Molecular engineering of largely π-extended metal-free sensitizers containing benzothiadiazole units: Approaching 10% efficiency dye-sensitized solar cells using iodine-based electrolytes

Article abstract of DOI:10.1016/j.dyepig.2016.04.020

Based on the structural motif of a robust organic dye, new largely π-extended organic sensitizers for dye-sensitized solar cells (DSSCs) have been designed to optimize the interactions between the dye/TiO₂ layer and the redox electrolyte. The m

Full text of DOI:10.1016/j.dyepig.2016.04.020

Accepted Manuscript
Molecular Engineering of Largely π-Extended Metal-Free Sensitizers Containing
Benzothiadiazole Units: Approaching 10% Efficiency Dye-Sensitized Solar Cells
Using Iodine-Based Electrolytes
Roberto Grisorio, Luisa De Marco, Roberto Giannuzzi, Giuseppe Gigli, Gian Paolo
Suranna
PII:
S0143-7208(16)30141-3
10.1016/j.dyepig.2016.04.020
DYPI 5198
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 3 February 2016
Revised Date: 6 April 2016
Accepted Date: 7 April 2016
Please cite this article as: Grisorio R, De Marco L, Giannuzzi R, Gigli G, Suranna GP, Molecular
Engineering of Largely π-Extended Metal-Free Sensitizers Containing Benzothiadiazole Units:
Approaching 10% Efficiency Dye-Sensitized Solar Cells Using Iodine-Based Electrolytes, Dyes and
Pigments (2016), doi: 10.1016/j.dyepig.2016.04.020.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Molecular Engineering of Largely π-Extended Metal-Free Sensitizers
Containing Benzothiadiazole Units: Approaching 10% Efficiency Dye-
Sensitized Solar Cells Using Iodine-Based Electrolytes
ab
c
c
Roberto Grisorio, Luisa De Marco,* Roberto Giannuzzi,
bd
ab
Giuseppe Gigli, Gian Paolo Suranna*
a
DICATECh – Dipartimento di Ingegneria Civile, Ambientale, del Territorio, Edile e di Chimica,
Politecnico di Bari, Via Orabona 4, 70125 Bari, Italy. E-mail: gianpaolo.suranna@poliba.it.
b
CNR NANOTEC – Istituto di Nanotecnologia, Polo di Nanotecnologia c/o Campus Ecotekne, via
Monteroni, 73100 Lecce, Italy.
c
Center for Biomolecular Nanotechnologies (CBN) Fondazione Istituto Italiano di Tecnologia (IIT),
Via Barsanti 1, Arnesano, 73010, Italy. E-mail: luisa.demarco@iit.it.
d
Dipartimento di Matematica e Fisica “E. De Giorgi”, Università del Salento, via per Arnesano,
73100 Lecce, Italy.
Abstract
Based on the structural motif of a robust organic dye, new largely π-extended organic sensitizers for
dye-sensitized solar cells (DSSCs) have been designed to optimize the interactions between the
dye/TiO layer and the redox electrolyte. The molecular tailoring was performed by i) positioning
2
an alkyl chain onto the π-spacer and ii) increasing the bulkiness of the electron-donor. A deep
theoretical investigation reveals that the main electronic transitions of the three dyes own only a
weak charge-transfer character. Notwithstanding these unusual photo-excitation dynamics
experimentally supported by solvatochromism, impressive photovoltaic performances are obtained

ACCEPTED MANUSCRIPT
for the three sensitizers, reaching, power conversion efficiencies up to 9.8% under 1.0 sun
illumination, that were remarkably higher than those shown by N719 standard. This result
–
represents one of the highest performances exhibited by fully organic sensitizers employing the I
–
/
I redox shuttle.
3
Keywords D–A–π–A sensitizer, dye-sensitized solar cell, density functional theory, iodine-based
electrolyte.
Introduction
The prospected low manufacturing cost and potential building integration of dye-sensitized
solar cells (DSSCs) continue to fuel a vivid research effort mainly devoted to the dye
optimization [1]. The state-of-the-art of DSSC efficiencies has until recently been a
prerogative of ruthenium-based complexes (~11%) [2] and zinc-porphyrin sensitizers
(
~13%) [3]. Although the vast majority of highly performing fully-organic sensitizers
currently exhibits lower performances (~10%) [4], a conspicuous share of the DSSC
scientific community remains intrigued by the molecular engineering potential of metal-free
dyes, also considering that the use of a low-abundance metal such as ruthenium as well as
the many low-yielding steps needed to obtain suitably functionalized porphyrins might
jeopardize the perspective of full-scale DSSC mass production. The great efforts aimed at
unravelling the potential of fully organic sensitizers has been recently realised in the
obtainment of devices with outstanding efficiencies (up to 12.5%) [5], sustaining the
research for more and better performing organic dyes.
For years, the molecular design of efficient sensitizers has been dominated by the rule of
putting into conjugation an electron-donating and electron-accepting unit (the latter suitably

ACCEPTED MANUSCRIPT
designed to act as the anchoring group for the inorganic semiconductor) via an electron-rich
π-segment (namely the D–π–A architecture) [6]. The consequent “push-pull” configuration
determines the photo-induced shift of the electron density towards the anchoring region
bound on the inorganic semiconductor, thereby promoting the electron transfer step [7]. The
molecular engineering of the π-bridge has been mainly devoted to the amplification of the
light-harvesting properties of the corresponding sensitizers; however, a mere conjugation
extension of the chromophoric core was proven as detrimental for device performance [8].
Recently, the incorporation of a further electron-withdrawing unit within the π-bridge
(
leading to a D–A–π–A architecture) has opened new routes for improving the light-
harvesting and stability of the corresponding organic sensitizers [9]. However, placing an
electron-deficient aromatic unit within the π-bridge leads to structure-property relationships
that cannot be trivially anticipated. By gathering electron density at the excited state of the
corresponding D–A–π–A sensitizer, the introduction of a further electron-withdrawing
system within the π-spacer could indeed prevent an efficient electron transfer onto the TiO2
surface [10]. Consequently, the position of the electron-accepting aromatic unit with respect
to the anchoring moiety as well as the introduction of functional groups causing distortion of
the chromophore backbone have to be accurately pondered when drawing the “molecular
blueprint” of a D–A–π–A sensitizer. Our role in perfectioning the D–A–π–A sensitizers has
mainly been focused on the π-bridge extension as well as on the judicious selection of low-
environmental impact structures and synthetic strategies [11], in the awareness that even
minor structural modification can exert a significant influence on the performance of a
selected class of organic dyes.
In this contribution, the properties and the photovoltaic performances of new D–A–π–A
fully organic sensitizers (G4 and G5, Figure 1) have been investigated. These dyes have
been assembled on the basis of the structural motif characterizing a robust organic dye (G3),
that achieved high power conversion efficiency (8.6%) and exhibited an excellent stability

ACCEPTED MANUSCRIPT
during device operation [11a]. The new sensitizers were conceived to gain a control at the
interface between the dye/TiO layer and the redox electrolyte, aiming at enhancing the
2
corresponding device photocurrent and photovoltage. The molecular tailoring of the G3
scaffold was consequently carried out by suitable positioning an alkyl chain onto the π-
spacer (G4 vs G3) and by increasing the bulkiness of the electron-donating unit (G5 vs G4
)
extending its conjugation. Remarkable power conversion efficiencies (PCE) up to 9.8%
were achieved by G5 (the best of the series) using iodine-based electrolytes; the obtained
PCE were found to be higher than those shown by traditional N719 standard dye
formulations [12].
R2
N
COOH
CN
R2
S
S
R1
N
N
S
R2 =
R2 =
R2 =
O
O
G3: R1 = H
G4: R1 = n-Hexyl
G5: R1 = n-Hexyl
O
Figure 1. Chemical structures of G3, G4 and G5.
Results and discussion
Synthesis and characterization
Differently from the previously reported synthesis of G3, which could be accomplished by
C-H activation reactions [11a], the syntheses of G4 and G5 posed additional complications
due to the unsymmetrical placement of the n-hexyl functionality with respect to the
benzothiadiazole core (Scheme 1). Concerning the electron-donating building block, 4-

