiation
time duration given the fact that we are unable to permanently
Table 2
Summary of the photovoltaic properties of sensitizers 1–3 with N3 sensitizer included for comparison.
Sensitizer
TiO2 film thicknessa (␮m)
VOC (mV)
JSC (mA/cm2)
FF (%)
ꢀ (%)
IPCEmax (%)
1
13 + 5
16
16 + 5
13 + 5
16
16 + 5
16 + 5
16 + 5
649 4.83
615 11.7
617 1.80
664 9.03
636 4.41
642 6.59
523 10.1
683 4.09
5.60 0.10
5.77 0.07
6.24 0.30
4.72 0.39
5.73 0.39
5.96 0.20
7.03 0.28
15.9 1.01
67.4 2.89
68.7 0.50
70.2 1.77
71.5 1.00
67.4 6.11
70.5 1.23
60.5 0.30
55.9 1.58
2.45 0.04
2.44 0.00
2.70 1.77
2.24 0.25
2.44 0.07
2.70 0.05
2.23 0.04
6.04 0.31
74.0 0.39
74.4 0.68
75.3 0.36
75.1 0.18
74.0 0.75
75.9 0.57
66.5 0.64
71.5 0.37
2
3
N3
a
Thickness of the transparent layer was either 13 or 16 ␮m while the scattering layer is 5 ␮m thick. Four identical solar cells were prepared and evaluated in each case.
The redox electrolyte consisted of 0.2 M LiI, 0.05 M I2, 0.7 M PMII and 0.5 M tbupy in acetonitrile solution.

62
A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
16
80
60
40
20
0
14
1
2
12
10
8
3
N3
1
2
3
N3
6
4
2
0
0
100 200 300 400 500 600 700
Photovoltage (mV)
400
500
600
700
800
Wavelength (nm)
Fig. 4. Current–voltage curves of sensitizers 1–3 and N3. TiO2 films consisted of a
16 ␮m transparent layer and a scattering layer is 5 ␮m thick. The redox electrolyte
consisted of 0.2 M LiI, 0.05 M I2, 0.7 M PMII and 0.5 M tbupy in acetonitrile solution.
Fig. 3. Photoaction spectra of sensitizers 1–3 and N3. TiO2 films consisted of a 16 ␮m
transparent layer and a scattering layer is 5 ␮m thick. The redox electrolyte consisted
of 0.2 M LiI, 0.05 M I2, 0.7 M PMII and 0.5 M tbupy in acetonitrile solution.
seal the DSSCs. Apart from an initial (negligible) decrease in all the
photovoltaic parameters in the first ca. 7 min, the performance indi-
cators exhibited stable values and suggest that these sensitizers
may withstand prolonged light-soaking. Prolonged light-soaking
is usually accompanied by temperature increases which result
in a faster rate of redox electrolyte diffusion and a downwar−d
shift in the conduction band edge potential relative to I−/I3
[50]. The former results in improvements in fill factor and the
latter in decreases in the open-circuit voltage, while observed
increases in the short-circuit current are generally a combination
of both effects. Similar experiments, performed at 50–55 ◦C over a
1000 h period, have been reported by Wang et al. [13] and it was
shown that the power conversion efficiency and other photovoltaic
properties stabilized over the period of the test, indicating that
metal-free organic sensitizers exhibit adequate long term photo-
stability, similar to the more frequently used coordination complex
dye-sensitizers.
All photovoltaic performances were obtained with TiO2 pho-
toanodes incorporating 16 ␮m of a transparent layer and a 5 ␮m
layer of scattering particles to yield the best photovoltaic per-
formance (Table 2). Optimization of the TiO2 film thickness is
necessary in order to ascertain the optimum range that would
yield the best photovoltaic performance and in this case involved
the use of transparent layer of TiO2 film (13 and 16 ␮m thick-
ness) with or without a second layer of a 5 ␮m scattering layer
on top of it (Table 2). Thicker films are expected to give better light
absorption from an optical point of view since the films will adsorb
more sensitizer molecules whose combined absorption proper-
ties will increase the light-harvesting property of the sensitized
film. The thickness of a film affects its mechanical strength, adher-
ence onto substrates and the interconnectivity between particles
and these pose a challenge when making thicker films. Inter-
connectivity between particles is essential for efficient electron
transport and influences resistance within the bulk semiconduc-
tor. Inefficient charge transport through the substrates results
in losses through various recombination pathways. The opti-
mal thickness of any given photoelectrode, on the other hand,
depends on the extinction coefficient of the adsorbed sensi-
tizer as well as on the semiconductor particle properties [51],
and these conditions need to be optimized to realize the best
performance for any given combination of semiconductor and
sensitizer.
