Fine tuning of the photovoltaic properties of triarylethylene-bridged dyes by altering the position and proportion of phenyl/thienyl groups

Article abstract of DOI:10.1016/j.dyepig.2020.108630

Four triarylethylene (TAE) π-spacers have been applied for Y-shaped dyes in dye-sensitized solar cells (DSSCs) and their effects on photovoltaic performance have been fully investigated. All photosensitizers inherit the ability of three-dimensional TAE to weaken the π-π stacking in aggregation state, while their suppression on charge recombination in electrolytes presents a high reliance on the position and proportion of phenyl/thienyl groups embedded in TAE moieties. In iodine electrolytes, benzene-rich triphenylethene-based dye JY65 achieves the best open-circuit voltage of 739 mV, sharply higher than that of 614 mV for trithienylethene (TTE)-based dye JY63. Moreover, dyes JY64 and D-1 with hetero-TAE-bridge tend to show better power convention efficiency (PCE) stemming from the higher short-circuit current density and moderate open-circuit voltage. Whereas in cobalt electrolytes, the TTE-bridged JY63 attains a superior PCE of 5.84%, inversely surpassing the value of 5.28% based on D-1 under the same condition. Therefore, tuning the TAE skeletons via altering the position and proportion of aryl groups in it would bring in a promising design for anti-stacking sensitizers to meet the varied demand in DSSCs.

Full text of DOI:10.1016/j.dyepig.2020.108630

Journal Pre-proof
Fine tuning of the photovoltaic properties of triarylethylene-bridged dyes by altering
the position and proportion of phenyl/thienyl groups
Yi-Qiao Yan, Yi-Zhou Zhu, Jun Han, Pan-Pan Dai, Mao Yan, Jian-Yu Zheng
PII:
S0143-7208(20)31327-9
DOI:
https://doi.org/10.1016/j.dyepig.2020.108630
DYPI 108630
Reference:
To appear in: Dyes and Pigments
Received Date: 8 April 2020
Revised Date: 2 June 2020
Accepted Date: 9 June 2020
Please cite this article as: Yan Y-Q, Zhu Y-Z, Han J, Dai P-P, Yan M, Zheng J-Y, Fine tuning of the
photovoltaic properties of triarylethylene-bridged dyes by altering the position and proportion of phenyl/
thienyl groups, Dyes and Pigments, https://doi.org/10.1016/j.dyepig.2020.108630.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Elsevier Ltd. All rights reserved.

Author statement
Yi-Qiao Yan: Conceptualization, methodology, writing original draft
Yi-Zhou Zhu: Data analysis, writing reviewing and editing
Jun Han: Investigation and experimental assistance
Pan-Pan Dai: Investigation and experimental assistance
Mao Yan: Experimental assistance
Jian-Yu Zheng: Supervision

Fine tuning of the photovoltaic properties of triarylethylene-bridged
dyes by altering the position and proportion of phenyl/thienyl groups
Yi-Qiao Yan, Yi-Zhou Zhu,* Jun Han, Pan-Pan Dai, Mao Yan, Jian-Yu Zheng*
State Key Laboratory and Institute of Elemento-Organic Chemistry, College of
Chemistry, Nankai University, Tianjin 300071, China
E-mail: zhuyizhou@nankai.edu.cn; jyzheng@nankai.edu.cn
Abstract
Four triarylethylene (TAE) π-spacers have been applied for Y-shaped dyes in
dye-sensitized solar cells (DSSCs) and their effects on photovoltaic performance have
been fully investigated. All photosensitizers inherit the ability of three-dimensional
TAE to weaken the π-π stacking in aggregation state, while their suppression on
charge recombination in electrolytes presents a high reliance on the position and
proportion of phenyl/thienyl groups embedded in TAE moieties. In iodine electrolytes,
benzene-rich triphenylethene-based dye JY65 achieves the best open-circuit voltage
of 739 mV, sharply higher than that of 614 mV for trithienylethene (TTE)-based dye
JY63. Moreover, dyes JY64 and D-1 with hetero-TAE-bridge tend to show better
power convention efficiency (PCE) stemming from the higher short-circuit current
density and moderate open-circuit voltage. Whereas in cobalt electrolytes, the
TTE-bridged JY63 attains a superior PCE of 5.84%, inversely surpassing the value of
5.28% based on D-1 under the same condition. Therefore, tuning the TAE skeletons
via altering the position and proportion of aryl groups in it would bring in a promising
design for anti-stacking sensitizers to meet the varied demand in DSSCs.
Keywords:
Dye-sensitized
solar
cells;
Triarylethylene;
Thiophene;
Aggregation-induced emission

1. Introduction
As a new generation of photovoltaic technology, dye-sensitized solar cells
(DSSCs) have attracted worldwide attention because of their advantages in low cost,
easy fabrication and high efficiency [1-3]. Tremendous efforts have been engaged
over the past two decades in exploring highly efficient sensitizers for DSSCs.
Compared with Ru-complex and porphyrin-based dyes, metal-free organic sensitizers
have recently drawn increasing interest owing to their additionally
ecological-friendliness and flexibility in structural manufacture [4-8]. Typical
metal-free organic dyes generally adopt donor-π spacer-acceptor (D-π-A) structure
where intramolecular charge transfer (ICT) from donor to acceptor can be effectively
realized through π-spacer after excitation. Structure of π-spacer has proven to be vital
for the photovoltaic performance of dye-sensitized devices [9-13].
Thiophene-like π-spacers, such as oligothiophene, thieno[3,2-b]thiophene and
dithieno[3,2-b:2',3'-d]pyrrole, have been widely applied in DSSCs owing to their good
electron delocalization, which could broaden the visible absorption and thus increase
the photocurrent [14-17]. However, the expanded planarity of thiophene-involved
dyes often makes them inevitable to endure both excited-state self-quenching resulted
from the aggravated π-π stacking in aggregation, and charge recombination between
the injected electrons into semiconductor and electrolytes, which will generally harm
the photovoltaic performance [18-20]. In order to overcome such drawbacks, many
strategies have been taken into account, such as reducing the planarity of the main
chain [21,22], introducing flexible alkyl chains to the lateral position of π-spacer [23],
incorporating crown ether to cap the thiophene rings[24] or wrapping the skeleton
with special frameworks like rotaxanes and spiros [25-27]. In short, establishing a
steric framework would still be one of the promising methods to modify the