ACCEPTED MANUSCRIPT
(
4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-N,N- bis((2-ethylhexyloxy)phen-4-yl)aniline it
was obtained following a literature procedure [11a]. Conversely, the preparation of the
electron-donating unit of G5 started with the reaction of bromo-triphenylamine with two
equivalents of N-iodo-succinimide (NIS) to afford the diiodo-derivative
coupling reaction between and two equivalents of 2-((2-ethylhexyloxy)phen-4-yl)-4,4,5,5-
tetramethyl-1,3,2-dioxaborolane gave the bromo-derivative prone to be functionalized
with bis(pinacolato)diboron affording the relevant boronic ester . Concerning the synthesis
of the chromophoric -bridge, the benzothiadiazole-containing derivative was reacted in a
Suzuki coupling with an equimolar amount of thienyl-boronic acid yielding the
unsymmetrically substituted compound . A selective bromination on the activated
thiophene ring with -bromo-succinimide (NBS) afforded the corresponding bromo-
derivative . A Suzuki reaction between and 4-formylphenyl boronic acid placed the
aldehyde functional group for the subsequent conversion into the anchoring moiety. A
further NBS bromination of the so obtained intermediate yielded the bromo-derivative
The synthetic procedures were completed by carrying out a Suzuki cross-coupling between
and the relevant boronic-esters of the electron donating groups, concluded by a Knovenagel
1. A Suzuki cross-
1
2
3
π
4
5
N
6
6
7
8.
8
condensation of the obtained aldehydes
9 and 10 with cyanoacetic acid affording G4 and
G5, respectively.

ACCEPTED MANUSCRIPT
Scheme 1. Synthetic approach followed for the preparation of G4 and G5. Reaction conditions: i)
NIS, DMF; ii) 2-((2-ethylhexyloxy)phen-4-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, Pd(PPh ) ,
3
4
1
M NaHCO , THF; iii) bis(pinacolato)diboron, PdCl (dppf), potassium acetate, dioxane; iv)
3
2
thiophen-2-yl boronic acid, Pd(PPh ) , 1M NaHCO , THF; v) NBS, DMF; vi) 4-formylphenyl
3
4
3
boronic acid, Pd(PPh ) , 1M NaHCO , THF; vii) 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-
3
4
3
N,N-bis((2-ethylhexyloxy)phen-4-yl)aniline, Pd(PPh ) , 1M NaHCO , THF; viii) 3, Pd(PPh ) , 1M
3
4
3
3 4
NaHCO , THF; ix) cyanoacetic acid, piperidine, CHCl .
3
3
The UV–Vis absorption of G3-5 in THF are depicted in Figure 2A highlighting that all
sensitizers exhibit two major absorption bands at 350–400 nm and 500–550 nm, often
simplistically ascribed to localized aromatic π–π* transitions or to intramolecular charge-
transfer (ICT) transitions. However, a deeper insight into the nature of these two absorption
bands has been provided by theoretical calculations (vide infra). Concerning the absorption
wavelengths, the dye structural tailoring seems to have a non-negligible effect on the
absorption profiles of the triad. The absorption peaks falling in the UV region are located at
3
78 (G3), 370 (G4) and 354 (G5) nm, with molar extinction coefficients of 41600, 33600

ACCEPTED MANUSCRIPT
and 48600 M cm respectively, whereas the lower energy bands peaked at 535 (G3), 527
–1
–1
(
G4) and 524 (G5) nm, with corresponding molar extinction coefficients of 38300, 27900
–1
–1
and 30100 M cm . Their absorption onsets in THF solution followed the trend observed
for the maxima (641, 635 and 624 nm for G3 G4 and G5, respectively), confirming
,
potentially significant light-harvesting properties for all dyes. Nevertheless, it is apparent
how the introduction of the alkyl chain onto the π-bridge induces a blue-shift of the
absorption maxima in G4 and G5 with respect to G3 with a corresponding lowering of the
relevant molar extinction coefficients.
Suitable investigation of the solvatochromic behavior of an organic dye can be useful to
predict the influence on the absorption profile consequent to the sensitizer immobilization
on TiO surface. The formation of titanium carboxylate upon dye adsorption process might
2
indeed cause a detrimental blue-shift of its absorption profile [13]. The absorption properties
of the dyes in their deprotonated form can be evaluated in solution using a large excess of
triethylamine (TEA) thus bypassing the complications due to scattering and aggregation that
can occur carrying out the investigation on titania [14]. As shown in Figure 2A, the G3-5
absorption profiles recorded in the presence of excess of TEA are very similar to the
corresponding ones recorded in THF (the blue-shift of the absorption bands being contained
within 5 nm) substantiating the assumption that the original light-harvesting properties of
these molecules are preserved from the deprotonation-induced blue-shift. It can also be
anticipated that the peculiar structural motifs of these D–A–π–A dyes lead to electronic
transitions with a low ICT character, suggesting us to carefully deepen their nature and the
relevant photo-excitation dynamics typical of the DSSC work cycle. This behaviour was
confirmed by recording the absorption spectra of the three sensitizers adsorbed onto a TiO2
substrate (Figure S1) that were nearly superimposable to those measured in THF solution.

ACCEPTED MANUSCRIPT
Figure 2. (A) Absorption profiles of G3-5 in THF and THF/triethylamine (3/1 v/v) solutions.
Anodic (B) and cathodic (C) cyclic voltammograms of G3, G4 and G5 as thin films with ferrocene
as the reference. (D) Calculated and experimental ionization potential, calculated and experimental
electron affinity, HOMO and LUMO energy levels of G3, G4 and G5.
Electrochemistry provides crucial insights into the thermodynamics of the electron injection
as well as of the dye regeneration by the redox shuttle. The electrochemical properties of the
three sensitizers were investigated by cyclic voltammetry (CV) aiming at determining the
energy levels of their frontier orbitals. The anodic and cathodic scans for the three
sensitizers are illustrated in Figure 2B and Figure 2C, respectively. The observed oxidation
potentials correspond to highest occupied molecular orbitals (HOMO) energy levels located
at –5.04, –5.02 and –5.24 eV for G3
, G4 and G5, respectively, implying a
–
–
3
thermodynamically feasible regeneration of the dyes by the I /I redox shuttle during the
DSSC work cycle. To confirm the reversibility of the redox events in the case of G4 and G5
,

ACCEPTED MANUSCRIPT
their electrochemical behavior was also studied in CH Cl solution (see Figure S2). It is
2
2
evident that G5 shows a downshifted HOMO energy value (~0.2 eV) compared to G3 and
G4, suggesting an important role of the electron-donating unit in determining the ionization
potential of the relevant sensitizer, as confirmed (vide infra) by theoretical calculations. It is
reasonable to suppose that the introduction of a further phenyl spacer (biphenyl vs phenyl)
reduces the mesomeric contribution to the electron-donating properties of the relevant
triarylamine, to which an important contribution is also provided by the alkoxy groups. This
effect produces the HOMO stabilization observed in G5 despite the increased number of
aromatic units involved in the electronic delocalization. By contrast, the effect of the n-hexyl
chain functionalization was mainly observed on the lowest unoccupied molecular orbital
(
LUMO) energy levels of the three sensitizers, that were found to be –3.24, –3.15 and –3.17
eV for G3 G4 and G5, respectively. The slight upshift of the LUMO energy values of G4
and G5 with respect to that of G3 can be reasonably ascribed to the distortion of the
,
π
-
bridge backbone (vide infra) induced by the introduction of the alkyl chain. The LUMO
energy values lie above the conduction band edge of TiO , ensuring sufficient driving force
2
for the electron injection from the excited dyes into the conduction band of the inorganic
semiconductor. On these grounds, all three sensitizers are qualified for potential application
in DSSCs.
Theoretical calculations
To gain insight into the photophysical and electronic properties of the three sensitizers, a
theoretical investigation was performed applying density functional theory (DFT) and time-
dependent DFT (TD-DFT) calculations on the geometrically optimized suitable molecular
models, at the B3LYP/6-31G(d,p) level of the theory using the Gaussian 09 program
packages [15]. Since the calculations were carried out including solvent effects (THF) using
non-equilibrium implementation of the conductor-like polarizable continuum model (C-

