Metal-free benzodithiophene-containing organic dyes for dye-sensitized solar cells

Article abstract of DOI:10.1002/ejoc.201200958

Two new metal-free organic dyes, CR29 and CR52, with high extinction coefficients in the visible spectral region between 400-650 nm, have been synthesized. The donor-acceptor structure of the dyes feature benzodithiophene moieties BDT₁ and BDT

Full text of DOI:10.1002/ejoc.201200958

FULL PAPER
DOI: 10.1002/ejoc.201200958
Metal-Free Benzodithiophene-Containing Organic Dyes for Dye-Sensitized
Solar Cells
Elena Longhi,[a] Alberto Bossi,[b] Gabriele Di Carlo,[a] Stefano Maiorana,[a]
Filippo De Angelis,*[c] Paolo Salvatori,[c] Annamaria Petrozza,[d] Maddalena Binda,[d]
Vittoria Roiati,[d,e] Patrizia Romana Mussini,[a] Clara Baldoli,*[b] and Emanuela Licandro*[a]
Keywords: Donor-acceptor systems / Dyes/pigments / Sensitizers / Solid state structures / Solar cells
Two new metal-free organic dyes, CR29 and CR52, with high
extinction coefficients in the visible spectral region between
400–650 nm, have been synthesized. The donor–acceptor
structure of the dyes feature benzodithiophene moieties
BDT1 and BDT as rigid π-conjugated spacers, which have so
far been very little studied for dye-sensitized solar cell
(DSSC) applications. DFT/TDDFT calculations have been
employed to guide the design of the chromophores as well
as to shed light on their electronic and optical properties.
Photophysical and electrochemical characterization studies
have been carried out to gather information on the charge
transfer processes occurring at the dye–semiconductor inter-
faces. Under standard AM 1.5 conditions, DSSC sensitized
with CR29 showed good conversion efficiencies: 5.14% in
the liquid electrolyte cell setup and 2.47% in the solid-state
DSSC.
pyridyl redox electrolyte.[3] Although transition metal dyes
exhibit high performances, challenging synthesis and com-
plex purification steps are often required. For these reasons,
in recent years, alternative metal-free organic dyes have at-
tracted great attention. In fact, such chromophores can be
prepared rather inexpensively by using well-established syn-
thetic methodologies and they present optical, electronic,
and electrochemical properties that can be modulated
through appropriate molecular design.[4]
The development of innovative, stable organic dyes with
optical absorptions extending into red and near-IR regions
of the solar spectrum is, at present, a “hot topic” of re-
search in this field. The most efficient metal-free organic
dyes are characterized by a donor-π-acceptor (D-π-A)
structure, in which the donor group (D) is an electron-rich
unit, linked through a conjugated π-bridge spacer to the
electron-acceptor group (A). In the cell, unit A is bound to
the semiconductor surface (TiO2), usually through a carb-
oxylic or cyanoacrylic acid function. The nature of the
conjugated bridging segment (π) has proved to be of signifi-
cant importance for controlling the light-harvesting per-
formance of the dye and a number of new π-conjugated
aromatic and heteroaromatic systems[4] have been investi-
gated. Among these, the use of several kinds of substituted
thiophene or thienothiophene derivatives as π-bridges, have
been reported to give remarkable efficiencies.
Introduction
Dye-sensitized solar cells (DSSC or Grätzel cells) are
photovoltaic devices that are the object of competitive and
extensive research aimed at improving efficiencies and re-
ducing costs.[1] One of the key components in a DSSC is
the dye, which is responsible for harvesting the sunlight and
injecting electrons into the mesoporous semiconducting ox-
ide, typically TiO2.
Metal-based photosensitizers have been used over the
last 20 years as the key components for DSSCs.[1e,2] So far,
record efficiencies of 12.3% under AM 1.5 conditions have
been reached by using ZnII porphyrin dyes and cobalt poly-
[a] Dipartimento di Chimica, Università degli Studi di Milano,
via C. Golgi 19, 20133 Milano, Italy
Fax: +39-02-50314139
E-mail: emanuela.licandro@unimi.it
Homepage: http://www.unimi.it/
[b] CNR – Istituto di Scienze e Tecnologie Molecolari,
via C. Golgi 19, 20133 Milano, Italy
Fax: +39-02-50314139
E-mail: clara.baldoli@istm.cnr.it
Homepage: http://www.istm.cnr.it/
[c] CNR – Istituto di Scienze e Tecnologie Molecolari,
via Elce di Sotto 8, 06123 Perugia, Italy
Fax: +39-075-5855606
E-mail: filippo@thch.unipg.it
Hoempage: http://www.istm.cnr.it/
[d] Center for Nano Science and Technology @ Polimi, Istituto
Italiano di Tecnologia,
via G. Pascoli 70/3, 20133 Milano, Italy
[e] Dipartimento di Fisica, Politecnico di Milano,
P.za Leonardo da Vinci 32, 20133 Milano, Italy
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201200958.
Benzo[1,2-b:4,3-bЈ]dithiophene (BDT) and benzo[1,2-
b:4,5-bЈ]dithiophene (BDT1; Figure 1) attracted our atten-
tion because of their stable, rigid, π-conjugated condensed-
polycyclic structures.[5] These features lead to unique elec-
84
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94

Benzodithiophene-Containing Organic Dyes
tronic properties such as high conductivity, field-effect mo-
bility, and tunable stacking in the solid state; moreover, ri-
gid structures hamper the roto-vibrational modes responsi-
ble for the deactivation of the excited states in functional
materials.[5c] In this context, BDT and BDT1 therefore look
to be important frameworks in functional organic materials
for optoelectronic devices.[6] Although BDT1 has already
been incorporated in conducting polymers employed in
BHJ solar cells,[7] to the best of our knowledge, only one
recent report utilizes a 3-bromo-4,8-diisopropoxy-BDT1 as
the π-conjugated spacer incorporated into a D-π-A DSSC
dye.[8]
Figure 2. Structures of CR29 and CR52.
Results and Discussion
1.1 Synthesis of the Chromophores
Figure 1. Structure and numbering scheme of BDT and BDT1.
The synthetic strategy envisioned for the preparation of
dyes CR29 and CR52 used two isomeric BDT1 and BDT
Due to the above considerations, with the aim of estab- dialdehyde derivatives 2 and 3 as starting materials
lishing a structure/performance correlation of dyes endowed (Scheme 1). In fact, by adjusting the reaction stoichiometry,
with BDT and BDT1, a systematic study of donor–acceptor it was possible to selectively react, step by step, the two
systems obtained by coupling BDT1 and BDT with tradi- formyl groups with two different reagents, thus obtaining
tional and new A and D groups is ongoing. As part of a the target compounds.
more general research program undertaken in our laborato-
In detail, known phosphonium salt 1[10] gave aldehyde 4
ries,[9] in this work we wanted to investigate the ability of through Wittig reaction with one of the two carbonyl
BDT1 and BDT, acting as π-bridge spacers, to enhance the groups of 2,6-benzo[1,2-b:4,5-bЈ]dithiophenedicarbaldehyde
overall performance of dyes. We report here the synthesis, (2).[11] Subsequent Knoevenagel condensation of 4 with cy-
the optical and electrochemical characterization, and the anoacetic acid, in acetonitrile solution and in the presence
photovoltaic properties of two novel BDT1- and BDT-based of piperidine, led to the target dye CR29 as a red solid.
chromophores, respectively named CR29 and CR52 (Fig- Similarly, CR52 was obtained as a red solid, starting from
ure 2), along with their DFT/TDDFT computational char- aldehyde 3 and following the same procedure used for
acterization. In both CR29 and CR52 the donor and the CR29. The molecular structure of both dyes was confirmed
1
acceptor fragments are constituted by the classical triaryl- by H and 13C NMR, and MS characterization. Reverse-
amino and cyanoacrylic groups. The triarylamine moiety phase HPLC analysis was employed to establish the purity
is linked to the BDT1 and BDT spacers through a double of those samples submitted to DSSC fabrication; being, in
bond.
both cases, higher than 98%.