thiophene-based π-bridges.
To construct three-dimensional backbones, triphenylethene (TPE) and
tetraphenylethene derivatives, well-known as aggregation-induced emission (AIE)
molecules, have aroused our concern because of the specially twisted configuration
which can diminish the intermolecular π-π stacking by forming intermolecular
C-H···π interactions between phenyl rings in adjacent molecules at aggregation state,
and thus inhibit the excited-state self-quenching of the luminogens [28, 29]. Actually,
TPE units have been adopted as hindrance groups to the peripheral position of
indoline donors and resulted in a positive impact on the photovoltaic performance of
DSSCs [30-32]. A theoretical support has also been given by Voorhis and coworkers
in 2017 [33], where the restriction of the intramolecular motions and twisted
configuration of the TPE part at high dye coverage could be coincidentally similar to
the aggregation mode of general AIE luminogens. In other words, the special
aggregation conformation caused by TPE units would also help to overcome the
unfavorable stacking mode of the dyes on semiconductor surface in DSSCs.
However, the investigation of TPE building block as the π-spacer is very limited.
According to Han and coworkers’ studies, the twisted dye D-1 shows obviously better
photovoltaic performance than its planar counterpart owing to the enhanced
suppression on the unfavorable π-π stacking and intermolecular energy transfer in
aggregation [34], preliminarily revealing the potential significance of TPE-like units
for the design of anti-stacking π-linker in dyes. On the other hand, benzene-involved
π-spacers have also been applied for sensitizers, e.g., fluorenes,
benzothieno[3,2-b]benzothiophene or di(1-benzothieno)[3,2-b:2′,3′-d]pyrrole and
achieved much promoted photovoltaic performance [35-38]. Overall, it would be
promising to put electron-rich thienyl unit and benzene-rich TPE entity together to

create effective π-spacers, where the propeller-like triarylethylene (TAE) backbone is
anticipated to weaken the intermolecular π-π stacking, meanwhile, the proportion of
phenyl and thienyl groups embedded within TAE is expected to tune its electronic
communication and steric conformation.
Fig.1. Molecular structure of dyes JY63-65 and D-1.
In this paper, TAE structure was selected as π-bridge skeleton to build up
Y-shaped dyes where triphenylamine (TPA) was chosen as electron donor while
cyanoacrylic acid worked as electron acceptor (Fig.1). Thiophene was adopted to
pursue a good electron delocalization and hence better light response. Phenyl ring was
further introduced to replace partial (near-acceptor/near-donor replacement) or all the
thienyl groups to form a hetero-TAE-bridge or benzene-rich TPE, no additional
decoration of the main skeleton was employed so as to clearly understand the
relationship between TAE-based π-spacer and device’s photovoltaic performance.

Br
Br
S
Br
oxalyl chloride
DMF / DCM
S
S
S
S
Br
Br
COCl
Br
COOH
AlCl3
DCM
O
1
2, 44%, two steps
S
S
P(OEt)3
S
(HCHO)n, HBr
HOAc
(EtO)2(O)P
Br
Br
Br
3
4, crude product
TPA
TPA
N
B(OH)2
TPA
TPA
Br
Br
X
X
Y
(EtO)2(O)P
, t-BuOK / THF
X
X
X
X
Br
Pd(PPh3)4, K2CO3
Toluene / H2O
or 4, LDA / THF
O
O
5: X = S, 93%
6: X = -CH=CH-, 88%
Br
7: X = S,
8: X = S,
Y = S, 76%
Y = -CH=CH-, 34%
9: X = -CH=CH-, Y = S, 95%
10: X = -CH=CH-, Y = -CH=CH-, 95%
TPA
TPA
TPA
TPA
X
Y
X
X
Y
X
(HO)2
B
S
CHO
CNCH2CO2H / Piperidine
Chloroform / CH3CN
TPA=
N
Pd(dppf)Cl2, K2CO3
Toluene / CH3OH
S
S
CHO
COOH
NC
11: X = S,
12: X = S,
Y = S, 80%
Y = -CH=CH-, 93%
JY63: X = S,
JY64: X = S,
Y = S, 75%
Y = -CH=CH-, 48%
13: X = -CH=CH-, Y = S, 68%
D-1: X = -CH=CH-, Y = S, 94%
14: X = -CH=CH-, Y = -CH=CH-, 88%
JY65: X = -CH=CH-, Y = -CH=CH-, 82%
Scheme 1. Synthetic routes for dyes JY63-65 and D-1.
2. Results and discussion
2.1. Synthesis
The synthetic routes of four TAE-based dyes are shown in Scheme 1. Briefly, the
initial TPA-substituted aryl methanone 5 and 6 were attained through Suzuki reaction
of 4-(diphenylamino)phenylboronic acid with corresponding brominated aryl
methanone. Afterwards, the precursors 7, 8, 9 and 10 prepared from resultant
methanone with aryl phosphonate through Wittig-Horner reaction, were reacted with
the 5-formylthiophen-2-yl-boronic acid via Suzuki reaction to afford the relevant
aldehydes 11, 12, 13 and 14, respectively. Finally, the target dyes were obtained
through Knoevenagel condensations with cyanoacetic acid. More details of the
synthesis are provided in Supporting Information.