These sensitizers exhibit excellent photon-to-current conver-
sion efficiencies in the blue region of the spectrum. Further work is
focusing on developing metal-free sensitizers with excellent photo-
conversion efficiencies in the lower energy region and assembling
the blue- and red-response sensitizers in a tandem arrangement
to boost the energy conversion efficiency. The sensitizers reported
herein exhibit unique photovoltaic behaviors and show promises
for further optimization and application in the growing field of
organic dye-sensitized solar cells.
a
η (%)
b
4
2
5
0
80
4
0 min
ff (%)
7 min
70
60
17 min
3
22 min
34 min
44 min
JSC (mA/cm2)
VOC (mV)
6
4
2
56 min
64 min
750
700
650
1
0
0
100
200
300
400
500
600
700
0
10
20
30
40
50
60
Photovoltage (mV)
Irradiation time (min)
Fig. 5. Preliminary light-soaking test results for DSSCs based on sensitizer 2 showing (a) current–voltage curves and (b) properties of the current–voltage curves as a function
of irradiation time.

A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
63
Fig. 6. HOMO–LUMO structures of the sensitizers 1–3 calculated using density functional theory.
3.5. Quantum chemical calculations
indicate that these organic compounds are promising candidates
in the development and further optimization of metal-free organic
sensitizers for application in DSSCs.
Density functional theory (DFT) calculations were per-
formed on the sensitizers 1–3 to get a deep insight into the
structure–photovoltaic cell performance relationships. The calcu-
lations were done on a B3LYP level for the geometry optimization.
In the ground state, the electron density is localized mainly on the
carbazole donor units, whereas in the excited state they move com-
pletely to the acceptor units near to the anchoring group, which
favor the electron injection from the sensitizer to the conduction
band of TiO2. From the electron distribution structures in Fig. 6, it is
obvious that the acceptor can withdraw almost all the electrons to
the phenyl unit next to the anchoring group in all sensitizers 1–3.
The HOMO electrons are delocalized almost from donor carbazole
to linker carbazole in sensitizer 1 but in sensitizer 2 it is not up to
linker carbazole, whereas in sensitizer 3 the electrons are localized
only in donor carbazole moiety. The absence of electronic com-
munication between the thiophene linker and the donor carbazole
in the HOMO on the one hand, and the linker and the accep-
tor cyanoacrylic acid on the other hand, in sensitizer 3 is closely
related to the relatively higher energy of the linker, compared to
the energies of the donor and acceptor groups. The electronic com-
munication that exists in sensitizers 1 and 2 between the donor
and the linker in the HOMO, and the acceptor and the linker in the
LUMO, is made possible by the relative stabilization of the energy
levels of the linker by the electron-withdrawing trifluoromethyl
group. These theoretical results show that the sensitizers can have
efficient electron injection from the LUMO to the conduction band
of TiO2 and improving the cell efficiency.
Acknowledgements
We thank Gayani Wasana Pinnawala-Arachchilage for synthe-
sis of sensitizer 3. We also thank Dr. Ravi M. Adhikari for useful
suggestions and Dr. Patrick Z. El-Khoury for quantum chemical cal-
culations.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jphotochem.2011.08.001.
References
[1] B. O’Regan, M. Grätzel, Nature 353 (1991) 737.
[2] A. Hagfeldt, M. Grätzel, Acc. Chem. Res. 33 (2000) 269.
[3] N. Vlachopoulos, P. Liska, J. Augustynski, M. Grätzel, Am. Chem. Soc. 110 (1988)
1216.
[4] M.K. Nazeeruddin, F.D. Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito,
B. Takeru, M. Grätzel, Am. Chem. Soc. 127 (2005) 16835.
[5] G. Sauve, M.E. Cass, G. Coia, S.J. Doig, I. Lauermann, K.E. Pomykal, N.S. Lewis,
Phys. Chem. B 104 (2000) 6821.
[6] S. Altobello, R. Argazzi, S. Caramori, C. Contado, S.D. Fre, P. Rubino, C. Chone, G.
Larramona, C.A. Bignozzi, Am. Chem. Soc. 127 (2005) 15342.
[7] A.C. Onicha, F.N. Castellano, Phys. Chem. C 114 (2010) 6831.
[8] A. Islam, H. Sugihara, K. Hara, L.P. Singh, R. Katoh, M. Yanagida, Y. Takahashi, S.
Murata, H. Arakawa, Inorg. Chem. 40 (2001) 5371.
[9] W.M. Campbell, K.W. Jolley, P. Wagner, K. Wagner, P.J. Walsh, K.C. Gordon,
L. Schmidt-Mende, M.K. Nazeeruddin, Q. Wang, M. Grätzel, D.L. Officer, Phys.
Chem. C 111 (2007) 11760.
[10] L. Xiao, Y. Liu, Q. Xiu, L. Zhang, L. Guo, H. Zhang, C. Zhong, Tetrahedron 66 (2010)
2835–2842.
4. Conclusions
[11] W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C. Pan, P. Wang,
Chem. Mater. 22 (2010) 1915.
[12] K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H.
Sugihara, H. Arakawa, Phys. Chem. B 107 (2003) 597.
[13] Z.-S. Wang, Y. Cui, K. Hara, Y. Dan-oh, C. Kasada, A. Shinpo, Adv. Mater. 19 (2007)
1138.
[14] Z.-S. Wang, Y. Cui, Y. Dan-oh, C. Kasada, A. Shinpo, K. Hara, Phys. Chem. C 111
(2007) 7224.
[15] R. Chen, X. Yang, H. Tian, L. Sun, Photochem. Photobiol. A 189 (2007) 295.
[16] R. Chen, X. Yang, H. Tian, X. Wang, A. Hagfeldt, L. Sun, Chem. Mater. 19 (2007)
4007.
[17] K. Sayama, S. Tsukagoshi, K. Hara, Y. Ohga, A. Shinpou, Y. Abe, S. Suga, H.
Arakawa, Phys. Chem. B 106 (2002) 1363.
[18] K. Sayama, K. Hara, N. Mori, M. Satsuki, S. Suga, S. Tsukagoshi, Y. Abe, H. Sugi-
hara, H. Arakawa, Chem. Commun. 1173 (2000).
[19] W.-H. Zhan, W.-J. Wu, J.-l. Hua, Y.-H. Jing, F.-S. Meng, H. Tian, Tetrahedron Lett.
48 (2007) 2461.
[20] X. Ma, J. Hua, W. Wu, Y. Jin, F. Meng, W. Zhan, H. Tian, Tetrahedron 64 (2008)
345.
Organic sensitizers with a carbazole as a donor, carbazole or
bithiophene as a ␲-conjugation linker and a cyanoacrylic acid group
as an acceptor have been synthesized as for DSSCs. Compared with
the sensitizer having bithiophene in the linker, the sensitizers con-
taining carbazole linker showed more efficiency. Sensitizers 1 and
2 produce a power conversion efficiency of 2.70% with a maximum
IPCE of 75% at 450 nm, while sensitizer 3 has a power conver-
sion efficiency of 2.23% with a maximum IPCE of 66% at 440 nm.
The sensitizers thus exhibit excellent photon-to-current conver-
sion efficiencies in the blue region of the spectrum and serve as
candidates for further optimization in tandem cells. These sensi-
tizers could be mixed with other sensitizers with optimized red
response in a tandem arrangement to achieve excellent photocon-
version efficiencies in the spectral regions of interest. All the results

64
A.C. Onicha et al. / Journal of Photochemistry and Photobiology A: Chemistry 223 (2011) 57–64
[21] H. Tian, X. Yang, R. Chen, Y. Pan, L. Li, A. Hagfeldt, L. Sun, Chem. Commun. 3741
(2007).
[36] M. Li, S. Tang, F. Shen, M. Liu, W. Xie, H. Xia, L. Liu, L. Tian, Z. Xie, P. Lu, M. Hanif,
D. Lu, G. Cheng, Y. Ma, Chem. Commun. 3393 (2006).
[22] T. Horiuchi, H. Miura, K. Sumioka, S. Uchida, Am. Chem. Soc. 126 (2004) 12218.
[23] L. Schmidt-Mende, U. Bach, R. Humphry-Baker, T. Horiuchi, H. Miura, S. Ito, S.
Uchida, M. Grätzel, Adv. Mater. 17 (2005) 813.
[37] K. Panthi, R.M. Adhikari, T.H. Kinstle, Phys. Chem. A 114 (2010) 4550.
[38] J. Song, F. Zhang, C. Li, W. Liu, B. Li, Y. Huang, Z. Bo, Phys. Chem. C 113 (2009)
13391.
[24] Z.-S. Wang, F.-Y. Li, C.-H. Huang, Phys. Chem. B 105 (2001) 9210.
[25] Y.-S. Chen, C. Li, Z.-H. Zeng, W.-B. Wang, X.-S. Wang, B.-W. Zhang, Mater. Chem.
15 (2005) 1654.
[39] Y. Liu, M. Nishiura, Y. Wang, Z. Hou, Am. Chem. Soc. 128 (2006) 5592.
[40] F. Dierschke, A.C. Grimsdale, K. Mullen, Synthesis (2003) 2470–2472.
[41] W.C. Barrette Jr., H.W. Johnson Jr., D.T. Sawyer, Anal. Chem. 56 (1984) 1890.
[42] S.-H.A. Lee, N.M. Abrams, P.G. Hoertz, G.D. Barber, L.I. Halaoui, T.E. Mallouk,
Phys. Chem. B 112 (2008) 14415.
[43] S. Hwang, J.H. Lee, C. Park, H. Lee, C. Kim, C. Park, M.-H. Lee, W. Lee, J. Park, K.
Kim, N.-G. Park, C. Kim, Chem. Commun. 4887 (2007).
[44] A.A. Bothner-By, J. Dadok, T.E. Johnson, J.S. Lindsey, Phys. Chem. 100 (1996)
17551.
[26] K. Kalyanasundaram, M. Grätzel, Coord. Chem. Rev. 177 (1998) 347.
[27] D. Liu, R.W. Fessenden, G.L. Hug, P.V. Kamat, Phys. Chem.
2583.
B 101 (1997)
[28] M.K.R. Fischer, S. Wenger, M. Wang, A. Mishra, S.M. Zakeeruddin, M. Grätzel, P.
Bäuerle, Chem. Mater. 22 (2010) 1836.
[29] N. Ikeda, T. Miyasaka, Chem. Commun. 1886 (2005).
[30] Z.-S. Wang, N. Koumura, Y. Cui, M. Takahashi, H. Sekiguchi, A. Mori, T. Kubo, A.
Furube, K. Hara, Chem. Mater. 20 (2008) 3993.
[45] A.J. Bard, L.R. Faulkner, Electrochemical Methods Fundamentals and Applica-
tions, John Wiley & Sons Inc., New York, 1980.
[31] X.-F. Wang, O. Kitao, H. Zhou, H. Tamiaki, S.-i. Sasaki, Phys. Chem. C 113 (2009)
7954.
[46] Z. Zhang, Z. Zhang, P. Chen, T.N. Murakami, S.M. Zakeeruddin, M. Grätzel, Adv.
Funct. Mater. 18 (2008) 341.
[32] N. Koumura, Z.-S. Wang, S. Mori, M. Miyashita, E. Suzuki, K. Hara, Am. Chem.
Soc. 128 (2006) 14256.
[33] D. Kim, J.K. Lee, S.O. Kang, J. Ko, Tetrahedron 63 (2007) 1913.
[34] C. Teng, X. Yang, C. Yuan, C. Li, R. Chen, H. Tian, S. Li, A. Hagfeldt, L. Sun, Org.
Lett. 11 (2009) 5542.
[47] M. Grätzel, Inorg. Chem. 44 (2005) 6841.
[48] L. Han, N. Koide, Y. Chiba, T. Mitate, Appl. Phys. Lett. 84 (2004) 2433.
[49] L. Han, N. Koide, Y. Chiba, A. Islam, R. Komiya, N. Fuke, A. Fukui, R. Yamanaka,
Appl. Phys. Lett. 86 (2005) 213501.
[50] B.C. O’Regan, J.R. Durrant, Phys. Chem. B 110 (2006) 8544.
[51] Z.-S. Wang, H. Kawauchi, T. Kashima, H. Arakawa, Coord. Chem. Rev. 248 (2004)
1381.
[35] Y.-S. Yen, Y.-C. Hsu, J.T. Lin, C.-W. Chang, C.-P. Hsu, D.-J. Yin, Phys. Chem. C 112
(2008) 12557.