Photophysical characteristics and photosensitizing abilities of thieno[3,2-b]thiophene-Based photosensitizers for photovoltaic and photocatalytic applications

Article abstract of DOI:10.1016/j.jphotochem.2020.112979

Six donor?π?acceptor (D–π–A) organic dyes based on thieno[3,2-b]thiophene π-linker comprising different alkyl side chains were synthesized and characterized. Their photophysical and electrochemical properties were studied and their corresponding performances as photosensitizers (PSs) in dye-sensitized solar cells (DSSCs) and photocatalytic hydrogen production systems were also investigated. A power conversion efficiency (PCE) of 5.25 % in DSSCs and a decent H₂ turnover number (TON) of 5170 (48 h) in platinized TiO₂ hydrogen production system were demonstrated. The results indicated that alkyl chain engineering is one of the crucial steps in designing a superior photosensitizer.

Full text of DOI:10.1016/j.jphotochem.2020.112979

Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
Contents lists available at ScienceDirect
Journal of Photochemistry & Photobiology, A: Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Photophysical characteristics and photosensitizing abilities of thieno[3,2-b]
thiophene-Based photosensitizers for photovoltaic and
photocatalytic applications
a
,
b
,
c
,
1
,
2
c
,
1
a
,
b
,
c
a,
b
c
Po-Yu Ho
, Yi Wang , Sze-Chun Yiu
, Yan-Yi Kwok , Chi-Ho Siu ,
a
,
b
,
c,
e
d,e,
Cheuk-Lam Ho
*, Lawrence Tien Lin Lee , Tao Chen
**
a
Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, PR China
PolyU Shenzhen Research Institute, Shenzhen, PR China
b
c
Department of Chemistry and Institute of Advanced Materials, Hong Kong Baptist University, Waterloo Road, Kowloon Tong, Hong Kong, PR China
Key Laboratory of Materials for Energy Conversion, Chinese Academy of Sciences, Department of Materials Science and Engineering, University of Science and
d
Technology of China, No. 96 Jinzhai Road, Hefei, Anhui Province, 230026, PR China
e
Department of Physics, The Chinese University of Hong Kong, Shatin, Hong Kong, PR China
A R T I C L E I N F O
A B S T R A C T
Six donorꢀ ꢀ acceptor (D–
Keywords:
π
π
–A) organic dyes based on thieno[3,2-b]thiophene
side chains were synthesized and characterized. Their photophysical and electrochemical properties were studied
and their corresponding performances as photosensitizers (PSs) in dye-sensitized solar cells (DSSCs) and pho-
tocatalytic hydrogen production systems were also investigated. A power conversion efficiency (PCE) of 5.25 %
π
-linker comprising different alkyl
Thieno[3,2-b]thiophene
Alkyl chain effects
Dye-sensitized solar cells
Photocatalytic hydrogen production
Photosensitizers
2 2
in DSSCs and a decent H turnover number (TON) of 5170 (48 h) in platinized TiO hydrogen production system
were demonstrated. The results indicated that alkyl chain engineering is one of the crucial steps in designing a
superior photosensitizer.
′
1
. Introduction
reported
D– –A–
a
′
triazatruxene-based donorꢀ πꢀ acceptorꢀ πꢀ acceptor
(
π
π
–A ) organic dye, which was claimed to overcome the hurdle of
electron-injection energy losses and achieved a champion power con-
version efficiency (PCE) of 13.6 % in a single dye-sensitized DSSC device
at that time [14].
Environmental problems such as global warming and energy crisis
have led to the urgency of utilization of renewable energy resources in
the past two decades [1]. Solar power is a promising clean energy source
for the community because of its tremendous amount of supply from the
sun [2]. Other than inorganic solar cells, dye-sensitized solar cell (DSSC)
has been regarded as another potential candidate to be adopted in large
scale application. In DSSCs, apart from the electrolytes and nano-
structured titania film, the role of photosensitizer (PS) is utterly signif-
icant to the device performance [3–5]. In recent years, organic dyes with
In addition to DSSCs, photocatalytic water splitting [15], which
converts water molecules (feedstock) into hydrogen (fuel) and oxygen
through a light-driven catalytic process, has been regarded as another
solar energy research direction, because nature provides us with abun-
dant amount of feedstock and the fuel is obtained without the discharge
of carbon-containing species during the energy conversion. In principle,
water splitting reaction is equivalent to the sum of two redox
half-reactions [16]. For the cathodic side, a workable system comprises
an PS for exciton formation, a pathway for charge separation, a catalyst
for hydrogen generation, aqueous protons and a sacrificial electron
high molar extinction coefficients ( ) have attracted attention since they
ε
are low-cost in production and flexible in molecular modifications.
Stable organic PSs are able to compete with polypyridyl ruthenium
complexes in terms of the device performance [6–13]. In 2019, Sun et al.
*
Corresponding author at: Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, PR China.
** Corresponding author at: Key Laboratory of Materials for Energy Conversion, Chinese Academy of Sciences, Department of Materials Science and Engineering,
University of Science and Technology of China, No. 96 Jinzhai Road, Hefei, Anhui Province, 230026, PR China.
E-mail addresses: cheuk-lam.ho@polyu.edu.hk (C.-L. Ho), tchenmse@ustc.edu.cn (T. Chen).
1
Both authors contributed equally to this work.
2
Present address: Ming Wai Lau Centre for Reparative Medicine, Karolinska Institutet, Hong Kong, (P.R. China).
https://doi.org/10.1016/j.jphotochem.2020.112979
Received 16 July 2020; Received in revised form 8 October 2020; Accepted 10 October 2020
Available online 20 October 2020
1
010-6030/© 2020 Elsevier B.V. All rights reserved.

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
donor (SED) for providing electrons [17]. Similar to the case in DSSCs,
PS also plays a pivotal role in the cathodic reaction and affects the ac-
tivity, stability and production cost of the photocatalytic systems.
Therefore, the performance of a wide variety of PSs has been investi-
gated in the past decade [18–24].
acid. All of the corresponding organic precursors were also character-
ized by 1H and C NMR spectroscopy. The singlet peak located at
around δ 10.0 ppm refers to the proton of aldehyde in the corresponding
precursors.
13
Herein, six new organic dyes based on thieno[3,2-b]thiophene
2
.2. Photophysical properties
π
-linker are reported (Fig. 1). These six compounds are composed of
cyanoacrylic acid which acts as an anchoring group attached to TiO as
well as an electron-withdrawing unit. For the -linker, thieno[3,2-b]
2
2 2
The UV/Vis absorption spectra of all the organic dyes in CH Cl so-
π
lution are shown in Fig. 2 and the corresponding data are summarized in
Table 1. All of the dyes exhibit two major distinct broad absorption
thiophene and the congener with one more 3-hexylthiophene group are
selected due to their strong intramolecular charge transfer (ICT) prop-
erty and intense absorption in visible light region as reported in the
literature [25–27]. 9-Substituted carbazoles with three different alkyl
chains (i.e., octyl, ethylhexyl and hexyl) are chosen to act as the donor
moiety because of the appropriate HOMO energy level and the ease of
alkyl chain alteration [28]. The synthetic procedures are simple and all
the products can be purified by column chromatography. The photo-
physical and electrochemical properties of the organic PSs were studied
and their corresponding performances in photocatalytic hydrogen pro-
duction as well as DSSCs were also investigated.
bands in the range of 300 ꢀ 600 nm. The
relatively high as compared to those of standard ruthenium dyes due to
the rigid and planar thieno[3,2-b]thiophene -linker [30]. The absorp-
tion bands at short wavelengths (centered at 301 ꢀ 352 nm) are ascribed
to the * transitions of chromophores. The low-energy broad ab-
ε
of these organic dyes are
π
π→π
sorption bands (centered at 459 ꢀ 502 nm) are attributed to the intra-
molecular charge transfer (ICT) from the donor to acceptor, hence
providing an efficient charge separation at the excited state. Dyes 8aꢀ 8c
show a significant red shift in the ICT transition as compared to 5aꢀ 5c,
which implies that embedding one more thiophene ring can increase the
light-harvesting ability efficiently. In addition, the
ε
of these dyes are in
was achieved
2
. Results and discussion
the order of 5a < 5b < 5c and 8a < 8b < 8c. The highest
ε
4
1
1
by 8c with the value of 5.81 × 10 M- cm- . No significant influence is
observed on the absorption wavelength regarding the nature of alkyl
chains. Although the absorption maxima of the ICT bands of 5aꢀ 5c
vary, the onsets of absorption peaks are almost the same.
2
.1. Synthesis and characterization
The structures of new organic photosensitizers are shown in Fig. 1.
The corresponding synthetic routes are depicted in Scheme 1. The key
intermediate 2,5-dibromothieno[3,2-b]thiophene 2 was prepared ac-
cording to the published procedures [29]. Compound 2 was then reacted
with n-butyllithium and dimethylformamide according to the general
Bouveault aldehyde synthetic procedure to get the compound 3 with
around 50 % yield. Palladium-catalyzed Suzuki coupling reaction was
then performed to react aryl bromide 3 with different alkylated
carbazole-based arylboronic acids to prepare organic dye precursors
2
.3. Electrochemical properties
Besides light harnessing ability, suitable energy-offsets of the dye
2
molecules between the nanostructured TiO , redox electrolytes and SED
are necessary for the high performance DSSC and photocatalytic system.
If the lowest unoccupied molecular orbital (LUMO) of the photosensi-
tizer is more positive than that of the conduction band edge of TiO
while the corresponding highest occupied molecular orbital (HOMO) is
2
4
aꢀ 4c. On the other hand, compound 6 was synthesized by Suzuki
coupling reaction using compound 2, followed by Vilsmeier-Haack re-
action. The organic dye precursors 7aꢀ 7c were obtained by Suzuki
coupling reactions with compound 6 utilizing corresponding arylbor-
onic acids. Finally, all the organic dyes 5aꢀ 5c and 8aꢀ 8c were prepared
by Knoevenagel condensation using cyanoacetic acid from the respec-
tive aldehyde precursors. All the target organic PSs were soluble in
common organic solvents, such as chloroform and tetrahydrofuran
ꢀ
ꢀ
more negative than the redox potential of redox couples (i.e., I /I ) and
3
SED, the electron injection and charge separation would be efficient at
the interface [31] in DSSCs and photocatalytic hydrogen production
system, respectively. To investigate such electron and charge transfer,
cyclic voltammetry (CV) was performed in PS solution using glassy
carbon electrode as the working electrode, platinum wire as the counter
+
electrode and Ag/Ag as the reference electrode in dichloromethane
1
13
(
(
THF) and were characterized by H and C NMR spectroscopies
containing 0.1 M tetrabutylammonium hexafluorophosphate at a scan
Figures S1ꢀ S12) and Matrix-assisted Laser Desorption Ionization
ꢀ 1
rate of 100 mV s . Under these conditions, the reversible oxidation
Time-of-flight (MALDI-TOF) mass spectrometry. In particular, the
downfield singlet peak located at around δ 8.5 ppm is a characteristic
peak of the proton on the carbon-carbon double bond of cyanoacrylic
+
potential (E1/2) of ferrocene was 0.25 V versus Ag/Ag . The corre-
sponding behaviors of the PSs (irreversible onset oxidation potential) in
CV are recorded in Figure S13 and tabulated in Table 2.
Fig. 1. The chemical structures of organic photosensitizers 5aꢀ 5c and 8aꢀ 8c.
2

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
Scheme 1. Synthetic routes of 5aꢀ 5c and 8aꢀ 8c. Ar-B(OH)
2
= 9-octylcarbazolylboronic acid or 9-(2-ethylhexyl)carbazolylboronic acid or 9-hexylcarbazolylboronic
acid; Ar’-B(OH) = (4-hexylthiophen-2-yl)boronic acid.
2
Table 2
Electrochemical data and energy levels of 5aꢀ 5c and 8aꢀ 8c.
onset
a
b
c
d
Dye
E
/V
HOMO /eV
E
g
/eV
E
ox* /V
LUMO /eV
Ox
5
5
5
8
8
8
a
b
c
a
b
c
0.64
0.60
0.59
0.49
0.47
0.50
ꢀ 5.19
ꢀ 5.15
ꢀ 5.14
ꢀ 5.04
ꢀ 5.02
ꢀ 5.05
2.32
2.35
2.31
2.12
2.13
2.12
ꢀ 1.68
ꢀ 1.75
ꢀ 1.72
ꢀ 1.63
ꢀ 1.66
ꢀ 1.62
ꢀ 2.87
ꢀ 2.80
ꢀ 2.83
ꢀ 2.92
ꢀ 2.89
ꢀ 2.93
a
b
c
onset
Calculated from ꢀ (E
+ 4.55).
Ox
g
Energy bandgap (E ) was determined from the onset of absorption spectrum.
onset
Eox* = E
– E
g
.
Ox
d
LUMO = HOMO + E
g
.
which ensures an efficient dye regeneration [33,34]. The first irrevers-
ible oxidation wave observed was ascribed to the removal of an electron
from the carbazole donor [35]. The embedded electron-rich 3-hexylthio-
phene in 8aꢀ 8c dramatically decreases the oxidation potential as
Fig. 2. UV/Vis absorption spectra of 5aꢀ 5c and 8aꢀ 8c in CH
2 2
Cl at 293 K.
compared to 5aꢀ 5c, due to the elongation of
π-conjugated length in the
former, hence rising the HOMO levels as well as reducing the
HOMO-LUMO gap.
Table 1
UV/Vis absorption data of 5aꢀ 5c and 8aꢀ 8c in CH
2 2
Cl at 293 K.
4
ꢀ 1
ꢀ 1
cm )
Dye
λ
max /nm (
ε
/10
M
λ
onset /nm
2
.4. Computational studies
5
5
5
8
8
8
a
b
c
a
b
c
302 (1.81)
301 (2.46)
302 (2.36)
309 (2.70)
309 (2.36)
309 (3.00)
352 (1.00)
349 (1.07)
351 (1.35)
472 (4.58)
459 (5.27)
470 (5.37)
500 (4.25)
499 (4.32)
502 (5.81)
534
528
537
585
582
585
To explain the structural properties of the dyes and understand the
possibility of charge transfer from donor to acceptor on electronic
excitation, the optimized structures of six dyes were calculated
geometrically using the density functional theory (DFT) with Gaussian
0
9 program. The optimized structures and electron densities of HOMOs
and LUMOs of the dyes are shown in Supporting information. From
Figures S14ꢀ S19, the electron density of the HOMO is mainly distrib-
uted along the carbazole donor to adjacent thieno[3,2-b]thiophene for
According to the results, the conduction band edge of TiO
2
(ꢀ 4.4 eV)
is more negative than LUMO of all our organic dyes (ranging from ꢀ 2.93
to ꢀ 2.80 eV), thereby facilitating an efficient electron injection [32]. On
the other hand, the HOMO energy levels of our dyes (ranging from
5
aꢀ 5c and to its adjoining hexylthiophene unit for 8aꢀ 8c, while the
ꢀ
ꢀ
electron density of the LUMO is mainly localized on the cyanoacrylic
acid group and adjacent functionalized thiophenes. This phenomenon
suggested that a well-inductive electron tendency from carbazole donor
ꢀ
5.19 to ꢀ 5.02 eV) are more negative than the redox potentials of I /I
3
electrolyte (ꢀ 4.45 eV) and ascorbic acid (SED, ꢀ 4.65 eV, pH ~ 4),
3

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
to the cyanoacrylic acceptor. It is clear that the HOMO ꢀ LUMO exci-
tation induced by light irradiation in our dyes could move the electron
distribution from the carbazole moiety to the cyanoacrylic moiety
effectively, and the photoinduced electrons can be efficiently transferred
from the dye to the TiO
2
surface in charge separation. In this system, the
-conjugation through the skeleton
planar structure can enhance the
π
backbone, increasing the degree of electronic resonance between donor
and acceptor moieties in the molecules and facilitating the electron
transfer from donor to cyanoacrylic acceptor. However, the planarity
increases the stacking of the dye molecules, resulting in more dye ag-
gregation and electron recombination. From the contour plots
(
Figures S14ꢀ S19), it is clearly shown that all the alkylated chains are
free of intramolecular steric hinderance, flexible and able to twirl to-
wards different directions. This could effectively suppress the intermo-
lecular stacking, therefore the geometric structure might effectively
retard the charge recombination and enhance the open-circuit photo-
voltage (Voc) in DSSC. In addition, the calculated HOMO and LUMO
from time-dependent density functional theory (TD-DFT) further
confirm that the differences in alkyl chains do not have significant in-
fluence on the HOMO and LUMO energy levels. The embedded thio-
phene unit in 8aꢀ 8c can elevate the HOMO energy level, resulting in the
enhancement of light absorption property for the corresponding com-
pounds by reducing the energy gap between HOMO/LUMO transition.
Fig. 3. Photocurrent density ꢀ voltage (Jꢀ V) plots obtained with 5aꢀ 5c,
8
aꢀ 8c and N719.
2
.5. Photovoltaic performance
DSSCs with an effective area of 0.20 cm2 were fabricated by
adsorbing our new organic photosensitizers onto nanocrystalline
anatase TiO as the photoanode and using the liquid electrolyte con-
sisting of 0.6 M 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.1
M LiI, 0.05 M I in a mixture of acetonitrile and 4-tert-butylpyridine
volume ratio, 1:1). All the DSSCs were tested under standard conditions
2
2
(
ꢀ
2
of AM 1.5 G, 100 mW cm . The basic device performance parameters,
such as short-circuit current density (Jsc), open-circuit voltage (Voc), fill
factor (FF) and power conversion efficiencies (PCEs), are collected in
Table 3. Figs. 3 and 4 demonstrate the photocurrent density ꢀ voltage
curves (J–V curves) and incident-photon-to-current conversion effi-
ciency (IPCE) spectra, respectively.
The highest PCE of 5.25 % (Jsc =12.32 mA cm 2, Voc = 0.680 V and
FF = 0.626) was achieved by 8a, attaining 66 % of the PCE of standard
ruthenium dye N719-based cell (PCE = 7.91 %), fabricated and
measured under the same condition. PCEs of DSSCs based on 5aꢀ 5c, 8b
and 8c are 3.95 %, 3.49 %, 2.91 %, 5.01 % and 4.58 %, respectively. The
IPCE values of 5aꢀ 5c and 8aꢀ 8c exceed 50 % in the range of 350 ꢀ 500
nm and 400 ꢀ 550 nm, respectively. Dyes 5aꢀ 5c, with a shorter
ꢀ
Fig. 4. Incident photon-to-current efficiency (IPCE) curves obtained with
aꢀ 5c, 8aꢀ 8c and N719.
5
which their hydrophobic property and steric hindrance may block the
approach of the hydrophilic electrolytes (i.e., iodide/triiodide ions) to
2
the hydrophilic surface of TiO , thereby inducing the charge recombi-
nation suppression and resulting in larger Voc values of 8aꢀ 8c as
compared to 5aꢀ 5c [36,37]. This alkyl chain effect is also profound for
the carbazole ring in which the Voc rises from 0.652 V to 0.672 V and
from 0.672 V to 0.692 V as the alkyl chain on carbazole changed from
hexyl chain to octyl chain (5c vs 5a) or 2-ethylhexyl chain (8c vs 8b),
respectively.
π
-conjugation length, have lower PCEs than those of 8aꢀ 8c due to the
narrower IPCE spectral response, which is also reflected from the lower
J
sc values of 5aꢀ 5c. The almost double Jsc values of dyes 8aꢀ 8c as
compared to those of 5aꢀ 5c can be ascribed to the following factors.
Firstly, the red-shifted ICT absorption bands of 8aꢀ 8c lead to a broader
harnessing of sunlight. In addition, the attached hexyl chain present in
In order to elucidate the interfacial charge recombination process in
DSSCs, electrochemical impedance spectroscopy (EIS) has been applied
under dark condition for the selected dyes 8aꢀ 8c with higher PCEs and
the results are recorded in Figure S20. An ideal EIS Nyquist plot shows
three semi-circles and these semicircles are referring to electrochemical
reactions among (I) Pt/electrolyte interface, (II) charge transfer at the
the
π
-linker of 8aꢀ 8c might weaken the intermolecular
π-stacked ag-
gregation, thus facilitating the electron injection process and increasing
the Jsc values. Moreover, these hexyl chains may play another role in
Table 3
TiO
2
/dye/electrolyte interface and (III) Warburg diffusion processes
Photovoltaic performance of DSSCs with different photosensitizers (5aꢀ 5c,
ꢀ
ꢀ
8
aꢀ 8c and N719) under AM 1.5 G sunlight illumination.
(I /I
3
), respectively. However, the impedance of a DSSC device will
ꢀ
2
rec /Ω cmꢀ 1
R
change with the device potential, therefore different shapes and patterns
of the curves regarding the impedance signal can be observed while the
applied bias voltage varies. In the Figure S20, the semi-circle revealed
the resistance of electron recombination (Rrec) of process (II) at the
interface for 8aꢀ 8c [38]. The calculated Rrec values are summarized in
Table 3. The Rrec (Ω/cm) value increases in the order of 8c (140.29) < 8a
Dye
Jsc /mA cm
Voc /V
FF
PCE /%
5
5
5
8
8
8
a
b
c
a
b
c
8.61
0.672
0.657
0.652
0.680
0.692
0.672
0.701
0.683
0.664
0.579
0.626
0.629
0.641
0.713
3.95
3.49
2.91
5.25
5.01
4.58
7.91
ꢀ
8.00
ꢀ
7.69
ꢀ
12.32
11.51
10.64
15.83
141.59
152.55
140.29
ꢀ
(
141.59) < 8b (152.55), therefore 8b is the most effective dye to sup-
N719
press charge recombination while 8c suffers from the most conspicuous
4

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
charge recombination. This conforms to their measured increasing order
of Voc and supports that 2-ethylhexyl chain, as compared to hexyl and
octyl chains, is more effective to weaken the interaction between elec-
Table 4
Photocatalytic H
2
generation data with different photosensitizers over 48 h.
a
b
c
d
e
Dye
H
2
TON
TOF
TOF
i
/
Activity
i
/μ
mol
i
AQY %
ꢀ
1
ꢀ 1
ꢀ 1 ꢀ 1
trolyte and TiO
2
in such a structure framework.
/mL
/h
h
g
h
5
a
10.5
7.22
13.7
11.1
11.7
15.5
1.06
3500
2410
4570
3700
3900
5170
354
72.9
50.2
95.2
77.1
81.3
108
7.4
124
155,000
112,000
194,000
148,000
171,000
185,000
15,000
3.97
2.87
5.00
3.83
4.42
4.79
0.35
2
.6. Light-driven hydrogen generation
5b
89.2
155
118
137
148
12.0
5
8
8
8
c
a
b
c
In view of the light-harvesting capability of 5a‒5c and 8a‒8c in
50–550 nm region demonstrated from the IPCE spectra (Fig. 4), we
4
anticipated that 5a‒5c and 8a‒8c could be utilized in light-driven H
2
Eosin
Y
production from water. Herein, we carried out the photocatalytic H
2
Blank
0.45
–
–
–
10,400
0.23
production investigation by adopting the same system to these six
Dꢀ ꢀ A organic dyes. Also, a reference dye Eosin Y, which has been
used to study light-driven H production in many different photo-
a
π
Turnover number (TON) of H
2
is calculated as number of mole of H
. And the dye-
loading value for 5a‒5c, 8a‒8c and Eosin Y were all measured to be close to 100
2
pro-
duced over the number of mole of PS attached to platinized TiO
2
2
catalytic systems was also examined for comparison [39–41].
%
.
2
Detailed procedures for the preparation of platinized TiO , attach-
b
Turnover frequency (TOF) is calculated per hour.
Initial turnover frequency (TOF ) in the first 16 h.
Initial photocatalytic activity (Activity ) is defined as number of micromole
evolved per gram of platinum loaded per hour.
Initial apparent quantum yield percentage (AQY ).
ment of PS onto the platinized TiO
are referred to the Experimental section. In short, each PS was loaded
onto the TiO
using sonication for 30 min, the ’active’ photocatalytic
reaction mixture was then retrieved by centrifugation and dried before
the H production experiments. The dye-loading percentage for each PS
2
and photocatalytic reaction studies
c
i
d
i
2
of H
e
2
i
2
was estimated by comparing the absorbance value of low-energy ab-
sorption peak before and after the dye adsorption and all these
carboxylate anchor-based PSs were found to attach onto the composite
completely. The photocatalytic reactions were carried out in 5 mL
aqueous ascorbic acid (AA, 0.5 M) solution at pH 4.0 with AA serving as
the sacrificial electron donor (SED). The photocatalytic system in 25 mL
reaction flasks was continuously radiated with green light-emitting di-
odes (ca. 520 nm) and the headspace of each sample was qualitatively
and quantitatively analyzed by gas chromatography (GC) by the end of
illumination. The GC analysis of hydrogen generation was conducted
after 16, 24, 40 and 48 h of irradiation using methane as an internal
standard to calibrate the hydrogen measurement. Hydrogen generation
curves (vs. time) for 5a‒5c, 8a‒8c, Eosin Y and blank are exhibited in
decrease of AA concentration and dye degradation [42,43]. PS 8c, with a
longer -linker, is considered to be more stable in the photolysis con-
π
dition and more persistent in photocatalytic efficiency than 5c
throughout the two-day experiment.
According to Table 4, 5c and 8c, both with a hexyl chain attached to
the carbazole group, show higher TON and TOF than other PSs with
octyl chains. This comparison indicates that the shorter alkyl chain on
the donor moiety may favor the photocatalytic reduction of aqueous
protons by promoting the interaction among the photocatalyst (i.e., dye-
sensitized platinized TiO
Besides, by comparing each set of the two molecules with the same
donor moiety but different -linkers, the PS with a longer -linker ex-
hibits a higher photocatalytic efficiency than that with a shorter
-bridge. This trend corresponds to the UV/Vis absorption spectra
Fig. 2) and IPCE spectra (Fig. 4), the increased photocatalytic effi-
2
nanoparticles), aqueous protons and AA [44].
π
π
Fig. 5 and the corresponding data (TON, TOF, TOF
AQY %) are tabulated in Table 4.
All Dꢀ ꢀ A organic dyes present in this study potently facilitate the
light-driven H production (> 7 mL over 48 h) while the system without
PS (blank) only produced ~ 0.5 mL of H . Overall, the TON and TOF of
i i
, Activity and
π
i
(
π
ciencies of 8a‒8c are attributed to the wider light absorption bands in
green light region, especially from 550 to 600 nm. In contrast, 5a‒5c
with blue-shifted absorption bands would lower the photons utilization
2
2
the six studied photosensitizers exceed the reference dye Eosin Y by
approximately 10 times and this shows a pronounced effect in protons
reduction, especially in consideration of the simplicity of their chemical
structures. Among the six organic dyes, the best TON of 5170 and TOF of
2
from this region, thus producing smaller amount of H in the same
duration. Therefore, 8c, which possesses a shorter alkyl chain on the
donor moiety and more intense light absorption in the green light re-
2
gion, displays the largest H evolution amount.
¡1
¡1
1
08 h in 48 h are achieved by 8c. The highest TOF
AQY of 5.0 % are attained by 5c. The apparent decrease in H
rate of the six dyes after 20 h of irradiation is probably ascribed to the
i
of 155 h and
i
2
evolution
2
.7. Conclusion
A series of Dꢀ πꢀ A organic photosensitizers based on thieno[3,2-b]
thiophene -linker has been designed and synthesized. All these PSs
π
were fully characterized by spectroscopic studies. The best performance
in DSSCs (PCE = 5.25 %) was achieved by dye 8a with a thieno[3,2-b]
thiophene-based -linker linked between an octyl chain-functionalized
π
carbazole donor and cyanoacrylic acid acceptor. For the H
tion, the largest H TON of 5170 (48 h) was achieved by the 8c-anchored
platinized TiO system, whereas 8c possesses hexyl chain-
functionalized carbazole group. From these results, it was found that
the modification of alkyl chain in donor moiety in the confined Dꢀ ꢀ A
framework is crucial to attain a significant variation in efficiency no
matter in light-driven H generation or DSSCs. Therefore, it could be
2
genera-
2
2
a
π
2
anticipated that alkyl chain engineering is one of the necessary step in
designing a superior photosensitizer, because the length, orientation and
bulkiness of alkyl chain(s) would dramatically change the interactions
between PS and other variable components at the charge separation
interface.
Fig. 5. Photocatalytic H
2
generation curves of dye-sensitized systems (5a‒5c,
8
a‒8c, Eosin Y and blank) under green LED irradiation (50 mW).
5

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
3
. Experimental section
by sputtering method at 15 mA for 90 s at a power of 150 W. Two holes
(
0.75 mm in diameter) were predrilled in the FTO glass for introducing
3
.1. Materials and reagents
the electrolyte. The dye-adsorbed TiO electrode and Pt-counter elec-
2
trode were assembled into a sandwich type cell and sealed with a hot-
◦
All reactions were performed under an inert nitrogen atmosphere
melt parafilm at about 100 C. The liquid electrolyte consisting of 0.6
with the use of a Schlenk line. Glassware was dried in oven prior to use.
Commercially available reagents were used without purification unless
otherwise stated. All the reagents for the chemical synthesis were pur-
chased from Sigma-Aldrich and Acros Organics. Solvents were purified
by distillation over appropriate drying agents. All reactions were
monitored by thin-layer chromatography (TLC) with Merck pre-coated
aluminum plates. Products were separated and purified by column
chromatography using silica gel (230 ꢀ 400 mesh) purchased from
Merck.
M 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.1 M LiI, 0.05 M
2
I in a mixture of acetonitrile and 4-tert-butylpyridine (volume ratio,
1:1) was introduced into the cell through the drilled holes at the back of
the counter electrode. At last, the holes were sealed by parafilm and
covering glass (0.1 mm thickness) at elevated temperature. The effective
2
areas of all the TiO
2
electrodes were ~ 0.20 cm . The current ꢀ voltage
(Jꢀ V) characteristics of the assembled DSSCs were measured by a
semiconductor characterization system (Keithley 236) at room temper-
ature in air under the spectral output from solar simulator (Newport
–2
9
4023A) using an AM 1.5 G filter with a light power of 100 mW cm .
3
.2. Instrumentation
IPCEs of DSSCs were recorded in the Solar Cell QE/IPCE Measurement
System (Zolix Solar Cell Scan 100) using dc mode. CHI 660D electro-
chemical workstation was used to characterize the electrochemical
properties of the DSSCs. Electrochemical impedance spectroscopy (EIS)
Matrix-assisted laser desorption ionization time-of-flight (MALDI-
TOF) mass spectra were recorded on an Autoflex Bruker MALDI-TOF
system. Proton and carbon NMR spectra were measured in CDCl
DMSO-d /THF-d on a Bruker Ultra-shield 400 MHz FT-NMR spec-
5
3
/
was recorded under dark conditions over a frequency range of 0.1 ꢀ 10
Hz with an ac amplitude of 10 mV and the parameters were calculated
from Z-View software (v2.1b, Scribner Associates, Inc.).
6
8
trometer and tetramethylsilane (TMS) was used as an internal standard
for calibrating the chemical shift. UV/Vis absorption spectra were
recorded on a Hewlett Packard 8453 spectrometer in CH
93 K. Electrochemical measurements were conducted on Potentiostat/
Galvanostat/EIS Analyzer model Parstat 4000 at a scan rate of 100 mV
2
Cl
2
solution at
3.5. Preparation of platinized TiO
2
2
For a 0.5 wt-% platinized TiO
to 1.6 g of titanium(IV) oxide nanopowder (anatase, < 25 nm particle
size, 99.7 % trace metals basis, Sigma-Aldrich) and 0.1 mL of H PtCl
2
sample, addition of 40 mL methanol
ꢀ
1
s
.
2
6
3
.3. Computational details
aqueous solution (8 wt. %) created slurry which was then subjected to
radiation from a 300 W coated Hg lamp (HF300PD, EYE Lighting) under
vigorously stirring for 24 h. The resulting crude product had a grayish
All the calculations in this work were carried out using Gaussian 09
package [45]. The ground state geometries of all the dyes in dichloro-
methane (
= 8.93) were optimized using the density functional theory
DFT) with hybrid functional PBE0 and the solvation effects were
2
colour. Platinized TiO was then retrieved by centrifugation at 3500 rpm
ε
for 5 min and washed three times with methanol. The obtained material
◦
(
was dried under vacuum at ~ 60 C in darkness for 8 h.
included using the integral equation formalism [46,47] (IEF) version of
the polarizable continuum [48,49] (PCM) model. The frequency calcu-
lations were then performed to confirm each optimized structure was
the real minimum without imaginary vibration frequency. The
time-dependent DFT (TD-DFT) was used to calculate vertically excited
singlet-state energies. The 6ꢀ 31 G(d) basis set was adopted. The ab-
sorption profiles were calculated using the Multiwfn program [50] as a
sum of Gaussian-shaped bands with full width at half maximum
3.6. Adsorption of photosensitizer onto platinized TiO
2
20 mg of the prepared platinized TiO was added to 2.5 mL of 50 M
2
μ
photosensitizer dichloromethane solution, the mixture was then evenly
sonicated for 30 min. The solution gradually became colourless while
the solid became orange/red in colour. The dye-loaded solid was then
retrieved by centrifugation at 3500 rpm for 5 min. The supernatant was
removed carefully using a dropper and the pellet at the bottom was dried
under vacuum for two to three hours in darkness. At the end, the whole
dried pellet was directly utilized in the photocatalytic reaction without
further characterization. The dye-loading percentage was estimated by
comparing the absorbance of low energy absorption peak between the
supernatant and original photosensitizer solution.
(
FWHM) equal to 0.667 eV.
3
.4. Fabrication and characterization of DSSCs
All the anode films of DSSCs were made under the same standard
manner and composed of a 12
diameter of 20 nm) and a 6 m thick scattering layer (TiO
with a diameter of 200 nm). Doctor-blade technique was utilized to
prepare photoanode (TiO ) films. First, a TiO paste (20 nm) layer of ~6
m was doctor-bladed onto a FTO conducting glass and then relaxed at
μ
m thick transparent layer (TiO
2
with
μ
2
nanoparticles
2
3.7. Light-driven H production studies
2
2
μ
A 25 mL pear-shaped flask was placed above stirrers and the flask
was sealed with a rubber septum. The photocatalytic reaction mixture
with a stir bar was then purged with a mixture of argon/methane (80:20
mol%) for 15 min. The methane present in the gas mixture was served as
an internal standard for GC analysis. The reaction mixture was steadily
stirred and directly radiated from the bottom with green (520 nm) light-
emitting diodes inside a just-fit container which blocks the stray light
from the environment. The light power was measured using a thermal
sensor and power meter (Model: BIM-7203ꢀ 0100 F & BIM-7001;
Hangzhou Brolight Technology Co., Ltd.) and giving an estimated
value of 50 mW for each reaction. At the end of the experiment, head-
spaces of the flasks were characterized by GC to examine the amount of
hydrogen produced. The amount of hydrogen evolved was determined
using GC (Agilent 6890 Series GC System with a molecular sieve 5 Å
column and TCD detector) by the end of radiation and were quantified
◦
room temperature for 3 min before heated at 150 C for 6 min; this
procedure was repeated once to achieve a film thickness of ~12 m and
the resulting surface was finally coated with a scattering layer (~6 m)
of TiO paste (200 nm). Then, the electrodes were gradually heated
μ
μ
2
◦
◦
◦
under an air flow at 275 C for 5 min, 325 C for 5 min, 375 C for 5 min,
◦
and 470 C for 30 min to remove polymers and generate a three
dimensional TiO
2
nanoparticle network. After that, the sintered films
◦
were soaked into 0.02 M TiF
4
aqueous solution for 45 min at 70 C,
◦
washed with deionized water, and further annealed at 450 C for 30 min.
◦
After cooling down to ~80 C, the electrodes were immersed into a 5 ×
ꢀ 4
1
0
M concentration of corresponding dye bath in acetonitrile/tert-
butyl alcohol (volume ratio, 1:1) and maintained in the dark for 15 h.
Afterwards, the electrodes were rinsed with ethanol to remove the non-
adsorbed dyes and dried in the air. Pt counter electrodes were prepared
6

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
referring to a calibration plot of integrated amount of hydrogen relative
to that of methane (Figure S21). In the course of a 48 -h radiation,
samples were taken at four different time points from the reaction
mixtures of each photosensitizer and blank. The LED radiation is
assumed to be monochromatic at emission maximum (520 nm) and the
corresponding apparent quantum yield values for each PS could be
estimated according to the equation shown below.
= 9.73 (s, 1H, CHO), 8.14 (d, 1H, J = 1.6 Hz, Ar), 7.98 (d, 1H, J = 7.6 Hz,
Ar), 7.64 (s, 1H, Ar), 7.54 ꢀ 7.52 (m, 1H, Ar), 7.40 ꢀ 7.36 (m, 1H, Ar),
7.31 ꢀ 7.26 (m, 2H, Ar), 7.21 ꢀ 7.13 (m, 2H, Ar), 4.00 ꢀ 3.92 (m, 2H,
alkyl), 1.95 ꢀ 1.89 (m, 1H, alkyl), 1.33 ꢀ 1.168 (m, 8H, alkyl), 0.81 ꢀ
0.74 (m, 6H, alkyl) ppm; 13C NMR (100 MHz, CDCl
): δ = 183.02 (CHO),
3
154.85, 147.28, 143.60, 141.45, 141.25, 137.35, 129.38, 126.35,
124.68, 124.06, 123.32, 122.60, 120.52, 119.46, 118.09, 114.26,
1
09.50, 109.43 (Ar), 47.51, 39.44, 31.02, 28.82, 24.42, 23.09, 14.09,
rate of H
2
production × 2
+
AQY(%) =
× 100%
(1)
10.94 (alkyl) ppm. HRMS (MALDI-TOF, m/z): [M ] 445.1529; calcd for
rate of incident photons
.8. Synthesis
,5-Dibromothieno[3,2-b]thiophene (2): Thieno[3,2-b]thio-
(
C
27
2
H27NOS ) 445.1534.