(b)
(a)
1.0
0.8
0.6
0.4
0.2
0.0
JY63
JY64
D-1
JY63
JY64
D-1
6
4
2
0
JY65
JY65
400
500
600
700
400
500
600
700
Wavelength / nm
Wavelength /nm
Fig. 2. UV-Vis absorption spectra of dyes JY63–65 and D-1 (a) in THF solutions and (b) on TiO2
films.
2.2. Photophysical properties
The UV-Vis absorption spectra of the dye JY63-65 in THF solution and on TiO2
films are shown in Fig. 2 and the detailed spectroscopic data are summarized in Table
1. The high energy absorption bands attributed to the π-π* transition lie around 350
nm similarly, while low energy absorption bands belongs to the ICT transition from
TPA donor to cyanoacrylic acid acceptor vary a lot from 430 to 490 nm, indicating
such modification on TAE-type π-spacer would considerably dominate the ICT
process. Specifically, the TTE-based dye JY63 presents the longest absorption λmax
while the TPE-based dye JY65 exhibits the shortest (488 nm for JY63 and 429 nm for
JY65). On the other hand, it is notable that dye D-1 where phenyl rings were
introduced at the near-donor position displays clearly red-shifted (23 nm) absorption
in comparison to dye JY64 where phenyl ring was incorporated at the near-acceptor
position of TAE, demonstrating that the former substitution mode is more beneficial
for ICT process (475 nm for D-1 and 452 nm for JY64). As a result, the ICT effect in
this class can be easily tuned by altering both the proportion and the position of
phenyl/thienyl rings in TAE moiety. After adsorbed onto TiO2 films, the ICT bands of
all these dyes present almost the same order as those in THF solution with about 50
nm bathochromical shifted onsets, which will favor the light harvesting ability of the
sensitizers in DSSCs.

Table 1
Optical and electrochemical properties of dyes JY63-65 and D-1.a
Eox/V
Ered/V
Dye
λ
max/nm
ε/104 M-1 cm-1
λTiO2/nm
E0-0/eV
(vs. NHE)
(vs. NHE)
488
452
475
429
3.53
3.55
3.43
5.12
461
447
453
405
1.91
2.08
2.05
2.25
-0.80
-0.91
-0.93
-1.10
JY63
JY64
D-1
1.11
1.17
1.12
1.15
JY65
a
λmax was attained from the absorption spectra in THF solution; Eox was determined on
dye-sensitized TiO2 films and calibrated with ferrocene (0.63 V vs. NHE) in THF containing 0.1
M tetra-n-butylammonium hexafluorophosphate (TBAPF6) as supporting electrolyte; E0–0 was
estimated from the onsets of UV-Vis spectra in THF solution; Ered = Eox−E0–0
.
Considering the fluorescence of dyes could be well reduced by the strong
electron-withdrawing cyanoacrylic acid groups owing to the solvent polarity effect
and ICT/TICT effect according to the previous studies [39-41], the aldehyde
precursors keeping similar main skeletons as relevant target dyes were selected
instead to estimate their intrinsic capabilities of suppressing the intermolecular energy
transfer in aggregation state. The fluorescence properties of aldehydes 11, 12, 13 and
14 were characterized in the diluted solutions of H2O/THF with different H2O
fractions (Fig S1 and S2). After the water fraction surpassing 50%, an obvious AIE
phenomenon could be observed in all these TAE-based derivatives, confirming that
theses TAE-based dyes could still maintain the beneficial anti π-π stacking ability of
the well-known twisted TPE building blocks.

-1.5
-1.0
-0.5
0.0
LUMO
-1.10
-0.93
-0.91
-0.80
-0.5
TiO
E
CB
2
0.4
-
-
0.5
I /I
3
1.0
1.11
1.12
1.15
1.17
HOMO
JY63
JY64
D-1
JY65
1.5
Fig. 3. HOMOs, LUMOs and energy gaps of dyes JY63–65 and D-1.
2.3. Electrochemical properties
Electrochemical properties of these four dyes-sensitized TiO2 films were carried
out by cyclic voltammetry (CV) calibrated with ferrocene (0.63 V vs. NHE) in THF
solution containing 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as
supporting electrolyte (Fig. S3). The schematic energy levels for JY63-65 are listed in
Fig. 3. The first oxidation potentials (EOX) relating to the highest occupied molecular
orbital (HOMO) levels of the dyes exhibit values around 1.14 V, rising from the
uniform electron donating ability of TPA donors. JY64 and JY65 present a somewhat
lower HOMO compared with JY63 and D1, respectively, indicating that replacing the
phenyl ring at the end of TAE which is close to acceptor by thiophene might be
advantageous for electron conjugation hence decrease the EOX. With the band gaps E0–
calculated from the onsets of UV-Vis spectra, the lowest unoccupied molecular
0
orbital (LUMO) levels were evaluated using Ered = E0-0 – Eox and present strong
correlation to the π-spacers. The TTE-involved JY63 gives the lowest LUMO while
benzene-rich JY65 exhibits the largest upshift. Dyes JY64 and D-1 featuring
hetero-TAE-bridge show medium LUMO levels and are very close to each other (–
0.91 and –0.93 V for JY64 and D-1, respectively). Notably, both HOMOs and
LUMOs of the four dyes can guarantee an efficient regeneration of oxidized dye by
iodine redox couples (0.4 V vs. NHE) and sufficient photo-excited electron injection