Scheme 1. Synthetic procedures for the synthesis of CR29 and CR52. Reagents and conditions: (i) DMF, K2CO3, 18-crown-6, room temp.
15 h. (ii) CNCH2CO2H, piperidine, CH3CN, 80 °C, 4 h.
Eur. J. Org. Chem. 2013, 84–94
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
85

F. De Angelis, C. Baldoli, E. Licandro et al.
FULL PAPER
1.2 Theoretical Calculations
of TDDFT excited state calculations performed with the
MPW1K functional on the two dyes are reported in
Table 2.
The molecular and electronic structure and the excited
state of CR29 and CR52 were investigated by performing
DFT/TDDFT calculations. The two dyes are characterized
by a planar arrangement between the donor N lone pair
and the acceptor cyanoacrylic group, which ensures a
strong conjugation across the donor-π-bridge assembly. A
schematic representation of the HOMO and LUMO of the
two dyes obtained by B3LYP/6-31G* calculations[12,13] is
displayed in Figure 3 along with their isodensity plots.
Table 1 reports the calculated frontier orbital energies in
acetonitrile environment.
Table 2. Computed excitation energies (eV, nm) and oscillator
strength (f), using the MPW1K xc functional, for the optical transi-
tion of CR29 and CR52 dyes, in acetonitrile solution.
CR29
State E [eV] λ [nm]
ƒ
Composition [%]
1
2
2.56
3.50
484
354
1.73
0.81
HǞL
79
40
23
19
58
H-2ǞL
HǞL+1
H-2ǞL
HǞL+1
3
3.62
342
0.15
CR52
State E [eV] λ [nm]
ƒ
Composition [%]
1
2
2.61
3.38
476
367
1.12
1.01
HǞL
H-1ǞL
HǞL
HǞL+1
H-1ǞL
HǞL+1
H-4ǞL
H-2ǞL
82
40
14
27
36
46
13
63
3
4
3.44
3.60
360
344
0.26
0.20
For both dyes we calculate a single, low-lying transition
with large oscillator strength (f), followed, at higher energy,
by a series of additional transitions. For CR29 and CR52
we calculate the low-energy excitations at 484 and 476 nm,
Figure 3. Energy diagram and isodensity plots of the HOMO and
LUMO of CR29 (left) and CR52 (right).
Table 1. Calculated HOMO–LUMO values for CR29 and CR52.[a]
HOMO
LUMO
CR29
CR52
–4.73
–4.70
–2.83
–2.74
[a] B3LYP/6-31G* in CH3CN solution.
Our results show the characteristic HOMO–LUMO
pattern of these D-π-A dyes, with the HOMOs essentially
confined to the donor moiety and partially spreading over
the conjugated π-bridge, whereas the LUMOs extend from
the linker to the cyanocrylic acid, with the largest compo-
nents being localized on the latter. The HOMO localization
in the outer molecular region ensures exposure of the oxid-
ized dye, generated after photoinduced electron injection
into the TiO2, to the dye regenerating medium, therefore
favoring rapid hole transfer from the oxidized chromo-
phore. The LUMO distribution close to the anchoring
group enhances the orbital overlap with the titanium mani-
fold of unoccupied states, assisting the electron injection
step. A comparison of the two dyes shows for CR52 a slight
destabilization of the LUMO with respect to CR29, due to
the reduced conjugation in the former, whereas the HOMO
is located at essentially the same energy in the two dyes.
Considering the HOMO–LUMO localization, this elec-
tronic structure pattern is not surprising, since the two dyes
share the same donor moiety and differ in the acceptor;
thus, a consequent slight increase of the HOMO–LUMO
Figure 4. Optical absorption spectra of CR29 (top) and CR52 (bot-
tom) in different solvents. All spectra are acquired in the presence
gap is observed on going from CR29 to CR52. The results of TFA.
86
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94

Benzodithiophene-Containing Organic Dyes
respectively, which is consistent with the experimental ab- Table 3. Experimental optical absorption coefficients of CR29 and
CR52 in a range of solvents in the presence of TFA.
sorption spectrum. These transitions, of HOMO to LUMO
character, are the typical charge transfer (CT) excitations of
this type of donor–acceptor dyes, leading to charge dis-
placement from the donor to the acceptor dye moiety. The
most notable difference between CR29 and CR52 is the AcCN
considerable decrease of the oscillator strength (1.73 vs.
1.12) for the charge transfer transition that is observed for
CR52, along with a slight blue-shifted CT transition (484
vs. 476 nm). This behavior, which is also seen in the experi-
mental absorption spectra (Figure 4), can be related to a
ЈbreakoutЈ of the conjugation, resulting from the different
spacer group, across the CR52 donor–acceptor moiety. For
both dyes, we assign the transitions in the UV region as
originating from π-π* excitations. It is worth noting that
TDDFT calculations nicely reproduce the redshift of the π-
π* transitions measured experimentally for CR52 compared
with CR29, which are found at ca. 400 nm in the former
and at ca. 370 nm in the latter, see Figure 4.
Solvent
CR29
CR52
λ [nm], [ε (105 cm–1 m–1)]
λ [nm], [ε (105 cm–1m–1)]
THF
477 [3.71], 373 [2.64]
480 [4.05], 379 [2.86]
480 [3.34], 379 [2.42]
461 [2.51], 392 [4.16]
464 [2.57], 391 [4.21]
465 [2.27], 394 [3.83]
484 [2.06], 397 [3.81]
489 [2.68], 399 [5.52]
EtOH
Toluene 496 [3.15], 379 [2.17]
CH2Cl2 502 [3.48], 380 [2.71]
higher transition probability of the CT band in CR29 is a
consequence of the better conjugation ability of the linear
BDT1 spacer than BDT. Indeed, as previously discussed, the
CT transition is of HOMO to LUMO character (Figure 3).
Whereas the LUMO surfaces delocalize similarly in both
dyes, the HOMO surfaces extend over the whole BDT1 moi-
ety in CR29 and to a noticeable lesser extent over BDT in
CR52. Therefore, the larger overlap between the two fron-
tier orbitals in CR29 translates into the higher transition
probability. Overall, CR29 displays a better overlap with the
visible spectrum, showing, at the same time, higher molar
absorptivities.
1.3. Optical and Electrochemical Characterization
Both CR29 and CR52 are highly luminescent only in tol-
uene solution (Figure 5). The broad featureless emission
profiles, which mirror the lowest energy absorption bands,
The molar absorptions of the two dyes in different sol-
vents are reported in Figure 4 (for full absorption profiles,
see Figure S7 in the Supporting Information). Compound
CR29, possibly due to its linear and planar structure, suffers
from aggregation and limited solubility except in tetra-
hydrofuran (THF) and EtOH solutions. We found the bene-
ficial effect of trifluoroacetic acid (TFA) in dissolving the
chromophore,[14] therefore the molar absorptions have been
recorded upon addition of TFA to all the CR29 solutions
and, for better comparisons, also to solutions of CR52 (in
the solutions, the TFA concentration was ca. 0.5–1ϫ10–3
m
with respect to the 1.2–2.0ϫ10–5 m dye concentration).
Compounds CR29 and CR52 display absorption onsets
around 650 nm, and two main transitions in the visible re-
gion, as predicted by quantum chemical calculation. In
both cases, the high energy transition, peaking around 370–
400 nm, shows a poor solvatochromic effect and has been
tentatively attributed to a π-π* transition of the conjugated
D-π-bridge system. This band is significantly shifted to
lower energy in CR52 (ca. 400 nm) compared with CR29
(ca. 375 nm). On the other hand, the low energy transition
(around 500 nm) is more sensitive to the solvent polarity
and has been assigned to a charge transfer (CT) transition,
as already described in the theoretical section. The CT band
shifts towards the red, in both dyes, following the order
THF, AcCN ≈ EtOH, toluene, CH2Cl2 (see Table 3; for
THF and EtOH spectra see Figure S7 in the Supporting
Information).
Corroborating the TDDFT calculations, the CR29 CT
band displays, on average, higher extinction coefficients
than
CR52
(respectively
being
ca.
3.5
and
2.5 ϫ 104 m–1 cm–1). Another interesting aspect that sub-
stantially differentiates the two chromophores is the relative
intensity of the π-π* and CT bands. The ratio of the two
Figure 5. Normalized absorption (solid lines) and emission (dashed
lines) spectra of dyes CR29 (top) and CR52 (bottom) in toluene
solution recorded in the presence of TFA (black lines) and Et3N
bands is roughly 2:1 in CR52 and 3:4 in the CR29 dye. The (grey lines).
Eur. J. Org. Chem. 2013, 84–94
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
87

F. De Angelis, C. Baldoli, E. Licandro et al.
FULL PAPER
and the large Stokes shift (ca. 130–140 nm) are indicative of the conjugate acid of the triarylamine group, which is
signatures of the CT nature of the emissive state.
–5, whereas that of TFA is 0.5.
The excitation spectra match, in shape and intensity, the
In all of the solvents (see Figure S5 and S6 in the Sup-
absorption spectra, hence the luminescence stems from ra- porting Information), we found that the CT absorption
diative emission of the molecular species. In all other sol- band in the carboxylate form blue shifts, on average, 35–
vents the emission is completely quenched, possibly sug- 50 nm compared to the protonated form (see Figure 5 for
gesting a strong interaction with the solvents that triggers toluene solution). This destabilization of the CT character
nonradiative relaxation pathways.[14]
is also evident in the emission spectra; in fact, CR29 emits
The optical properties of CR29 and CR52 were also in- at 624 nm in the protonated form and at 577 nm in the de-
vestigated upon addition of Et3N to the solutions (Fig- protonated form; CR52 emits respectively at 620 and
ure 5). A blueshift in absorption maximum was observed in 580 nm.
this case when compared to the absorption spectra in the
The electron transfer properties of chromophores CR29
presence of TFA. This could be related to the deproton- and CR52 have been studied by cyclic voltammetry together
ation/protonation equilibrium of the carboxylic acid group. with those of the corresponding parent molecules BDT1
In fact, the addition of base shifts the dye from its free and BDT. Selected results are reported in Table 4, see also
carboxylic acid form (COOH) to the carboxylate form Figures 6 and 7.
(COO–), drastically reducing the strength of the acceptor
The angular BDT and the linear BDT1 fragments feature
moiety and, at the same time, the donor–acceptor interac- similar CV patterns, with chemically irreversible first oxi-
tion within the dye.[14] The protonation of the amine unit dation and first reduction peaks (Figure 6), indicating that
by the TFA can be excluded on the basis of the pKa value the corresponding radical cation and radical anion undergo
Table 4. Selected CV features for chromophores CR52 and CR29 and parent molecules BDT and BDT1, together with linear α,α-bithio-
phene (T2) as a benchmark, and corresponding HOMO and LUMO energy levels and gaps.
EIc,onset
EIc,p
EIa,onset EIa,p
EIIa,p
ELUMO
EHOMO
Eg
ELUMO
EHOMO
Eg
V(Fc+|Fc) V(Fc+|Fc)
V(Fc+|Fc) V(Fc+|Fc) V(Fc+|Fc) (onset) [eV] (onset) [eV] (onset) [eV] (max) [eV] (max) [eV] (max) [eV]
[a]
T2
BDT[b]
–2.84
–3.00
0.88
1.10
≈3.6
3.79
–1.96
–1.80
–5.68
–5.90
3.72
4.10
–2.82
0.97
–1.98
–5.77
[b]
BDT1
–2.75
–2.04
–1.86
–2.93
–2.13
–1.96
0.87
0.15
0.14
1.01
0.28
0.26
–2.05
–2.76
–2.94
–5.67
–4.95
–4.94
3.62
2.19
2.00
–1.87
–2.67
–2.84
–5.81
–5.08
–5.06
3.94
2.41
2.22
CR52[c]
CR29[c]
0.84
0.85
[a] From Refs.[15a] (anodic peak, in CH2Cl2 + 0.1 m TBAPF6) and Refs.[20] (cathodic peak, in dimethylacetamide + 0.1 m TBABr), here
both referred to Fc+|Fc. [b] In CH3CN + 0.1 m TBAP, at 0.2 Vs–1, referred to Fc+|Fc (this work). [c] In CH2Cl2 0.1 m TBAP, at 0.2 Vs–1,
referred to Fc+|Fc (this work).
Figure 6. CV features for BDT and BDT1 in CH3CN + 0.1 m TBAP, at 0.2 Vs–1 scan rate, with ohmic drop compensation.
88
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94

Benzodithiophene-Containing Organic Dyes
Figure 7. CV features for chromophores CR52 and CR29 in CH2Cl2 + 0.1 m TBAP, on GC, at 0.2 Vs–1 scan rate, with ohmic drop
compensation.
chemical reactions. BDT1 has its first oxidation peak at localized on the electron-rich triarylamino site, as evidenced
1.01 V and first reduction at –2.93 V. In contrast, angular by DFT calculations, and are consistent with literature
BDT oxidizes at 1.10 V and reduces at –3.00 V. In this re- data.[19] The localized nature of the electron transfer, to-
spect the π conjugation is more effective in the linear isomer gether with the distance from the BDT1 or BDT building
than in the angular isomer. The conjugation efficiency of block explains the nearly equal oxidation potential in the
BDT1 appears comparable to, or slightly lower than, the two chromophores.
linear 2,2Ј-bithiophene (Table 4).[15]
The first reduction accounts for a monoelectronic, elec-
From the onset and peak potentials for the first re- trochemically reversible peak at –1.96 V for CR29 and
duction and first oxidation, LUMO and HOMO energies –2.13 V for CR52, which are considerably less negative po-
can be estimated from the equations reported in the Sup- tentials than in the parent BDT molecules. The observed
porting Information (Equations S1a, S1b, S2a, S2b),[16,17] reductions correspond to localized processes involving the
and are ultimately based on the absolute value for the nor- electron-poor terminal double bond carrying a cyanoacrylic
mal hydrogen electrode (NHE), which was critically as- group. The immediately adjacent BDT1/BDT moiety affects
sessed in a fundamental review paper.[18] The resulting val- these processes such that the reduction peak of CR29 is
ues, together with the corresponding energy gaps Eg, are significantly less negative than that of CR52. Moreover, the
reported in Table 4 together with their counterparts ob- reduction peak appears to be electrochemically reversible
tained by absorption spectroscopy.
The electrochemical Eg values show good consistency but is chemically reversible only in the linear form.
with the spectroscopic data, adopting the onset criterion. Remarkably, the lower Eg value of BDT1 with respect to
(i.e., accounting for a fast electron transfer) in both cases,
It is worth noting that, although the electrochemical and that of BDT is reflected in the corresponding dyes, such
spectroscopic Eg values are often correlated, they refer to that CR29 has a smaller Eg value than CR52. The 0.19 eV
different processes (respectively electron transfer from/to difference between the two dyes mostly arise from the differ-
the molecule vs. electron promotion between different ence in the reduction peak potentials (i.e., in LUMO en-
states), which can be affected to different extents by solvent ergy). This feature is in perfect agreement with the trend
interactions and other parameters.
(not the absolute values) of the theoretical computations.
Chromophores CR29 and CR52, which only differ in This is less evident in the spectroscopic approach, possibly
having an angular or a linear BDT linker, have similar CV as a consequence of the width and complexity of the first
patterns. In particular, as shown in Figure 7, two nearly absorption band.
merging two-peak systems, partially chemically reversible,
It is worth noting that the calculated HOMO–LUMO
can be observed for oxidations, followed by a complex mul- energies (Table 1) and trends are fully consistent with elec-
tielectron irreversible shoulder on the background. First trochemical measurements (Table 4), showing a HOMO
oxidations (0.26 V for CR29 and 0.28 V for CR52) must be (LUMO) onset at –4.94/–4.95 (–2.94/–2.76) eV (Table 4) for
Eur. J. Org. Chem. 2013, 84–94
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
89

F. De Angelis, C. Baldoli, E. Licandro et al.
FULL PAPER
CR29 and CR52, respectively, which are comparable to the
calculated values at –4.73/–4.70 (–2.83/–2.74) eV.
With these photovoltaic parameters, efficiencies of 5.14
and 3.63% are calculated for the CR29 and CR52 dyes. In
both cases, these results were obtained by using chenod-
eoxycholic acid (CDCA) in the dye solutions as an anti-
aggregant agent.[21] Indeed, it has been reported that or-
ganic dyes often suffer from aggregation issues, leading to
poor charge injection and consequent reduction in the effi-
ciency of the device.[22] We also observed these problems
with the CR29 dye. In fact, without using CDCA, a maxi-
mum of 1.1% efficiency was obtained, with poor Jsc and
Voc values of respectively 3.5 mA/cm2 and 0.56 V. Another
important additive that can be used to optimize cell per-
formance is the LiI in the electrolyte solutions. It has been
noted that Li ions are able to downshift the TiO2 conduc-
tion band, facilitating the charge injection process and then
improving the generated photocurrent.[23] For the CR29
dye, without using LiI in the electrolyte solution, we ob-
tained an efficiency of 3.35%, which is considerably lower
with respect to the optimized cell (see the Supporting Infor-
mation). We measured a Jsc value of 7.04 mA, leading to a
decrease in the efficiency, whereas the Voc (0.71 V) and FF
(0.67) values were not influenced by the additive. The cell
experimental results are encouraging when compared to so-
lar cells made with the prototype N719 ruthenium dye un-
der similar conditions (i.e., employing the sensitization pro-
cedure optimized for N719), the output efficiency yield of
which was 6.36% (see the Supporting Information). It is
also worth noting here the importance of careful selection
of the spacer unit in a donor–acceptor dye architecture. In
fact, as already anticipated by the theoretical and spectro-
scopic analysis of the chromophore, the use of the BDT1
conjugated spacer allows stabilization of the CT transition
together with an improvement of its absorption cross-sec-
tion, which is translated, in terms of device performance, in
an improvement of thee solar cell photocurrent.
1.4. Solar Cell Performance and Charge Generation
Processes
Liquid electrolyte based DSSC were fabricated and
tested under AM 1.5 conditions (Figure 8a). CR29 gave the
best results in terms of photocurrent (10.65 mA/cm2) and
open-circuit voltage (0.71 V) with respect to CR52
(7.97 mA/cm2 and 0.67 V), whereas the same FF values are
found (0.68, see Table 5).
The most efficient molecule, CR29, has also been tested
in a solid state device in which a molecular hole transporter,
2,20,7,70-tetrakis(bis-p-methoxyphenylamino)-9,90-spirobi-
fluorene (spiro-OMe TAD) was used instead of the liquid
electrolyte. The solar cell, although not fully optimized in
terms of device architecture (i.e., lithium salts and tert-bu-
tylpyridine concentration, device thickness), clearly shows
the full potential of this metal-free dye. Figure 8b shows the
device characteristic compared to the output of a device
that uses the conventional N719 ruthenium dye, and in
Table 5 the figures of merit of the solar cells are reported.
Furthermore, in this case, it is clear that the strong enhance-
ment in photocurrent drives the improvement in the devices
performances, which is a result of an enhanced ability of
the fully organic dye to harvest sunlight, together with good
charge generation processes.
Some preliminary studies have been carried out by using
time-resolved photoluminescence spectroscopy to investi-
gate the charge generation processes (i.e., electron transfer
from the dye to the TiO2 and hole transfer to the spiro
OMeTAD).[24] All samples present the same additives used
in the actual device (CDCA, Li-TFSI and 4-tert-butylpyr-
idine) because it is known that they can strongly influence
Figure 8. Current density vs. voltage under AM1.5 simulated sun-
light (100 mW/cm2) illumination for: (a) CR29 (black line) and
CR52 (dashed line) in liquid DSSC. A commercial electrolyte con-
taining 1-butyl-3-methylimidazolium iodide, iodine, guanidinium
thiocyanate, and tert-butylpyridine in a mixture of valeronitrile and
acetonitrile, with 0.06 m LiI added, was used. (b) CR29 (black line)
and N719 (dashed line) in solid state DSSC.
Table 5. Photovoltaic performance of CR29 and CR52.
Dye
Jsc [mA/cm2]
Voc [V]
FF
η [%]
CR29[a]
CR29[b]
CR52[a]
N719[b]
10.65
5.65
7.97
2.10
0.71
0.78
0.67
0.77
0.68
0.54
0.68
0.56
5.14
2.47
3.63
0.95
[a] Liquid electrolyte DSSCs. [b] Solid State DSSCs.
90
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94

Benzodithiophene-Containing Organic Dyes
the metal oxide and dye energetics and the kinetics of the formed with the dye in contact with a high band gap metal
main device working mechanisms.[25] In Figure 9a the pho- oxide, therefore, the charge-transfer process starts from a
toluminescence (PL) emission from the dye chemisorbed on bound singlet exciton photogenerated on the dye, whereas,
a TiO2 and ZrO2 mesoporous substrates is reported. The in a working device, the dye can be already oxidized thanks
latter is a high band gap oxide that does not allow for elec- to efficient electron injection to the TiO2 layer.
tron transfer from the dye to the metal oxide. The decays
have been fitted by a stretched exponential function to ex-
Conclusions
tract the average electron injection lifetime on the TiO2 (τinj
≈ 15 ps) and the average exciton lifetime on the ZrO2 (τobs
≈ 71 ps) and finally estimate the electron injection efficiency
(ηinj) according to the following expression.[24]
We have reported on the design and synthesis of two
novel dyes, CR29 and CR52, as photosensitizers for DSSCs.
The dyes are endowed with two isomeric benzodithiophene
units (BDT1 and BDT) as π-spacers. Their synthesis has
been achieved in good yields by using standard synthetic
procedures and the dyes have been analytically charac-
terized. On the basis of DFT calculations, UV/Vis absorp-
tion and emission studies, and cyclovoltammetric charac-
terization, it has been possible to show that the BDT1 π-
bridge appears to be more suitable than BDT in donor–
acceptor molecules on the basis of its better conjugation,
to effectively couple the triarylamine donor unit and the
cyanoacetic acceptor moiety. DSSC fabricated from the
BDT1-containing chromophore (CR29) exhibits promising
performances, especially in the solid state device in which
the use of a high extinction coefficient dye is a demanding
requirement. SS-DSSCs fabricated and sensitized with
CR29 exceed the performance of a solar cell using a stan-
dard ruthenium complex N719.
1/τinj
ηinj
=
1/τinj + 1/τobs
Further developments and exploitation of this work will
consider the use of less conventional donor and acceptor
groups, to establish, in a homogeneous series of organic
dyes based on BDT1, the structure–performance relation-
ships.
Experimental Section
Reagents and Methods: All reagents and solvents were obtained
from highest grade commercial sources and used without further
purification unless otherwise stated. Column chromatography was
carried out using silica gel 60 (70–230 mesh, Merck). 1H NMR
spectra were acquired with a Bruker AVANCE DRX-400, a Bruker
AC300, or an AMX 300 MHz spectrometer; chemical shifts (δ) are
reported in parts per million relative to the solvent residual peak
([D6]acetone, [D6]DMSO, CDCl3). IR spectra were recorded with
a Fourier Bruker Vector 22 FT; UV spectra were recorded with a
Jasco V-520 or Agilent 8453 UV/Vis spectrophotometer in a range
of λ from 190 to 800 nm at room temperature. Steady-state emis-
sion and excitation spectra were measured using a Horiba scientific
FluoroLog 3. HRMS spectra were recorded with a Bruker Dalton-
ics ICR-FTMS APEX II. Melting points were obtained with a
Büchi B-540 melting point apparatus and are uncorrected. HPLC
analyses were performed with an Agilent 1100 series equipped with
a PDA detector and a reverse-phase ZORBAX Eclipse XBD-C18
column (4.6ϫ 150 mm, 5 μm particle size); samples were analyzed
with 1 mL/min acetonitrile/water (80:20) with 0.1% TFA. CR29
and CR52 samples for cell fabrication were assessed by HPLC
analysis and the purity was estimated to be over 98%.
Figure 9. Time-resolved photoluminescence (PL) spectra from (a)
CR29 dye adsorbed on mesoporous films of ZrO2 and TiO2; (b)
CR29 adsorbed on mesoporous films of ZrO2 with and without
spiro-oMeTAD infiltrated. All samples present the same additives
used in the fabrication of the solid state DSSC. Decays are taken
at 700 nm, the peak of the PL spectra. Excitation at 450 nm.
We calculate an electron injection efficiency of 76%. In
Figure 9b the PL emission from the dye chemisorbed on a
ZrO2 substrate is shown with and without spiro
OMeTAD to estimate the hole transfer efficiency as de-
scribed above. The noninjecting substrate is used to disen-
tangle the hole-transfer and electron-transfer mechanisms
in the PL quenching experiment. A hole-transfer efficiency
of the order of 53% has been found. The hole-transfer yield
({4-[Bis(4-methoxyphenyl)amino]phenyl}methyl)triphenylphosphon-
ium Bromide (1): Obtained with a slightly modified literature pro-
is likely underestimated because measurements were per- cedure.[26] A solution of 4-[bis(4-methoxyphenyl)amino]benzyl
Eur. J. Org. Chem. 2013, 84–94
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
91

F. De Angelis, C. Baldoli, E. Licandro et al.
FULL PAPER
alcohol (374.5 mg, 1.12 mmol) and PPh3·HBr[27] (421.5 mg,
1.23 mmol) in CHCl3 (6 mL) was heated to reflux whilst stirring
for 4 h. The solvent was then evaporated and the residue was
treated with Et2O (10 mL). The phosphonium salt, precipitated
from the solution as a white solid, was filtered to give 1 (700 mg,
95%). 1H NMR (300 MHz, [D6]acetone): δ = 3.76 (s, 6 H), 5.30
(d, J = 14.2 Hz, 2 H), 6.52 (d, J = 8.4 Hz, 2 H), 6.87 (d, J = 8.9 Hz,
(3;[29] 0.36 mmol) in DMF (6 mL). A crude dark-orange product
was obtained that was purified by silica gel column chromatog-
raphy (CH2Cl2/hexane, 9:1) to afford 5 (76 mg, 46%) as a red solid;
m.p. 160–162 °C. 1H NMR (300 MHz,CDCl3): δ = 3.81 (s, 6 H),
6.83–6.91 (m, 6 H), 7.01 (d, J = 15.9 Hz, 1 H), 7.03–7.09 (m, 4 H),
7.13 (d, J = 15.9 Hz, 1 H), 7.32 (d, J = 8.7 Hz, 2 H), 7.5 (s, 1 H),
7.70 (d, J = 8.7 Hz, 1 H), 7.80 (d, J = 8.7 Hz, 1 H), 8.26 (s, 1 H),
4 H), 6.88–6.95 (m, 2 H), 6.99 (d, J = 8.9 Hz, 4 H), 7.70–7.76 (m, 10.12 (s, 1 H) ppm. 13C NMR (75 MHz, CDCl3): δ = 55.51, 114.83,
6 H), 7.81–7.91 (m, 9 H) ppm. 31P NMR (121 MHz, CDCl3): δ = 117.15, 118.81, 119.40, 119.9, 122.54, 126.97, 127.37, 127.52,
23.09 ppm.
128.17, 131.64, 133.28, 135.74, 136.97, 140.57, 143.03, 146.10,
148.95, 156.34, 184.09 ppm. IR (nujol): ν = 1646 (νCO) cm–1
.
˜
6-(2-{4-[Bis(4-methoxyphenyl)amino]phenyl}ethenyl)benzo[1,2-b:4,5-
bЈ]dithiophen-2-carbaldehyde (4): A solution of 1 (100 mg,
0.15 mmol) in DMF (4 mL) was slowly added, whilst stirring at
room temperature, to a slurry of 2,6-benzo[1,2-b:4,5-bЈ]dithio-
phenedicarbaldehyde (2;[28] 50 mg, 0.20 mmol), 18-crown-6 ether
(3 mg), and anhydrous K2CO3 (32 mg, 0.30 mmol) in DMF (4 mL).
The resulting orange solution was stirred overnight at room tem-
perature. The solution was then poured into water (10 mL) and
extracted with dichloromethane (4ϫ 10 mL). The organic phase
was washed with water (20 mL), dried with Na2SO4, and filtered.
After evaporation of the solvent, the crude dark-orange product
was purified by silica gel column chromatography (CH2Cl2/hexane,
9:1) to afford 4 (51 mg, 63%) as a red solid; m.p. 212–214 °C (dec).
1H NMR (300 MHz, [D6]DMSO): δ = 3.73 (s, 6 H), 6.72 (d, J =
8.6 Hz, 2 H), 6.92 (d, J = 8.9 Hz, 4 H), 6.97 (d, J = 16.4 Hz, 1 H),
7.05 (d, J = 8.9 Hz, 4 H), 7.39 (d, J = 16.4 Hz, 1 H), 7.44 (d, J =
8.6 Hz, 2 H), 7.45 (s, 1 H), 8.41 (s, 2 H), 8.6 (s, 1 H), 10.11 (s, 1
H) ppm. 13C NMR (75 MHz, CDCl3): δ = 55.18, 114.96, 117.15,
118.16, 118.38, 119.06, 119.99, 121.68, 124.35, 126.95, 127.37,
127.58, 127.73, 127.86, 129.71, 132.53, 135.53, 139.60, 143.12,
HRMS (EI): calcd. for C33H25NO3S2 547.12758; found 547.127588.
U V / Vi s ( C H 3 C N, 6 . 3 ϫ 1 0 – 5 m ) : λ m a x ( ε, m – 1 c m – 1 ) =
299 (1.84ϫ104), 354 (2.72ϫ104), 439 (3.42ϫ104) nm.
2-Cyano-3-[7-(2-{4-[bis(4-methoxyphenyl)amino]phenyl}ethenyl)-
benzo[1,2-b:4,3-bЈ]dithiophen-2-yl]acrylic Acid (CR52): Synthesized
following the same procedure used for CR29, starting from 5
(70 mg, 0.13 mmol) to afford CR52 (46 mg, 55%) as a red solid;
m.p. 198–200 °C. 1H NMR (300 MHz, [D7]DMF): δ = 3.80 (s, 6
H), 6.74 (d, J = 8.5 Hz, 2 H), 6.95 (d, J = 9 Hz, 4 H), 6.97 (d, J =
16.4 Hz, 1 H), 7.08 (d, J = 9 Hz, 4 H), 7.42 (d, J = 16.4 Hz, 1 H),
7.47 (d, J = 8.5 Hz, 2 H), 7.45 (s, 1 H), 8.35 (s, 2 H), 8.47 (s, 1 H),
8.58 (s, 1 H), 8.64 (s,1 H) ppm. 13C NMR (75 MHz, [D7]DMF): δ
= 55.78, 115.67, 116.84, 119.57, 119.73, 121.23, 123.60, 127.97,
128.56, 128.76, 132.17, 133.59, 135.92, 136.50, 140.76, 146.84,
147.93, 149.85, 157.32, 164.13 ppm. IR (nujol): ν = 3393 (νOH),
˜
2342 (νCN), 1560 (νCO) cm–1. HRMS (ESI): calcd. for
C36H25N2O4S2 [M – H]– 613.12612; found 613.12504. MS (ESI):
m/z = 614.3 [M]–, 570.4 (M – CO2), 555.4 [M – CH3]. UV/Vis
(EtOH, 3.18ϫ10–5 m): λmax (ε,
m =
–1 cm–1) 391 (3.8ϫ104),
441 (2.7ϫ104) nm; THF (3.21ϫ10–5 m): λmax (ε, m–1 cm–1) =
394 (4.2ϫ104), 460 (2.6ϫ104) nm; toluene (3.03ϫ10–5 m): λmax (ε,
148.35, 155.97, 185.95 ppm. IR (nujol): ν = 1670 (νCO) cm–1
.
˜
HRMS (EI): calcd. for C33H25NO3S2 547.127588; found
547.126650. UV/Vis (CH3CN): λmax (ε, m–1 cm–1) = 298 (1.8ϫ104),
344 (2.1ϫ104), 451 (3.6ϫ104) nm.
m
–1 cm–1)
=
399 (3.8ϫ104), 478 (2.1ϫ104) nm; CH3CN
(2.97ϫ10–5 m): λmax (ε, m–1 cm–1) = 393 (4.2ϫ104), 451 (2.6ϫ104)
nm; CH2Cl2 (3.21ϫ10–5 m): λmax (ε, m–1 cm–1) = 399 (5.6ϫ104);
486 (2.74ϫ104) nm.
2-Cyano-3-[6-(2-{4-{bis(4-methoxyphenyl)amino}phenyl}ethenyl)-
benzo[1,2-b:4,5-bЈ]dithiophen-2-yl]propenoic Acid (CR29):
Cyanoacetic acid (20.0 mg, 0.24 mmol) and piperidine (5 mg,
0.05 mmol) were added to a suspension of 4 (70 mg, 0.13 mmol) in
CH3CN (10 mL). The reaction mixture was stirred for 4 h at 80 °C.
The solvent was then evaporated and the residue was taken up with
a mixture HCl 0.1M (6 mL) and hexane (4 mL) and filtered to
afford CR29 (70 mg, 95%) as a red solid; m.p. 258–260 °C (dec).
1H NMR (300 MHz, [D6]DMSO): δ = 3.76 (s, 6 H), 6.74 (d, J =
8.5 Hz, 2 H), 6.90–6.96 (m, 5 H), 7.05 (d, J = 9 Hz, 4 H), 7.36–
7.46 (m, 3 H), 7.96 (s, 1 H), 8.23 (s, 2 H), 8.37 (s, 2 H) ppm. 13C
NMR (75 MHz, CDCl3): δ = 55.13, 101.17, 114.88, 115.84, 116.44,
118.26, 118.50, 119.02, 119.08, 121.51, 126.83, 126.95, 127.27,
127.55, 127.68, 127.81, 132.05, 135.06, 135.28, 135.35, 136.97,
139.06, 139.33, 140.89, 146.50, 147.12, 148.63, 155.98, 163.09 ppm.
Electrochemistry: Chromophores CR29 and CR52, and the corre-
sponding parent molecules BDT1 and BDT were characterized by
cyclic voltammetry at potential scan rates typically ranging from
0.05 to 2 Vs–1, in a 4 cm3 minicell with 0.5–0.75 mm solutions in
CH2Cl2 (chromophores) and CH3CN (parent molecules) with 0.1 m
TBAP as the supporting electrolyte, deaerated by N2 purging. In
particular, parent molecules required solvent CH3CN to allow ob-
servation of the first reduction peaks, located at very negative po-
tentials. On the other hand, the chromophores were hardly soluble
in CH3CN; however, in this case, the cathodic potential window of
CH2Cl2, granting good solubility, was sufficient, albeit narrower
than in CH3CN, on account of the much more positive reduction
peak potentials.
IR (nujol): ν = 3402 (νOH), 2360 (νCN), 1569 (νCO) cm–1. HRMS
The experiments were carried out with an AUTOLAB PGSTAT
12(8) potentiostat of EcoChemie (Utrecht, The Netherlands) run
by a PC with the GPES 4.9 software of the same manufacturer.
˜
(ESI): calcd. for C36H25N2O4S2 [M – H]– 613.12557; found
613.12579. UV/Vis (CH2Cl2, 6ϫ10–5 m): λmax (ε, m–1 cm–1) = 370
(2.2 ϫ 104), 460 (2.1 ϫ 104) nm; EtOH (5 ϫ 10–5 m): λmax (ε,
The working electrode was a glassy carbon GC disk embedded in
Teflon (Amel, 0.071 cm2). The optimized polishing procedure in-
volved surface treatment with a synthetic diamond powder of 1 mm
diameter (Aldrich) on a DP-Nap wet cloth (Struers).
m
–1 cm–1) = 368 (7.2ϫ103), 443 (1.2ϫ104) nm; THF (5ϫ10–5 m):
λmax (ε, m–1 cm–1) = 375 (2.9 ϫ104), 478 (3.7 ϫ 104) nm; toluene
(5.9ϫ10–5 m): λmax (ε, m–1 cm–1) = 371 (1.5ϫ104), 440 (1.3ϫ104)
nm; CH3CN (5.7ϫ 10–5 m): λmax (ε, m–1 cm–1) = 371 (3.6 ϫ104),
469 (3.9ϫ104) nm.
The operating reference electrode was aqueous saturated calomel
(SCE), but the experimental peak potentials have been normalized
vs. the Fc+|Fc redox couple (the intersolvental redox potential ref-
erence currently recommended by IUPAC),[16,17] having a redox po-
tential of 0.39 V (in CH3CN) or of 0.495 V (in CH2Cl2) vs. the
operating SCE reference electrode. The counter electrode was a
7-(2-{4-[Bis(4-methoxyphenyl)amino]phenyl}ethenyl)benzo[1,2-b:4,3-
bЈ]dithiophen-2-carboxy aldehyde (5): Following the same procedure
used for the synthesis of 4, phosphonium salt 1 (0.30 mmol) was
treated with benzo[1,2-b:4,3-bЈ]-2,7-dithiophenedicarbaldehyde
92
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94

Benzodithiophene-Containing Organic Dyes
platinum disk embedded in glass. The ohmic potential drop was
compensated for by the positive feedback technique.[18]
water bath and oven-baked for 1 h at 70 °C. After rinsing with
water, ethanol, and drying in air, they were subsequently baked at
550 °C for 45 min in air, then cooled to 70 °C and finally intro-
duced into a dye solution [0.3 mm in CH3CN/THF (10:1) contain-
ing 3 mm 3a,7a-dihydroxy-5b-cholic acid (CDCA)] for 16 h. The
Spiro-OMeTAD was dissolved in chlorobenzene at a concentration
of 90 mg/mL; a solution of lithium bis(trifluoromethylsulfonyl)-
imide salt (Li-TFSI) (0.04M in CH3CN) and 4-tert-butylpyridine
was added to the Spiro-OMeTAD solution and spin-coated at
700 rpm for 60 s. Spin coater rotation was activated 45 seconds
after dispensing the solution onto the substrate to promote effective
infiltration of Spiro-OMeTAD inside the mesoporous TiO2 layer.
The films were then placed in a thermal evaporator where 80 nm
thick silver electrodes were deposited through a shadow mask un-
der high vacuum (10–6 mbar), to obtain pixels of an active area of
ca. 0.08 cm2.
Computational Methods: Calculations on the structure and simu-
lated UV/Vis absorption spectra were performed using density
functional theory (DFT), and its time-dependent formulation
(TDDFT), as implemented in Gaussian program suite.[30] The
ground-state geometry for the protonated dyes were optimized in
gas-phase, within the B3LYP functional[31] using a 6-31G* basis
set.[12] At the optimized ground-state geometries TDDFT (B3LYP
and MPW1K/6-31G*)[13] excited state calculations were performed
in the gas phase as well as in acetonitrile, adopting the nonequilib-
rium conductor-like polarizable continuum model (CPCM).[32]
This computational setup has previously been shown to adequately
describe the electronic and optical properties of similar push-pull
dyes.[33,34]
Liquid Electrolyte Solar Cell Assembly: Fluorine-doped tin oxide
(FTO) glass (TEC-15/2.2 mm thickness, Solaronix) was used for
transparent conducting electrodes. The substrate was first cleaned
in a ultrasonic bath using a detergent solution, acetone, and eth-
anol (each step 15 min long). The FTO glass plates were immersed
into a 40mm aqueous TiCl4 solution at 70 °C for 30 min and
washed with water and ethanol. The Dyesol 18NR-AO TiO2 paste
was spread on the FTO glass plates by using the doctor blade tech-
nique. The TiO2 coated electrodes (active area 0.2 cm2) were grad-
ually 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 40mm TiCl4 solution, rinsed
with water and ethanol. The electrodes were heated at 500 °C for
30 min and after cooling (80 °C) were immersed for six hours in
sensitizing baths [CH3CN solutions of the CR29/52 dyes in 0.2 mm
concentration, containing 3 mm 3a,7a-dihydroxy-5b-cholic acid
(CDCA)].
Photovoltaic Measurements: Photovoltaic measurements were re-
corded under 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. J–V curves were
obtained by applying an external bias to the cell and measuring the
generated photocurrent 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.[35]
Time-Resolved Photoluminescence: Time-resolved photolumines-
cence spectra were recorded with a Hamamatsu Picosecond Fluo-
rescence Lifetime system based on a Strek camera with 2 ps time
response. Excitation (450 nm) was provided by a 150 fs ultrafast
laser oscillator (Chamaleon, Coherent Inc.); emisson signals were
detected at 700 nm. All samples were measured at room tempera-
ture in a vacuum chamber (10–4 mbar).
Counter electrodes were prepared by coating with a drop of
[H2PtCl6] solution (2 mg of Pt in 1 mL of ethanol) a FTO plate
(TEC 15/2.2 mm thickness, Solaronix) 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 iono-
mer film (Surlyn, 25 μm thickness, Dyesol).
Supporting Information (see footnote on the first page of this arti-
cle): Extended calculations, simulated absorption spectra, ad-
ditional solar cell parameters, UV absorption of BDT and BDT1,
optical absorptions upon acid and base treatment of CR29 and
CR52. UV absorption of CR29 on TiO2.
An Iolitech ES-0004 HP electrolyte was used, containing 1-butyl-
3-methylimidiazolium iodide, iodine, guanidinium thiocyanate, and
tert-butylpyridine in a mixture of valeronitrile and acetonitrile, with
0.06M LiI added. The hole was then sealed by applying a Surlyn
patch and a cover glass and finally a conductive Ag-based paint
was deposed at the electrical contacts.
Acknowledgments
V. R. thanks Marcelo Alcocer and Ajay Ram Srimath Kandada for
support in the transient optical spectroscopy experiments. P. S. and
F. D. A. thank MIUR-PRIN 2008 and CNR-EFOR for financial
support. E. L. and S. M. thank the Fondazione Cariplo (grant
number 2008-2205) and the University of Milan for a Ph.D. fellow-
ship (E. L.). C. B. thanks CNR-PM.P04.012 for financial support.
The authors thank Rossana Colombo for some experimental work.
Solid-State Solar Cell Assembly: FTO-coated glass sheets (15 Ω/sq
Pilkington) were etched with zinc powder and HCl (2 m) to obtain
the required electrode pattern. The sheets were then washed with
soap (Hellmanex at 2% in water), deionized water, acetone, meth-
anol, and finally treated under an oxygen plasma for 10 min to
remove the last traces of organic residues. The FTO sheets were
subsequently coated with a compact layer of TiO2 (100 nm) by
aerosol spray pyrolysis deposition at 450 °C, using oxygen as the
carrier gas.
[1] a) B. O’Regan, M. Grätzel, Nature 1991, 353, 737–740; b) M.
Grätzel, J. Photochem. Photobiol. C: Photochem. Rev. 2003, 4,
145–153; c) N. S. Lewis, Science 2007, 315, 798–801; d) M.
Graetzel, Acc. Chem. Res. 2009, 42, 1788–1798; e) A. Hagfeldt,
G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev. 2010,
110, 6595–6663; f) Md. K. Nazeeruddin, E. Baranoff, M.
Grätzel, Sol. Energy Mater. Sol Cells 2011, 85, 1172–1178.
[2] a) Md. K. Nazeeruddin, C. Klein, P. Liska, M. Grätzel, Coord.
Chem. Rev. 2005, 249, 1460–1467; b) Y. Chiba, A. Islam, Y.
Watanabe, R. Komiya, N. Koide, L. Han, Jpn. J. Appl. Phys.
Part 2 2006, 45, L638–L640; c) F. Gao, Y. Wang, D. Shi, J.
Zhang, M. Wang, X. Jing, R. Humphry-Baker, P. Wang, S. M.
Zakeeruddin, M. Grätzel, J. Am. Chem. Soc. 2008, 130, 10720–
10728.
Standard Dyesol TiO2 paste was previously diluted 1:3.5 in ethanol
and ultrasonicated until complete mixing. The paste was then ap-
plied by means of the doctor blade technique using scotch tape and
a pipette on the TiO2 compact layer coated FTO sheets to get a
TiO2 average thickness of 1.8 μm.
The sheets were then slowly heated to 550 °C (ramped over 1.5
hours) and baked at this temperature for 30 min in air. After cool-
ing, the slides were cut to size and soaked in 15 mm TiCl4 in a
Eur. J. Org. Chem. 2013, 84–94
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
93

F. De Angelis, C. Baldoli, E. Licandro et al.
FULL PAPER
[3] A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran,
Md. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M.
Zakeeruddin, M. Grätzel, Science 2011, 334, 629–633.
[4] a) M. Amaresh, K. Markus, R. Fischer, P. Bäuerle, Angew.
Chem. 2009, 482, 474–2499; b) Z. Ning, H. Tian, Chem. Com-
mun. 2009, 375, 483–5495; c) G. Li, B. K.-J. Jiang, P. Bao, Y.-
F. Li, S.-L. Li, L.-M. Yang, New J. Chem. 2009, 33, 868–876;
d) W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F.
Wang, C. Pan, P. Wang, Chem. Mater. 2010, 22, 1915–1925 and
ref. 4 therein e) G. Li, K.-J. Jiang, P. Bao, Y.-F. Li, S.-L. Lia,
L.-M. Yang, New J. Chem. 2009, 33, 868–876; f) X. Cheng, M.
Liang, S. Sun, Y. Shi, Z. Ma, Z. Sun, S. Xue, Tetrahedron 2012,
68, 5375–5385.
[5] a) B. Wex, B. R. Kaafarani, E. O. Danilov, D. C. Neckers, J.
Phys. Chem. A 2006, 110, 13754–13758; b) K. Takimiya, Y.
Konda, H. Ebata, N. Niihara, T. Otsubo, J. Org. Chem. 2005,
70, 10569–10571; c) Y. Liu, X. Wan, F. Wang, J. Zhou, G.
Long, J. Tian, Y. Chen, Adv. Mater. 2011, 23, 5387–5391.
[6] a) Y. Zhang, S. K. Hau, H.-L. Yip, Y. Sun, O. Acton, A. K.-Y.
Jen, Chem. Mater. 2010, 22, 2696–2698; b) Y. He, Y. Zhou, G.
Zhao, J. Min, X. Guo, B. Zhang, M. Zhang, J. Zhang, Y. Li,
F. Zhang, O. Inganaes, J. Polym. Sci. Part A 2010, 48, 1822–
1829; c) T. T. M. Dang, S.-J. Park, J.-W. Park, D.-S. Chung,
C. E. Park, Y.-H. Kim, S.-K. Kwon, J. Polym. Sci. Part A 2007,
22, 455277–5284; d) C. Kim, T. J. Marks, A. Facchetti, M.
Schiavo, A. Bossi, S. Maiorana, E. Licandro, F. Todescato, S.
Toffanin, M. Muccini, C. Graiff, A. Tiripicchio, Org. Electron.
2009, 22, 101511–1520; e) U. Giovannella, C. Botta, A. Bossi,
E. Licandro, S. Maiorana, J. Appl. Phys. 2006, 100, 083107.
[7] a) L. Huo, J. Hou, Polym. Chem. 2011, 2, 2453–2461; b) Q.
Shi, H. Fan, Y. Liu, J. Chen, L. Ma, W. Hu, Z. Shuai, Y. Li,
X. Zhan, Macromolecules 2011, 44, 4230–4240; c) S. C. Price,
A. C. Stuart, W. You, Macromolecules 2010, 43, 4609–4612; d)
Y.-J. Cheng, S.-H. Yang, C.-S. Hsu, Chem. Rev. 2009, 109,
5868–5923.
[8] a) X. Hao, M. Liang, X. Cheng, X. Pian, Z. Sun, S. Xue, Org.
Lett. 2011, 13, 5424–5427; b) After acceptance of this manu-
script, a study concerning the use of substituted BDT in push-
pull dye structures was published, see: P. Gao, H. N. Tsao, M.
Grätzel, M. K. Nazeeruddin, Org. Lett. 2012, 14, 4330–4333.
[9] M. Pizzotti, R. Ugo, F. Tessore, S. Maiorana, E. Licandro, E.
Longhi, C. Baldoli, A. Orbelli Biroli, F. De Angelis, Patent
Appl. 2011, MI2011A000181; S. Maiorana, E. Licandro, E.
Longhi, C. Baldoli, F. De Angelis, Patent Appl. 2011,
MI2011A000184.
[10] L. R. Dalton, A. K.-Y. Jen, T. Londergan, W. B. Carlson, G.
Phelan, D. Huang, D. Casmier, T. Ewy, N. Buker, WO/2002/
008215, PCT/US2001/023339, 2002.
[11] G. Kossmehl, P. Beimling, G. Manecke, Makromol. Chem.
1983, 184, 627–650.
[17]
R. S. Ashraf, M. Shahid, E. Klemm, M. Al-Ibrahim, S.
Sensfuss, Macromol. Rapid Commun. 2006, 27, 1454–1459.
S. Trasatti, Pure Appl. Chem. 1986, 58, 955–966.
S. Amthor, B. Noller, C. Lambert, Chem. Phys. 2005, 316, 141–
152.
[18]
[19]
[20]
[21]
K. Meerholz, J. Heinze, Electrochim. Acta 1996, 41, 1839–1842.
a) A. Kay, M. Grätzel, J. Phys. Chem. 1993, 97, 6272–6277; b)
K. Hara, Y. Dan-Oh, C. Kasada, Y. Ohga, A. Shinpo, S. Suga,
K. Sayama, H. Arakawa, Langmuir 2004, 20, 4205–4210; c) X.
Jiang, T. Marinado, E. Gabrielsson, D. P. Hagberg, L. Sun, A.
Hagfeldt, J. Phys. Chem. C 2010, 114, 2799–2805.
a) S. Tatay, S. A. Haque, B. O’Regan, J. R. Durrant, W. J. H.
Verhees, J. M. Kroon, A. Vidal-Ferran, P. Gavina, E. Palo-
mares, J. Mater. Chem. 2007, 17, 3037–3044; b) Z. S. Wang, Y.
Cui, Y. Dan-Oh, C. Kasada, A. Shinpo, K. Hara, J. Phys.
Chem. C 2007, 111, 7224–7230; c) M. Pastore, F. De Angelis,
ACS Nano 2010, 4, 556–562.
a) G. Redmond, D. Fitzmaurice, J. Phys. Chem. 1993, 97,
1426–1430; b) K. Fredin, J. Nissfolk, G. Boschloo, A. Hagfeldt,
J. Electroanal. Chem. 2007, 609, 55–60.
a) Q. Yu, Y. Wang, Z. Yi, N. Zu, J. Zhang, M. Zhang, P. Wang,
ACS Nano 2010, 4, 6032–6038; b) A. Abrusci, R. S. S. Kumar,
M. Al-Hashimi, M. Heeney, A. Petrozza, H. J. Snaith, Adv.
Funct. Mater. 2011, 21, 2571–2579.
[22]
[23]
[24]
[25]
[26]
A. Listorti, C. Creager, P. Sommeling, J. Kroon, E. Palomares,
A. Fornelli, B. Breen, P. R. F. Barnes, J. R. Durrant, C. Law,
B. O’Regan, Energy Environ. Sci. 2011, 4, 3494–3501.
L. R. Dalton, A. K.-Y. Jen, T. Londergan, W. B. Carlson, G.
Phelan, D. Huang, D. Casmier, T. Ewy, N. Buker, 2002, WO/
2002/008215, PCT/US2001/023339.
[27] A. Hercouret, M. Le Corre, Synthesis 1988, 157–158.
[28] G. Kossmehl, P. Beimling, G. Manecke, Makromol. Chem.
1983, 184, 627–650.
[29] J. Larsen, K. Bechgaard, Acta Chem. Scand. 1996, 50, 71–76.
[30] 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. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, 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. Starov-
erov, 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. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, 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,
rev. A.1, Gaussian, Inc., Wallingford CT, 2009.
[12] V. A. Rassolov, J. A. Pople, M. A. Ratner, T. L. Windus, J.
Chem. Phys. 1998, 109, 1223–1229.
[13] B. J. Lynch, P. L. Fast, M. Harris, D. G. Truhlar, J. Phys. Chem.
A 2000, 104, 4811–4815.
[14] A. Baheti, P. Tyagi, K. R. J. Thomas, Y.-C. Hsu, J. T’suen Lin,
J. Phys. Chem. C 2009, 113, 8541–8547; D. Kumar, K. R. J.
Thomas, C.-P. Lee, K.-C. Ho, Org. Lett. 2011, 13, 2622–2625.
[15] a) C. Ma, E. Mena-Osteriz, T. Debaerdemaeker, M. M. Wienk,
R. A. J. Janssen, P. Bauerle, Angew. Chem. 2007, 119, 1709; An-
[31] P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623–11627.
[32] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem.
2003, 24, 669–681.
[33] A. Abbotto, N. Manfredi, C. Marinzi, F. De Angelis, E. Mos-
coni, J.-H. Yum, Z. Xianxi, M. K. Nazeeruddin, M. Grätzel,
Energy Environ. Sci. 2009, 2, 1094–1101.
gew. Chem. Int. Ed. 2007, 46, 1679–1683; b) A. Bossi, L. Falci- [34] M. Pastore, E. Mosconi, F. De Angelis, M. Grätzel, J. Phys.
ola, C. Graiff, S. Maiorana, C. Rigamonti, A. Tiripicchio, E.
Licandro, P. R. Mussini, Electrochim. Acta 2009, 54, 5083–
5097.
Chem. C 2010, 114, 7205–7212.
[35] S. Ito, M. K. Nazeeruddin, P. Liska, P. Comte, R. Charvet, P.
Pechy, M. Jirousek, A. Kay, S. M. Zakeeruddin, M. Grätzel,
Prog. Photovolt Res. Appl. 2006, 14, 589–601.
Received: July 18, 2012
[16] W.-Y. Wong, X.-Z. Wang, Z. He, A. B. Djurisic, C.-T. Yip, K.-
Y. Cheung, H. Wang, C. S. K. Mak, W.-K. Chan, Nat. Mater.
2007, 6, 6521–527.
Published Online: November 22, 2012
94
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 84–94