ACCEPTED MANUSCRIPT
PCM), in order to reduce the computational load, the 2-ethyl-hexyl chains were replaced by
methyl groups, in consideration of their minor influence on the molecules conformation. The
analysis of the optimized geometry (Figure S3) of the three sensitizers revealed a near planar
conformation of the
thiophene units, while higher dihedral angles were observed at the linkage between
thiophene and benzene units. The introduction of the -hexyl chain onto the thiophene unit
induced a pronounced distortion of the corresponding thiophene/benzene dihedral angle
~37°) in G4 and G5 without, however, any apparent influence on the electronic distribution
π-bridge with low dihedral angles between benzothiadiazole and
n
(
of the frontier orbitals. The calculated electron density distributions of the main energy
levels are shown in Figure S4. The HOMOs are principally distributed on the electron-rich
triarylamine group as well as on part of the chromophoric
experimental observation previously reported that the electron binding energy for the
HOMOs is not sensitive to the structural modifications involving the -spacer extension, but
only those inherent the electron-donating unit. Conversely, the LUMOs of the three
π-bridge, confirming the
π
sensitizers are partially localized on the
group.
π-bridge and in the proximity of the anchoring
The ionization potentials (the absolute value of the HOMO energy level) of G3, G4 and G5
were calculated as the difference between the total energies of the (N–1)-electron and (N)-
electron states [16]. All energies were calculated starting from the geometrically optimized
structures of the sensitizer neutral states. The calculations returned ionization potentials
following the same trend observed by CV measurements (see Figure 2D). With the same
approach, the electron affinities (the absolute value of the LUMO energy level) of G3-5
were obtained calculating the difference between the total energies of the (N)-electron and
(
N+1)-electron states (Figure 2D).
The optical behavior of the three sensitizers was deepened by investigating the vertical
S →S excitation energies at the TD-DFT level of theory using the CAM-B3LYP functional
0
n

ACCEPTED MANUSCRIPT
[
17] including solvent effects (THF) with the C-PCM simulation. It was found that the
calculated optical data in terms of both excitation energies and oscillator strength are in good
agreement with the experimental measurements (Figure S5–S7). The same considerations
can be made for the calculation of the absorption spectra of the deprotonated form of the
three sensitizers, substantiating the accurateness of the theoretical calculations and the
corresponding interpretation on the basis of the solvatochromic measurements. According to
the calculation (Table 1), the S →S excitations of G3-5 sensitizers are predominantly
0
1
described as HOMO→LUMO transitions with additional and significant HOMO–
→LUMO and HOMO→LUMO+1 contributions. On the other hand, the S →S excitations
1
0
2
can principally be described as HOMO–1→LUMO transitions with additional and
significant HOMO→LUMO, HOMO→LUMO+1 and HOMO–2→LUMO character.
Dye
G3
Transition
→S
λ(nm)
531.3
f
configuration
S
1.75
H→L (56%); H–1→L (33%); H→L+1 (9%)
H–1→L (43%); H→L+1 (26%); H–2→L (24%);
H→L (8%)
H→L (56%); H–1→L (32%); H→L+1 (9%)
H–1→L (45%); H→L+1 (25%); H–2→L (23%);
H→L (7%)
H→L (49%); H–1→L (39%); H→L+1 (7%)
H–1→L (35%); H→L+1 (24%); H–2→L (23%);
H–1→L+1 (10%); H→L (8%)
0
1
S0→S2
→S
S0→S2
→S
S0→S2
396.8
519.3
388.2
511.0
384.5
0.51
1.59
0.52
1.66
0.51
G4
G5
S
0
1
S
0
1
Table 1. Excitation energies (l), oscillator strength (f) and configuration of the main calculated
transitions (S →S and S →S ) of G3, G4 and G5.
0
1
0
2
Therefore, the TD-DFT calculations describe the fundamental absorption peak as an
electronic transition composed of different contributions, complicating an interpretation
straightforwardly based on the electron density shift upon photo-excitation of the dye. In
cases such as this, it is more opportune to analyze the individual excited states using natural
transition orbitals (NTOs), consisting of a simple orbital picture of the TD-DFT calculations
obtained by representing the electronic transition (exciton) in terms of two molecular

ACCEPTED MANUSCRIPT
orbitals (describing electron and hole respectively) thus allowing to straightforwardly
monitor the electron density shift from the ground state (hole) to the excited state (electron).
The calculated NTOs of the three sensitizers (Figure 3) visually describe the photo-dynamics
of the main excitation (S →S ) implying a partial shift of the electron density towards the
0
1
acceptor unit of the dyes, allowing in principle the electron transfer from the excited dye to
the semiconductor conduction band at the dye/titania interface. However, contrary to what
was observed in the case of G3, the π-backbone distortion in G4 and G5 seems to impair the
photo-induced shift of the electronic density towards the anchoring group, thereby resulting
to be mostly localized around the electron-withdrawing benzothiadiazole unit upon photo-
excitation. Since the hole/electron NTOs of the S →S excitation are mainly localized on the
0
1
chromophoric π-bridge of the three sensitizers, the low ICT character of these electronic
transitions emerging from the solvatochromic behavior previously described finds a deeper
rationale. Also the higher energy S →S excitation (the relevant NTOs are shown in Figure
0
2
S8) only promoting a densification of the electron density in proximity of the anchoring
group during the photo-excitation can be classified as a low ICT character transition, thereby
also partially contributing to photon-to-current conversion in DSSC under ambient solar
illumination. On these bases, however, it is reasonable to suppose that the electron transfer
process at the dye/TiO interface is not only dependent on the involved electronic transition
2
(
S →S or S →S excitation) but also relatively slower with respect to the very wide range
0 1 0 2
of organic sensitizers endowed with strong ICT absorption bands.