into TiO2 conduction band (–0.5 V vs. NHE).
2.4. Theoretical calculations
To gain insight into the molecular structure and electron distribution, the
geometries of dyes JY63-65 were optimized by density functional theory (DFT)
calculations at the B3LYP/6-31G* level (Table 2 and Fig. 4). For the near-donor aryl
ring π1 and near-acceptor aryl ring π3 which locate at the different sides of olefin
center in TAE bridge, the dihedral angle α (between π1 and the olefin plane) increases
in the order of JY63<JY64<D-1<JY65 while the dihedral angle γ (between the olefin
plane and π3) rises in the sequence of JY63<D-1<JY64<JY65, indicating that
replacing the thiophene with phenyl ring can induce a slightly enlarged distortion. For
the near-donor ring π2, all the dihedral angles β (between π2 and the olefin plane)
exhibit a considerable large distortion beyond 50°, which would play a crucial role in
building up the three-dimensional configuration. With respect to charge distribution,
the HOMOs in these four dyes mainly delocalize at the TPA donor parts and partial
TAE-based π-bridges, while the LUMOs diffuse on the cyanoacrylic acid acceptors
and also a portion of TAE moieties. Specifically, dye JY63 displays more overlap
between the HOMOs and LUMOs, aroused from the higher co-planarity of TTE
π-spacer. By contrast, the larger distortion of TPE π-linker provides dye JY65 a
remarkable charge separation. When coming to dyes JY64 and D-1 with
hetero-TAE-bridge, a medium scale between the charge overlap and separation can be
observed. Clearly, such type of charge distribution of four Y-shape dyes fits the need
for charge transfer from the excited sensitizers to the conduction band of TiO2.

Table 2
Dihedral angles in the geometry optimized dyes JY63-65 and D-1. a
N
N
1
2
3
S
NC
COOH
Dyes
JY63
JY64
D-1
α
β
γ
24.9°
26.7°
29.9°
32.7°
55.8°
50.7°
56.1°
52.1°
15.6°
25.5°
20.0°
28.5°
JY65
a
α/β/γ refers to the dihedral angles between the aromatic ring π1/π2/π3 and the central
ethylene in TAE moiety, respectively.
JY63
JY64
D-1
JY65
LUMO
HOMO
Fig. 4. The frontier molecular orbitals of JY63-65 and D-1 optimized at the B3LYP/6-31G* level.
2.5.Photovoltaic performance
The photovoltaic performances of DSSCs based on dyes JY63-65 were

measured under standard AM 1.5G illumination (100 mW·cm−2). The photocurrent
density-voltage (J–V) curves of the dyes in iodine electrolytes are plotted in Fig. 5a,
and detailed photovoltaic data are collected in Table 3. Surprisingly, the TTE-based
dye JY63 which has the broadest absorption range does not give the highest
short-circuit current density (JSC). In detail, the JSC of 8.75 mA cm-2 for dye JY63 is
only slightly higher than that of 8.13 mA cm-2 for TPE-based dye JY65 featuring the
narrowest light absorption, but dramatically lower than dyes JY64 and D-1 containing
hetero-TAE-bridge, implying that the photocurrent would not be dominated only by
the light harvesting region under this condition. Specifically, dye D-1 where the
phenyl rings are incorporated at the donor side displays superior JSC of 10.57 mA cm-2
than that of dye JY64 where the replacement occurs at the near-acceptor position,
agreeing with its relatively bathochromic absorption band. With respect to
open-circuit voltage (VOC), the TTE-contained dye JY63 endures the sharply lowest
VOC of 614 mV, dooming its power conversion efficiency (PCE) to only 3.67%. By
contrast, benzene-rich TPE-based dye JY65 attains the highest VOC of 739 mV,
leading the PCE up to 4.01%. With the good balance of VOC and JSC, two
hetero-TAE-embedded dyes show superior PCE than TTE-based dye JY63 and
TPE-based dye JY65, and dye D-1 gets the best PCE of 5.20%. In total, all the
benzene-blended TAE π-bridges work better than the TTE-based π-spacer on the
-
photovoltaic performance of DSSCs using I-/I3 redox shuttle. The order of VOC values
stands in accordance with the number of phenyl rings in TAE units, but the JSC shows
less reliance on the proportion of thienyl groups in π-spacers.

(a) 12
(b)100
JY63
JY64
D-1
JY63
JY64
D-1
10
80
JY65
JY65
8
6
4
2
0
60
40
20
0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
400
500
600
700
Voltage / V
Wavelength / nm
(c)
(d)100
12
JY63/iodine
D-1/iodine
JY63/cobalt
D-1/cobalt
80
10
8
60
40
20
0
6
JY63/iodine
D-1/iodine
JY63/cobalt
D-1/cobalt
4
2
0
400
500
600
700
0.0
0.2
0.4
0.6
0.8
Wavelength / nm
Voltage / V
Fig. 5. The J–V curves (a) and IPCE curves (b) of DSSCs based on JY63-65 and D-1 in iodine
electrolytes, the J–V curves (c) and the IPCE curves (d) of DSSCs based on JY63 and D-1 in
iodine and cobalt electrolytes.
The incident photon-to-current conversion efficiency (IPCE) spectra of DSSCs
-
using I-/I3 redox shuttle are presented in Fig. 5b. All these four TAE-based dyes cover
the IPCE response region in agreement with the tendency of their UV-Vis absorption
spectra. The bathochromic shift in IPCE response region can well explain the
promotion of JSC in the sequence of JY65<JY64<D-1 where the IPCE values
approximately approach to 70%. Different from that, due to the significantly
decreased IPCE less than 56%, TTE-based dye JY63 which has the broadest
absorption and most bathochromic IPCE response can only give an inferior JSC.

