Fluorene-containing organic photosensitizers for dye-sensitized solar cells

Article abstract of DOI:10.1002/cplu.201200127

Seven new donor-spacer-acceptor metal-free organic dyes containing a fluorene and/or thiophene unit as a conjugated spacer are prepared and their physicochemical properties as well as photovoltaic performance are studied. The electron transfer from donor to acceptor upon excitation of these dyes is sensitive to the conjugated spacers. If a stronger donor moiety such as (ethylenedioxy)thiophene is used as a spacer, such as in F-E1 dye, the dye molecule has a better charge separation and therefore high efficiency. Furthermore, for the same donor and acceptor, fluorene is a better conjugated spacer than the bithiophene moiety because of its nonplanar conformation, which avoids the electron-hole recombination of dye upon photoexcitation. Short molecules with planar conformation aggregate easily, and the corresponding device reveals low open-circuit voltage Voc. On the other hand, dye molecules with long alkyl side chains (large size) adsorb less on the TiO2 electrode, which results in a smaller short-circuit current density Jsc. The difference in the efficiency of these dyes is small except for F-1 (owing to its short conjugation length), and electron density difference map analyses reveal that photoexcitation transfer of electron density from the donating moieties toward the acceptor/anchoring units occurs for all molecules. The small difference in the efficiency is attributed partly to the degree of effective charge transfer upon excitation, not totally to the absorption profile of the dye molecules. Furthermore, time-dependent DFT simulated absorption spectra are in excellent agreement with experimental spectra when the dyes are in their protonated states in THF solution. The transition probability calculated with the electron density difference analyses combined with the absorption data and photovoltaic performance provide useful information for designing organic sensitizers for dye-sensitized solar cells.

Full text of DOI:10.1002/cplu.201200127

DOI: 10.1002/cplu.201200127
Fluorene-Containing Organic Photosensitizers for Dye-
Sensitized Solar Cells
Chun-Guey Wu,*[a, b] Ming-Fong Chung,[b] Hui-Hsu Gavin Tsai,*[a] Chun-Jui Tan,[a] Su-
Chien Chen,[a, b] Chiao-Hsin Chang,[a] and Ting-Wen Shih[a]
Seven new donor–spacer–acceptor metal-free organic dyes
containing a fluorene and/or thiophene unit as a conjugated
spacer are prepared and their physicochemical properties as
well as photovoltaic performance are studied. The electron
transfer from donor to acceptor upon excitation of these dyes
is sensitive to the conjugated spacers. If a stronger donor
moiety such as (ethylenedioxy)thiophene is used as a spacer,
such as in F-E1 dye, the dye molecule has a better charge sep-
aration and therefore high efficiency. Furthermore, for the
same donor and acceptor, fluorene is a better conjugated
spacer than the bithiophene moiety because of its nonplanar
conformation, which avoids the electron–hole recombination
of dye upon photoexcitation. Short molecules with planar con-
formation aggregate easily, and the corresponding device re-
veals low open-circuit voltage Voc. On the other hand, dye mol-
ecules with long alkyl side chains (large size) adsorb less on
the TiO2 electrode, which results in a smaller short-circuit cur-
rent density Jsc. The difference in the efficiency of these dyes is
small except for F-1 (owing to its short conjugation length),
and electron density difference map analyses reveal that pho-
toexcitation transfer of electron density from the donating
moieties toward the acceptor/anchoring units occurs for all
molecules. The small difference in the efficiency is attributed
partly to the degree of effective charge transfer upon excita-
tion, not totally to the absorption profile of the dye molecules.
Furthermore, time-dependent DFT simulated absorption spec-
tra are in excellent agreement with experimental spectra when
the dyes are in their protonated states in THF solution. The
transition probability calculated with the electron density dif-
ference analyses combined with the absorption data and pho-
tovoltaic performance provide useful information for designing
organic sensitizers for dye-sensitized solar cells.
Introduction
The problems of global warming and the fast depletion of
fossil fuels mean that renewable energy sources such as solar
energy have attracted mounting interest.[1–7] In particular, the
dye-sensitized solar cell (DSSC), one type of solar cell, has
drawn a great deal of scientific, technical, and industrial atten-
tion because of its low cost of manufacture and high efficiency.
mine,[20] thiophene,[21] oligoene,[22] porphyrin,[23] and pery-
lene,[24] have been developed, and up to 10% efficiency has
been attained.[25] The structure of the organic dyes employed
in DSSCs typically consists of an electron donor, an anchoring
acceptor, and a p-conjugated spacer for tuning the spectral
coverage. Remarkable progress has been made, and the struc-
ture-related photovoltaic performance has also been reviewed
recently.[26]
The DSSC, which is
a photoelectrochemical photovoltaic
device invented by Michael Grꢀtzel and Brian O’Regan in 1991,
is also known as the Grꢀtzel cell.[8] Ruthenium-based com-
plexes such as N3/N719[9] developed by the Grꢀtzel group and
CYC-B11[4] developed by our group are the most promising
ruthenium-based sensitizers for DSSC applications, and have
achieved an excellent light-to-electricity conversion efficiency
of over 11% under standard air mass (AM) 1.5 illumination at
100 mWcmꢀ2. Nevertheless, ruthenium-based sensitizers, al-
though with excellent efficiency, have the drawback of high
cost and low natural abundance. Therefore, metal-free organic
dyes (one of the alternatives) have also received much atten-
tion.[5,8,10–16]
In metal-free organic dyes, the p-conjugation spacer can
redshift the absorption band but also induces unfavorable
p-stacked dye aggregation. To avoid the aggregation from
[a] C.-G. Wu, H.-H. G. Tsai, C.-J. Tan, S.-C. Chen, C.-H. Chang, T.-W. Shih
Department of Chemistry, National Central University
Jhong-Li City, Tao-Yuan County 32001 (Taiwan (R.O.C.))
E-mail: t610002@ncu.edu.tw
hhsai@ncu.edu.tw
[b] C.-G. Wu, M.-F. Chung, S.-C. Chen
Research Center for the New Generation PhotoVoltaics
National Central University
In general, the efficiency of DSSCs based on organic dyes is
lower than that of devices sensitized with Ru complexes. Nev-
ertheless, the high molar absorption coefficient, variety of
structure, environmentally friendly characteristic, and essential-
ly unlimited resources are the advantages of organic dyes.
