Assessment of new gem-silanediols as suitable sensitizers for dye-sensitized solar cells

Article abstract of DOI:10.1016/j.jorganchem.2012.10.012

Four novel gem-silanediol-containing organic dyes featuring a highly conjugated backbone have been synthesized in order to investigate their potential as active materials for photovoltaics. After spectroscopic characterization, the compounds showing the b

Full text of DOI:10.1016/j.jorganchem.2012.10.012

Journal of Organometallic Chemistry 723 (2013) 198e206
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Assessment of new gem-silanediols as suitable sensitizers for dye-sensitized
solar cells
Gabriella Barozzino Consiglio a b, Fabio Pedna a, Cosimo Fornaciari a, Fabrizia Fabrizi de Biani c,
,
,
b
Gabriele Marotta d, Paolo Salvatori d, Riccardo Basosi c, Filippo De Angelis d, Alessandro Mordini a
,
Maria Laura Parisi c, Maurizio Peruzzini b, Gianna Reginato b, Maurizio Taddei e, Lorenzo Zani f
,
*
a Dipartimento di Chimica “Ugo Schiff”, Università degli Studi di Firenze, Via della Lastruccia 13, 50019 Sesto Fiorentino, Florence, Italy
b Istituto di Chimica dei Composti Organometallici (CNR-ICCOM), Via Madonna del Piano 10, 50019 Sesto Fiorentino, Florence, Italy
c Dipartimento di Chimica, Università degli Studi di Siena, Via A. de Gasperi 2, 53100, Siena, Italy
d Istituto di Scienze e Tecnologie Molecolari (CNR-ISTM), c/o Dipartimento di Chimica, Università degli Studi di Perugia, Via Elce di Sotto 8, 06123 Perugia, Italy
e Dipartimento Farmaco-Chimico-Tecnologico, Università degli Studi di Siena, Via A. Moro 2, 53100 Siena, Italy
f Istituto per la Sintesi Organica e la Fotoreattività (CNR-ISOF), Via Piero Gobetti 101, 40129 Bologna, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 September 2012
Accepted 4 October 2012
Four novel gem-silanediol-containing organic dyes featuring a highly conjugated backbone have been
synthesized in order to investigate their potential as active materials for photovoltaics. After spectro-
scopic characterization, the compounds showing the best light harvesting and electrochemical proper-
ties were applied as sensitizers in dye-sensitized solar cells (DSSCs). Interestingly, photovoltaic cells built
Keywords:
using the new silanediol dyes showed low power conversion efficiencies (h), comparable to those ob-
Dye-sensitized solar cells
Organic sensitizer
gem-Silanediol
Cyclic voltammetry
TD-DFT calculations
tained with silicon-based sensitizers having simple azobenzene moieties as the light-harvesting units.
Such values are mostly due to unsatisfactory photocurrent densities; a computational study suggested
that the latter can be justified considering the insufficient degree of charge transfer taking place during
photoexcitation of the silicon-containing sensitizers, which is likely to make electron injection into the
TiO2 layer less efficient.
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
cyanoacrylic acid or rhodanine-3-acetic acid fragments as
anchoring groups [5,6]; a smaller number of dyes has been
In recent years, dye-sensitized solar cells (DSSCs) have attracted
much attention as efficient and potentially low-cost alternatives to
traditional silicon-based photovoltaic cells [1]. Initially, sensitizers
employed in such devices have been mostly metallorganic
compounds, in particular complexes of ruthenium with bipyridine
or terpyridine ligands [2,3]; DSSCs fabricated with such dyes have
reported, which are attached to the semiconductor by other func-
tionalities, such as carboxylate [2], phosphonate [7], sulfonate [8],
catechol [9] and pyridine [10] groups.
Recently, a series of simple sensitizers having silicon-based
anchoring groups was described by Unno, Hanaya and co-
workers [11]. The choice of such anchoring functions was based
on the fact that they form very strong bonds with a number of
metal oxides, including titania [12], therefore potentially allowing
the fabrication of highly stable photovoltaic cells. In those studies,
the photovoltaic performances of DSSCs built with dyes 1aec,
containing one or two azobenzene units, were compared with
that of an analogous device fabricated using 4-carboxyazobenzene
(2) as the sensitizer (Fig. 1).
It was found that the power conversion efficiencies of devices
containing compounds 1aec were higher than that of the cell
containing compound 2, thus suggesting that dyes possessing
a silicon anchoring function would yield more efficient solar cells
than the corresponding carboxylic acids. However, the azobenzene
unit is not an ideal chromophore for photovoltaic applications: first
reached solar-to-electric power conversion efficiencies
(h)
exceeding 11% [4]. More recently, a large number of metal-free
organic dyes have also been introduced, and the corresponding
solar cells have often displayed respectable power conversion
efficiencies [5].
In a dye-sensitized solar cell, efficient attachment of the sensi-
tizer to the semiconductor layer (most often constituted by TiO2)
is required to achieve smooth electron transfer between these
two elements. Most of the organic dyes reported so far possess
* Corresponding author. Tel.: þ39 051 6398301; fax: þ39 051 6398349.
E-mail address: lorenzo.zani@isof.cnr.it (L. Zani).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2012.10.012

G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
199
2. Results and discussion
The synthesis of silicon-containing substances S1 and S2 was
inspired by that of compound D5, with a suitable modification in
the last step to allow the introduction of the silanediol moiety
(Scheme 1).
Aldehyde 3 [19] was first subjected to a Wittig reaction with
2-thienylmethyltriphenylphosphonium chloride (4) [20] to afford
alkene 5 as a mixture of E/Z diastereoisomers in good yield.
A subsequent iodine-catalyzed equilibration yielded compound 5
as single E diastereoisomer, in 61% overall yield starting from
aldehyde 3.
Compound 5 was selectively metalated in position 2 of the
thiophene ring employing nBuLi, and the resulting thienyllithium
derivative was reacted with an excess of the appropriate aryl- or
alkyltrichlorosilane, to give compounds S1e2 after hydrolysis.
Whereas the reaction with phenyltrichlorosilane proceeded
smoothly to furnish silanol S1 in good yield, problems were
encountered in the purification of tert-butyl derivative S2, which
was therefore isolated in lower yield.
At this stage, preparation of gem-silanediol 6 (Fig. 3) was
attempted by reaction of compound 5 with nBuLi and tetra-
chlorosilane. Unfortunately, such reaction proved unsuccessful.
Access to compound 6 would have been highly interesting due
to the presence on its structure of two photoactive systems. The
reaction attempted for its preparation was similar to those used for
the synthesis of compounds S1e2, using tetrachlorosilane in place
of an alkyl- or aryltrichlorosilane, but its stoichiometry proved hard
to control and a mixture of products was consistently obtained,
preventing isolation of compound 6 in pure form.
Compounds S1e2 were found to have absorption maxima below
400 nm, which suggested an insufficient light-harvesting capacity
in the visible region (vide infra). To overcome this problem, we
reasoned that an increase in the conjugation length of the system
could have a beneficial effect, causing a red shift in the light
absorption of the resulting compounds. Thus, silanols S3e4,
featuring an extra thiophene ring and a further double bond
compared to S1e2, were prepared (Scheme 2).
Starting from common intermediate 5, treatment with nBuLi in
THF at ꢀ78 ꢁC followed by quench with N,N-dimethylformamide
and aqueous work-up afforded aldehyde 7 in good yield. The latter
compound was once more subjected to a Wittig reaction with
phosphonium salt 4 to give olefin 8 as a mixture of diastereoiso-
mers, which were equilibrated to yield compound 8 as single (E,E)
diastereoisomer. Unfortunately, in this case the Wittig reaction/
equilibration sequence proceeded in lower yield compared to
preparation of compound 5, possibly due to the lower reactivity of
the highly conjugated aldehyde 7 compared to aldehyde 3 in the
previous sequence of reactions.
Finally, compound (E,E)-8 was reacted with nBuLi and the
appropriate aryl- or alkyltrichlorosilane in a manner analogous to
compound 5 (see above), to afford silanols S3e4 in moderate yields
after quenching with water.
Fig. 4 shows that compounds S1e4 give rise to two structured
absorption bands in the UV and visible regions of the spectrum.
Fig. 1. Organic sensitizers 1 and 2, featuring azobenzene units [11].
of all, it has a short conjugation length, which results in absorption
maxima just above 300 nm; secondly a large part of the solar
energy absorbed by the dye can be lost due to cis/trans isomeriza-
tion around the nitrogen-nitrogen double bond [13], thereby
reducing the efficiency of the charge-transfer process toward the
semiconductor. The efficiencies of DSSCs fabricated using sensi-
tizers 1aec and 2 were indeed very low; therefore, a meaningful
assessment of the potential of silicon-containing groups as an
anchoring moieties for DSSC sensitizers was difficult, since the low
efficiency could be attributed to the use of an unsuitable
chromophore.
Meanwhile, Grätzel et al. showed [14] that sensitizers contain-
ing a trialkoxysilane anchoring group indeed yielded less efficient
DSSCs compared to the previously described dye JK-2 having
a cyanoacrylic anchoring function [15]. In particular, lower values of
photocurrent were measured, which were attributed to a low
density of absorbed siloxanes. This result is in contrast with the
findings of Unno et al., who measured comparable absorption
densities for compounds 1aec and 2 on TiO2 based on the corre-
sponding UV spectra [11]. Such discrepancy was surprising, given
that both groups attribute the attachment of the silicon-bearing
dyes on TiO2 surface to the formation of SieOeTi bonds [11,14].
Finally, it should be mentioned that disilanylene polimers carrying
silicon anchoring groups were also recently applied as DSSC
sensitizers, giving moderate power conversion efficiencies [16].
Prompted by these findings, we decided to independently assess
the viability of silanols as sensitizers for efficient and stable DSSCs.
To this end, we synthesized a series of novel silanol-containing
organic dyes whose structure was based on a previously reported
sensitizer and measured the photovoltaic performances of the
corresponding solar cells.
The structural model sensitizer was the well-known tripheny-
lamino-thienyl derivative D5 (Fig. 2) [17,18]. This substance was
selected because it had already been shown to yield solar cells of
good efficiency and has a rather simple structure which could easily
be modified in order to accommodate a silanol anchoring group.
Herein, we describe the synthesis and characterization of
compounds S1e4 (Fig. 2). In addition, we will report the results of
photovoltaic measurements conducted on the corresponding dye-
sensitized solar cells, as well as the results of TD-DFT studies that
provide a possible explanation of the observed performances.
Fig. 2. Structure of the organic dye D5 together with compounds S1e4 prepared in this study.

200
G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
Scheme 1. Preparation of silicon-containing compounds S1 and S2.
More in detail, silanols S1e2 have a first band around 302 nm and
behavior. They undergo two close subsequent oxidation processes,
the second of which is complicated by a further chemical reaction,
as shown by the presence of a third peak in Fig. 6. The first oxida-
tion process, observed at þ0.70 V and þ0.78 V vs. Ag/AgCl NaCl 3M
(corresponding to þ0.91 V and þ0.99 V vs. NHE) for S3 and S4,
respectively, can be assigned to the reversible removal of one
electron from the triphenylamine portion of the compounds. This
assignment was confirmed by the UVeVis spectroelectrochemical
characterization of S4 (vide infra). The second oxidation (observed
at þ0.86 V and þ0.88 V vs. Ag/AgCl NaCl 3M for S3 and S4,
respectively) is, on the contrary, assigned to removal of one elec-
tron from the bis-thiophene backbone. The values of the first
oxidation potentials are more positive than the Iꢀ/I3ꢀ redox poten-
tial (þ0.4 V vs. NHE), thus assuring dye regeneration by the Iꢀ/I3ꢀ
redox couple in a DSSC. The electrochemical behavior of S3e4 was
not affected by the nature of the working electrode material (Glassy
Carbon, ITO, Au). Considering the optical band gap (2.60e2.61 eV),
we calculated the excited state oxidation potential (E*ox) of the dyes
to be ꢀ1.70 V and ꢀ1.61 V (vs. NHE) for S3 and S4, respectively
(Table 1): such values are largely more negative than that of the
conduction band of TiO2 (ꢀ0.5 V) and should guarantee smooth
electron injection from the dye to the semiconductor. This is
a consequence of the relatively large band gap observed for the
silanediol-containing compounds, and it indicates that the latter
have a higher-lying LUMO compared to most cyanoacrilic-based
dyes [5,18]: such difference could help electron injection toward
the semiconductor, but is also responsible for a blue-shift of their
maximum absorption wavelength (vide supra).
a second, stronger band below 400 nm, at 377 and 376 nm, respec-
tively, with molar extinction coefficients of 2.2e2.4 ꢂ 104 Mꢀ1 cmꢀ1
.
This result indicates that the region where such compounds absorb
light is not appreciably influenced by the organic substituent on the
silicon atom (Ph vs. tBu). The molar extinction coefficients of S1e2
are comparable to those of several other organic sensitizers [5],
suggesting that silanols with red-shifted absorption could be used for
the construction of working DSSCs.
It is widely accepted that a photosensitizer to be employed in
dye-sensitized solar cells should be able to absorb light efficiently
in the visible and near-infrared region of the solar spectrum (where
the photon flux is maximized) in order to give rise to well-
performing photovoltaic devices [1]. Clearly, compounds S1e2
displaying absorption maxima barely within visible wavelengths
do not seem to fulfill such requirement. On the contrary, S3e4
display a first absorption band at around 350 nm and absorption
maxima at 425e427 nm, red-shifted of approximately 50 nm
compared to S1e2, providing a beneficial effect for the efficiency of
the resulting solar cells. Moreover, their molar extinction coeffi-
cients were much larger than those of the previous compounds,
with values in the 3.6e3.7 ꢂ 104
M
ꢀ1 cmꢀ1 range.
As far as the emission spectra are concerned, silanols S1e2
showed maxima at comparable wavelengths of 464e470 nm;
once again, compounds S3e4 displayed a notable red-shift of the
corresponding emission, with maxima for both species at 528 nm
(Fig. 5). Thus, compounds S3e4 are characterized by smaller Stokes
shifts (4480e4590 cmꢀ1) compared to those shown by compounds
S1e2 (5044e5249 cmꢀ1). The optical bandgaps of silanols S1e4
(corresponding to the zero-zero transition energy, E0e0) were esti-
mated from the intersection between their normalized absorption
and emission spectra, and turned out to be in the 2.60e2.98 eV
range.
The electrochemical characteristics of compounds S3e4 are
presented in Table 1.
To further characterize the spectroscopic features and the redox
behavior of the new silanediol compounds we carried out spec-
troelectrochemical measurements by observing the UVeVis spec-
tral changes of compound S4 upon oxidation in CH2Cl2 solution
(Fig. 7).
The optical properties of compounds S1e4 are summarized in
Table 1.
Due to the results of the photochemical measurements, we
decided to carry out the subsequent characterization only on those
compounds that showed superior light-harvesting properties,
namely S3e4. The ground-state oxidation potentials (Eox) of S3e4
were evaluated by means of cyclic voltammetry in CH2Cl2 solu-
tion (Fig. 6) [21]. The two compounds have very similar redox
As previously noted, the UVeVis spectrum of S4 shows also in
this solvent two structured bands, at 301 and 436 nm, respectively.
The first one is probably due to the electronic
p / p* transition
between orbitals centered on the triphenylamine fragment, while
the second one could be caused by intermolecular charge transfer
between the triphenylamine and the thiophene moieties; such an
absorption pattern has been previously described for organic DSSC
sensitizers [22]. As shown in Fig. 7, removal of the first electron
from S4 alters its UVeVis spectrum and, predictably, both these
bands decrease. At the same time, a couple of bands appear at lower
energies (centered at 688 and 772 nm) and one isosbestic point is
maintained at 490 nm, indicating the stability of the S4þ radical
cation. The appearance of this group of bands, due to the HOMO/
(HOMO-1) / SOMO transitions in triarylamine radical cations, is
a known phenomenon [23,24] and confirms our interpretation of
the redox behavior.
Fig. 3. Compound 6 whose synthesis failed.

G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
201
Scheme 2. Preparation of silicon-containing compounds S3 and S4.
The spectroscopic and electrochemical data reported above
suggest that silanediols S3e4 are compounds suitable to be used as
photosensitizers in dye-sensitized solar cells. Accordingly, we
proceeded to build a few devices incorporating S3 and S4 as the
photoactive components. The power conversion efficiency of the
cells was measured under AM 1.5 G simulated solar irradiation at
room temperature, using a Iꢀ/Iꢀ3 electrolyte (see Section 5 for
details). The corresponding JeV curves are displayed in Fig. 8, while
the relevant photovoltaic parameters are summarized in Table 2.
As can be seen from the table, both dyes S3 and S4 yielded
working solar cells; compound S3 showed superior photovoltaic
properties compared to S4, which resulted in a higher power
conversion efficiency. In both cases, however, efficiencies were low
compared to those given by typical organic sensitizers having
cyanoacrylic anchoring groups, mostly because of much smaller
short-circuit currents (see for example D5 in Table 2).
better Jsc values; indeed, the values obtained in this study are much
closer to those reported for azobenzene dyes 1aec in the works
published by Unno et al. [11].
The low photocurrent densities could be due to an insufficient
absorption of the dyes on the TiO2 surface. To verify this possibility,
we measured the density of absorbed dye S3 on titanium dioxide
after a typical sensitization procedure (see Section 5 for details),
and found a value of approx. 5.0 ꢂ 10ꢀ8 mol cmꢀ2, which is indeed
relatively low compared to values found for most organic dyes. On
the other hand, such value is quite close to those reported by
Grätzel for his trialkoxysilane dyes (6.2e7.3 ꢂ 10ꢀ8 mol cmꢀ2) [14],
thereby suggesting that a low amount of absorbed dye is not
responsible for the different performances of compounds S3e4.
3. Computational study
Interestingly, the trialkoxysilane dyes reported by Grätzel,
which have a similar backbone, gave higher efficiencies than those
obtained with dyes S3e4 [14], the difference being mostly due to
To help explaining the present results, we carried out density
functional theory (DFT) calculations using the Gaussian09 software
package [25], in an attempt to gather some information on the
electronic structure of the sensitizers.
Fig. 4. UVeVis absorption spectra of compounds S1e4 in MeOH solution. Concentra-
tions: S1e2, 1.12 ꢂ 10ꢀ4 M; S3e4, 5.6 ꢂ 10ꢀ5 M.
Fig. 5. Normalized emission spectra of compounds S1e4 in MeOH solution (irradiation
at maximum absorption wavelength).

202
G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
Table 1
Summary of the optical and electrochemical properties of compounds S1e4.
a
Compound
lmax(abs) (nm)a
ε (Mꢀ1 cmꢀ1
)
lmax(em) (nm)
E0e0 (eV)b
Eox (V)c
E*ox (V)c
S1
S2
S3
S4
377
376
425
427
22,300
24,200
36,900
37,000
470
464
528
528
2.95
2.98
2.61
2.60
þ0.91 (þ0.70)
þ0.99 (þ0.78)
ꢀ1.70
ꢀ1.61
a
UVeVis spectra were recorded in MeOH at the following concentrations: S1e2, 1.12 ꢂ 10ꢀ4 M; S3e4, 5.6 ꢂ 10ꢀ5 M.
b
c
Estimated from the intersection of the normalized absorption and emission spectra.
Potentials vs. NHE (potentials vs. Ag/AgCl NaCl 3M are given in parentheses).
First of all, the ground-state geometries of compounds S1 and S3
compounds is mainly localized on the conjugated scaffold of the
molecules, with a sizeable contribution coming from the terminal
diphenylamino-donor groups. In the case of the LUMO, however,
the electron density is also substantially found on the middle part of
the two molecular structures; despite the fact that the coefficients
seem to be slightly larger on the terminal thiophene rings, the
contribution arising from the silanol oxygen atoms looks negligible.
TD-DFT computations conducted at the MPW1K/6-31G* level
both in vacuo and in acetonitrile solution, indicated that the exci-
tation process for both dyes is mainly due to the HOMO / LUMO
transition, accounting for approx. 80% of the total, and that the
oscillator strength for dye S3 is much larger than that of dye S1; this
result is consistent with the molar extinction coefficients measured
for both substances and reported in Table 1. The values obtained for
the maximum absorption wavelengths were 364 (380) nm for S1
and 437 (457) nm for S3, in vacuo (in solution), in good agreement
with the experimental values reported in Table 1, 377 and 425 nm
for S1 and S3, respectively.
were optimized in vacuo at the B3LYP/6-31G* level. Subsequently,
we performed calculations at the B3LYP and MPW1K/6-31G* level
on the same compounds, both in the gas phase and in acetonitrile
solution, in order to determine the frontier orbital energies and the
corresponding electronic distribution.
The results are reported in Table 3. It can be observed that the
HOMOeLUMO energy gap computed for compound S1 is larger
than that relative to compound S3 both in vacuum and in aceto-
nitrile (3.25 eV vs. 2.69 eV and 3.16 eV vs. 2.63 eV, respectively); this
result is in agreement with the relative magnitude of the optical
band gap values derived from spectroscopic measurements (vide
supra), and can be explained in terms of the more extended
conjugation present on the molecular scaffold of S3 compared to S1.
As far as the electronic distribution is concerned (Fig. 9), it
appears that, in line with our expectations, the HOMO of both
Based on these findings, the actual charge-transfer nature of the
photoexcitation process appears questionable. As a consequence,
efficient electronic communication between the dye molecules and
the semiconductor surface could be hampered; the low efficiency
of the electron injection process could therefore serve as an
explanation of the poor power conversion efficiencies observed for
the cells built using sensitizers S3e4.
4. Conclusions
Despite the remarkable results obtained so far with fully organic
compounds as light-harvesting materials for dye-sensitized solar
cells [5], the search for new and more efficient sensitizers is still
a very active field of research. In particular, optimization of the
anchoring moiety of the dyes could give access to more powerful
and, above all, more stable photovoltaic devices. From this point
of view, the silanediol function could represent a promising
Fig. 6. Semiderivative deconvoluted cyclic voltammograms recorded at a platinum
electrode of a CH2Cl2 solution of S3 (a, 2.0 mM) and S4 (b, 2.8 mM). [NBu4][PF6] (0.2 M)
was used as the supporting electrolyte. Scan rates: 0.2 V sꢀ1 (solid line), 0.02 V sꢀ1
(dashed line).
Fig. 7. UVeVis spectra recorded upon stepwise oxidation of S4 in an OTTLE cell in
CH2Cl2 solution; supporting electrolyte: [NBu4][PF6] (0.2 M).

G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
203
Table 3
Energies of the lowest unoccupied and highest occupied KohneSham orbitals of
compounds S1 and S3 in vacuum and in CH3CN solution with the B3LYP functional.
Energy in eV.
S1
S3
Vac
AcN
Vac
AcN
LD2
LD1
L
H
H L 1
H L 2
ꢀ0.34
ꢀ0.58
ꢀ1.50
ꢀ4.75
ꢀ5.62
ꢀ6.58
ꢀ0.36
ꢀ0.50
ꢀ1.67
ꢀ4.83
ꢀ5.76
ꢀ6.71
ꢀ0.54
ꢀ0.95
ꢀ1.94
ꢀ4.63
ꢀ5.34
ꢀ6.12
ꢀ0.44
ꢀ1.05
ꢀ2.09
ꢀ4.72
ꢀ5.34
ꢀ6.25
5. Materials and methods
5.1. General synthetic remarks
Fig. 8. JeV characteristics of DSSCs sensitized with compounds S3 (A) and S4 (B).
All air-sensitive reactions were performed using Schlenk tech-
niques [26]. Tetrahydrofuran (THF) was distilled from sodium
benzophenone ketyl radical, CH2Cl2 and N,N-dimethylformamide
were stored under nitrogen over 4 Å molecular sieves. Petroleum
ether was the 40e60 ꢁC boiling fraction. Compound D5 [17], alde-
hyde 3 [19] and 2-thienylmethyltriphenylphosphonium chloride
(4) [20] were prepared according to literature methods, while all
the other chemicals employed were commercially available
and used as received. Thin layer chromatography was carried out
on aluminum-supported Merck 60 F254 plates; flash column
chromatography [27] was performed using Merck Kieselgel 60
(300e400 mesh) as the stationary phase.
alternative to the usual cyanoacrylic acid functional group, due to
the strong bonds it forms with metal oxides, and in particular with
titania.
In this work, we synthesized a series of novel silanol-bearing
compounds S1e4, inspired by the molecular structure of organic
dye D5, and assessed their optical and photoelectrochemical
properties, as well as their capacity to serve as photoactive mate-
rials for DSSCs.
Solar cells built with dyes S3e4 had power conversion effi-
ciencies in the 0.2e0.3% range, with S3 being the most effective
sensitizer. Such values are relatively close to those previously ob-
tained with silicon-containing sensitizers having a simple azo-
benzene group as their light-harvesting unit [11]. Therefore,
attachment of an extended organic chromophore, displaying
a more intense and red-shifted absorption spectrum, to the gem-
silanediol moiety does not seem to improve the photovoltaic
properties of the resulting sensitizers.
Computational studies suggested that the photoexcitation
process in compound S3 happens with a smaller degree of charge
transfer compared to a cyanoacrylic dye, which should reduce
the efficiency of the electron injection process from the dye to the
nanocrystalline semiconductor, thereby possibly explaining the
low conversion efficiencies observed with the photosensitizers
described in the present study.
1H-NMR spectra were recorded at 200, 300, or 400 MHz, and
13C-NMR spectra were recorded at 50.3, 75.5, or 100.6 MHz,
respectively, on Varian Gemini, Varian Mercury and Bruker Avance
series instruments. Chemical shifts were referenced to the residual
solvent peak (CHCl3,
d d 77.0 ppm for
7.26 ppm for 1H-NMR and
13C-NMR). ESI-MS spectra were obtained by direct infusion using
a Thermo Scientific LCQ-FLEET instrument and are reported in the
form m/z (intensity relative to base ¼ 100). IR spectra were recor-
ded using a PerkineElmer Spectrum BX FT-IR spectrometer and are
reported in cmꢀ1. UVeVis spectra were recorded with a Varian Cary
400 spectrometer, and fluorescence spectra were recorded with
a Varian Eclipse instrument, irradiating the sample at the wave-
length corresponding to maximum absorption in the UV spectrum.
As a consequence, we conclude that the gem-silanediol moiety
does not appear to be a suitable replacement for the cyanoacrylic
anchoring group in DSSC sensitizers. In this context, we would like
to point out that the primary objective of the study was not to
obtain cells of very high photovoltaic efficiency, but rather to verify
the previous findings using rationally designed compounds.
However, the promising results obtained with trialkoxysilane
sensitizers and the possibility to form strong bonds with titanium
dioxide should continue to foster research aimed at finding efficient
silicon-containing photosensitizers.
5.2. (E)-2-[2-(4-diphenylaminophenyl)vinyl]thiophene (5)
In a round-bottom two-necked flask under an inert atmosphere
of nitrogen was introduced phosphonium salt 4 (2.0 g, 5.1 mmol,
1.4 eq.), which was suspended in anhydrous THF (36 mL). The
suspension was cooled down to ꢀ78 ꢁC and potassium tert-butoxide
(0.49 g, 4.4 mmol,1.2 eq.) was then added; the resulting mixture was
stirred at ꢀ78 ꢁC for 1 h, after which aldehyde 3 (1.18 g, 4.3 mmol,
1.0 eq.) was added. The reaction mixture was warmed to rt and
stirred overnight. After a TLC check (petroleum ether/EtOAc 50:1)
had shown the disappearance of the starting material, the reaction
was quenched with water. The two layers were separated, and the
aqueous layer was washed with CH2Cl2 (3 ꢂ 50 mL). The combined
organic layers were dried over Na2SO4 and filtered through a short
pad of silica, eluting with CH2Cl2, affording 1.5 g (4.24 mmol) of crude
olefin 5 as a mixture of E and Z diastereoisomers.
Table 2
Summary of the photovoltaic parameters of the cells built using S3e4 dyes
compared with a standard organic dye (D5).
Compounda
Jsc (mA/cm2)
Voc (V)
ff
h (%)
S3
S4
D5
0.95
0.66
9.91
0.48
0.43
0.61
0.62
0.58
0.71
0.3
0.2
4.3
Crude compound 5 was dissolved in CHCl3 (68 mL), and iodine
(0.11 g, 0.43 mmol, 0.1 eq.) was added at rt. The resulting red-brown
mixture was stirred overnight at rt. The organic layer was then
washed with satd. aq. sodium metabisulphite (50 mL) and
water (50 mL); the aqueous phase was then extracted with CHCl3.
a
JeV curves measured under AM 1.5
G
simulated solar irradiation
(100 mW cmꢀ2) at room temperature; film thickness approx. 10
mm; active area
0.2 cm2.

204
G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
Fig. 9. Wave function plots for compounds S1 and S3 in CH3CN solution.
The combined organic layers were dried over Na2SO4 and filtered
to give the crude product as a yellow oil. Purification by flash
column chromatography (petroleum ether/EtOAc 50:1) afforded
pure compound (E)-5 as a yellow solid (0.92 g, 2.6 mmol, 61%).
1.0 eq.), tert-butyltrichlorosilane (80 mg, 0.42 mmol, 2.0 eq.), anhy-
drous THF (10 mL total). After work-up, drying and filtration, crude
product S2 was obtained. The latter was purified by recrystallization
from CH2Cl2/hexane followed by flash column chromatography
(petroleum ether/EtOAc 3:1) to afford pure silanol S2 as a brown
amorphous solid (13 mg, 0.03 mmol, 13%). 1H-NMR (300 MHz;
1H-NMR (300 MHz, CDCl3):
d
(ppm) 6.88 (d, J ¼ 15.9 Hz, 1H),
6.98e7.01 (m, 6H), 7.10e7.13 (m, 4H), 7.16e7.18 (m, 2H;), 7.24e7.30
(m, 4H), 7.33e7.35 (m, 2H). 13C-NMR (50.3 MHz, CDCl3):
(ppm)
d
CDCl3):
(m, 4H), 7.09e7.14 (m, 4H), 7.15e7.29 (m, 7H), 7.31e7.35 (m, 2H).
13C-NMR (50.3 MHz; CDCl3):
(ppm) 25.2, 29.5, 119.2, 122.9, 123.1,
d
(ppm) 1.45 (s, 9H), 6.95 (d, J ¼ 15.9 Hz, 1H), 7.01e7.07
120.1,123.0,123.4,123.8,124.4,125.5,127.1,127.5,127.8,129.2,131.0,
143.2, 147.2, 147.5. ESI-MS: m/z 354 [M þ H]þ.
d
124.4, 124.5, 124.7, 126.3, 127.0, 129.1, 130.5, 137.0, 147.1, 147.2, 149.3.
ESI-MS: m/z 472 [M þ H]þ.
5.3. (E)-[2-[2-(4-diphenylaminophenyl)vinyl]thienyl](phenyl)
silandiol (S1)
5.5. (E)-5-[2-(4-diphenylaminophenyl)vinyl]thiophen-2-al (7)
In a Schlenk tube under an inert atmosphere of nitrogen was
placed nBuLi (sol. 1.6 M in hexane, 0.2 mL, 19 mg, 0.32 mmol,
1.6 eq.). Hexane was partially removed under vacuum and the
concentrated solution was cooled down to ꢀ78 ꢁC. Anhydrous THF
(5 mL) was then added, followed by olefin 5 (70 mg, 0.20 mmol,
1.0 eq.), dissolved in anhydrous THF (5 mL). The resulting solution
was stirred at ꢀ78 ꢁC for 1 h and then at 0 ꢁC for 0.5 h, during which
it became dark. Phenyltrichlorosilane (0.064 mL, 0.40 mmol,
2.0 eq.) was subsequently added; the reaction mixture was warmed
to rt and stirred overnight. The reaction was then checked by TLC
(petroleum ether/EtOAc 3:1) and quenched with water (10 mL). The
solution was diluted with CH2Cl2 (10 mL) and the layers were
separated. The aqueous layer was extracted with CH2Cl2 (10 mL);
the combined organic layers were dried over Na2SO4 and filtered to
afford the crude product. The latter was purified by recrystalliza-
tion from CH2Cl2/hexane followed by washing with pentane to give
silanol S1 as a brown amorphous solid (63 mg, 0.13 mmol, 64%).
In a Schlenk tube under an inert atmosphere of nitrogen was
placed nBuLi (sol. 1.6 M in hexane, 3.9 mL, 0.40 g, 6.26 mmol,
2.0 eq.). Hexane was partially removed under vacuum and the
concentrated solution was cooled down to ꢀ78 ꢁC. Anhydrous THF
(20 mL) was then added, followed by compound 5 (1.14 g,
3.23 mmol, 1.0 eq.), dissolved in anhydrous THF (12 mL). The
resulting solution was stirred at ꢀ78 ꢁC for 1 h and then at 0 ꢁC for
0.5 h, during which it became dark. N,N-dimethylformamide
(1.18 g, 16.1 mmol, 5.0 eq.) was added, and the resulting mixture
was warmed to rt and stirred overnight. The reaction was then
checked by TLC (petroleum ether/EtOAc 30:1) and quenched with
10% aq. HCl (100 mL). The aqueous layer was extracted with CH2Cl2
(3 ꢂ 50 mL). The combined organic layers were dried over Na2SO4,
filtered and concentrated under vacuum to give the crude reaction
product. The latter was purified by flash column chromatography
(gradient: petroleum ether/EtOAc 30:1 to 3:1) to afford pure
aldehyde 7 as a yellow solid (1.07 g, 2.81 mmol, 87%). 1H-NMR
1H-NMR (400 MHz; CDCl3):
(m, 6H), 7.07e7.21 (m, 6H), 7.27e7.30 (m, 2H), 7.32e7.38 (m, 3H),
7.71e7.74 (m, 2H). 13C-NMR (100 MHz; CDCl3):
(ppm) 121.0, 124.2,
d (ppm) 6.86e6.90 (m, 4H), 6.96e7.00
(300 MHz; CDCl3):
d (ppm) 7.01e7.14 (m, 13H), 7.25e7.38 (m, 4H),
d
7.64 (d, J ¼ 6.3 Hz, 1H), 9.83 (s, 1H).
The analytical data are in agreement with those reported in the
literature [18].
124.3, 125.6, 127.8, 128.3, 128.7, 129.7, 130.4, 131.1, 132.4, 135.2,
135.6, 137.2, 138.0, 148.7, 148.8, 150.6. ESI-MS: m/z 492 [M þ H]þ.
5.4. (E)-[2-(2-(4-diphenylaminophenyl)vinyl)thienyl](tert-butyl)
5.6. (E,E)-2-[((5-(4-diphenylaminophenyl)vinyl)thien-2-yl)vinyl]
silandiol (S2)
thiophene (8)
The title compound was prepared according to a procedure
analogous to that used for the synthesis of compound S1 (see above),
with tert-butyltrichlorosilane as the electrophile. The following
compounds and quantities were used: nBuLi (sol. 1.6 M in hexane,
0.2 mL, 19 mg, 0.32 mmol, 1.6 eq.), olefin 5 (74 mg, 0.21 mmol,
The title compound was prepared according to a procedure analo-
gous to that used for the synthesis of compound 5 (see above),
employing aldehyde 7instead of aldehyde 3. The following compounds
and quantities were used: aldehyde 7 (1.07 g, 2.81 mmol, 1.0 eq.),
2-thienylmethyltriphenylphosphonium chloride (4,1.53 g, 3.88 mmol,

G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
205
1.38 eq.), potassium tert-butoxide (0.41 g, 3.65 mmol, 1.3 eq.), anhy-
drous THF (total 140 mL). After work-up, drying and filtration, crude
product 10 was obtained as a mixture of diastereoisomers (1.22 g,
2.65 mmol). The latter was treated with iodine (0.071 g, 0.028 mmol,
0.1 eq.) in CHCl3 (40 mL) according to the same procedure reported
previously for olefin 5 (see above). After work-up, the crude product
was purified by flash column chromatography (petroleum ether/EtOAc
50:1), to give compound (E,E)-8 as a yellow/brown solid (0.61 g,
Fluka and used as received. Cyclic voltammetry was performed in
a three-electrode cell containing a platinum working electrode, an
Ag/AgCl NaCl (3M) and a platinum auxiliary electrode. All the
potential values are referred to the Ag/AgCl electrode. Under the
present experimental conditions, the one-electron oxidation of
ferrocene occurs at Eꢁ0 ¼ þ0.42 V. UVeVis spectroelectrochemical
measurements were carried out using a PerkineElmer Lambda 2
UVeVis spectrophotometer and a Optically Transparent Thin Layer
spectroelectrochemical (OTTLE) cell in quartz glass with an optical
path-length of 1 mm, equipped with a Pt-minigrid working elec-
trode, a Pt auxiliary electrode and an Ag/AgCl NaCl (3M) reference
electrode. A nitrogen-saturated CH2Cl2 solution of the compound
under study was used with [NBu4][PF6] (0.2 M) as the supporting
electrolyte. The in situ spectroelectrochemistry has been recorded
by collecting spectra in the spectral window 250e1000 nm during
1.32 mmol, 47%). 1H-NMR (300 MHz; CDCl3):
d (ppm) 6.84
(d, J ¼ 15.6 Hz, 1H), 6.88e6.91 (m, 2H), 6.98e7.07 (m, 9H), 7.09e7.13
(m, 4H), 7.18 (d, J ¼ 4.8 Hz, 1H), 7.23e7.29 (m, 4H), 7.32 (d, J ¼ 8.8 Hz,
2H). 13C-NMR (50.3 MHz; CDCl3):
d (ppm) 120.1, 121.3, 121.5, 123.1,
123.3, 124.3, 124.5, 126.0, 126.5, 127.0, 127.1, 127.6, 128.1, 129.2, 130.9,
140.8, 142.2, 142.4, 147.40, 147.44. ESI-MS: m/z 462 [M þ H]þ.
5.7. (E,E)-[2-(((5-(4-diphenylaminophenyl)vinyl)thien-2-yl)vinyl)
thien-2-yl](phenyl)silandiol (S3)
the stepwise oxidation of the compound (Ei ¼ þ0.55 V, E ¼ 50 mV,
D
D
t ¼ 1 min). A BAS 100A electrochemical analyser was used as the
polarising unit during all the experiments.
The title compound was prepared according to a procedure
analogous to that used for the synthesis of compound S1 (see above),
employing compound 8 as the starting material instead of 5. The
following compounds and quantities were used: nBuLi (sol. 1.6 M in
hexane, 0.06 mL, 6.1 mg, 0.10 mmol, 1.5 eq.), (E,E)-8 (30 mg,
0.065 mmol, 1.0 eq.), phenyltrichlorosilane (28 mg, 0.13 mmol,
2.0 eq.), anhydrous THF (total 4 mL). After work-up, drying and
filtration, crude product S3 was obtained. The latter was purified by
precipitation from Et2O/pentane to afford pure silanol S3 as a brown
amorphous solid (14 mg, 0.024 mmol, 37%). 1H-NMR (400 MHz;
5.10. Solar cells fabrication and characterization
FTO glass (TEC-15/2.2 mm thickness, Solaronix) was used for
transparent conducting electrodes. The substrate was first cleaned
in an ultrasonic bath using a detergent solution, followed by acetone
and ethanol (each step 15 min long). The FTO glass plates were
immersed into a 40 mM aqueous TiCl4 solution at 70 ꢁC for 30 min
and washed with water and ethanol. Dyesol 18NR-AO TiO2 paste
was spread on the FTO glass plates by doctor blading. The TiO2
CDCl3):
7.25e7.35 (m, 8H), 7.38e7.47 (m, 3H), 7.70e7.75 (m, 2H). 13C-NMR
(100 MHz; CDCl3): (ppm) 120.2, 120.8, 120.9, 123.0, 123.3, 123.5,
124.8, 126.6, 126.7, 127.2, 127.4, 127.8, 128.1, 128.5, 129.4, 129.6, 130.9,
131.1, 134.4, 137.9, 140.8, 142.9, 147.6, 149.7. IR (neat)
¼ 3406, 3068,
d
(ppm) 3.35 (bs, 2H), 6.84e6.90 (m, 2H), 6.96e7.15 (m,12H),
coated electrodes (thickness approx. 10 m
m, active area 0.2 cm2)
were gradually heated under air flow at 325 ꢁC for 5 min, at 375 ꢁC
for 5 min, at 450 ꢁC for 15 min, and 500 ꢁC for 15 min. After the
sintering process, the TiO2 film was treated with 40 mM TiCl4
solution, rinsed with water and ethanol. The electrodes were heated
at 500 ꢁC for 30 min and after cooling (80 ꢁC) were immersed into
sensitizing baths (anhydrous THF solutions of the S3/S4 dyes in
0.7 mM concentration). The sensitization process was conducted at
room temperature for 4 h and then at 50 ꢁC for further 18 h.
Counter electrodes were prepared by coating a pre-drilled FTO
plate (TEC-15/2.2 mm thickness, Solaronix) with a drop of H2PtCl6
solution (2 mg of Pt in 1 mL of ethanol) and heating at 400 ꢁC for
15 min. The TiO2 sensitized photoanode and Pt counter electrode
were assembled into a sealed sandwich-type cell by a hot-melt
d
n
2924, 1650, 1589, 1508, 1492, 1428, 1328, 1280, 1124, 1073 cmꢀ1
.
ESI-MS: m/z 598 [M ꢀ H]ꢀ (negative channel).
5.8. (E,E)-[2-(((5-(4-diphenylaminophenyl)vinyl)thien-2-yl)vinyl)
thien-2-yl](tert-butyl)silandiol (S4)
The title compound was prepared according to a procedure
analogous to that used for the synthesis of compound S1 (see above),
employing compound 8 as the starting material instead of 5, and
tert-butyltrichlorosilane as the electrophile instead of phenyl-
trichlorosilane. The following compounds and quantities were used:
nBuLi (sol. 1.6 M in hexane, 0.09 mL, 9.3 mg, 0.14 mmol, 1.5 eq.),
(E,E)-8 (43 mg, 0.093 mmol, 1.0 eq.), tert-butyltrichlorosilane (35 mg,
0.19 mmol, 2.0 eq.), anhydrous THF (total 5.5 mL). After work-up,
drying and filtration, crude product S4 was obtained. The latter
was purified by flash column chromatography (petroleum ether/
EtOAc 3:1) to afford pure silanol S4 as a brown amorphous solid
ionomer film (Surlyn, 25 mm thickness, Dyesol).
The cells were vacuum-filled with a few drops of Iolitech
ES-0004 HP electrolyte, containing 1-butyl-3-methylimidiazolium
iodide, iodine, guanidinium thiocyanate and tert-butylpyridine in
a mixture of valeronitrile and acetonitrile, with 0.1 M LiI added.
Then, the hole drilled on the back of the counter electrode was
sealed by using an additional Surlyn patch and a cover glass and
finally a conductive Ag-based paint was deposed at the electrical
contacts.
(16 mg, 0.028 mmol, 30%). 1H-NMR (400 MHz; CDCl3):
d (ppm) 1.04
(s, 9H), 6.84 (d, J ¼ 16.0 Hz, 1H), 6.88e6.91 (m, 2H), 7.02e7.07
Photovoltaic measurements were recorded by means of an AM
1.5 solar simulator equipped with a Xenon lamp (LOT-ORIEL LS
0106). The power of incoming radiation, set at 100 mW/cm2, was
checked by a piranometer. JeV curves were obtained by applying
an external bias to the cell and measuring the generated photo-
current with a Keithley model 2400 digital source-meter, under
the control of dedicated LabTracer 2.0 software. A black shading
mask was employed to avoid overestimation of the measured
parameters.
(m, 7H), 7.10e7.12 (m, 5H), 7.24e7.28 (m, 6H), 7.32 (dd, J ¼ 8.1 Hz,
J ¼ 3.4 Hz, 1H). 13C-NMR (100 MHz; CDCl3):
d (ppm) 25.7, 29.8, 120.2,
120.8, 122.9, 123.3, 123.5, 124.7, 126.7, 127.2, 127.4, 127.7, 128.5, 129.4,
131.0, 131.7, 137.5, 137.6, 140.8, 142.8, 147.6, 149.0. IR (neat)
n
¼ 3416,
3058, 2951, 2927, 2854, 1651, 1590, 1507, 1493, 1329, 1281, 1175,
1072 cmꢀ1. ESI-MS: m/z 580 [M þ H]þ.
5.9. Electrochemical and spectroelectrochemical measurements
Electrochemical measurements were performed in a dichloro-
methane solution containing [NBu4][PF6] 0.2 M as the supporting
electrolyte. Anhydrous 99.9% dichloromethane was obtained from
Aldrich. Electrochemical grade [NBu4][PF6] was purchased from
5.11. Measurement of the density of absorbed dye on TiO2
A nanocrystalline TiO2 electrode (surface area 0.88 cm2) similar
to those used for the photovoltaic measurements (see Section 5.10)

206
G. Barozzino Consiglio et al. / Journal of Organometallic Chemistry 723 (2013) 198e206
was immersed in a 7.0 ꢂ 10ꢀ4 M solution of dye S3 in THF at 50 ꢁC
for 16 h. The stained dark yellow electrode was removed from the
solution, dried under a stream of nitrogen and immersed in 6 mL of
a 0.01 M KOH solution in MeOH/THF 2:1 at rt for 5 h; after this time
full discoloration of the electrode was observed. The absorbance of
the resulting yellow solution was measured by UVeVis spectros-
copy and compared to that of a standard 5.0 ꢂ 10ꢀ5 M solution of
compound S3 in the same solvent/base mixture. The amount of dye
present in the unknown solution was calculated and divided by the
electrode surface area, yielding a density value of approximately
(b) C. Houarner, E. Blart, P. Buvat, F. Odobel, Photochemical and Photobio-
logical Sciences 4 (2005) 200e204;
(c) F. Odobel, E. Blart, M. Lagrée, M. Villieras, H. Boujtita, N. El Murr,
S. Caramori, C.A. Bignozzi, Journal of Materials Chemistry 13 (2003) 502e510.
[8] F.Y. Li, J. Zheng, C.H. Huang, L.P. Jin, J.Y. Zhuang, J.Q. Guo, X.S. Zhao, T.T. Liu,
Journal of Physical Chemistry B 104 (2000) 5090e5095.
[9] C.R. Rice, M.D. Ward, M.K. Nazeeruddin, M. Grätzel, New Journal of Chemistry
24 (2000) 651e652.
[10] Y. Ooyama, S. Inoue, T. Nagano, K. Kushimoto, J. Ohshita, I. Imae,
K. Komaguchi, Y. Harima, Angewandte Chemie International Edition 50 (2011)
7429e7433.
[11] (a) K. Kakiage, Y. Nakada, T. Kogure, M. Yamamura, T. Kyomen, M. Unno,
M. Hanaya, Silicon Chemistry 3 (2008) 303e305;
5.0 ꢂ 10ꢀ8 mol cmꢀ2
.
(b) M. Unno, K. Kakiage, M. Yamamura, T. Kogure, T. Kyomen, M. Hanaya,
Applied Organometallic Chemistry 24 (2010) 247e250;
(c) K. Kakiage, M. Yamamura, E. Fujimura, T. Kyomen, M. Unno, M. Hanaya,
Chemistry Letters 39 (2010) 260e262.
5.12. Computational studies
[12] C.L. Lin, M.Y. Yeh, C.H. Chen, S. Sudhakar, S.J. Luo, Y.C. Hsu, C.Y. Huang, K.C. Ho,
T.Y. Luh, Chemistry of Materials 18 (2006) 4157e4162.
Geometry optimizations of compounds S1eS3 were performed
in vacuo using the B3LYP functional, with a 6-31G* basis set using
the Gaussian09 (G09) program package.
TD-DFT calculations were performed using the Gaussian03
(G03) program package, in vacuo and in CH3CN solution, with both
the B3LYP and MPW1K functionals and 6-31G* basis set.
[13] E.W.G. Diau, Journal of Physical Chemistry
references therein.
A 108 (2004) 950e956. and
[14] C. Baik, D. Kim, M.S. Kang, S.O. Kang, J. Ko, M.K. Nazeeruddin, M. Grätzel, Journal
of Photochemistry and Photobiology A: Chemistry 201 (2009) 168e174.
[15] S. Kim, J.K. Lee, S.O. Kang, J. Ko, J.H. Yum, S. Fantacci, F. De Angelis, D. Di Censo,
M.K. Nazeeruddin, M. Grätzel, Journal of the American Chemical Society 128
(2006) 16,701e16,707.
[16] (a) J. Ohshita, J. Matsukawa, M. Hara, A. Kunai, S. Kawajima, Y. Ooyama,
Y. Harima, M. Kakimoto, Chemistry Letters 37 (2008) 316e317;
(b) D. Tanaka, J. Ohshita, Y. Ooyama, T. Mizumo, Y. Harima, Journal of
Organometallic Chemistry (2012). http://dx.doi.org/10.1016/j.jorganchem.
2012.08.019.
[17] D.P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt, L. Sun,
Chemical Communications (2006) 2245e2247.
[18] D.P. Hagberg, T. Marinado, K.M. Karlsson, K. Nonomura, P. Qin, G. Boschloo,
T. Brinck, A. Hagfeldt, L. Sun, Journal of Organic Chemistry 72 (2007) 9550e9556.
[19] X. Tang, W. Liu, J. Wu, C.S. Lee, J. You, P. Wang, Journal of Organic Chemistry 75
(2010) 7273e7278.
[20] (a) Y. Xin, X. Wu, Y. Shen, Journal of Fluorine Chemistry 40 (1988) 15e22;
(b) A.G. Shilabin, A.A. Entezami, European Polymer Journal 36 (2000)
2005e2020.
[21] Fig. 6 shows the semi-derivative deconvoluted voltammograms of (a) S3 and
(b) S4. This mathematical treatment was adopted to accentuate peak sepa-
ration. See: J.M. Pedrosa, M.T. Martín, J.J. Ruiz, L. Camacho Journal of Elec-
troanalytical Chemistry 523 (2002) 160e168.
Acknowledgements
L.Z. and M.L.P. are grateful to Regione Toscana within the
POR-FSE 2007e2013 program for
(FOTOSENSORG project). P.S. and F.D.A. thank MIUR-PRIN 2008 for
financial support.
a post-doctoral fellowship
References
[1] (a) A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chemical Review 110
(2010) 6595e6663;
(b) K. Kalyanasundaram (Ed.), Dye-sensitized Solar Cells, EPFL Press, Lausanne,
2010.
[2] (a) B. O’Regan, M. Grätzel, Nature 353 (1991) 737e740;
(b) M.K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mueller,
P. Liska, N. Vlachopoulos, M. Grätzel, Journal of the American Chemical Society
115 (1993) 6382e6390.
[22] (a) M. Liang, W. Xu, F. Cai, P. Chen, B. Peng, J. Chen, Z. Li, Journal of Physical
Chemistry C 111 (2007) 4465e4472;
(b) J. Song, F. Zhang, C. Li, W. Liu, B. Li, Y. Huang, Z. Bo, Journal of Physical
Chemistry C 113 (2009) 13391e13397.
[23] C. Lambert, G. Nöll, Journal of the American Chemical Society 121 (1999)
8434e8442.
[3] (a) M.K. Nazeeruddin, P. Pechy, M. Grätzel, Chemical Communications (1997)
1705e1706;
(b) M.K. Nazeeruddin, P. Pechy, T. Renouard, S.M. Zakeeruddin, R. Humphry-
Baker, P. Comte, P. Liska, L. Cevey, E. Costa, V. Shklover, L. Spiccia, G.B. Deacon,
C.A. Bignozzi, M. Grätzel, Journal of the American Chemical Society 123 (2001)
1613e1624.
[24] S. Dapperheld, E. Steckhan, K.H. Grosse Brinkhaus, T. Esch, Chemische Berichte
124 (1991) 2557e2567.
[25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi,
J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin,
R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski,
G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas,
J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1,
Gaussian Inc., Wallingford, CT, 2009.
[4] C.Y. Chen, M. Wang, J.Y. Li, N. Pootrakulchote, L. Alibabaei, C.H. Ngoc-le,
J.D. Decoppet, J.H. Tsai, C. Grätzel, C.G. Wu, S.M. Zakeeruddin, M. Grätzel, ACS
Nano 3 (2009) 3103e3109.
[5] (a) Y. Ooyama, Y. Harima, European Journal of Organic Chemistry (2009)
2903e2934;
(b) A. Mishra, M.K.R. Fischer, P. Bäuerle, Angewandte Chemie International
Edition 48 (2009) 2474e2499.
[6] (a) T. Horiuchi, H. Miura, K. Sumioka, S. Uchida, Journal of the American
Chemical Society 126 (2004) 12218e12219;
(b) S. Ito, S.M. Zakeeruddin, R. Humphry-Baker, P. Liska, R. Charvet, P. Comte,
M.K. Nazeeruddin, P. Pechy, M. Takata, H. Miura, S. Uchida, M. Grätzel,
Advanced Materials 18 (2006) 1202e1205.
[26] W. Schlenk, E. Bergmann, Liebigs Annalen der Chemie 464 (1928) 22e34.
[27] W.C. Still, M. Kahn, A. Mitra, Journal of Organic Chemistry 43 (1978)
2923e2925.
[7] (a) P. Wang, C. Klein, J.E. Moser, R. Humphry-Baker, N.L. Cevey-Ha, R. Charvet,
P. Comte, S.M. Zakeeruddin, M. Grätzel, Journal of Physical Chemistry B 108
(2004) 17553e17559;