Table 3
Photovoltaic performance of JY63-65 and D-1.a
Dye
JY63
JY64
D-1
Electrolytes
V
OC[mV]
JSC[mA cm-2]
8.75±0.20
9.32±0.14
10.57±0.18
8.13±0.13
11.41±0.06
9.74±0.21
FF[%]
PCE[%]
3.67±0.03
4.38±0.08
5.20±0.05
4.01±0.02
5.84±0.03
5.28±0.02
614±5
685±5
721±5
739±5
736±10
817±10
68.3±0.8
68.6±1.2
68.3±1.2
66.8±0.6
69.5±1.2
66.4±0.6
Iodine
JY65
JY63
D-1
Cobalt
a An average value was obtained from three devices.
(a)
(b)
350
60
50
40
30
20
10
0
JY63
JY64
D-1
JY63
JY64
D-1
300
250
200
150
100
50
JY65
JY65
0
0
200
400
600
800
10-1
100
101
102
103
104
105
Z' / Ω
Frequency / Hz
50
(c) 160
(d)
JY63
D-1
JY63
D-1
140
120
100
80
40
30
20
10
0
60
40
20
0
10-1
100
101
102
103
104
105
0
100
200
300
400
500
Z' / Ω
Frequency / Hz
Fig. 6. Electrochemical impedance spectra (a) Nyquist plots and (b) Bode phase plots for the
DSSCs based on JY63-65 and D-1 in iodine electrolytes measured at −0.60 V bias in the dark, (c)
Nyquist plots and (d) Bode phase plots for the DSSCs based on JY63 and D-1 in cobalt
electrolytes measured at −0.72 V bias in the dark.
2.6. Electrochemical impedance spectroscopy
Electrochemical impedance spectroscopy (EIS) analysis in iodine electrolytes
was determined in the dark condition under the bias of –0.60 V within a frequency

range of 0.1 Hz-100 kHz. The EIS Nyquist and Bode phase plots are displayed in Fig.
6, and the detailed data are summarized in Table 4. The larger semicircle in Nyquist
plots represents the charge-transfer resistance at the TiO2/dye/electrolyte interface
(RCT). A smaller RCT means a faster charge recombination and thus induces a lower
VOC. The fitted RCT witnesses an distinguishable increase in the order of JY63＜JY64
＜D-1＜JY65, which is in accordance with the trend of VOC. Although all these
TAE-based dyes can maintain the three-dimensional structures, the capability of
-
preventing the charge recombination between I-/I3 redox couples and the injected
electron into the TiO2 presents high reliance on the position and proportion of the
phenyl groups blended in TAE moiety. TTE-based dye JY63 gives a relative low VOC
because of the insufficient hindrance to the electrolyte penetration through the dye
layer, which might be caused by the general sulfur-iodine interactions throughout the
whole chain. However, replacing only one of the thiophenes in TTE unit with phenyl
ring near the acceptor in dye JY64 can clearly promote the VOC, indicating that a
slightly increased distortion induced by the phenyl group with larger size can
effectively resist the charge back transfer from the semiconductor surface to the
electrolytes. On the other hand, the D-1 where the phenyl groups incorporated closely
to the donor position can further stem the electrolytes caused by the much
strengthened shielding effect. Finally, TPE-based dye JY65 can apparently minimize
the charge recombination, owing to the largest volume constructed by benzene-rich
TPE unit. Totally speaking, all the benzene-blended TAEs present better charge
barriers than TTE counterpart in iodine electrolytes. The model of the TAE units
-
blocking the I-/I3 redox shuttle near TiO2 surface is inferred in Fig. S4. Besides the
interfacial RCT, electron lifetime relates also to the VOC value, and a longer electron
lifetime stands for a lower dark current and thus a higher photovoltage. The electron

lifetime (τ) can be calculated from the peak frequency at lower frequency region in the
Bode phase plots by adopting the equation τ =1/2πf. The lifetime also meets an
elevation trend of JY63＜JY64＜D-1＜JY65, which can further explain the variation
of the VOC in DSSCs.
Table 4
Parameters obtained by fitting the impedance spectra of the DSSCs based on JY63-65 and D-1.
Dye
JY63
JY64
D-1
Electrolytes
Bias/V
RCT/Ω
64.2
f/Hz
22.459
7.931
7.080
5.482
71.72
25.76
τ/ms
7.1
325.2
413.1
753.5
88.9
20.1
22.5
29.0
2.22
6.18
Iodine
-0.60
JY65
JY63
D-1
Cobalt
-0.72
308.1
As investigated above, the presence of phenyl groups around the ethylene core
can effectively hinder the charge recombination, and an appropriate proportion of
thienyl groups embedded in TAE π-spacer can improve the photovoltaic performance
owing to the expansion of light harvesting region. Nevertheless, the absence of phenyl
group in TAE π-bridge leads to the unexpected loss on photocurrent in iodine
electrolytes to some extent. To deeply understand the charge barrier built by
2+/3+
TAE-spacers, the Co(phen)3
redox shuttle (0.61 V vs. NHE) was chosen for
further experiments. Dye JY63 which holds the intriguingly broadest absorption but
presents inferior JSC, and D-1 which covers the second broadest light harvesting but
exhibits the highest JSC in iodine electrolytes, were selected as the subjects. The J–V
curves of the DSSCs are shown in Fig. 5c and photovoltaic performances are
-
summarized in Table 3. In contrast to the results in I-/I3 redox couple, a superior JSC