5
-(9-Hexyl-9H-carbazol-3-yl)thieno[3,2-b]thiophene-2-carbal-
1
3
dehyde (4c): 0.13 g; 82 % yield. H NMR (400 MHz, CDCl
3
): δ = 9.88 (s,
1
H, CHO), 8.30 (d, 1H, J = 1.6 Hz, Ar), 8.14 (d, 1H, J = 8 Hz, Ar), 7.78
2
(s, 1H, Ar), 7.70 ꢀ 7.67 (m, 1H, Ar), 7.56 ꢀ 7.52 (m, 1H, Ar), 7.46 ꢀ
7.42 (m, 2H, Ar), 7.37 ꢀ 7.28 (m, 2H, Ar), 4.26 (t, 2H, J = 8 Hz, alkyl),
phene 1 (1 g, 7.131 mmol) was dissolved in THF (40 mL) at room
◦
temperature and the solution was cooled to 0 C. NBS (2.547 g, 14.262
1.91 ꢀ 1.83 (m, 2H, alkyl), 1.40 ꢀ 1.28 (m, 6H, alkyl), 0.95 ꢀ 0.92 (m,
◦
13
mmol) was then added in ten portions at 0 C. The reaction mixture was
3H, alkyl) ppm; C NMR (100 MHz, CDCl
3
): δ = 183.02 (CHO), 154.86,
slowly warmed to room temperature and was stirred for 12 h. The
mixture was then poured into cold water (100 mL) and extracted with
diethyl ether (3 × 80 mL). The organic layers were washed with water
147.27, 143.58, 140.99, 140.76, 137.35, 129.38, 126.37, 124.71,
124.09, 123.36, 122.63, 120.58, 119.47, 118.16, 114.28, 109.22,
109.14 (Ar), 43.23, 31.58, 28.96, 26.97, 22.58, 14.06 (alkyl) ppm.
+
(
100 mL) twice and dried over Na
2
SO
4
. Removal of solvents gave 2 as a
HRMS (MALDI-TOF, m/z): [M ] 417.1125; calcd for (C25
H23NOS
2
)
1
pure white solid (2.1 g, 7.047 mmol, 99 % yield). H NMR (400 MHz,
417.1121.
CDCl
3
): δ = 7.29 (s, 2H, Ar) ppm; 13C NMR (100 MHz, CDCl
3
): δ =
5-(5-Bromothieno[3,2-b]thiophen-2-yl)-3-hexylthiophene-2-
carbaldehyde (6): A mixture of 2 (1 g, 3.355 mmol), (4-hexylthiophen-
2-yl)boronic acid (0.711 g, 3.355 mmol), Pd(PPh (193 mg, 0.168
mmol) and 2 M K CO (6 mL) in THF (30 mL) was heated to reflux under
a N atmosphere overnight. The reaction mixture was poured into water,
followed by extraction using ethyl acetate. The organic layer was dried
over anhydrous Na SO . The solvent was then removed under reduced
+
1
38.29, 121.76, 113.62 (Ar) ppm. HRMS (MALDI-TOF, m/z): [M ]
2
95.7957; calcd for (C
6
H
2
Br
S
2 2
) 295.7965.
3 4
)
5
-Bromothieno[3,2-b]thiophene-2-carbaldehyde (3): To a two-
2
3
necked round-bottom flask containing 2 (0.7 g, 2.348 mmol) in 20 mL
dry THF, n-BuLi (0.978 mL, 2.348 mmol, 2.4 M in hexane) was added
2
◦
dropwisely at ꢀ 78 C. After the reaction mixture was stirred for 15 min,
2
4
dimethylformamide (0.271 mL, 3.532 mmol) was added. The reaction
mixture was then warmed up gradually to room temperature and was
stirred for 30 min. To the reaction mixture, an excess amount of 2 M HCl
pressure. In order to remove the residue of palladium catalyst, the
organic fraction was filtered by short column chromatography on silica
2 2
gel using a 1:3 mixture of CH Cl and hexane as eluent to afford the
(
aq) (10 mL) was introduced, while the mixture was stirred for another 1
filtered crude product which was immediately subjected to the Vils-
h. The reaction was quenched by pouring it into water, followed by
extraction using ethyl acetate. The organic layer was dried over anhy-
meier–Haack reaction. Next, to a two-necked round-bottom flask con-
taining dry DMF (0.936 g, 12.818 mmol), POCl
3
(1.961 g, 12.818 mmol)
◦
drous Na
sure. The crude product was subjected to flash column chromatography
with hexane/CH Cl (2:1) as eluent. Product 3 (0.284 g, 1.149 mmol)
was obtained as a pale-yellow oil in 49 % yield. H NMR (400 MHz,
SO
2 4
and the solvent was then removed under reduced pres-
was added dropwisely by syringe at 0 C, Vilsmeier-Haack reagent was
produced. A solution of the filtered crude product (0.494 g) in dry
2 2
CH Cl (20 mL) was added dropwise to the prepared Vilsmeier-Haack
2
2
1
reagent under N
2
atmosphere. The mixture was stirred for 15 min at
1
3
◦
◦
CDCl
3
): δ = 9.98 (s, 1H, CHO), 7.85 (s, 1H, Ar), 7.36 (s, 1H, Ar) ppm;
C
0 C and overnight at 65 C. After cooling, the reaction was poured into
aqueous NaOH solution slowly in an ice bath with stirring for 30 min,
NMR (100 MHz, CDCl
3
): δ = 183.28 (CHO), 144.72, 144.54, 139.35,
+
1
28.20, 122.91, 120.95 (Ar) ppm. HRMS (MALDI-TOF, m/z): [M ]
the resulting mixture was then extracted with CH
organic fraction was dried over anhydrous Na SO and the solvent was
removed under reduced pressure. A yellow oil was yielded, which was
purified by column chromatography on silica gel with hexane/CH Cl
(3:2 v/v) as the eluent to give 6 as a yellow oil (0.228 g, 0.552 mmol, 16
2 2 2
Cl and H O. The
2
45.8811; calcd for (C
General synthetic procedure of 4a-4c: A mixture of 3 (94 mg, 0.38
mmol), (9-octylcarbazolyl)/ (9-(2-ethylhexyl)carbazoyl)/ (9-hex-
ylcarbazolyl)boronic acid (0.76 mmol), Pd(PPh (22 mg, 0.019 mmol)
and 2 M K CO (2 mL) in THF (20 mL) was heated to reflux under a N
7
H
3
BrOS
2
) 245.8809.
2
4
2
2
3 4
)
1
% yield). H NMR (400 MHz, CDCl
3
): δ = 9.98 (s, 1H, CHO), 7.36 (d, 1H,
2
3
2
atmosphere overnight. The reaction mixture was poured into water,
followed by extraction using ethyl acetate. The organic layer was dried
J = 0.4 Hz, Ar), 7.20 (d, 1H, J = 0.8 Hz, Ar), 7.05 (s, 1H, Ar), 2.91 (t, 2H,
J = 7.6 Hz, alkyl), 1.72 ꢀ 1.65 (m, 2H, alkyl), 1.42 ꢀ 1.31 (m, 6H, alkyl),
1
3
over anhydrous Na
pressure. The residue was purified by column chromatography on silica
gel using a 1:1 mixture of CH Cl and hexane as eluent to afford 4a-4c as
orange red solids, respectively.
-(9-Octyl-9H-carbazol-3-yl)thieno[3,2-b]thiophene-2-carbal-
2
SO
4
and the solvent was removed under reduced
0.908 ꢀ 0.873 (m, 3H, alkyl) ppm; C NMR (100 MHz, CDCl ): δ =
3
181.45 (CHO), 153.83, 145.63, 140.09, 138.45, 137.23, 136.24, 126.68,
122.18, 117.46, 115.31 (Ar), 31.55, 31.34, 29.00, 28.57, 22.55, 14.06
2
2
+
(alkyl) ppm. HRMS (MALDI-TOF, m/z): [M ] 411.9629; calcd for
5
3
(C17H17BrOS ) 411.9625.
1
dehyde (4a): 0.112 g; 66 % yield. H NMR (400 MHz, CDCl
3
): δ = 9.96
General synthetic procedure of 7a¡7c: A mixture of 6 (76 mg,
(
s, 1H, CHO), 8.38 (s, 1H, Ar), 8.15 (d, 1H, J = 8 Hz, Ar), 7.93 (s, 1H, Ar),
0.184 mmol), (9-octylcarbazolyl)/ (9-(2-ethylhexyl)carbazolyl)/ (9-
7
4
.77 (d, 1H, J = 8.4 Hz, Ar), 7.58 (s, 1H, Ar), 7.53 ꢀ 7.43 (m, 4H, Ar),
.32 (t, 2H, J = 7.2 Hz, alkyl), 1.89 ꢀ 1.88 (m, 2H, alkyl), 1.35 ꢀ 1.19
hexylcarbazolyl) boronic acid (0.368 mmol), Pd(PPh
mmol) and 2 M K CO (2 mL) in THF (20 mL) was heated to reflux under
a N atmosphere for overnight. The reaction mixture was poured into
water, followed by extraction using ethyl acetate. The organic layer was
dried over anhydrous Na SO . Then, the solvent was removed under
reduced pressure and the residue was purified by column chromatog-
3 4
) (11 mg, 0.009
2
3
1
3
(
m, 10H, alkyl), 0.88 ꢀ 0.84 (m, 3H, alkyl) ppm; C NMR (100 MHz,
2
CDCl
3
):
δ
=
183.05 (CHO), 154.95, 147.34, 143.70, 141.03,
1
1
2
40.84,137.45, 129.37, 126.38, 124.80, 124.21, 123.44, 122.64,
20.58, 119.48, 118.32, 114.44, 109.29, 109.14 (Ar), 43.31, 31.79,
2
4
9.37, 29.17, 29.00, 27.31, 22.61, 14.07 (alkyl) ppm. HRMS (MALDI-
2 2
raphy on silica gel using a 1:1 mixture of CH Cl and hexane as eluent to
+
TOF, m/z): [M ] 445.1532; calcd for (C27
H
27NOS
2
) 445.1534.
afford 7a–7c as red solids.
5
-(9-(2-Ethylhexyl)-9H-carbazol-3-yl)thieno[3,2-b]thiophene-
3-Hexyl-5-(5-(9-octyl-9H-carbazol-3-yl)thieno[3,2-b]thiophen-
1
1
2
-carbaldehyde (4b): 0.15 g; 89 % yield. H NMR (400 MHz, CDCl
3
): δ
2-yl)thiophene-2-carbaldehyde (7a): 72 mg; 64 % yield. H NMR (400
7

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
MHz, CDCl
3
): δ = 9.91 (s, 1H, CHO), 8.24 (d, 1H, J = 1.6 Hz, Ar), 8.06 (d,
H, J = 7.6 Hz, Ar), 7.65 ꢀ 7.63 (m, 1H, Ar), 7.42 ꢀ 7.38 (m, 3H, Ar),
.34 ꢀ 7.31 (m, 2H, Ar), 7.19 ꢀ 7.18 (m, 1H, Ar), 6.99 (s, 1H, Ar), 4.21
1H, C = CHꢀ ), 8.34 ꢀ 8.28 (m, 2H, Ar), 8.06 (s, 1H, Ar), 7.863 (dd, 1H, J
= 8.8, 1.6 Hz, Ar), 7.67 (d, 1H, J = 8.4 Hz Ar), 7.60 (d, 1H, J = 8 Hz, Ar),
7.53 ꢀ 7.49 (m, 1H, Ar), 7.28 ꢀ 7.25 (m, 1H, Ar), 4.30 (d, 2H, J = 7.2
Hz, alkyl), 2.02 ꢀ 1.99 (m, 1H, alkyl), 1.35 ꢀ 1.16 (m, 8H, alkyl), 0.86
1
7
(
t, 2H, J = 7.2 Hz, alkyl), 2.84 (t, 2H, J = 7.6 Hz, alkyl), 1.81 ꢀ 1.78 (m,
H, alkyl), 1.65 ꢀ 1.61 (m, 2H, alkyl), 1.33 ꢀ 1.16 (m, 16H, alkyl), 0.85
0.77 (m, 6H, alkyl) ppm; 13C NMR (100 MHz, CDCl
): δ = 181.42
CHO), 154.01, 149.79, 146.63, 140.98, 140.97, 140.43, 138.30,
1
3
2
(t, 3H, J = 7.2 Hz, alkyl), 0.78 (t, 3H, J = 6.8 Hz, alkyl) ppm; C NMR
ꢀ
3
(100 MHz, THF-d
8
): δ = 162.84 (COOH), 154.22, 147.69, 146.08,
(
141.11, 140.91, 137.40, 136.34, 129.84, 125.66, 124.49, 123.47,
123.03, 122.27, 119.93, 118.80, 117.41, 115.42, 109.23, 108.94, 97.42
1
1
3
2
36.20, 135.69, 126.20, 126.14, 125.36, 124.04, 123.39, 122.69,
20.57, 119.28, 118.37, 117.88, 113.99, 109.18, 109.03 (Ar), 43.27,
1.80, 31.58, 31.37, 29.38, 29.18, 29.04, 29.01, 28.60, 27.32, 22.62,
(Ar), 67.42, 46.60, 38.98, 30.46, 28.30, 22.57, 12.97, 9.82 ppm (alkyl).
+
HRMS (MALDI-TOF, m/z): [M ] 512.1590; calcd for (C30
H
28
N
2
O
S
2 2
)
+
2.57, 14.08 (alkyl) ppm. HRMS (MALDI-TOF, m/z): [M ] 611.2350;
512.1592.
calcd for (C37
-(5-(9-(2-Ethylhexyl)-9H-carbazol-3-yl)thieno[3,2-b]thio-
phen-2-yl)-3-hexylthiophene-2-carbaldehyde (7b): 85 mg; 75 %
H
41NOS
3
) 611.2350.
(E)-2-Cyano-3-(5-(9-hexyl-9H-carbazol-3-yl)thieno[3,2-b]thio-
1
5
phen-2-yl)acrylic acid (5c): 150 mg; 99 % yield; red solid. H NMR
(400 MHz, DMSO-d
6
): δ = 8.60 (d, 1H, J = 1.6 Hz, Ar), 8.56 (s, 1H, C =
1
yield. H NMR (400 MHz, CDCl
3
): δ = 9.85 (s, 1H, CHO), 8.15 (d, 1H, J =
CHꢀ ), 8.33 ꢀ 8.28 (m, 2H, Ar), 8.07 (s, 1H, Ar), 7.88 ꢀ 7.86 (m, 1H, Ar),
7.73 (d, 1H, J = 8.8 Hz, Ar), 7.65 (d, 1H, J = 8 Hz, Ar), 7.53 ꢀ 7.49 (m,
1H, Ar), 7.28 ꢀ 7.24 (m, 1H, Ar), 4.44 (t, 2H, J = 7.2 Hz, alkyl), 1.81 ꢀ
1.77 (m, 2H, alkyl), 1.30 ꢀ 1.20 (m, 6H, alkyl), 0.81 (t, 3H, J = 6.8 Hz,
1
.6 Hz, Ar), 8.00 (d, 1H, J = 7.6 Hz, Ar), 7.54 (dd, 1H, J = 8.4, 1.6 Hz,
Ar), 7.40 ꢀ 7.36 (m, 1H, Ar), 7.29 ꢀ 7.26 (m, 3H, Ar), 7.21 (d, 1H, J =
8
2
1
9
1
1
1
3
1
.4 Hz, Ar), 7.17 ꢀ 7.15 (m, 1H, Ar), 6.90 (s, 1H, Ar), 4.00 ꢀ 3.97 (m,
H, alkyl), 2.76 (t, 2H, J = 7.6 Hz, alkyl), 1.96 ꢀ 1.90 (m, 1H, alkyl),
.59 ꢀ 1.53 (m, 2H, alkyl), 1.29 ꢀ 1.13 (m, 14H, alkyl), 0.82 ꢀ 0.75 (m,
alkyl) ppm; 13C NMR (100 MHz, THF-d
): δ = 162.96 (COOH), 154.22,
8
147.63, 145.99, 140.70, 140.48, 137.38, 136.33, 129.73, 125.66,
124.47, 123.47, 123.05, 122.29, 119.94, 118.77, 117.42, 115.48,
113.96, 108.94, 108.67, 97.56 (Ar), 42.31, 31.18, 28.49, 26.36, 22.07,
H, alkyl) ppm; 13C NMR (100 MHz, CDCl
): δ = 181.35 (CHO), 153.99,
3
49.74, 146.61, 141.42, 140.97, 140.87, 138.22, 136.17, 135.53,
26.15, 126.03, 125.28, 123.91, 123.31, 122.68, 120.51, 119.26,
18.30, 117.66, 113.85, 109.41, 109.32 (Ar), 47.49, 39.45, 31.61,
1.36, 31.03, 29.07, 28.85, 28.59, 24.43, 23.09, 22.61, 14.12, 14.08,
+
12.97 ppm (alkyl). HRMS (MALDI-TOF, m/z): [M ] 484.1284; calcd for
24 2 2 2
(C28H N O S ) 484.1279.
(E)-2-Cyano-3-(3-hexyl-5-(5-(9-octyl-9H-carbazol-3-yl)thieno
+
0.93 (alkyl) ppm. HRMS (MALDI-TOF, m/z): [M ] 611.2353; calcd for
[3,2-b]thiophen-2-yl)thiophen-2-yl)acrylic acid (8a): 79 mg; 99 %
1
(
C
37
H
41NOS
3
) 611.2350.
yield; dark purple solid. H NMR (400 MHz, DMSO-d
6
): δ = 8.54 (s, 1H,
3
-Hexyl-5-(5-(9-hexyl-9H-carbazol-3-yl)thieno[3,2-b]thiophen-
C = CHꢀ ), 8.29 ꢀ 8.26 (m, 2H, Ar), 7.99 (s, 1H, Ar), 7.92 (s, 1H, Ar),
7.82 ꢀ 7.80 (m, 1H, Ar), 7.68 ꢀ 7.62 (m, 2H, Ar), 7.53 ꢀ 7.47 (m, 2H,
Ar), 7.26 ꢀ 7.23 (m, 1H, Ar), 4.41 (t, 2H, J = 6.8 Hz, alkyl), 2.80 (t, 2H, J
= 7.2 Hz, alkyl), 1.79 ꢀ 1.76 (m, 2H, alkyl), 1.63 ꢀ 1.60 (m, 2H, alkyl),
1
2
-yl)thiophene-2-carbaldehyde (7c): 80 mg; 74 % yield. H NMR (400
): δ = 10.01 (s, 1H, CHO), 8.35 (d, 1H, J = 2 Hz, Ar), 8.16 (d,
H, J = 7.6 Hz, Ar), 7.77 ꢀ 7.74 (m, 1H, Ar), 7.53 ꢀ 7.50 (m, 3H, Ar),
.46 ꢀ 7.42 (m, 2H, Ar), 7.31 ꢀ 7.28 (m, 1H, Ar), 7.10 (s, 1H, Ar), 4.33
MHz, CDCl
3
1
7
1
3
1.30 ꢀ 1.18 (m, 16H, alkyl), 0.89 ꢀ 0.79 (m, 6H, alkyl) ppm; C NMR
(
t, 2H, J = 7.2 Hz, alkyl), 2.97 ꢀ 2.93 (m, 2H, alkyl), 1.92 ꢀ 1.89 (m, 2H,
(100 MHz, THF-d
8
): δ = 163.34 (COOH), 154.83, 149.59, 144.99,
alkyl), 1.74 ꢀ 1.72 (m, 2H, alkyl), 1.44 ꢀ 1.31 (m, 12H, alkyl), 0.95 ꢀ
141.92, 140.90, 140.63, 140.02, 138.08, 135.38, 128.49, 125.48,
125.00, 124.95, 123.16, 122.97, 122.33, 119.88, 118.58, 116.89,
115.65, 113.57, 108.79, 108.57, 96.69 (Ar), 42.27, 31.40, 31.16, 30.70,
0
1
1
1
3
.89 (m, 6H, alkyl) ppm; 1 C NMR (100 MHz, CDCl
3
3
): δ = 181.44 (CHO),
54.03, 149.79, 146.64, 140.98, 140.97, 140.43, 138.30, 136.20,
35.61, 126.21, 126.16, 125.36, 124.05, 123.39, 122.69, 120.57,
19.28, 118.39, 117.90, 114.01, 109.18, 109.03 (Ar), 43.28, 31.58,
29.00, 28.81, 28.55, 28.24, 26.72, 22.12, 22.09, 13.04 ppm (alkyl).
+
HRMS (MALDI-TOF, m/z): [M ] 678.2401; calcd for (C40
H
42
N
2
O
S
2 3
)
1.38, 29.03, 28.98, 28.61, 26.98, 22.57, 22.56, 14.09, 14.02 (alkyl)
678.2408.
+
ppm. HRMS (MALDI-TOF, m/z): [M ] 583.2031; calcd for
(E)-2-Cyano-3-(5-(5-(9-(2-ethylhexyl)-9H-carbazol-3-yl)thieno
[3,2-b]thiophen-2-yl)-3-hexylthiophen-2-yl)acrylic acid (8b): 80
mg; 85 % yield; dark purple solid. 1H NMR (400 MHz, DMSO-d
35 3
(C H37NOS ) 583.2037.
General synthetic procedure of 5a¡5c and 8a¡8c: A mixture of
6
): δ =
each dye precursors (4aꢀ 4c and 8aꢀ 8c) and cyanoacetic acid (10 M
8.53 (s, 1H, C = CHꢀ ), 8.28 ꢀ 8.25 (m, 2H, Ar), 7.93 (s, 1H, Ar), 7.89 (s,
1H, Ar), 7.81 ꢀ 7.78 (m, 1H, Ar), 7.66 ꢀ 7.57 (m, 2H, Ar), 7.52 ꢀ 7.48
(m, 2H, Ar), 7.27 ꢀ 7.23 (m, 1H, Ar), 4.27 (d, 2H, J = 7.2 Hz, alkyl), 2.77
(t, J = 7.2 Hz, alkyl), 2.01 ꢀ 1.98 (m, 1H, alkyl), 1.61 ꢀ 1.59 (m, 2H,
equivalents) in acetic acid (8 mL) was refluxed in the presence of
ammonium acetate (25 mg) overnight under a N
cooling, the reaction mixture was washed with water and extracted with
CHCl . The solvent was then removed under reduced pressure and the
crude compound was purified by column chromatography on silica gel
eluting with CHCl followed by CHCl /MeOH (10:1, v/v) to give the
corresponding products 5aꢀ 5c and 8aꢀ 8c.
E)-2-Cyano-3-(5-(9-octyl-9H-carbazol-3-yl)thieno[3,2-b]thio-
phen-2-yl)acrylic acid (5a): 125 mg; 97 % yield; red solid. 1H NMR
400 MHz, DMSO-d
): δ = 8.59 (d, 1H, J = 1.6 Hz, Ar), 8.57 (s, 1H, C =
CHꢀ ), 8.33 ꢀ 8.28 (m, 2H, Ar), 8.06 (d, 1H, J = 0.4 Hz, Ar), 7.87 ꢀ 7.85
m, 1H, Ar), 7.71 (d, 1H, J = 8.4 Hz, Ar), 7.64 (d, 1H, J = 8.4 Hz, Ar),
.53 ꢀ 7.48 (m, 1H, Ar), 7.28 ꢀ 7.24 (m, 1H, Ar), 4.43 (t, 2H, J = 6.8 Hz,
alkyl), 1.80 ꢀ 1.77 (m, 2H, alkyl), 1.26 ꢀ 1.18 (m, 10H, alkyl), 0.83 ꢀ
.79 (m, 3H, alkyl) ppm; 13C NMR (100 MHz, THF-d
): δ = 162.92
COOH), 154.25, 147.65, 146.00, 140.73, 140.52, 137.42, 136.36,
2
atmosphere. After
1
3
3
alkyl), 1.32 ꢀ 1.14 (m, 14H, alkyl), 0.87 ꢀ 0.86 (m, 9H, alkyl) ppm;
C
NMR (100 MHz, THF-d
8
): δ = 164.38 (COOH), 154.20, 149.31, 144.41,
3
3
141.37, 140.98, 140.71, 140.37, 138.01, 135.54, 128.71, 125.44,
124.97, 124.87, 123.12, 122.88, 122.30, 119.85, 118.58, 118.26,
116.78, 116.22, 113.52, 108.98, 108.80, 98.14 (Ar), 46.51, 38.93,
31.21, 30.65, 30.42, 28.59, 28.27, 22.55, 22.11, 13.09, 12.99, 9.83 ppm
(
+
(
6
(alkyl). HRMS (MALDI-TOF, m/z): [M ] 678.2403; calcd for
42 2 2 3
(C40H N O S ) 678.2408.
(
(E)-2-Cyano-3-(3-hexyl-5-(5-(9-hexyl-9H-carbazol-3-yl)thieno
7
[3,2-b]thiophen-2-yl)thiophen-2-yl)acrylic acid (8c): 81 mg; 91 %
1
yield; dark purple solid. H NMR (400 MHz, DMSO-d
6
): δ = 8.55 (s, 1H,
0
8
C = CHꢀ ), 8.30 ꢀ 8.26 (m, 2H, Ar), 8.02 (s, 1H, Ar), 7.94 (s, 1H, Ar),
7.84 ꢀ 7.81 (m, 1H, Ar), 7.66 (d, 1H, J = 8.4 Hz, Ar), 7.63 (d, 1H, J = 8.4
Hz, Ar), 7.54 (s, 1H, Ar), 7.52 ꢀ 7.47 (m, 1H, Ar), 7.27 ꢀ 7.23 (m, 1H,
Ar), 4.42 (t, 2H, J = 7.2 Hz, alkyl), 2.81 (t, J = 7.6 Hz, alkyl), 1.80 ꢀ 1.75
(
1
1
2
29.73, 125.67, 124.51, 123.50, 123.09, 122.31, 119.95, 118.78,
17.46, 115.45, 114.01, 108.97, 108.69, 97.58 (Ar), 42.33, 31.40,
8.99, 28.80, 28.54, 26.73, 22.11, 13.02 ppm (alkyl). HRMS (MALDI-
(m, 2H, alkyl), 1.64 ꢀ 1.59 (m, 2H, alkyl), 1.35 ꢀ 1.21 (m, 12H, alkyl),
+
13
TOF, m/z): [M ] 512.1585; calcd for (C30
H
28
N
2
O
S
2 2
) 512.1592.
0.89 ꢀ 0.79 (m, 6H, alkyl) ppm; C NMR (100 MHz, THF-d ): δ =
8
(
E)-2-Cyano-3-(5-(9-(2-ethylhexyl)-9H-carbazol-3-yl)thieno
163.23 (COOH), 154.90, 149.63, 145.03, 141.99, 140.93, 140.66,
140.06, 138.12, 135.39, 128.52, 125.50, 125.08, 124.98, 123.19,
123.00, 122.34, 119.88, 118.64, 118.60, 116.94, 115.58, 113.64,
[
3,2-b]thiophen-2-yl)acrylic acid (5b): 145 mg; 84 % yield; red solid.
1
H NMR (400 MHz, DMSO-d
6
): δ = 8.60 (d, 1H, J = 1.2 Hz, Ar), 8.57 (s,
8

P.-Y. Ho et al.
Journal of Photochemistry & Photobiology, A: Chemistry 406 (2021) 112979
1
2
08.82, 108.58, 96.64 (Ar), 42.28, 31.21, 31.17, 30.74, 28.55, 28.53,
[21] B. Zheng, R.P. Sabatini, W.-F. Fu, M.-S. Eum, W.W. Brennessel, L. Wang, D.
W. McCamant, R. Eisenberg, Proc. Natl. Acad. News Physiol. Sci. 112 (2015)
E3987–E3996.
8.25, 26.39, 22.09, 13.05, 12.99 ppm (alkyl). HRMS (MALDI-TOF, m/
+
z): [M ] 650.2099; calcd for (C38
H
38
2
N O
2
3
S ) 650.2095.
[
[
[
22] P.-Y. Ho, B. Zheng, D. Mark, W.-Y. Wong, D.W. McCamant, R. Eisenberg, Inorg.
Chem. 55 (2016) 8348–8358.
23] J. Warnan, J. Willkomm, J. Ng, R. Godin, S. Prantl, J. Durrant, E. Reisner, Chem.
Sci. 8 (2017) 3070–3079.
Declaration of Competing Interest
24] P.-Y. Ho, Y. Wang, S.-C. Yiu, W.-H. Yu, C.-L. Ho, S. Huang, Org. Lett. 19 (2017)
1
048–1051.
[25] M. Xu, R. Li, N. Pootrakulchote, D. Shi, J. Guo, Z. Yi, S.M. Zakeeruddin, M. Gr
a¨ tzel,
The authors report no declarations of interest.
P. Wang, J. Phys. Chem. C 112 (2008) 19770–19776.
[
[
[
[
[
26] G. Zhang, H. Bala, Y. Cheng, D. Shi, X. Lv, Q. Yu, P. Wang, Chem. Commun.
Acknowledgements
(Camb.) (2009) 2198–2200.
27] G. Zhang, Y. Bai, R. Li, D. Shi, S. Wenger, S.M. Zakeeruddin, M. Gratzel, P. Wang,
Energy Environ. Sci. 2 (2009) 92–95.
C.-L. Ho thanks the Hong Kong Research Grants Council (PolyU
28] N. Koumura, Z.-S. Wang, S. Mori, M. Miyashita, E. Suzuki, K. Hara, J. Am. Chem.
Soc. 128 (2006) 14256–14257.
1
23021/17 P), the Science, Technology and Innovation Committee of
29] M. Elbing, A. Garcia, S. Urban, T.-Q. Nguyen, G.C. Bazan, Macromolecules 41
Shenzhen Municipality (JCYJ20180306173720000) and the Hong Kong
Polytechnic University (1-BE0Q and G-YBYZ) for their financial support.
(
2008) 9146–9155.
30] C.-Y. Chen, M. Wang, J.-Y. Li, N. Pootrakulchote, L. Alibabaei, C.-h. Ngoc-le, J.-
D. Decoppet, J.-H. Tsai, C. Gr ¨a tzel, C.-G. Wu, S.M. Zakeeruddin, M. Gr a¨ tzel, ACS
Nano 3 (2009) 3103–3109.
Appendix A. Supplementary data
[
31] F.-R. Dai, Y.-C. Chen, L.-F. Lai, W.-J. Wu, C.-H. Cui, G.-P. Tan, X.-Z. Wang, J.-
Ts. Lin, H. Tian, W.-Y. Wong, Chem. Asian J. 7 (2012) 1426–1434.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.jphotochem.2020.
[32] T. Bessho, E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H. Imai,
U. Rothlisberger, M.K. Nazeeruddin, M. Gr ¨a tzel, J. Am. Chem. Soc. 131 (2009)
5
930–5934.
1
12979.
[
[
33] G. Boschloo, A. Hagfeldt, Acc. Chem. Res. 42 (2009) 1819–1826.
34] X. Li, J. Yu, J. Low, Y. Fang, J. Xiao, X. Chen, J. Mater. Chem. A 3 (2015)
2
485–2534.
References
[
35] Y.-C. Chen, Y.-H. Chen, H.-H. Chou, S. Chaurasia, Y.S. Wen, J.T. Lin, C.-F. Yao,
Chem. Asian J. 7 (2012) 1074–1084.
[
[
[
1] N. Robertson, Angew. Chem. Int. Ed. 45 (2006) 2338–2345.
[36] Z. Ning, H. Tian, Chem. Commun. (Camb.) (2009) 5483–5495.
[37] S.-H. Lin, Y.-C. Hsu, J.T. Lin, C.-K. Lin, J.-S. Yang, J. Org. Chem. 75 (2010)
7877–7886.
2] W.-Y. Wong, C.-L. Ho, Acc. Chem. Res. 43 (2010) 1246–1256.
3] C.-H. Siu, L.T.L. Lee, P.-Y. Ho, C.-L. Ho, T. Chen, S. Suramitr, S. Hannongbua,
Z. Xie, M. Wei, W.-Y. Wong, Chem. Asian J. 12 (2017) 332–340.
[38] J. He, G. Benk o¨ , F. Korodi, T. Polívka, R. Lomoth, B. Åkermark, L. Sun, A. Hagfeldt,
V. Sundstr o¨ m, J. Am. Chem. Soc. 124 (2002) 4922–4932.
[
[
4] C.-H. Siu, L.T.L. Lee, S.-C. Yiu, P.-Y. Ho, P. Zhou, C.-L. Ho, T. Chen, J. Liu, K. Han,
W.-Y. Wong, Chem. Eur. J. 22 (2016) 3750–3757.
[39] Q. Li, Z. Jin, Z. Peng, Y. Li, S. Li, G. Lu, J. Phys. Chem. C 111 (2007) 8237–8241.
[40] Y. Li, C. Xie, S. Peng, G. Lu, S. Li, J. Mol. Catal. A Chem. 282 (2008) 117–123.
[41] Z. Jin, X. Zhang, Y. Li, S. Li, G. Lu, Catal. Commun. 8 (2007) 1267–1273.
[42] T. Shimidzu, T. Iyoda, Y. Koide, J. Am. Chem. Soc. 107 (1985) 35–41.
[43] L.L. Tinker, N.D. McDaniel, P.N. Curtin, C.K. Smith, M.J. Ireland, S. Bernhard,
Chem. Eur. J. 13 (2007) 8726–8732.
5] P.-Y. Ho, C.-H. Siu, W.-H. Yu, P. Zhou, T. Chen, C.-L. Ho, L.T.L. Lee, Y.-H. Feng,
J. Liu, K. Han, Y.H. Lo, W.-Y. Wong, J. Mater. Chem. C 4 (2016) 713–726.
6] T. Horiuchi, H. Miura, S. Uchida, Chem. Commun. (Camb.) (2003) 3036–3037.
7] T. Horiuchi, H. Miura, K. Sumioka, S. Uchida, J. Am. Chem. Soc. 126 (2004)
[
[
1
2218–12219.
[
[
8] T. Kitamura, M. Ikeda, K. Shigaki, T. Inoue, N.A. Anderson, X. Ai, T. Lian,
S. Yanagida, Chem. Mater. 16 (2004) 1806–1812.
[44] N. Manfredi, B. Cecconi, V. Calabrese, A. Minotti, F. Peri, R. Ruffo, M. Monai,
I. Romero-Ocana, T. Montini, P. Fornasiero, A. Abbotto, Chem. Commun. (Camb.)
52 (2016) 6977–6980.
9] K.R.J. Thomas, J.T. Lin, Y.-C. Hsu, K.-C. Ho, Chem. Commun. (Camb.) (2005)
4
098–4100.
[45] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M.
J. Bearpark, J. Heyd, E.N. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi,
J. Normand, K. Raghavachari, A.P. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, N.J. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin,
R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski,
[
[
[
[
[
10] D.P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt, L. Sun, Chem.
Commun. (Camb.) (2006) 2245–2247.
11] S.-L. Li, K.-J. Jiang, K.-F. Shao, L.-M. Yang, Chem. Commun. (Camb.) (2006)
2
792–2794.
12] Z.S. Wang, Y. Cui, K. Hara, Y. Dan-oh, C. Kasada, A. Shinpo, Adv. Mater. 19 (2007)
1
138–1141.
13] W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C. Pan, P. Wang,
Chem. Mater. 22 (2010) 1915–1925.
14] L. Zhang, X. Yang, W. Wang, G.G. Gurzadyan, J. Li, X. Li, J. An, Z. Yu, H. Wang,
B. Cai, A. Hagfeldt, L. Sun, ACS Energy Lett. 4 (2019) 943–951.
15] T. Hisatomi, J. Kubota, K. Domen, Chem. Soc. Rev. 43 (2014) 7520–7535.
16] A.J. Bard, M.A. Fox, Acc. Chem. Res. 28 (1995) 141–145.
¨
G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.
B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian, Inc., Wallingford, CT,
USA, 2009.
[
[
[
[
[
[
17] W.T. Eckenhoff, R. Eisenberg, Dalton Trans. 41 (2012) 13004–13021.
18] Y. Sun, J. Sun, J.R. Long, P. Yang, C.J. Chang, Chem. Sci. 4 (2013) 118–124.
19] B.F. DiSalle, S. Bernhard, J. Am. Chem. Soc. 133 (2011) 11819–11821.
20] T. Lazarides, T. McCormick, P. Du, G. Luo, B. Lindley, R. Eisenberg, J. Am. Chem.
Soc. 131 (2009) 9192–9194.
[46] B. Mennucci, E. Canc `e s, J. Tomasi, J. Phys. Chem. B 101 (1997) 10506–10517.
[47] E. Canc `e s, B. Mennucci, J. Tomasi, J. Chem. Phys. 107 (1997) 3032–3041.
[48] S. Miertu ˇs , E. Scrocco, J. Tomasi, Chem. Phys. 55 (1981) 117–129.
[49] R. Cammi, J. Tomasi, J. Comput. Chem. 16 (1995) 1449–1458.
[50] T. Lu, F. Chen, J. Comput. Chem. 33 (2012) 580–592.
9