ACCEPTED MANUSCRIPT
Figure 3. Calculated natural transition orbitals (NTOs) describing the photo-excitation dynamics of
the S →S transition in G3, G4 and G5.
0
1
Photovoltaic properties
In view of their favourable optical and electrochemical properties, the photovoltaic
–
–
3
performances of G3-5 were evaluated using the I /I couple as the redox shuttle and
comparing them with those of the ruthenium-based sensitizer N719 as the reference dye. In
order to avoid the undesirable dye aggregation on the TiO surface and to suppress
2
detrimental recombination processes leading to dark current [18], several tests using
different amounts of chenodeoxycholic acid (CDCA) were carried out for the dyeing of the
TiO electrodes, as summarized in Table 2 along with the relevant device figures of merit.
2
The appropriate amount of CDCA significantly enhanced the dye performances: their power
conversion efficiency (PCE) was, in fact, increased from 4.1% to 8.6% in the case of G3
(
using 30 mM CDCA), from 5.1% to 9.5% in the case of G4 (using 10 mM CDCA) and
from 5.5% to 9.8% in the case of G5 (using 10 mM CDCA). These results markedly

ACCEPTED MANUSCRIPT
outperform those obtained using the very well-known ruthenium-based N719 reference dye
which reached a PCE of 8.7% in DSSC fabricated with the same architecture. Furthermore,
the performances of G3-5 sensitizers were also recorded under intensities simulating diffuse
sunlight (0.66 and 0.33 sun) observing that the efficiencies were even increased, reaching, in
the case of the most performing dye G5, a PCE of 10.3% under 0.33 sun (Figure 4A). These
results can be interpreted considering that, at reduced sun intensity, mass-transport
limitations of the redox mediators are remarkably alleviated.
It is apparent how G4 and G5 exhibited the better performances using lower CDCA
amounts compared to those employed in the case of G3. Reasonably, the presence of the n-
hexyl chain in G4 and G5 effectively contributes to space the sensitizer molecules onto the
TiO surface preventing the analogous effect exerted by the CDCA molecules at the expense
2
of the photoanode active area, as supported by dye loading measurements (Table S1).
A significant aspect emerges from the analysis of the spectra of monochromatic incident
photon-to-current conversion efficiencies (IPCEs) recorded for the DSSCs. Notwithstanding
the presence of large TiO scattering nanoparticles in the photoanode broadening the shape
2
of the IPCE spectra, these (Figure 4B) seem to be in agreement with the light-harvesting
range of sensitizers, although not following the UV–Vis absorption profiles previously
discussed. It is rational to ascribe this behavior to the different electron injection efficiency
of the two electronic transitions characterizing the "camel-back" absorption profiles of G3-5
sensitizers, which was suggested by our theoretical calculations. Notwithstanding the
broadest light-harvesting response exhibited by G3-based devices, their IPCE spectrum
showed the highest values (~75%) at wavelengths below 550 nm. By contrast, G4 and G5
maintained a much higher (~85%) IPCE value across the visible wavelengths ranging from
4
50 to 600 nm. Although G4 showed a slightly broader IPCE profile with respect to that of
G5, better conversion efficiencies below 500 nm were observed for G5, plausibly deriving
from its higher molar extinction coefficients (vide supra) with respect to G4 in this region.

ACCEPTED MANUSCRIPT
Figure 4. (A) J–V curves of DSSCs based on the three sensitizers G3, G4 and G5 under 1.0, 0.66
and 0.33 sun illumination. (B) Incident photon-to-current conversion efficiency (IPCE) behaviour
of the same cells.
To deepen the relationship between dye molecular structure and photovoltaic performance,
impedance spectroscopy (IS) measurements were performed on the CDCA-optimized
DSSCs at different bias potentials under 1.0 sun irradiation. In the framework of our
investigation, the IS measurements principally aim at associating the differences in devices
performance with the shift of the TiO conduction band (EC) or with the recombination rate,
2
both aspects having a strong influence on the net charge-transfer at the TiO /dye interface
2
during the DSSC work cycle [19]. IS spectra were analyzed through the well-known
equivalent circuit proposed by Bisquert [20]. Figure 5A and Figure 5B show the capacitance
(
C ) and recombination resistance (R ) plotted versus voltage drop (V ) at the sensitized
ꢀ ct corr
electrode. The Vcorr in the sensitized electrode is obtained after correcting the applied
potential (V ) for the drop of total series resistance V (contacts, counter electrode,
app
series
electrolyte diffusion) collected from impedance spectroscopy data as Vcorr = Vapp – Vseries
21]. The chemical capacitance (C ) represented in Figure 4A provides information about
[
ꢀ
the position of the conduction band [22]. From the analysis of C , we could
ꢀ

ACCEPTED MANUSCRIPT
straightforwardly identify the upward shift of the conduction band edge of titania in the case
of the DSSC fabricated with the G3 dye with respect to G4 and G5. On the other hand, a
–
lower ECB for G4- and G5-based devices should decrease the energy difference from the I
–
3
/
I redox potential, which would result in a lower open-circuit voltage (V ) with respect to
OC
G3. On the contrary, G5-based cells show the highest V , due to their higher charge-
OC
transfer resistance (R , Figure 5B).
ct
Figure 5. EIS of CDCA-optimized DSSCs at different bias potentials under 1.0 sun irradiation.
Capacitance (A) and recombination resistance (B) with respect to the Fermi level voltage (removing
the effect of series resistance, V ). Capacitance (C) and recombination resistance (D) replotted
corr
with respect to the equivalent common conduction band voltage (V ) using G5 as reference dye,
ecb
so that distance between Fermi level and conduction band is the same in all cases.

ACCEPTED MANUSCRIPT
To gain more insight into the recombination resistance behavior, the R was determined on
ct
the basis of a similar number of injected electrons, i.e. at the same energy gap between the
electron Fermi level and TiO conduction band. To this purpose, the voltage scale in Figure
2
5
C has been shifted, so that chemical capacitances coincide. In this graph, the voltage is
obtained as Vecb = Vcorr - ∆E , where “ecb” stands for “common equivalent conduction
C
band” and ∆E is the shift of the conduction band with respect to G5 taken as reference
C
sample [22b]. In Figure 5D, the recombination resistance is plotted using the same scale.
This procedure equalizes the conduction band of titania for all the cells and allows a better
comparison of the charge-transfer rate. Under the same equivalent conduction band level,
the R values of G4 and G5 are clearly confirmed as higher than those of G3 (Figure 5D),
ct
indicating that the parasitic charge recombination between injected electrons into TiO from
2
the excited sensitizer and the oxidized form of the redox shuttle occurs in a higher extent in
G3 with respect to G4 and G5. This results can be ascribed to the insulating effect of the n-
hexyl chain on the π-bridge of G4 and G5 dyes which suppresses the dark current.
Moreover, the steric hindrance of the G5 triarylamine-based donor group further hinders
charge recombination, explaining the higher VOC and the higher short-circuit current density
(
JSC) observed for the G5-based photovoltaic cell.

ACCEPTED MANUSCRIPT
2
Dye
CDCA
PCE (%)
FF
VOC (mV)
J (mA/cm )
Sun
SC
N719
G3
G3
G3
G3
G3
G3
G4
G4
G4
G4
G4
G4
G5
G5
G5
G5
G5
G5
0 mM
0 mM
8.67
4.14
7.44
8.64
8.94
9.20
7.85
5.11
6.84
9.49
9.57
9.83
9.52
5.49
7.68
9.77
9.90
10.26
9.45
0.66
0.69
0.74
0.72
0.73
0.74
0.71
0.66
0.68
0.68
0.70
0.73
0.70
0.67
0.68
0.68
0.69
0.72
0.70
751
666
661
663
660
661
663
693
702
704
702
701
708
714
722
721
720
715
735
17.51
9.00
15.21
18.11
12.25
6.21
16.67
11.16
14.24
19.76
12.85
6.34
19.23
11.54
15.53
19.80
13.15
6.58
1.0
1.0
1.0
20 mM
30 mM
30 mM
30 mM
40 mM
0 mM
1.0
0.66
0.33
1.0
1.0
1.0
5 mM
10 mM
10 mM
10 mM
15 mM
0 mM
1.0
0.66
0.33
1.0
1.0
1.0
5 mM
10 mM
10 mM
10 mM
15 mM
1.0
0.66
0.33
1.0
18.40
–
–
Table 2. Photovoltaic performance of DSSCs based on G3, G4 and G5 using the I /I redox
3
shuttle.
Conclusions
Taking into consideration that even the slightest structural modification conceived for an
organic sensitizer can exert a strong influence on the relevant DSSC performances, based on
the structural motif of a robust D–A–π–A organic dye (G3), new largely π-extended organic
sensitizers (G4 and G5) have been designed with the main aim of favourably modulating the
interactions between the dye/TiO layer and the employed redox shuttle. Complying with the
2
guidelines traditionally followed to retard disadvantageous charge recombination and to
inhibit detrimental interactions between dye molecules onto the titania surface, the
molecular tailoring of the G3 scaffold was performed by suitably positioning an alkyl chain
onto the π-spacer (G4 vs G3) and increasing the dimension of the electron-donating unit (G5