of 11.41 mA cm-2 has been obtained for TTE-based dye JY63 against D-1 of 9.74 mA
cm-2. This variation of the JSC is further explained by the IPCE spectra displayed in
Fig. 5d, the maximum value of dye JY63 witnesses an approximate 10% elevation
ranging from 350 to 700 nm in comparison to that in iodine electrolytes, while dye
D-1 meets a slight decrease in whole absorption region. The above results could
attribute to the larger volume of Co(phen)32+/3+ mediator and less sulfur-electrolyte
interaction, demonstrating that the charge barrier established by the propeller-like
TTE unit is more suitable for blocking the penetration of cobalt complex than iodine
mediator [42,43]. Benefitted from the obviously enhanced JSC, TTE-based JY63
achieves higher PCE of 5.84% than that of 5.28% for D-1. The EIS Nyquist and Bode
phase plots are presented in Fig. 6, and the detailed data are listed in Table 4. The
order of RCT and τ matching with the sequence of corresponding VOC (JY63＜D-1)
indicates that the benzene-involved TAE π-spacer could still act as an ideal building
blocks to suppress the charge recombination in cobalt electrolytes.
3. Conclusion
In summary, four different TAEs were applied as π-bridge to construct
photosensitizer for DSSCs, and the π-bridge effect on the photovoltaic performance
has been fully investigated. In addition to the uniformly three-dimensional
configuration of TAE, both the proportion and position of phenyl/thienyl groups in
TAE moieties have been observed to pose a significant impact on the devices’
photovoltaic performance. In iodine electrolytes, the order of VOC agrees well with the
increasing proportion of phenyl rings in TAE-bridge. Suffering from the undesirable
charge recombination, TTE-based dye JY63 gives a poor VOC of 614 mV and
mediocre JSC of 8.75 mA cm−2 although it shows the broadest light harvesting. On the
contrary, benefitted from the largest volume of benzene-rich TPE-bridge working as

an ideal barrier to minimize the back charge transfer, dye JY65 achieved the highest
OC of 739 mV. A hetero-TAE-bridge has proven to be more effective to maintain the
V
balance between the suppression on charge recombination and the enhancement on
photocurrent, leaving JY64 and D-1 with better PCEs. Whereas in cobalt electrolytes,
JY63 with less-twisted TTE also can effectively retard the charge recombination and
thus make promoted photocurrent into reality. As a result, JY63 achieved higher PCE
of 5.84% than that of 5.28% for D-1 with hetero-TAE. In total, the three-dimensional
TAE-bridge provides a facile tool to tune the photovoltaic properties of dyes and a
balance design should be carefully taken in account to meet the individual
requirements in device.
4. Experimental
4.1. Materials and instruments
All the synthesis procedures were carried out under the argon atmosphere.
Solvents for reaction and other sections were re-distilled before use. The 1H NMR and
13C NMR spectra were characterized on Bruker 400 MHz spectrometer adopting TMS
as the internal standard. HR-MS data were collected on a Varian 7.0T FTMS. UV−Vis
spectra were scanned on a Varian Cary 300 Conc UV-Visible spectrophotometer.
Cyclic voltammetry, J−V curves and EIS experiments were detected on a Zennium
electrochemical workstation (Zahner Corporation).
4.2. Fabrication and characterization of DSSCs
The photoanodes were manufactured by screen printing the TiO2 paste
(Heptachroma Corporation) onto FTO conductive glasses (Nippon Sheet Glass, Japan,
15 Ω/sq) with 20 nm nanoparticles for transparent layer and then 200 nm
nanoparticles for scattering layer, respectively. Specially, the transparent layers for
photoanode in iodine electrolytes were printed for three times while for photoanode in

cobalt electrolytes were printed for one time. The active area of photoanode is 0.196
cm2. After that, the photoanodes were heated by a ramped temperature profile
(gradually heated to 450 °C and kept for 15 min, then heated to 500 °C and kept for
30 min). After cooling down to room temperature, the obtained photoanode were
immersed in a TiCl4 aqueous solution (40 mM) at 70 °C for 30 min and sintered again
at 500 °C for 40 min. At last, the resulting photoanodes were immersed into 0.3 mM
dye solution in THF for 12 h under dark. The Pt electrodes were prepared by
spin-coating the 0.02 M H2PtCl6 in iso-propanol onto the FTO conductive glasses and
then heated at 400 °C for 30 min. The iodine electrolytes consisted of 0.6 M
1-propyl-3-methylimidazolium iodide (PMII), 0.1 M LiI, 0.03 M I2, 0.1 M GuSCN
and 0.5 M 4-tert-butylpyridine (TBP) in acetonitrile/valeronitrile (v/v = 85/15). The
cobalt electrolytes were composed of 0.25 M [Co(II)(phen)3](PF6)2, 0.05 M
[Co(III)(phen)3](PF6)3, 0.5 M TBP and 0.1 M LiTFSI in acetonitrile. The DSSCs were
illuminated under a solar simulator (CHF-XM-500W, Trusttech Co. Ltd.) under 100
mW cm−2 irradiation calibrated by a standard silicon solar cell (91150V, Newport
Corporation). The IPCE was detected by a commercial setup (QTest Station 2000
IPCE Measurement System, CROWNTECH, USA).
4.3. Theoretical calculation methods
All the theoretical calculations were accomplished by the Gaussian 09 program
package. The ground-state geometries were fully optimized at the B3LYP/6-31G*
level in vacuum. The electron density distributions on HOMOs and LUMOs were also
calculated at the same level.
Notes
The authors declare no competing financial interest.