Therefore, various organic dyes for DSSCs, such as molecules
based on coumarin,[17] cyanine,[18] hemicyanine,[19] triaryla-
Jhong-Li 32001 (Taiwan (R.O.C.))
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201200127. It contains the detailed syn-
thetic procedures and structure characterizations of F-series dyes, the
fabrication and photovoltaic performance measurements of the corre-
sponding DSSC devices, the cyclic voltammograms and absorption and
emission spectra of F-series dyes measured in THF, and some of the the-
oretical calculation data.
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
a strong intermolecular p–p interaction and maintain the elec-
trochemical and optical properties, steric structures such as an
alkyl-substituted aromatic ring were introduced to inhibit the
possible intermolecular aggregation and retard the electron
transfer from TiO2 to the oxidized dye and electrolyte. Fluorene
derivatives (well-known hole-transporting materials[27] with
good stability[28]) were used as the spacer in donor–spacer–ac-
ceptor organic dyes.[29–33] Furthermore, the hydrogen on the
9-position of the fluorene unit can be substituted by alkyl
chains with various lengths to suppress the charge recombina-
tion and intermolecular interac-
Results and Discussion
Structure design and optical properties of dye molecules
The molecular structure of F-1, F-T1, F-DT1, F-DOT1, F-E1,
F-ET1, and F-E2 as well as their synthesis mechanisms are dis-
played in Schemes 1–7. The detailed preparation procedures
and characterization are summarized in the Supporting Infor-
mation. The framework of these dye molecules is donor–
spacer–acceptor/anchor, in which the donor is triphenylamine
tion as well as to improve the
solubility of the dye molecules.
So far no single rule for the
structure-related
photovoltaic
performance fits all types of
dyes because the efficiency of
the DSSC depends not only on
the optical properties but also
on the molecular conformation,
frontier orbital energy levels,
donor–acceptor electronic transi-
tion probability, and adsorption
behavior of dye molecules on
TiO2 as well as coadsorbent and
additive. Although there are
a large number of factors deter-
mining the efficiency of DSSCs,
the geometric and electronic
structures of the sensitizers play
an important role. Therefore
a number of theoretical calcula-
tions using density functional
theory (DFT) and time-depen-
dent DFT (TD-DFT) were per-
formed to study the structure,
electronic absorption spectra,
and energy levels of sensitizers
both in solution[11,13,15,16,33,34] and
on TiO2 clusters.[3,10,35–38] The re-
Scheme 1. Synthetic procedure for F-1 dye. DMF=N,N-dimethylformamide, THF=tetrahydrofuran.
Scheme 2. Synthetic procedure for F-T1 dye.
sults give good agreements with experimental data, thus pro-
viding useful information to aid the design of more sophisti-
cated and appropriate dye structures to be used in DSSCs.
Herein, we report the physicochemical properties and pho-
tovoltaic performance of seven metal-free organic dyes
(named F-1, F-T1, F-DT1, F-DOT1, F-E1, F-ET1, and F-E2,
F-series) constructed with a fluorene unit, which is widely used
in organic sensitizers. Furthermore, DFT and TD-DFT were per-
formed to calculate the molecular geometries, molecular orbi-
tals, and electronic transitions as well as the UV/Vis spectra of
these dyes. It is known that the changes in molecular structure
and conjugation length as well as the protonation state and
environment can result in changing of the physical and optical
properties of dye molecules. Therefore, the calculations were
performed at dye molecules in the protonated and deproton-
ated states in THF solvent.
(TPA), the acceptor is the commonly used cyanoacrylic acid,
and fluorene was used as (or a part of) the conjugated spacer.
Other spacers, such as thiophene, bithiophene (T-T), and 3,4-
(ethylenedioxy)thiophene (EDOT) motifs, are designed and
used to tune the photochemical properties of dye molecules.
Taking F-1 dye as a reference, the F-T1 and F-E1 molecules
have one thiophene or one EDOT unit besides fluorene as a
p-conjugated spacer between their donor and acceptor, re-
spectively. Thiophene is known as an ideal moiety in tuning
the optical properties of the sensitizers.[4,38] Therefore, F-DT1
and F-ET1 molecules have a spacer with longer conjugation
length such as T-T or EDOT–thiophene (EDOT-T) introduced be-
tween the donor and fluorene, respectively. Furthermore, to in-
crease the solubility and to retard the aggregation of dye mol-
ecules and charge recombination between electrons in TiO2
and the electrolyte, alkyl chains were introduced: F-DOT1 dye
has a similar molecular framework to F-DT1, except it has two
&2
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

Photosensitizers for Solar Cells
higher than 60000 at their lmax
.
The UV/Vis absorption spectra of
dyes in THF and adsorbed on
TiO2 are displayed in Figure 1
and the relative optical data are
summarized in Table 1. The fea-
tures of the adsorption patterns
(between 270 and 750 nm, ne-
glecting the signal close to
250 nm that appears in all dyes)
for F-series dyes in THF are not
the same. F-1 has only one ab-
sorption peak with lmax at
340 nm whereas F-T1, F-DT1,
and F-DOT1 have a peak cen-
tered at the visible-light region
with a higher-energy shoulder.
On the other hand, both F-E1
and F-E2 have three well-defined
absorption peaks and F-ET1 has
one absorption peak with two
shoulders. These absorption
peaks correspond to the p–p*
transition, intramolecular charge
transfer of the molecule, or a cer-
tain moiety in the dye molecule,
although it is hard to assign by
just looking at the molecular
structures shown in Schemes 1
to 7. The detailed electronic
transitions of each dye molecule
based on the calculated results
will be discussed later.
Scheme 3. Synthetic procedure for F-DT1 dye.
All F-series dyes display
a broad peak if adsorbed on
TiO2. The broad absorption for
the intramolecular p–p* charge-
transfer transition is a typical be-
havior in organic dye with
donor-conjugated
spacer–ac-
ceptor architecture. Compared
to the dyes in THF, the absorp-
tion profiles of the dyes ad-
sorbed on TiO2 films are slightly
broader and their lmax blueshifts
Scheme 4. Synthetic procedure for F-DOT1 dye.
long octyl chains located at the two thiophene groups. The
F-E2 molecule has the conjugated spacer of F-E1 replaced by
the T-T group designed to probe the property of fluorene. Mul-
tiple well-known synthesis steps were used to prepare the dye
molecules with reasonable yields (see the Supporting Informa-
tion). The synthetic intermediates and final products were puri-
fied by chromatography on silica gel using suitable eluents or
a recrystallization method, and were identified with 1H NMR
spectroscopy. The dye molecules were further identified by
mass spectrometry and elemental analysis.
Table 1. Optical data of F-series organic dyes.
Dye
lmax
e at lmax
[mꢀ1 cmꢀ1
lEM
lmax on TiO2
[nm]
[nm]
]
[nm]
F-1
F-T1
F-DT1
F-DOT1
F-E1
402
427
436
423
443
510
454
53957
88752
97276
100245
71472
56425
81220
512
548
532
539
544
617
546
351
395
404
389
410
455
417
F-E2
F-ET1
All dyes were soluble in THF, and the color of the solution
was orange, red, or deep red with an absorption coefficient
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&
3
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
the iodide/triiodide couple used
in the liquid electrolyte. The oxi-
dation potential of the dye mol-
ecules is related to their conju-
gation length (lmax): a longer
conjugation length results in
a
lower oxidation potential
except for F-E2 dye, which has
the longest conjugation length
but a high oxidation potential
relative to F-ET1. It seems that
besides the conjugation length,
fluorene and thiophene units in
the dye molecules have a differ-
ent effect on the HOMO energy
level of the dye molecules. The
Scheme 5. Synthetic procedure for F-E1 dye.
higher oxidation potential of
F-E2 relative to F-ET1 indicates
that the fluorene unit can raise
the HOMO level more effectively
than the thiophene unit if they
are attached to the acceptor
unit of the dye molecule. This
issue will be discussed below
with the theoretical calculation
data. The optical transition ener-
gies E0–0 determined from the in-
terception between the absorp-
tion and emission spectra are
2.19–2.71 eV. Neglecting the en-
tropy changes during excitation,
the LUMO energy level is ꢀ1.34–
ꢀ1.68 V versus NHE, which is
more negative than the poten-
Scheme 6. Synthetic procedure for F-E2 dye. NBS=N-bromosuccinimide.
tial of the TiO2 (anatase) conduc-
tion band edge (ꢀ0.50 V versus
NHE). Thus upon photoexcita-
32–55 nm, which suggests an H-aggregation.[39] The broadened
absorption can be attributed to the interaction between the
carboxylate group and the surface TiIV ions, which may lead to
an increase in the delocalization of the p* orbital of the dye
molecule. F-1 and F-ET1 dyes have more serious aggregation
than the others, which suggests molecules with a more planar
structure or short chain length will aggregate more seriously if
adsorbed on TiO2.
tion, all dyes self-assembled onto the TiO2 film can inject elec-
trons into the TiO2 conduction band and then the reduction of
oxidized dyes by the electrolyte will occur spontaneously in
this regenerative photovoltaic device.
Table 2. Electro-optical data of F-series sensitizers.
Dye
E(D+/D)[a]
E0–0
HOMO
LUMO
[V vs. NHE]
[eV]
[V vs. NHE]
[V vs. NHE]
Oxidation potential and frontier orbital energy level of the
dye molecules
F-1
F-T1
F-DT1
F-DOT1
F-E1
0.54
0.46
0.42
0.49
0.41
0.36
0.29
2.71
2.57
2.52
2.56
2.49
2.19
2.44
1.03
0.95
0.91
0.98
0.90
0.85
0.78
ꢀ1.68
ꢀ1.62
ꢀ1.61
ꢀ1.58
ꢀ1.59
ꢀ1.34
ꢀ1.66
The oxidation potential and energy level of the frontier orbitals
for the F-series dye molecules listed in Table 2 were calculated
from the cyclic voltammograms (see the Supporting Informa-
tion) and band-gap energy (E0–0) determined by the intercep-
tion of the absorption and emission spectra (see also the Sup-
porting Information). The oxidation potentials of the F-series
dyes are 0.29–0.54 V versus the normal hydrogen electrode
(NHE), which are all more positive than the redox potential of
F-E2
F-ET1
[a] The oxidation potential of the dye molecule. HOMO (eV)=E(D+/D)ꢀE(Fc/
) +4.8, LUMO=HOMOꢀE0–0, in which E0–0 is the energy of the intercep-
Fc+
tion between UV/Vis and photoluminescence spectra of the dye mole-
cule. Fc/Fc+ =ferrocene/ferrocenium.
&4
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

Photosensitizers for Solar Cells
in DSSCs (the detailed DSSC
device fabrication and photovol-
taic performance measurements
can be found in the Supporting
Information). The current densi-
ty–voltage (J–V) curves and the
corresponding incident photon-
to-current conversion efficiency
(IPCE) plots for the DSSC devices
sensitized by F-series dyes are
presented in Figure 2 (the J–V
curves shown are the best
amongst five devices measured
for each dye), and the photovol-
taic parameters are summarized
in Table 3. Moreover, the quanti-
ty of the dye molecules assem-
bled on the practical TiO2 elec-
trode is also an important pa-
rameter for the photovoltaic per-
formance of
a
DSSC. The
amount of dye adsorbed on TiO2
is also listed in Table 3.
The dye loading was estimat-
ed by facilely desorbing the dye
Scheme 7. Synthetic procedure for F-ET1 dye.
Figure 2. a) J–V curves of DSSCs based on F-series dyes. b) The correspond-
ing IPCE spectra.
Figure 1. UV/Vis absorption spectra of dyes (a) in THF and (b) adsorbed on
TiO2.
molecules from the TiO2 electrode using a basic solution and
measuring their absorption spectra as reported before.[40] The
quantity of dye molecules adsorbed on the TiO2 electrode is
quite different. F-DOT1 with a long octyl chain attached to the
thiophene ring occupies a large space; therefore, fewer mole-
Photovoltaic performance of DSSC devices based on the
F-series dyes
Preliminary photovoltaic performance tests were undertaken
to evaluate the potential of these F-series dyes for application
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&
5
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
adsorbed on TiO2, see Table 1) of F-E2 dye may also cause
charge recombination between electrons in TiO2 and the oxi-
dized dye, which results in lower Voc and fill factor values.
Therefore, the conversion efficiency of the F-E2-based DSSC is
smaller than that of the F-E1-based device.
Table 3. Photovoltaic performance of DSSCs based on F-series sensitizers
and the corresponding dye loading in TiO2 electrodes.
Dye
Voc
[V]
Jsc
[mA]
FF[a]
h
Dye load
[mole]
[%]
F-1
F-T1
F-DT1
F-DOT1
F-E1
F-E2
F-ET1
N719
0.64
0.68
0.69
0.68
0.67
0.61
0.64
0.66
8.23
10.77
14.33
12.55
13.97
12.66
13.70
18.13
0.60
0.63
0.56
0.63
0.62
0.56
0.56
0.58
3.1
4.6
5.6
5.4
5.8
4.4
4.9
6.9
0.6ꢃ10ꢀ8
0.64ꢃ10ꢀ8
1.03ꢃ10ꢀ8
0.33ꢃ10ꢀ8
1.18ꢃ10ꢀ8
1.65ꢃ10ꢀ8
0.77ꢃ10ꢀ8
1.81ꢃ10ꢀ8
Molecular geometries, frontier molecular orbitals, and elec-
tronic transitions as well as UV/Vis spectra of dyes calculat-
ed with DFT and TD-DFT
The ground-state molecular geometries of the seven studied
F-series sensitizers in their protonated and deprotonated states
in THF were optimized at the B3LYP/6-31G(d,p) level. Figure 3
displays selected structural parameters of the optimized dyes,
which were mainly constituted by p-conjugated, rigid aromatic
units (benzene, thiophene, EDOT, fluorene); therefore the geo-
metries are affected little by their environment. The largest
structural changes in the molecules after incorporating the dif-
ferent spacers were the bond lengths and the dihedral angles
of the two adjacent rigid aromatic rings. These two structural
parameters are particularly important because they affect the
degree of p conjugation of each dye molecule, and thus deter-
mine the adsorption spectra and the energy levels of the fron-
tier molecular orbitals of the molecules. Moreover, the degree
of p conjugation also affects the coupling strength between
the donor and acceptor moieties and, consequently, the proba-
bility of electron transfer upon photoexcitation. For conven-
ience, herein we abbreviate these two structural parameters
with reference to their two adjacent rings. For example, the di-
hedral angle between the benzyl ring on the TPA unit and the
fluorene ring (F) is abbreviated as t(TPA-F) and the length of
the bond connecting its two adjacent rings is abbreviated as
d(TPA-F); the dihedral angle between the fluorene ring and the
anchoring (A) group is denoted as t(F-A) and the length of the
bond connecting the fluorene ring and the anchoring group is
denoted as d(F-A).
[a] FF=fill factor.
cules are adsorbed on the TiO2 surface relative to the other
dyes. On the other hand, F-E2 with a less bulky T-T moiety
connected to the anchoring group adsorbed 1.5 times more
molecules on the TiO2 electrode relative to F-E1 with methyl
fluorene.
As listed in Table 3, the IPCE of the device based on F-DT1
dye is better than the F-T1-sensitized device, which is also
better than that based on F-1 dye. Extending the conjugation
length (adding one more thiophene unit) of the spacer not
only redshifts the lmax of the dye assembled on TiO2 but also
increases the molar absorption coefficient of the dye molecule.
Therefore the efficiency of F-T1-based devices is 1.5 times that
for the device based on F-1 dye if the number of molecules
adsorbed on the TiO2 anode is similar. However, the marginal
utility diminished when another thiophene unit was added to
the spacer, and therefore the efficiency of the F-DT1-sensitized
device was only 1% higher than that of the F-T1-based device.
Furthermore, the higher efficiency of the F-DT1 dye may also
be because more F-DT1 molecules were adsorbed on the TiO2
electrode than F-T1 molecules, which resulted in a higher
short-circuit current density Jsc and therefore better efficiency.
Adding alkyl chains on the thiophene spacer (such as F-DOT1)
did not significantly avoid aggregation (F-DOT1- and F-DT1-
based devices have a similar open-circuit voltage Voc value) but
increased the size of the dye molecule. Less F-DOT1 dye was
adsorbed on the TiO2 electrode, which resulted in a low Jsc.
The efficiency of F-DOT1-based devices is slightly lower than
that of devices based on F-DT1 dye.
In F-E1 and F-ET1, addition of a thiophene unit in the
spacer redshifts the lmax and increases the molar absorption
coefficient but the device based on F-ET1 has lower conver-
sion efficiency than the F-E1-based device, which may be
a result of the smaller number of F-ET1 molecules adsorbed
on the TiO2 electrode. Finally, the photovoltaic performance of
F-E1 and F-E2 dyes indicated that the fluorene motif is
a better spacer/acceptor than T-T, because F-E2 has a longer
conjugation length and more molecules adsorbed on the TiO2
electrode but has lower conversion efficiency than F-E1. F-E2
with a T-T spacer has a lower oxidation potential (high HOMO
energy level): this disfavors the electron transfer from electro-
lyte to the oxidized dye, which results in lower Jsc. Moreover,
serious aggregation (owing to the larger number of molecules
For F-1, the benzene ring on the TPA unit and the fluorene
ring are nonplanar, with a value of t(TPA-F) of approximately
338 regardless of the molecular state; the value of d(TPA-F)
was approximately 1.48 ꢂ, longer than a standard carbon–
carbon double bond (1.34 ꢂ) and shorter than a standard
carbon–carbon single bond (1.547 ꢂ). These results indicate
that the electron-donating TPA group is partially conjugated
with the fluorene unit in the molecular backbone of F-1. We
observed similar characteristics for the dihedral angles and
bond lengths between the benzene rings on the TPA units and
their neighboring rings for the other dye molecules. On the
other hand, the fluorene ring of F-1 was nearly coplanar with
the anchoring group and the value of d(F-A) was shorter than
the value of d(TPA-F). These results indicated that the anchor-
ing group and its adjoined fluorene ring were highly conjugat-
ed in F-1. Again, the other dye molecules exhibited similar fea-
tures. Interestingly, if the protonated F-1 was deprotonated,
the bonds of the anchoring groups generally elongated. In par-
ticular, the bond between the C=O group and its neighboring
carbon atom elongated from 1.490 ꢂ in the protonated state
to 1.542 ꢂ in the deprotonated state, which suggests the char-
&6
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

Photosensitizers for Solar Cells
Figure 3. Selected molecular geometries for seven F-series dye molecules. The bond lengths are shown in ꢂngstroms and are labeled around their corre-
sponding bond. The dihedral angles are shown in degrees and are labeled around their corresponding central bond; their four consecutive atoms are high-
lighted in bold. The data correspond to the deprotonated state in THF. X=TMA (tetramethylammonium).
acter of a carbon–carbon single bond. We also observed these
structural changes for the other dye molecules. These results
imply that if the dye molecules were deprotonated, the C=O
groups were less conjugated with the molecular framework,
thereby decreasing the effective conjugation length, consistent
with the blueshift when dye adsorbed on TiO2.
The structures of F-T1 and F-E1 feature one thiophene and
EDOT group, respectively, introduced into the structure of F-1;
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&
7
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
Figure 3. (Continued).
the values of t(TPA-T) in F-T1 and t(TPA-EDOT) in F-E1 were
less than the value of t(TPA-F) in F-1. As expected, the values
of d(TPA-T) in F-T1 and d(TPA-EDOT) in F-E1 were less than the
value of d(TPA-F) in F-1. The value of t(T-F) in F-T1 (ca. 228)
was slightly larger than that of t(F-EDOT) in F-E1 (ca. 158).
These results allowed us to determine that the degree of con-
jugation between the TPA unit and the other rings followed
the order TPA-EDOTꢁTPA-T>TPA-F, and that the degree of
conjugation between fluorene units and the other rings fol-
lowed the order F-EDOT>F-T>F-TPA.
phene unit. In contrast, the T-T unit in F-DT1 was twisted with
values of t(T-T) of approximately 158. These results suggest
that the degree of conjugation of the EDOT-T unit was greater
than that of the T-T unit. The twisting angles of EDOT-T (<58)
and T-T (ꢁ158) were less than that (358) of two adjacent ben-
zene rings,[10] thus indicating that EDOT and thiophene were
better p-conjugated spacers than benzene rings. Not surpris-
ingly, the values of t(TPA-T) and t(T-T) for F-DOT1 were larger
than the corresponding dihedral angles in the other studied
molecules owing to steric repulsion arising from the bulky
octyl chains. For F-E2, the value of t(T-T) was close to 1808,
which suggests that thiophene is a better p-conjugated spacer
than a benzene ring.[10]
The structure of F-ET1 features one thiophene ring intro-
duced between the EDOT and fluorene groups of F-E1. Inter-
estingly, the EDOT and thiophene rings were nearly coplanar.
F-DT1 possesses a molecular framework similar to that of
F-ET1, except for the presence of the EDOT instead of a thio-
For the UV/Vis spectra of the dyes in THF, we first discuss
the relationship between (trends in) the experimental UV/Vis
&8
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

Photosensitizers for Solar Cells
spectra and (their relationships to) the molecular geometries
calculated from B3LYP/6. The experimental UV/Vis spectra indi-
cate that the EDOT unit has a stronger ability to induce a red-
shift in adsorption than a thiophene unit. The geometrical
analysis discussed above revealed that the EDOT unit adopted
a more planar structure with the fluorene unit in F-E1 than the
thiophene unit in F-T1, thereby resulting in the greater red-
shifted absorption of F-E1. Furthermore, F-E2 possesses a
p-conjugation length similar to that of F-E1 and shorter than
that of F-ET1. Nevertheless, among the seven dyes studied,
F-E2 provided the most redshifted value of lmax up to 510 nm
in THF solution. This result can be understood by considering
molecular geometries. As discussed above, the structure of
F-E2 is more planar than those of the other dyes, particularly
in the spacer region. The more planar structure of F-E2 leads
to a greater effective p-conjugation length and, consequently,
Table 4 summarizes the calculated electronic absorptions for
the F-series dye molecules in THF in their protonated and de-
protonated states. The UV/Vis spectra were calculated at the
CAM-B3LYP/6-311+ +G(d,p) level in the THF environment
based on the optimized structure obtained at the B3LYP/6-
31G(d,p) level. Diffuse basis sets were used to consider the dif-
fuse character of the electronic cloud in the excited state. The
calculated electronic absorptions of the F-series dye molecules
in their protonated states in THF agreed well with the experi-
mental observations; the smallest absolute deviation from the
experimental value was a 2 nm redshift for F-ET1 and the larg-
est absolute deviation was a 22 nm redshift for F-E2. In con-
trast, the values of lmax for the F-series dye molecules in THF in
their deprotonated states deviated more widely from their ex-
perimental values; all absolute deviations were greater than
14 nm, with the largest absolute deviation (>40 nm) occurring
for F-1. Based on the calculated UV/Vis spectra, the studied
F-series dye molecules may adopt protonated states in THF. In
the protonated states, HOMO!LUMO, HOMOꢀ1!LUMO, and
HOMO!LUMO+1 transitions were the major components of
the electronic transitions. These transitions were mainly p!p*
transitions. The major orbitals associated with these transitions
are also provided in Table 4. The calculated electronic struc-
longer lmax
.
If a sensitizer containing one protonated carboxylic acid
group acting as an anchoring group is adsorbed on a TiO2
electrode, the proton on the anchoring group can be trans-
ferred to the TiO2 surface, thus decreasing the Fermi level of
the TiO2 electrode.[11] Therefore, it is important to investigate
the degree of protonation of the dye molecules in solution.
Table 4. Experimental and computational electronic absorptions of F-series sensitizers in THF.[a]
Dyes
Experimental
(in THF)
Calculated (protonated)
Calculated (deprotonated, TMA)
lmax [nm]
lmax [nm] Deviation Transition[b]
lmax [nm] Deviation Transition[a]
from exp.
from exp.
F-1
402
396
419
437
456
439
ꢀ6
ꢀ8
ꢀ6
+2
+3
HOMO!LUMO (49%)
HOMOꢀ1!LUMO (39%)
HOMO!LUMO+1 (6%)
362
391
407
435
422
ꢀ40
HOMO!LUMO (49%)
HOMOꢀ1!LUMO (37%)
HOMO!LUMO+1 (9%)
F-T1
F-E1
F-ET1
F-DT1
427
443
454
436
HOMO!LUMO (45%)
HOMOꢀ1!LUMO (29%)
HOMO!LUMO+1 (16%)
ꢀ36
ꢀ36
ꢀ19
ꢀ14
HOMO!LUMO (51%) HOMOꢀ1!LUMO (21%)
HOMO!LUMO+1 (20%)
HOMO!LUMO (54%)
HOMOꢀ1!LUMO (22%)
HOMO!LUMO+1 (15%)
HOMO!LUMO (57%)
HOMO!LUMO+1 (20%)
HOMOꢀ1!LUMO (15%)
HOMO!LUMO (43%)
HOMO!LUMO+1 (31%)
HOMOꢀ1!LUMO (14%)
HOMO!LUMO (53%)
HOMO!LUMO+1 (28%)
HOMOꢀ1!LUMO (10%)
HOM!LUMO (36%)
HOMO!LUMO (52%)
HOMO!LUMO+1 (24%)
HOMOꢀ1!LUMO (13%)
HOMO!LUMO+1 (29%)
HOMOꢀ1!LUMO (20%)
HOMOꢀ2!LUMO (5%)
F-DOT1
F-E2
423
510
432
532
+9
HOMO!LUMO (40%)
HOMOꢀ1!LUMO (22%)
HOMO!LUMO+1 (20%)
HOMOꢀ2!LUMO (8%)
406
482
ꢀ17
ꢀ28
HOMO!LUMO (48%)
HOMO!LUMO+1 (22%)
HOMOꢀ1!LUMO (18%)
+22
HOMO!LUMO (68%)
HOMOꢀ1!LUMO (21%)
HOMO!LUMO+1 (5%)
HOMO!LUMO (74%)
HOMOꢀ1!LUMO (15%)
HOMO!LUMO+1 (5%)
[a] Electronic absorptions were calculated at the TD-CAM-B3LYP/6-311+ +G(d,p) level, based on the optimized geometries calculated at the B3LYP/6-
31G(d,p) level. [b] Transitions with contribution greater than 5% are listed.
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&
9
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
Figure 4. Calculated electronic structures of the protonated F-series dye molecules in THF (only show HOMO, HOMOꢀ1, LUMO, and LUMO+1 orbitals).
tures of the protonated F-series dye molecules in THF are dis-
played in Figure 4. Only HOMO, HOMOꢀ1, LUMO, and LUMO+
1 orbitals are shown.
EDDM of the transitions clearly illustrated that the transition
promoted the transfer of electron density from the electron-
donating moiety to the acceptor/anchoring side. Table 5 sum-
marizes the electron densities of the F-series dye molecules in
their protonated states in THF. The anchor/acceptor underwent
a greater net increase in electron density than the conjugated
spacer, fluorene or thiophene or EDOT, upon excitation. For ex-
ample, the net increase in electron density at the anchor of
F-1 after the transition was 46%, whereas that at the fluorene
Figure 5 presents the simulated UV/Vis spectra and the elec-
tron density difference maps (EDDMs) of the transitions for the
protonated F-ET1 in THF solution; those for the rest of the
dyes can be found in the Supporting Information. The simulat-
ed UV/Vis spectra were consistent with the experimental spec-
tra for all seven F-series dyes in their protonated state. The
&10
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

Photosensitizers for Solar Cells
probably a result of the ineffective charge transfer (small net
changes in electron density) upon excitation. Finally, by com-
paring the net changes in electron density of F-E1 and F-E2
dyes, it is clear that fluorene is a better unit (compared to T-T)
for driving electron transfer from the TPA donor to the anchor-
ing ligand upon photoexcitation.
Conclusion
We have reported the synthesis and photovoltaic performance
of seven newly synthesized TPA–fluorene-based donor–spacer–
acceptor metal-free organic dyes. It was shown that extending
the conjugation length of the spacer increases the efficiency of
the dye; nevertheless the marginal utility diminishes as the
conjugation length increases. Short molecules with a planar
conformation show serious aggregation and thus a low effi-
ciency of the corresponding device. Molecules with long alkyl
side chains avoid aggregation but a smaller number of mole-
cules are adsorbed on the TiO2 electrode, which results in
a low Jsc value. The difference in the efficiency for the seven
dyes prepared is small except for F-1 (owing to its short lmax),
because EDDM analyses revealed that photoexcitation transfer
of electron density from the donating moieties of the dye mol-
ecules toward their acceptor/anchoring side units occurs for all
dye molecules. Nevertheless, the small difference in the effi-
ciency can be attributed partly to the degree of effective
charge transfer upon excitation. In this respect, fluorene is
a better conjugated spacer for use in organic dyes than bithio-
phene. Furthermore, TD-DFT simulated absorption spectra
were in excellent agreement with the experimental spectra,
which suggests this is a useful tool for helping to design so-
phisticated and appropriate organic sensitizers for DSSCs.
Figure 5. Comparison of the calculated (blue line) and experimental (green
line) UV/Vis absorption spectra of the protonated F-ET1 molecules in THF.
Blue vertical line: calculated excitation energies and oscillator strengths. The
absorption spectra were simulated through Gaussian convolution with a full
width at half maximum (FWHM) of 0.33 eV. EDDMs of the transitions are pro-
vided next to the spectra. Cyan indicates a decrease in electron density;
light blue indicates an increase.
Table 5. Characteristics of electron densities of protonated organic
F-series dyes in THF solution before and after transition.
Dye
State
Percent contribution [%]
TPA Spacer-1[a] Spacer-2[a] Fluorene[b] Anchor
before
after
net change ꢀ56
61
5
–
–
–
–
–
–
33
43
+10
6
52
+46
F-1
before
after
net change ꢀ52 ꢀ10
57
5
20
10
–
–
–
20
40
+20
3
45
+42
F-T1
before
after
net change ꢀ48 ꢀ16
54
6
25
9
–
–
–
19
40
+21
2
45
+43
F-E1
Experimental Section
before
after
net change ꢀ38 ꢀ13
44
6
25
12
18
13
ꢀ5
11
33
+22
1
36
+35
F-ET1
F-DT1
Materials
All reagents were obtained from commercial sources and used as
received unless specified. Solvents were dried over sodium or CaH2
before use. Unless otherwise specified, all reactions were per-
formed under an Ar atmosphere using standard Schlenk tech-
niques. All chromatographic separations were performed on silica
gel (60M, 240–400 mesh) columns. The structures of the dyes and
their intermediates were identified by 1H NMR spectroscopy. The
structure of the final product was further confirmed by fast atom
bombardment mass spectrometry (FAB-MS) and elemental analysis.
before
after
net change ꢀ42 ꢀ6
48
6
19
13
17
14
ꢀ3
14
32
+18
2
35
+33
before
F-DOT1 after
45
3
20
8
18
12
ꢀ6
15
37
+22
2
41
+39
net change ꢀ42 ꢀ12
before
after
48
3
23
9
17
19
8
34
4
36
F-E2
net change ꢀ45 ꢀ14
+2
+26
+32
[a] Spacer-1 and spacer-2 are the thiophene and/or EDOT rings located
between the TPA and fluorene units. [b] The fluorene unit in F-E2 was
substituted by a thiophene moiety.
Physicochemical studies
1H NMR spectra were recorded with a Bruker 300 MHz NMR spec-
trometer in CDCl3 or [D6]DMSO. FAB-MS spectra were obtained
using a JMS-700 HRMS instrument. UV/Vis spectra of dyes in THF
and adsorbed on TiO2 (4 mm thickness) films were measured using
a Cary 300 Bio spectrometer. Photoluminescence spectra (in THF
solutions, Figure S9) were obtained using a Hitachi F-4500 spectro-
photometer in the laboratory atmosphere at room temperature.
Cyclic voltammetric measurements were performed in a single-
compartment, three-electrode cell with an indium tin oxide (ITO)
glass working electrode and a Pt wire counter electrode. The refer-
unit was only 10%. The net changes in electron density of the
anchor/acceptor for F-1, F-T1, and F-E1 are close (42–46%) but
the efficiency is F-E1>F-T1>F-1, which is a result of the dif-
ference in conjugation length (F-E1>F-T1>F-1). F-E2 has the
longest conjugation length and adsorbed a large number of
molecules on the TiO2 electrode but with low efficiency. This is
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&
11
&
ÞÞ
These are not the final page numbers!

C.-G. Wu, H.-H. G. Tsai et al.
ence electrode was Ag/Ag+ and the supporting electrolyte was
0.1m LiClO4 in THF using an Autolab system (PGSTAT 30, Autolab,
Eco-Chemie, the Netherlands); the scan rate was 50 mVsꢀ1 and the
ferrocene/ferrocenium redox couple was used as a calibration stan-
dard. Elemental analysis was performed with a Heraeus CHN-O-S
Rapid-F002 analysis system. The thickness of the TiO2 film was
measured by a profilometer (Dektak3, Veeco/Sloan Instruments
Inc., USA).
[14] J. V. Ros-Lis, R. Martꢅnez-MꢆÇez, F. Sancenꢇn, J. Soto, M. Spieles, K.
Rurack, Chem. Eur. J. 2008, 14, 10101.
[15] D. Jacquemin, E. A. Perpete, G. E. Scuseria, I. Ciofini, C. Adamo, J. Chem.
Theory Comput. 2008, 4, 123.
[16] T. Edvinsson, C. Li, N. Pschirer, J. Schçneboom, F. Eickemeyer, R. Sens, G.
Boschloo, A. Herrmann, K. Mꢈllen, A. Hagfeldt, J. Phys. Chem. C 2007,
111, 15137.
[17] K. Hara, Y. Dan-oh, C. Kasada, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H.
Arakawa, Langmuir 2004, 20, 4205.
[18] A. Ehret, L. Stuhl, M. T. Spitler, J. Phys. Chem. B 2001, 105, 9960.
[19] Z.-S. Wang, F.-Y. Li, C.-H. Huang, Chem. Commun. 2000, 2063.
[20] N. Satoh, T. Nakashima, K. Yamamoto, J. Am. Chem. Soc. 2005, 127,
13030.
[21] K. R. Justin Thomas, Y.-C. Hsu, J. T. Lin, K.-M. Lee, K.-C. Ho, C.-H. Lai, Y.-M.
Cheng, P.-T. Chou, Chem. Mater. 2008, 20, 1830.
[22] T. Kitamura, M. Ikeda, K. Shigaki, T. Inoue, N. A. Anderson, X. Ai, T. Lian,
S. Yanagida, Chem. Mater. 2004, 16, 1806.
[23] 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. Grꢀtzel,
D. L. Officer, J. Phys. Chem. C 2007, 111, 11760.
[24] Y. Shibano, T. Umeyama, Y. Matano, H. Imahori, Org. Lett. 2007, 9, 1971.
[25] W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C. Pan, P.
Wang, Chem. Mater. 2010, 22, 1915.
[26] a) A. Mishra, M. K. R. Fischer, P. Bauerle, Angew. Chem. 2009, 121, 2510;
Angew. Chem. Int. Ed. 2009, 48, 2474; b) J. Preat, D. Jacquemin, E. A. Per-
pete, Energy Environ. Sci. 2010, 3, 891.
[27] G. Zhou, N. Pschirer, J. C. SchEoneboom, F. Eickemeyer, M. Baumgarten,
K. Mullen, Chem. Mater. 2008, 20, 1808.
[28] H. Choi, S. O. Kang, J. Ko, G. H. Gao, H. S. Kang, M. S. Kang, M. K. Nazeer-
uddin, M. Gratzel, Angew. Chem. 2009, 121, 6052; Angew. Chem. Int. Ed.
2009, 48, 5938.
Computational methods
The ground-state molecular geometry was optimized by Becke’s
three-parameter exchange functional,[41] the Lee–Yang–Parr gradi-
ent-corrected correlation functional,[42] and the 6-31G(d, p) basis
set,[43] as implemented in the Gaussian 09 program.[44] The conduc-
tor-like polarizable continuum model (C-PCM)[45] was used to ac-
count for the THF solvation effect. For the convenience of calcula-
tions, the long alkyl substituents were replaced by methyl groups.
Tetramethylammonium (TMA) was used to model the counter ion.
TD-DFT calculations were performed to calculate the UV/Vis spec-
tra. To account for the intramolecular charge-transfer excitations,
the coulomb-attenuating method (CAM)[46] was applied to calculate
the UV/Vis spectra of dyes in solution (calculated at the TD-CAM-
B3LYP/6-311+ +G(d,p) level) based on their corresponding opti-
mized ground-state geometries calculated at the B3LYP/6-31G(d,p)
level. The electron density difference map (EDDM) of electronic
transitions was calculated using the Gaussian program ver-
sion 2.5.5.[47]
[29] W. Li, Y. Wu, X. Li, Y. Xie, W. Zhu, Energy Environ. Sci. 2011, 4, 1830.
[30] J. Zhang, H.-B. Li, S.-L. Sun, Y. Geng, Y. Wu, Z.-M. Su, J. Mater. Chem.
2012, 22, 568.
[31] C.-H. Chen, Y.-C. Hsu, H. H. Chou, K. R. Justin Thomas, J. T. Lin, C.-P. Hsu,
Chem. Eur. J. 2010, 16, 3184.
Acknowledgements
[32] S. Higashijima, Y. Inoue, H. Miura, Y. Kubota, K. Funabiki, T. Yoshidac, M.
Matsui, RSC Adv. 2012, 2, 2721.
[33] W.-S. Chao, K.-H. Liao, C.-T. Chen, W.-K. Huang, C.-M. Lanc, E. W.-G. Diau,
Chem. Commun. 2012, 48, 4884.
[34] R. Ma, P. Guo, L. Yang, L. Guo, X. Zhang, M. K. Nazeeruddin, M. Grꢀtzel,
J. Phys. Chem. A 2010, 114, 1973.
[35] R. Jose, A. Kumar, V. Thavasi, K. Fujihara, S. Uchida, S. Ramakrishna,
Appl. Phys. Lett. 2008, 93, 023125.
Financial support from the National Science Council, Taiwan,
R.O.C., is gratefully acknowledged. We are grateful to the Nation-
al Center for High-Performance Computing and the V’ger com-
puter cluster at the National Center University of Taiwan for com-
puter time and facilities.
[36] F. De Angelis, S. Fantacci, A. Selloni, M. K. Nazeeruddin, M. Grꢀtzel, J.
Am. Chem. Soc. 2007, 129, 14156.
[37] F. De Angelis, A. Tilocca, A. Selloni, J. Am. Chem. Soc. 2004, 126, 15024.
[38] J.-G. Chen, C.-Y. Chen, S.-J. Wu, J.-Y. Li, C.-G. Wu, K.-C. Ho, Sol. Energy
Mater. Sol. Cells 2008, 92, 1723.
[39] a) C. Peyratout, L. Daehne, Phys. Chem. Chem. Phys. 2002, 4, 3032; b) Z.-
S. Wang, K. Hara, Y. Dan-oh, C. Kasada, A. Shinpo, S. Suga, H. Arakawa,
H. Sugihara, J. Phys. Chem. B 2005, 109, 3907.
Keywords: density
conversion · sensitizers · solar cells · organic dyes
functional
calculations
·
energy
[1] S. Ardo, G. J. Meyer, Chem. Soc. Rev. 2009, 38, 115.
[2] M. Grꢀtzel, Acc. Chem. Res. 2009, 42, 1788.
[3] F. De Angelis, S. Fantacci, A. Sgamellotti, Theor. Chim. Acta 2007, 117,
1093.
[4] C.-Y. Chen, S.-J. Wu, C.-G. Wu, J.-G. Chen, K.-C. Ho, Angew. Chem. 2006,
118, 5954; Angew. Chem. Int. Ed. 2006, 45, 5822.
[5] A. Mishra, M. K. R. Fischer, P. Bꢀuerle, Angew. Chem. 2009, 121, 2510;
Angew. Chem. Int. Ed. 2009, 48, 2474.
[6] P. V. Kamat, J. Phys. Chem. C 2007, 111, 2834.
[7] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev. 2010,
110, 6595.
[40] C. Y. Chen, J. G. Chen, S. J. Wu, J. Y. Li, C. G. Wu, K. C. Ho, Angew. Chem.
2008, 120, 7452; Angew. Chem. Int. Ed. 2008, 47, 7342.
[41] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[42] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[43] G. A. Petersson, M. A. Al-Laham, J. Chem. Phys. 1991, 94, 6081.
[44] M. J. T. Frisch, et al. Gaussian, Inc., Wallingford CT, 2009 See the Sup-
porting Information for the full citation.
[8] B. O’Regan, M. Grꢀtzel, Nature 1991, 353, 737.
[9] M. K. Nazeeruddin, F. De Angelis, S. Fantacci, A. Selloni, G. Viscardi, P.
Liska, S. Ito, B. Takeru, M. Grꢀtzel, J. Am. Chem. Soc. 2005, 127, 16835.
[10] T. Yakhanthip, S. Jungsuttiwong, S. Namuangruk, N. Kungwan, V. Pro-
marak, T. Sudyoadsuk, P. Kochpradist, J. Comput. Chem. 2011, 32, 1568.
[11] M. Pastore, E. Mosconi, F. De Angelis, M. Grꢀtzel, J. Phys. Chem. C 2010,
114, 7205.
[45] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24,
669.
[46] T. Yanai, D. P. Tew, N. C. Handy, Chem. Phys. Lett. 2004, 393, 51.
[47] N. M. O’Boyle, A. L. Tenderholt, K. M. Langner, J. Comput. Chem. 2008,
29, 839.
[12] J.-L. Brꢄdas, J. E. Norton, J. Cornil, V. Coropceanu, Acc. Chem. Res. 2009,
42, 1691.
Received: May 22, 2012
[13] Y.-S. Yen, Y.-C. Hsu, J. T. Lin, C.-W. Chang, C.-P. Hsu, D.-J. Yin, J. Phys.
Chem. C 2008, 112, 12557.
Published online on && &&, 0000
&12
&
www.chempluschem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2012, 00, 1 – 13
ÝÝ
These are not the final page numbers!

FULL PAPERS
Catch the sun: Seven new donor–
spacer–acceptor metal-free organic dyes
containing a fluorene and/or thiophene
unit as a conjugated spacer have been
prepared and their photovoltaic perfor-
mance studied (see a selection of struc-
tures). Theoretical calculations suggest
that the small difference in the efficien-
cy of these sensitizers is attributable
partly to the degree of effective charge
transfer upon excitation, not totally to
the absorption profile of the dye mole-
cules.
C.-G. Wu,* M.-F. Chung, H.-H. G. Tsai,*
C.-J. Tan, S.-C. Chen, C.-H. Chang,
T.-W. Shih
&& – &&
Fluorene-Containing Organic
Photosensitizers for Dye-Sensitized
Solar Cells
ChemPlusChem 2012, 00, 1 – 13
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chempluschem.org
&13
&
ÞÞ
These are not the final page numbers!