ACCEPTED MANUSCRIPT
vs G4). A detailed TD-DFT study has revealed that the main S →S electronic transitions of
0
1
the three sensitizers is endowed with only a weak charge-transfer character, being the
hole/electron densities mainly localized on the largely -extended bridge, and implying,
π
upon photo-excitation, a limited shift of the electron density towards the anchoring group.
Notwithstanding these unusual photo-excitation dynamics, the photovoltaic performances of
G3, G4 and G5 can be regarded as impressive, reaching, in the case of G5, power
conversion efficiencies up to 9.8%, ranking among the highest reported for metal-free
organic sensitizers employing iodine-based electrolytes.
Experimental Section
General remarks on dye synthesis and characterization: All reactants and raw materials
were purchased from standard commercial sources and used without any further purification.
All solvents used were carefully dried and freshly distilled according to standard laboratory
practice. All manipulations were carried out under inert dinitrogen atmosphere. The
synthesis of 4-bromo-7-(4-hexylthiophen-2-yl)benzo[2,1,3]thiadiazole (4) [8b], 2-((2-
ethylhexyloxy)phen-4-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane [11a] and 4-(4,4,5,5-
tetramethyl-1,3,2-dioxaborolan-2-yl)-N,N-bis((2-ethylhexyloxy)phen-4-yl)aniline [23] have
been carried out adapting literature procedures. Column chromatography were carried out on
a Biotage Isolera One instrument using K-Sil cartridges. The NMR spectra were recorded on
a Bruker Avance 700 MHz instrument. Elemental analyses were obtained on a EuroVector
CHNS EA3000 elemental analyser on the basis of three replicates. UV–Vis spectra were
recorded on a Jasco V-670 instrument. FT-IR measurements were carried out on a JASCO
FT/IR 4200 spectrophotometer. The cyclic voltammetry experiments were carried out using
a Metrohm Autolab PGSTAT 302-N potentiostat. In a typical experiment, the samples were
drop cast on a platinum working electrode from THF solutions of 1 mg/mL concentration.
Measurements have been performed at a 50 mV/s scan rate in acetonitrile solution

ACCEPTED MANUSCRIPT
containing tetrabutylammonium tetrafluoroborate (0.1 M) as supporting electrolyte at 25 °C.
+
Potentials were measured versus Ag/Ag as quasi-reference electrode. The potential of the
+
+
Ag/Ag electrode was calibrated against Fc/Fc after each experiment. After calibration
+
against Fc/Fc , the HOMO and LUMO energy levels were calculated according to the
following equations:
onset
onset
+
EHOMO (eV) = – [Eox
– E (Fc/Fc ) + 4.8]
1/2
+
ELUMO (eV) = – [Ered
– E (Fc/Fc ) + 4.8]
1/2
+
+
onset
ox
onset
where E (Fc/Fc ) is the half-wave potential of the Fc/Fc couple, while E
and Ered
1/2
represent the onset potentials of the first oxidation and reduction event, respectively. The
analyses of the ground-state structures were carried out by density functional theory (DFT)
calculations. The B3LYP function was used in conjunction with the 6-31G(d,p) basis set.
Time-dependent DFT (TD-DFT) was applied for the assessment of excited-state transition
energies. All theoretical calculations have been carried out using the Gaussian09 program
package on isolated molecules.
DSSC fabrication and photovoltaic measurement: Fluorine-doped tin oxide (FTO, 15
ꢁ
/sq, provided by Solaronix S.A.) glass plates were first cleaned in a detergent solution
using an ultrasonic bath for 15 min, and then rinsed with water and ethanol. Double-layer
electrodes (overall thickness 19 ꢀm) were prepared as follows: a layer of commercial
colloidal titania paste (Dyesol 18NR-T) was deposited onto the FTO glass and gradually
heated in an oven in air to obtain a ~14 ꢀm transparent nano-crystalline film; the
temperature gradient was programmed as follows: 170°C (40 min), 350°C (15 min) and
4
30°C (30 min). This procedure was repeated for the scattering layer (5 µm) constituted by
Solaronix D/SP colloidal paste. The double-layer electrode was eventually sintered at 450°C

ACCEPTED MANUSCRIPT
2
for 30 min. The thickness and the active area (0.16 cm ) of the sintered photo-anodes was
measured using a profilometer (Veeco Dektak 150 Surface Profiler). The dye loading was
performed by keeping the electrodes for 14 h and under dark in 0.2 mM THF solutions of
G3, G4 or G5 dyes containing a known amount of chenodeoxycholic acid (CDCA). The
reference photo-anodes were prepared analogously by dyeing the electrodes with 0.2 mM
solutions
of
bis(tetrabutylammonium)-cis-di(thiocyanato)-N,N′-bis(4-carboxylato-4′-
carboxylic acid-2,2′-bipyridine) ruthenium(II) (N719, purchased by Solaronix S.A.) in a
mixture of acetonitrile and tert-butyl alcohol (1:1 v/v). The counter-electrodes were prepared
by sputtering a 50 nm Pt layer on a hole-drilled cleaned FTO plate. In a typical device
construction procedure, the photo-anode and the counter-electrode were faced and
assembled using a suitably cut 50
µm thick Surlyn® hot-melt gasket for sealing. The iodine
redox electrolyte (0.1 M LiI, 0.05 M I , 0.6 M 1-methyl-3-propylimidazolium iodide, and
2
0
.5 M tert-butylpyridine in dry acetonitrile) was vacuum-injected into the space between the
electrodes through pre-drilled holes on the back of the counter electrode. The holes were
®
eventually sealed using Surlyn hot melt film and a cover glass. Photocurrent-voltage
measurements were performed using a Keithley unit (Model 2400 Source Meter). A
Newport AM 1.5 Solar Simulator (Model 91160A equipped with a 1000 W Xenon arc lamp)
2
serving as a light source. The light intensity was calibrated to 100 mW/cm using as
reference a Si solar cell. The incident photon-to-current conversion efficiency (IPCE) was
measured by the DC method using a computer-controlled xenon arc lamp (Newport, 140 W,
6
7005) coupled with a monochromator (Newport Cornerstore 260 Oriel 74125). The light
intensity was measured by a calibrated silicon UV-photodetector (Oriel 71675) and the short
circuit currents of the DSSCs were measured by using a dual channel optical power/energy
meter (Newport 2936-C). Single layer transparent photoelectrodes (5 ꢀm thick) sensitized
for 4 h in the best performing dye bath solutions were used for absorption spectra
measurement of the dyes on TiO surface. Electrochemical impedance spectroscopy (EIS)
2

ACCEPTED MANUSCRIPT
was performed by an AUTOLAB PGSTAT 302N (Eco Chemie B.V.) in a frequency range
between 100 kHz and 10 mHz. The impedance measurements were carried out at different
voltage biases under 1.0 sun illumination. The resulting impedance spectra were fitted by
using ZView (Scribner Associates) software.
Synthetic procedures: 4-Bromo-N,N-bis(4-iodophenyl)aniline (
1
).
2.25 g, 10.00 mmol) was added portionwise to a solution of 4-bromo-N,N-diphenylaniline
1.62 g, 5.00 mmol) in DMF (50 mL) kept at 0 °C. After the addition, the obtained reaction
N-iodo-succinimide
(
(
mixture was warmed to room temperature and allowed to react overnight before quenching
with water (50 mL). The compound was extracted with diethyl ether (3 × 50 mL) and the
combined organic phases were washed with water (50 mL) and dried over Na SO . After
2
4
solvent removal, the crude product was purified by column chromatography (SiO2,
petroleum ether 40-60 °C/CH Cl = 10/1 v/v) to afford (2.05 g, 71%) as a off-white solid.
7.56 (d, J = 8.8 Hz, 4H); 7.39 (d, J = 8.9 Hz, 2H); 6.96 (d, J
1
2
2
1
H NMR (700 MHz, CDCl3):
δ
13
1
=
8.9 Hz, 2H); 6.84 (d, J = 8.8 Hz, 4H) ppm. C{ H} NMR (176 MHz, CDCl ): δ 146.7,
3
1
3
45.9, 138.5, 132.6, 126.0, 125.9, 116.2, 86.5 ppm. Elem. Anal. calcd for C H BrI N: C
18
12
2
7.53, H 2.10, N 2.43; found: C 37.57, H 2.06, N 2.46.
4
2
-Bromo-N,N-bis(4'-(2-ethylhexyloxy)-1,1'-biphenyl-4-yl)aniline (
2
). Compound
1
(1.15 g,
.00 mmol), 2-((2-ethylhexyloxy)phen-4-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (1.46
g, 4.40 mmol) and Pd(PPh ) (0.12 g, 0.10 mmol) were put into a three-necked 50 mL round
3
4
bottom flask. Degassed THF (15 mL) and 1M NaHCO aqueous solution (10 mL) were
3
added to the flask and the obtained mixture was refluxed for 24 h. After cooling down the
reaction to room temperature, methylene chloride (50 mL) was added to the mixture and the
resulting organic phase was separated, washed with water (50 mL) and dried over Na SO .
2
4
After solvent removal, the crude product was purified by column chromatography (SiO2,

ACCEPTED MANUSCRIPT
petroleum ether 40-60 °C/CH Cl = 10/1 v/v) to obtain (1.04 g, 71%) as a white solid. H
1
2
2
2
NMR (700 MHz, CDCl3):
δ
7.51 (d, J = 8.5 Hz, 4H); 7.47 (d, J = 8.5 Hz, 4H); 7.37 (d, J =
8
3
.7 Hz, 2H); 7.16 (d, J = 8.5 Hz, 4H); 7.04 (d, J = 8.7 Hz, 2H); 6.98 (d, J = 8.5 Hz, 4H);
.92-3.87 (m, 4H); 1.79-1.73 (m, 2H); 1.56-1.32 (m, 16H); 0.98-0.91 (m, 12H) ppm.
13
1
C{ H} NMR (176 MHz, CDCl ):
δ 158.8, 146.9, 145.9, 135.9, 132.8, 132.2, 128.8, 127.7,
3
1
27.5, 125.2, 124.6, 114.9, 70.7, 39.4, 30.6, 29.1, 23.9, 23.1, 14.1, 11.1 ppm. Elem. Anal.
calcd for C H BrNO : C 75.39, H 7.43, N 1.91; found: C 75.47, H 7.36, N 1.96.
46
54
2
4
4
-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)-N,N-bis-(4'-(2-ethylhexyloxy)-1,1'-biphen-
-yl)aniline ( ). Compound (1.00 g, 1.36 mmol), bis(pinacolato)diboron (1.04 g, 4.08
3
2
mmol), PdCl (dppf) (0.05 g, 0.07 mmol) and potassium acetate (0.20 g, 2.04 mmol) were
2
put into a three-necked 50 mL round bottom flask. Degassed dioxane (25 mL) was added to
the flask and the obtained mixture was stirred at 80 °C for 24 h. After cooling down the
reaction to room temperature, methylene chloride (50 mL) was added to the mixture and the
resulting organic phase was separated, washed with water (50 mL) and dried over Na SO .
2
4
After solvent removal, the crude product was purified by column chromatography (SiO2,
1
petroleum ether 40-60 °C/CH Cl = 2/1 v/v) to obtain
3
(0.80 g, 74%) as a white solid. H
2
2
NMR (700 MHz, CDCl3):
δ
7.74 (d, J = 8.2 Hz, 2H); 7.54 (d, J = 8.9 Hz, 4H); 7.50 (d, J =
.9 Hz, 4H); 7.21 (d, J = 8.9 Hz, 4H); 7.16 (d, J = 8.2 Hz, 2H); 7.00 (d, J = 8.9 Hz, 4H);
.95-3.89 (m, 4H); 1.81-1.76 (m, 2H); 1.60-1.41 (m, 16H); 1.38 (s, 12H); 1.00-0.93 (m,
8
3
1
1
1
3
1
2H) ppm. C{ H} NMR (176 MHz, CDCl ):
δ
158.8, 150.4, 146.0, 136.0, 135.9, 132.9,
3
27.8, 127.5, 125.1, 122.0, 114.8, 83.6, 70.6, 39.4, 30.6, 29.1, 25.0, 23.9, 23.2, 14.1, 11.1
ppm. Anal. calcd for C H BNO : C 80.08, H 8.53, N 1.80; found: C 79.99, H 8.56, N 1.78.
52
66
4
4
-(4-Hexyl-thiophen-2-yl)-7-(thiophen-2-yl)benzo[2,1,3]thiadiazole (
5). Compound 4 (5.15
g, 13.50 mmol), thiophen-2-yl boronic acid (1.90 g, 14.85 mmol) and Pd(PPh ) (0.31 g,
3
4
0
.27 mmol) were put into a three-necked 250 mL round bottom flask. Degassed THF (50

ACCEPTED MANUSCRIPT
mL) and 1M NaHCO aqueous solution (40 mL) were added to the flask and the obtained
3
mixture was refluxed overnight. After the reaction was allowed to reach room temperature,
the compound was extracted with diethyl ether (50 mL) and the resulting organic phase was
washed with water (50 mL) and dried over Na SO . After solvent removal, the crude
2
4
product was purified by column chromatography (SiO , petroleum ether 40-60 °C/CH Cl =
2
2
2
1
5
3
7
7
3
1
/1 v/v) yielding 5 (4.05 g, 78%) as a red solid. H NMR (700 MHz, CDCl ): δ 8.12 (dd, J =
3
.6, 1.1 Hz, 1H); 8.00 (d, J = 1.4 Hz, 1H); 7.87 (d, J = 7.5 Hz, 1H); 7.85 (d, J = 7.5 Hz, 1H);
.46 ( d, J = 5.0 Hz, 1H); 7.22 (dd, J = 5.0, 3.6 Hz, 1H); 7.06 (d, J = 1.4 Hz, 1H); 2.71 (t, J =
.7 Hz, 2H); 1.75-1.70 (m, 2H); 1.45-1.39 (m, 2H); 1.37-1.33 (m, 4H); 0.92 (t, J = 7.0 Hz,
13
1
H) ppm. C{ H} NMR (176 MHz, CDCl ):
δ
152.7, 152.6, 144.4, 139.4, 138.9, 129.1,
3
28.0, 127.5, 126.8, 126.3, 125.8, 125.5, 121.6, 31.7, 30.7, 30.5, 29.1, 22.6, 14.1 ppm. Anal.
calcd for C H N S : C 62.46, H 5.24, N 7.28; found: C 62.49, H 5.26, N 7.24.
20
20
2 3
4
-(5-Bromo-4-hexylthiophen-2-yl)-7-(thiophen-2-yl)benzo[2,1,3]thiadiazole (
of -bromo-succinimide (1.78 g, 10.00 mmol) in DMF (60 mL) was added dropwise to a
solution of compound (3.85 g, 10.00 mmol) in DMF (90 mL) kept at 0 °C. After the
6). A solution
N
5
addition, the obtained reaction mixture was warmed to room temperature and allowed to
react for further 2 h before quenching with water (100 mL). The compound was extracted
with diethyl ether (3 × 100 mL) and the combined organic phases were washed with water
(
50 mL) and dried over Na SO . After solvent removal, the crude product was purified by
2 4
column chromatography (SiO , petroleum ether 40-60 °C/CH Cl = 5/1 v/v) to afford
6
2
2
2
1
(
3.48 g, 75%) as a red solid. H NMR (700 MHz, CDCl ):
δ
8.13 (dd, J = 3.6, 1.1 Hz, 1H);
.86 (d, J = 7.5 Hz, 1H); 7.79 (s, 1H); 7.78 (d, J = 7.5 Hz, 1H); 7.47 (dd, J = 5.0, 1.1 Hz,
H); 7.23 (dd, J = 5.0, 3.6 Hz, 1H); 2.70 (t, J = 7.7 Hz, 2H); 1.74-1.70 (m, 2H); 1.46-1.39
3
7
1
13
1
(
m, 2H); 1.38-1.32 (m, 4H); 0.93 (t, J = 7.0 Hz, 3H) ppm. C{ H} NMR (176 MHz,
CDCl3): 152.6, 152.3, 143.0, 139.3, 138.6, 128.1, 128.0, 127.6, 126.9, 126.1, 125.7, 125.3,
δ

ACCEPTED MANUSCRIPT
1
25.0, 111.5, 31.6, 29.7, 29.6, 29.0, 22.6, 14.1 ppm. Anal. calcd for C H BrN S : C 51.83,
20
19
2 3
H 4.13, N 6.04; found: C 51.79, H 4.16, N 6.04.
4
-(3-Hexyl-5-(7-(thiophen-2-yl)benzo[2,1,3]thiadiazol-4-yl)thiophen-2-yl)benzaldehyde (
7
).
Compound
6 (3.20 g, 6.90 mmol), 4-formylphenyl boronic acid (1.14 g, 7.59 mmol) and
Pd(PPh ) (0.16 g, 0.14 mmol) were put into a three-necked 100 mL round bottom flask.
3
4
Degassed THF (30 mL) and 1M NaHCO aqueous solution (20 mL) were added to the flask
3
and the resulting mixture was refluxed for 24 h. After cooling down the reaction to room
temperature, methylene chloride (50 mL) was added to the mixture and the resulting organic
phase was separated, washed with water (50 mL) and dried over Na SO . After solvent
2
4
removal, the crude product was purified by column chromatography (SiO , petroleum ether
2
1
4
0-60 °C/CH Cl = 3/1 v/v) to obtain
7
(2.53 g, 75%) as a red solid. H NMR (700 MHz,
2
2
CDCl3):
δ 10.08 (s, 1H); 8.15 (dd, 3.6, 1.1 Hz, 1H); 8.04 (s, 1H); 7.97 (d, J = 8.3 Hz, 2H);
7
.89 (s, 2H); 7.72 (d, J = 8.3 Hz, 2H); 7.48 (dd, J = 5.0, 1.1 Hz, 1H); 7.23 (dd, J = 5.0, 3.6
Hz, 1H); 2.80 (t, J = 7.7 Hz, 2H); 1.74-1.69 (m, 2H); 1.45-1.38 (m, 2H); 1.38-1.32 (m, 4H);
13
1
0
.93 (t, J = 7.0 Hz, 3H) ppm. C{ H} NMR (176 MHz, CDCl ): δ 191.6, 152.7, 152.6,
3
1
41.3, 140.8, 139.3, 138.7, 135.1, 130.5, 130.0, 129.5, 128.0, 127.6, 126.9, 125.7, 125.6,
25.5, 31.6, 30.9, 29.2, 29.1, 22.6, 14.0 ppm. Anal. calcd for C H N OS : C 66.36, H 4.95,
1
27
24
2
3
N 5.73; found: C 66.29, H 4.96, N 5.74.
4
-(5-(7-(5-Bromothiophen-2-yl)benzo[2,1,3]thiadiazol-4-yl)-3-hexylthiophen-2-
yl)benzaldehyde ( ). A solution of -bromo-succinimide (0.91 g, 5.12 mmol) in DMF (30
mL) was added dropwise to a solution of compound (2.50 g, 5.12 mmol) in DMF (45 mL)
8
N
7
kept at 0 °C. After the addition, the obtained reaction mixture was warmed to room
temperature and allowed to react for further 2 h before quenching with water (50 mL). The
compound was extracted with diethyl ether (3 × 50 mL) and the combined organic phases
were washed with water (50 mL) and dried over Na SO . After solvent removal, the crude
2
4
product was purified by column chromatography (SiO , petroleum ether 40-60 °C/CH Cl =
2
2
2

ACCEPTED MANUSCRIPT
1
3
1
/1 v/v) to afford
8
(2.35 g, 81%) as a red solid. H NMR (700 MHz, CDCl ):
δ
10.08 (s,
3
H); 8.05 (s, 1H); 7.97 (d, J = 7.9 Hz, 2H); 7.88 (d, J = 7.4 Hz, 1H); 7.83-7.80 (m, 2H); 7.72
(
d, J = 7.9 Hz, 2H); 7.17 (d, J = 4.0 Hz, 1H); 2.79 (t, J = 7.7 Hz, 2H); 1.77-1.71 (m, 2H);
1
3
1
1
.43-1.36 (m, 2H); 1.34-1.28 (m, 4H); 0.93 (t, J = 7.0 Hz, 3H) ppm. C{ H} NMR (176
191.6, 152.4, 152.4, 141.4, 140.7, 140.6, 138.5, 138.1, 135.1, 130.7, 130.6,
30.1, 129.5, 127.3, 125.9, 125.4, 125.2, 125.1, 114.7, 31.6, 30.9, 29.7, 29.2, 22.6, 14.0
ppm. Anal. calcd for C H BrN OS : C 57.14, H 4.08, N 4.94; found: C 57.19, H 4.06, N
MHz, CDCl3):
δ
1
27
23
2
3
5
.00.
-(5-(7-(5-(4-(N,N-Bis-(4-(2-ethylhexyloxy)-phenyl)-amino-phenyl)-thiophen-2-
).
Tetramethyl-1,3,2-dioxaborolan-2-yl)-N,N-bis((2-ethylhexyloxy)phen-4-yl)aniline (0.61 g,
.97 mmol), compound (0.50 g, 0.88 mmol) and Pd(PPh ) (0.05 g, 0.04 mmol) were put
4
yl)benzo[2,1,3]thiadiazol-4-yl)-3-hexylthiophen-2-yl)benzaldehyde
(
9
4-(4,4,5,5-
0
8
3
4
into a three-necked 50 mL round bottom flask. Degassed THF (15 mL) and 1M NaHCO3
aqueous solution (10 mL) were added to the flask and the resulting mixture was refluxed for
2
4 h. After cooling down the reaction to room temperature, methylene chloride (50 mL) was
added to the mixture and the resulting organic phase was separated, washed with water (50
mL) and dried over Na SO . After solvent removal, the crude product was purified by
2
4
1
column chromatography (SiO , CH Cl ) to obtain
9
(0.74 g, 85%) as a purple solid. H
10.10 (s, 1H); 8.16 (d, J = 3.8 Hz, 1H); 8.05 (s, 1H); 7.99 (d, J =
.5 Hz, 2H); 7.93-7.88 (m, 2H); 7.74 (d, J = 8.5 Hz, 2H); 7.52 (d, J = 8.5 Hz, 2H); 7.31 (d, J
2
2
2
NMR (700 MHz, CDCl3):
δ
8
=
3.8 Hz, 1H); 7.11 (d, J = 8.7 Hz, 4H); 6.97 (d, J = 8.5 Hz, 2H); 6.88 (d, J = 8.7 Hz, 4H);
3
0
1
1
.87-3.84 (m, 4H); 2.79 (t, J = 7.7 Hz, 2H); 1.79-1.72 (m, 4H); 1.57-1.30 (m, 24H); 0.99-
13
1
.89 (m, 15H) ppm. C{ H} NMR (176 MHz, CDCl ):
δ 191.9, 156.0, 152.6, 152.5, 148.7,
3
46.3, 141.3, 140.9, 140.2, 138.9, 137.7, 137.0, 135.0, 130.3, 130.1, 129.5, 129.1, 126.9,
26.5, 126.4, 125.8, 125.7, 124.9, 122.7, 120.1, 115.3, 70.7, 39.4, 31.6, 30.9, 30.6, 29.7,

ACCEPTED MANUSCRIPT
2
7
4
9.3 29.2, 29.1, 23.9, 23.2, 22.6, 14.2, 14.1, 11.1 ppm. Anal. calcd for C H N O S : C
61
69
3
3 3
4.12, H 7.04, N 4.25; found: C 74.19, H 7.06, N 4.28.
-(5-(7-(5-(4-(N,N-Bis-(4'-(2-ethylhexyloxy)-1,1'-biphen-4-yl)-amino-phenyl)-thiophen-2-
(0.76
(0.50 g, 0.88 mmol) and Pd(PPh ) (0.05 g, 0.04 mmol) were
yl)benzo[2,1,3]thiadiazol-4-yl)-3-hexylthiophen-2-yl)benzaldehyde (10). Compound
g, 0.97 mmol), compound
3
8
3
4
put into a three-necked 50 mL round bottom flask. Degassed THF (15 mL) and 1M NaHCO3
aqueous solution (10 mL) were added to the flask and the resulting mixture was refluxed for
2
4 h. After cooling down the reaction to room temperature, methylene chloride (50 mL) was
added to the mixture and the resulting organic phase was separated, washed with water (50
mL) and dried over Na SO . After solvent removal, the crude product was purified by
2
4
1
column chromatography (SiO , CH Cl ) yielding 10 (0.87 g, 87%) as a purple solid. H
2
2
2
NMR (700 MHz, CDCl3):
δ
10.10 (s, 1H); 8.18 (d, J = 3.8 Hz, 1H); 8.07 (s, 1H); 7.99 (d, J =
8
.2 Hz, 2H); 7.93-7.91 (m, 2H); 7.74 (d, J = 8.2 Hz, 2H); 7.64 (d, J = 8.4 Hz, 2H); 7.55 (d, J
=
8.3 Hz, 4H); 7.52 (d, J = 8.3 Hz, 4H); 7.39 (d, J = 3.5 Hz, 1H); 7.25 (d, J = 8.3 Hz, 4H);
7
2
.21 (d, J = 8.4 Hz, 2H); 7.00 (d, J = 8.3 Hz, 4H); 3.94-3.88 (m, 4H); 2.82 (t, J = 7.7 Hz,
13
1
H); 1.81-1.73 (m, 4H); 1.57-1.30 (m, 24H); 0.99-0.89 (m, 15H) ppm. C{ H} NMR (176
191.8, 158.8, 152.6, 147.5, 145.9, 141.4, 140.8, 138.9, 137.8, 137.7, 135.8,
35.0, 132.8, 130.4, 130.1, 129.6, 129.5, 129.0, 128.1, 128.0, 127.7, 127.5, 126.7, 126.2,
MHz, CDCl3):
δ
1
1
2
3
3
25.7, 125.1, 125.0, 124.8, 123.6, 123.3, 114.8, 70.6, 39.4, 31.7, 30.9, 30.5, 29.3, 29.2, 29.1,
3.8, 23.2, 22.6, 14.2, 14.1, 11.1 ppm. Anal. calcd for C H N O S : C 76.87, H 6.80, N
73
77
3
3 3
.68; found: C 76.79, H 8.76, N 3.65.
-(4-(5-(7-(5-(4-(N,N-Bis-(4-(2-ethylhexyloxy)-phenyl)-amino-phenyl)-thiophen-2-
yl)benzo[2,1,3]thiadiazol-4-yl)-3-hexylthiophen-2-yl)phenyl)-2-cyanoacrylic acid
(G4).
Compound (0.49 g, 0.50 mmol), cyanoacetic acid (0.09 g, 1.00 mmol) and piperidine (0.04
9
g, 0.50 mmol) were put into a three-necked 100 mL round bottom flask. Degassed
chloroform (50 mL) was introduced into the flask and the resulting mixture was refluxed for

ACCEPTED MANUSCRIPT
4
h. After cooling down the reaction to room temperature, the solvent was evaporated and
the residue was purified by column chromatography (SiO , CH Cl /MeOH = 20/1 v/v)
2
2
2
1
yielding G4 (0.25 g, 47%) as a purple solid. H NMR (700 MHz, DMSO-d ):
δ
8.29 (s, 1H),
6
8
1
5
1
.18-8.08 (m, 6H), 7.73 (d, J = 8.4 Hz, 2H), 7.57 (d, J = 8.4 Hz, 2H), 7.48 (d, J = 3.7 Hz,
H), 7.05 (d, J = 8.6 Hz, 4H), 6.94 (d, J = 8.6 Hz, 4H), 6.81 (d, J = 8.1 Hz, 2H), 3.84 (d, J =
.8 Hz, 4H); 2.74 (t, J = 6.8 Hz, 2H); 1.71-1.65 (m, 4H); 1.49-1.22 (m, 24H); 0.93-0.82 (m,
5H) ppm. Anal. calcd for C H N O S : C 72.83, H 6.68, N 5.31; found: C 72.77, H 6.65,
64
70
4
4 3
N 5.28. IR (KBr): νmax 3445, 3041, 2954, 2924, 2854, 2223, 1697, 1588, 1504, 1236, 829
–
1
cm .
3
2
-(4-(5-(7-(5-(4-(N,N-Bis-(4'-(2-ethylhexyloxy)-1,1'-biphen-4-yl)-amino-phenyl)-thiophen-
-yl)benzo[2,1,3]thiadiazol-4-yl)-3-hexylthiophen-2-yl)phenyl)-2-cyanoacrylic acid (G5).
Compound 10 (0.57 g, 0.50 mmol), cyanoacetic acid (0.09 g, 1.00 mmol) and piperidine
0.04 g, 0.50 mmol) were put into a three-necked 100 mL round bottom flask. Degassed
chloroform (50 mL) was introduced into the flask and the resulting mixture was refluxed for
h. After cooling down the reaction to room temperature, the solvent was evaporated and
the residue was purified by column chromatography (SiO , CH Cl /MeOH = 20/1 v/v) to
(
4
2
2
2
1
obtain G5 (0.27 g, 45%) as a purple solid. H NMR (700 MHz, DMSO-d ):
δ
8.22-8.12 (m,
H), 8.10-8.03 (m, 3H), 7.75-7.67 (m, 5H), 7.64-7.57 (m, 7H), 7.16 (d, J = 7.8 Hz, 4H), 7.12
d, J = 8.1 Hz, 2H), 7.01 (d, J = 7.8 Hz, 4H), 3.89 (d, J = 5.9 Hz, 4H); 2.77 (t, J = 6.9 Hz,
H); 1.73-1.66 (m, 4H); 1.48-1.23 (m, 24H); 0.92-0.81 (m, 15H) ppm. Anal. calcd for
C H N O S : C 75.59, H 6.51, N 4.64; found: C 75.59, H 6.56, N 4.65. IR (KBr): νmax
6
5
(
2
76
78
4
4 3
–1
3
435, 3032, 2955, 2923, 2854, 2217, 1597, 1491, 1244, 820 cm .
Acknowledgements