Acknowledgements
We are grateful to the National Natural Science Foundation of China (Nos.
21572108, 21971128 and 21935007) for their financial support.
Reference
[1] O’Regan B, Gratzel M. A low-cost, high-efficiency solar cell
colloidal TiO2 films. Nature 1991;353:737−40.
̈based on dye-sensitized
[2] Hagfeldt A, Boschloo G, Sun L, Kloo L, Pettersson H. Dye-sensitized solar cells. Chem Rev
2010;110:6595−663.
[3] Zhang S, Yang X, Numata Y, Han L. Highly efficient dye-sensitized solar cells: progress and
future challenges. Energy Environ Sci 2013;6:1443−64.
[4] Yella A, Lee H-W, Tsao HN, Yi C, Chandiran AK, Nazeeruddin MK, Diau EW-G, et al.
Porphyrin-sensitized solar cells with cobalt (II/III)-based redox electrolyte exceed 12% efficiency.
Science 2011;334:629−34.
[5] Mathew S, Yella A, Gao P, Humphry-Baker R, Curchod BFE, Ashari-Astani N, et al.
Dye-sensitized solar cells with 13% efficiency achieved through the molecular engineering of
porphyrin sensitizers. Nat Chem 2014;6:242−7.
[6] Eom YK, Kang SH, Choi IT, Yoo Y, Kim J, Kim HK. Significant light absorption
enhancement by
a
single heterocyclic unit change in the π-bridge moiety from
thieno[3,2-b]benzothiophene to thieno[3,2-b]indole for high performance dye-sensitized and
tandem solar cells. J Mater Chem A 2017;5:2297−308.
[7] Yao Z, Zhang M, Wu H, Yang L, Li R, Wang P. Donor/acceptor indenoperylene dye for
highly efficient organic dye-sensitized solar cells. J Am Chem Soc 2015;137:3799−802.
[8] Kakiage K, Aoyama Y, Yano T, Oya K, Fujisawa J-i, Hanaya M. Highly-efficient
dye-sensitized solar cells with collaborative sensitization by silyl-anchor and carboxy-anchor dyes.
Chem Commun 2015;51:15894−7.
[9] Zhou N, Prabakaran K, Lee B, Chang SH, Harutyunyan B, Guo P, et al. Metal-free
tetrathienoacene sensitizers for high-performance dye-sensitized solar cells. J Am Chem Soc
2015;137:4414−23.

[10] Cai N, Wang Y, Xu M, Fan Y, Li R, Zhang M, et al. Engineering of push-pull thiophene dyes
to enhance light absorption and modulate charge recombination in mesoscopic solar cells. Adv
Funct Mater 2013;23:1846−54.
[11] Xu M, Zhang M, Pastore M, Li R, Angelis FD, Wang P. Joint electrical, photophysical and
computational studies on D-π-A dye sensitized solar cells: the impacts of dithiophene
rigidification. Chem Sci 2012;3:976−83.
[12] Cai N, Li R, Wang Y, Zhang M, Wang P. Organic dye-sensitized solar cells with a cobalt
redox couple: influences of π-linker rigidification and dye-bath solvent selection. Energy Environ
Sci 2013;6:139−47.
[13] Feng Q, Jia X, Zhou G, Wang Z-S. Embedding an electron donor or acceptor into
Naphtho[2,1-b:3,4-b’]dithiophene based organic sensitizers for dye-sensitized solar cells. Chem
Commun 2013;49:7445−7.
[14] Zhang X-H, Cui Y, Katoh R, Koumura N, Hara K. Organic dyes containing
thieno[3,2-b]indole donor for efficient dye-sensitized solar cells.
2010;114:18283−90.
J Phys Chem C
[15] Liu J, Li R, Si X, Zhou D, Shi Y, Wang Y, Jing X, Wang P. Oligothiophene dye-sensitized
solar cells. Energy Environ Sci 2010;3:1924−8.
[16] Yang H-Y, Yen Y-S, Hsu Y-C, Chou H-H, Lin JT. Organic dyes incorporating the
dithieno[3,2-b:2’,3’-d]thiophene moiety for efficient dye-sensitized solar cells. Org Lett
2010;12:16−9.
[17] Qian X, Zhu Y-Z, Chang W-Y, Song J, Pan B, Lu L, et al. Benzo[a]carbazole-based donor−π–
acceptor type organic dyes for highly efficient dye-sensitized solar cells, ACS Appl Mater
Interfaces 2015;7:9015−22.
[18] Feng Q, Zhang Q, Lu X, Wang H, Zhou G, Wang Z-S. Facile and selective synthesis of
oligothiophene-based sensitizer isomers: an approach toward efficient dye-sensitized solar cells.
ACS Appl Mater Interfaces 2013;5:8982−90.
[19] Shen P, Liu Y, Huang X, Zhao B, Xiang N, Fei J, et al. Efficient triphenylamine dyes for solar
cells: effects of alkyl-substituents and π-conjugated thiophene unit. Dyes Pigments
2009;83:187−97.

[20] Baumann A, Cheema H, Sabuj MA, McNamara LE, Zhang Y, Peddapuram A, et al. Iodine
binding with thiophene and furan based dyes for DSCs. Phys Chem Chem Phys
2018;20:17859−70.
[21] Dong H, Liang M, Zhang C, Wu Y, Sun Z, Xue S. Twisted fused-ring thiophene organic
dye-sensitized solar cells. J Phys Chem C 2016;120:22822−30.
[22] Cheng H, Wu Y, Su J, Wang Z, Ghimire RP, Liang M, et al. Organic dyes containing
indolodithienopyrrole unit for dye-sensitized solar cells. Dyes Pigments 2018;149:16−24.
[23] Gao H-H, Qian X, Chang W-Y, Wang S-S, Zhu Y-Z, Zheng J-Y, Oligothiophene-linked D-π-A
type phenothiazine dyes for dye-sensitized solar cells. J Power Sources 2016;307:866−74.
[24] Lee M-W, Kim J-Y, Lee D-H, Ko MJ. Novel D-π-A organic dyes with
thieno[3,2-b]thiophene-3,4-ethylenedioxythiophene unit as a π-bridge for highly efficient
dye-sensitized solar cells with long-term stability. ACS Appl Mater Interfaces 2014;6:4102−8.
[25] Hua T, Huang Z-S, Cai K, Wang L, Tang H, Meier H, et al. Phenothiazine dye featuring
encapsulated insulated molecular wire as auxiliary donor for high photovoltage of dye-sensitized
solar cells by suppression of aggregation. Electrochim Acta 2019;302:225−33.
[26] Hua T, Zhang K, Huang Z-S, Wang L, Tang H, Meier H, et al. Effect of structural engineering
of π-spacers on anti-aggregation of D–A–π–A dyes. J Mater Chem C 2019;7:10379−88.
[27] Wu Z-S, Guo W-J, Zhang J, Liu Y-D, Song X-C, Jiang Y, et al. Novel
4,4’-bis(alkylphenyl/alkyloxyphenyl)-2,2’-bithiophene bridged cyclic thiourea functionalized
triphenylamine sensitizers for efficient dye-sensitized solar cells. Sol Energy 2019;186:1–8.
[28] Mei J, Leung NLC, Kwok RTK, Lam JWY, Tang BZ. Aggregation-induced emission:
together we shine, united we soar! Chem Rev 2015;115:11718−940.
[29] Zhang G-F, Wang H, Aldred MP, Chen T, Chen Z-Q, Meng X, et al. General Synthetic
Approach toward General synthetic approach toward geminal-substituted tetraarylethene
fluorophores with tunable emission properties: x-ray crystallography, aggregation-induced
emission and piezofluorochromism. Chem Mater 2014;26:4433−46.
[30] Ito S, Zakeeruddin SM, Humphry-Baker R, Liska P, Charvet R, Comte P, et al.
High-efficiency organic-dye-sensitized solar cells controlled by nanocrystalline-TiO2 electrode
thickness. Adv Mater 2006;18:1202–5.

[31] Jin MY, Kim B-M, Jung HS, Park J-H, Roh D-H, Nam DG, et al. Indoline-based molecular
engineering for optimizing the performance of photoactive thin films. Adv Funct Mater
2016;26:6876–87.
[32] Park J-H, Nam DG, Kim B-M, Jin MY, Roh D-H, Jung HS, et al. Planar D-D-π-A organic
sensitizers for thin film photoanodes. ACS Energy Lett 2017;2:1810−7.
[33] Vaissier V, Voorhis TV. Geometry of molecular motions in dye monolayers at various
coverages. J Phys Chem C 2017;121:12562−8.
[34] Numata Y, Islam A, Chen H, Han L. Aggregation-free branch-type organic dye with a twisted
molecular architecture for dye-sensitized solar cells. Energy Environ Sci 2012;5:8548–52.
[35] Zhang H, Iqbalb Z, Chena Z-E, Hong Y. Synthesis of novel sensitizers with a linear
conjugated di(1-benzothieno)[3,2-b:2’,3’-d]pyrrole unit for dye-sensitized solar cells. Dyes
Pigments 2019;162:89–96.
[36] Aumaitre C, Rodriguez-Seco C, Jover J, Bardagot O, Caffy F, Kervella Y, et al. Visible and
near-infrared organic photosensitizers comprising isoindigo derivatives as chromophores:
synthesis, optoelectronic properties and factors limiting their efficiency in dye solar cells. J Mater
Chem A 2018;6:10074–84.
[37] Capodilupo AL, Fabiano E, Marco LD, Ciccarella G, Gigli G, Martinelli C, et al.
Benzothieno[3,2-b]benzothiophene-based organic dyes for dye-sensitized solar cells. J Org Chem
2016;81:3235–45.
[38] Shen P, Tang Y, Jiang S, Chen H, Zheng X, Wang X, et al. Efficient triphenylamine-based
dyes featuring dual-role carbazole, fluorene and spirobifluorene moieties. Org Electron
2011;12:125–35.
[39] Chen C, Liao J-Y, Chi Z, Xu B, Zhang X, Kuang D-B, et al. Effect of polyphenyl-substituted
ethylene end-capped groups in metal-free organic dyes on performance of dye-sensitized solar
cells. RSC Adv 2012;2:7788–97
[40] Tong H, Dong Y, Häußler M, Hong Y, Lam JWY, Sung HH-Y, et al. Molecular packing and
aggregation-induced emission of 4-dicyanomethylene-2,6-distyryl-4H-pyran derivatives. Chem
Phys Lett 2006;428:326−30.
[41] Hu R, Lager E, Aguilar-Aguilar A, Liu J, Lam JWY, Sung HHY, et al. Twisted intramolecular

charge transfer and aggregation-induced emission of BODIPY derivatives. J Phys Chem C
2009;113:15845−53.
[42] Yang J, Ganesan P, Teuscher J, Moehl T, Kim YJ, Yi C, et al. Influence of the donor size in
D−π−A organic dyes for dye-sensitized solar cells. J Am Chem Soc 2014;136:5722−30.
[43] Zhang M, Liu J, Wang Y, Zhou D, Wang P. Redox couple related influences of π-conjugation
extension in organic dye-sensitized mesoscopic solar cells. Chem Sci 2011;2:1401–6.

Highlights:
1. Effect of TAE-spacer on photovoltaic performance has been systematically
explored.
2. All the TAE-based dyes show capability to weaken the π-π stacking in aggregation.
3. In iodine electrolytes, benzene-blended TAE-based dyes exhibit better
performance.
4. In cobalt electrolytes, thiophene-rich JY63 inversely achieves superior PCE.

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: