Double Fence Porphyrins that are Compatible with Cobalt(II/III) Electrolyte for High-Efficiency Dye-Sensitized Solar Cells

Article abstract of DOI:10.1002/anie.202013964

A series of new double fence porphyrin dyes bJS1–bJS3, with eight long alkoxyl chains attached to four β-phenyl groups, have been designed and synthesized. The single fence meso-substituted counterparts mJS1–mJS3 were also prepared as reference dyes. Dyes bJS1–bJS3 and mJS1–mJS3 exhibit power conversion efficiencies of 8.03–10.69 % and 2.33–6.69 %, respectively. Based on photovoltaic studies, the remarkable cell performance of double fence porphyrin sensitizers can be attributed to reduced dye aggregation and a decreased charge-recombination rate. Notably, porphyrins bJS2 and bJS3 exhibit better efficiency than the benchmark YD2-o-C8 (9.83 % in this work), demonstrating that the double fence structure is a promising design strategy for efficient porphyrin sensitizers in high-performance DSSCs.

Full text of DOI:10.1002/anie.202013964

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Double Fence Porphyrins that are Compatible with CoII/III
Electrolyte for High Efficiency Dye-Sensitized Solar Cells
Authors: Ching-Chin Chen, Jia-Sian Chen, Vinh Son Nguyen, Tzu-
Chien Wei, and Chen-Yu Yeh
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202013964
Link to VoR: https://doi.org/10.1002/anie.202013964

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Double Fence Porphyrins that are Compatible with CoII/III
Electrolyte for High Efficiency Dye-Sensitized Solar Cells
Ching-Chin Chen,[a][b] Jia-Sian Chen,[a] Vinh Son Nguyen,[b] Tzu-Chien Wei,*[b] Chen-Yu Yeh*[a]
[a]
[b]
Dr. C.-C. Chen, J.-S. Chen, Prof. C.-Y. Yeh
Department of Chemistry, i-Center for Advanced Science and Technology (i-CAST), Innovation and Development Center of Sustainable Agriculture
(IDCSA)
National Chung Hsing University
No. 145, Xingda Rd., South Dist., Taichung City 402, Taiwan
E-mail: cyyeh@dragon.nchu.edu.tw
Dr. C.-C Chen, V. S. Nguyen, Prof. T.-Z. Wei
Department of Chemical Engineering
National Tsing-Hua University
No. 101, Sec. 2, Guangfu Rd., East Dist., Hsinchu City 300, Taiwan
E-mail: tcwei@mx.nthu.tw
Supporting information for this article is given via a link at the end of the document.
Abstract: A series of new double fence porphyrin dyes bJS1−bJS3
with eight long alkoxyl chains attached to the four -phenyl groups
have been designed and synthesized. The single fence meso-
substituted counterparts mJS1−mJS3 were also prepared as
reference dyes. Dyes bJS1−bJS3 and mJS1−mJS3 exhibit power
conversion efficiency of 8.03%−10.69% and 2.33%−6.69%,
respectively. The remarkable cell performance of double fence
porphyrin sensitizers can be attributed to reduced dye aggregation
and decreased charge recombination rate based on photovoltaic
studies. It should be noted that porphyrins bJS2 and bJS3 exhibit
better efficiency than the benchmark YD2-o-C8 (9.83% in this work),
demonstrating that the double fence structure is a promising design
extensively studied to enhance photovoltaic performance of
DSSCs in the past decades.[5] As sunlight absorber in DSSCs, the
sensitizer is a crucial component, of which its properties can be
tuned by molecular modification to achieve high efficiency of the
devices. In general, there are four types of sensitizers including
ruthenium-based complexes, metal-free dyes, porphyrins, and
natural pigments.[6] To date, the benchmark PCE of DSSCs was
reported to be > 12% using porphyrin-based photosensitizers
such as SM315, SGT-021, and GY50 under illumination of
standard AM 1.5G simulated sunlight (1000 W/m2).[5d,7]
Porphyrins and their derivatives have intense Soret band (400
~ 500 nm) and moderate Q-band (500 ~ 600 nm) in the visible
strategy for efficient porphyrin sensitizers in high performance DSSCs. region[8] and the absorption broadness and strength of porphyrins
can be readily tailored by molecular engineering to enhance their
light-harvesting ability. The superior absorption properties of
porphyrins make them promising sensitizers for DSSCs. However,
porphyrin-based sensitizers had been suffering from low device
Introduction
performance (PCE < 7.1%) until YD-series of porphyrins were
The use of renewable energy resources has been an important
issue for sustainable development of human society. Among all
sustainable energy resources, solar power is the most abundant
one that can provide more than enough energy for global energy
demand. Thus, conversion of solar energy to electricity or
chemical energy has attracted much attention over the past
decades. Currently, crystalline silicon solar cells which show high
reported.[9] With the particular donor--acceptor (D--A)
molecular structure to enhance light-harvesting and electron
injection efficiency, YD2 reached a PCE as high as 11%, which is
comparable with the most efficient ruthenium-based dyes such as
N719 and black dye for DSSCs.[10] Later, further modification of
dye YD2 was performed by introduction of long alkoxy groups to
the ortho-positions of meso-phenyls of the porphyrin ring to afford
dye YD2-o-C8 (Figure 1), which is compatible with CoII/IIItris(2,2’-
power conversion efficiency (PCE) up to 25% hold 90% of global
market share of commercial solar cells.[1] However, the material-
bipyridyl) redox shuttle, leading to a remarkable PCE of 11.9% for
and energy-cost process as well as long energy payback time for
the device.[5j] Sterically hindered long alkoxy chains not only
manufacturing of silicon-based solar cells have prompted
reduce aggregation of dye molecules but also suppress charge
researchers to develop new types of photovoltaic technology.[2]
recombination between TiO2 and electrolyte, leading to
Dye-sensitized solar cells (DSSCs) which belong to a new
generation of photovoltaic techniques were invented by Grätzel in
1991.[3] Compared to silicon-based solar cells, DSSCs have
enhancement of short-circuit photocurrent (JSC) and open-circuit
voltage (VOC), thus improved PCE of the device. Meanwhile, to
improve the cell performance, various strategies for devices of
several advantages such as feasible fabrication process, low
porphyrins other than the D--A structural design have been
production cost, environmental friendliness, colorfulness, high
reported, such as co-sensitization,[5j,11] broadening absorption
efficiency under dim light and high-flexibility.[4] The device of a
spectra,[5e,12] modification of anchoring groups,[13] donor and
DSSC consists of a working electrode, a sensitizer, electrolytes,
acceptor unit,[7a,14] and co-adsorption.[15] All these molecular
and
a counter electrode. These components have been
design strategies showed more or less improvement on
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 1. Structure of efficient porphyrin-based sensitizers.
PCE of DSSCs. However, introduction of long-chain alkoxy
groups and D--A structure is the essential requirements of an
efficient porphyrin dye as cobalt electrolytes are used. Compared
to dye GY50 , the slightly better performance of SM315 indicates
that introduction of more alkoxy groups to a bulky donor moiety is
a useful strategy to further improve PCE and similar results were
also reported for XW- and SGT- series of porphyrins.[14a,16] Based
on the molecular structures, all these efficient sensitizers share a
common porphyrin core structure with the porphyrin ring being
wrapped by four long alkoxyl chains, which would form a
hydrophobic environment on the surface of TiO2 to impede the
approach of charged electrolytes. In such a case, the meso-
phenyls with long alkoxyl chains can be considered as an effective
“fence” to suppress dye aggregation and charge recombination.
To further attenuate charge recombination and molecular
aggregation, we designed novel “double fence” porphyrin
sensitizers bJS1, bJS2 and bJS3 with four -phenyl rings
substituted with long alkoxyl groups at the ortho-positions as
shown in Figure 1, and the architecture of single and double fence
porphyrin dyes are depicted in Figure 2. In addition, anthracene-
containing triarylamino group was employed as a donor with dual
functions, electron-donating and absorption-broadening.[17]
Incorporation of benzotriazole (BTA) and benzothiadiazole (BTD)
unit in acceptor moiety for bJS2 and bJS3, respectively, causes
further red-shift of the absorption, thus enhancing light-harvesting
capability. Introduction of BTD and BTA unit in their single fence
analogues mJS dyes were also synthesized as references to
better understand the effect of -substituted alkoxy-phenyl groups.
Based on the photovoltaic studies, double fence porphyrins
exhibited much higher JSC and Voc than single fence porphyrins
using CoII/IIItris(2,2’-bipyridyl) as redox electrolyte. Thus, bJS2
and bJS3 achieved remarkable PCE of 10.69% and 10.42% while
mJS2 and mJS3 only exhibited PCE of 3.48% and 2.33% under
illumination of AM 1.5G simulated sunlight. It is noteworthy that
porphyrins bJS2 and bJS3 show higher efficiency as compared
to benchmark YD2-o-C8 ( = 9.83%) under similar conditions.
Figure 2. Architecture of (a) single fence and (b) double fence porphyrins.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Scheme 1. (a) Chemical structures of donor moiety, acceptor moiety, and porphyrin precursors. (b) Synthetic route of novel -substituted porphyrin. (i) n-BuLi,
trimethylborate, THF. (ii) Pd(PPh3)4, 2 M Na2CO3(aq), 1,4-dioxane. (iii) Dimethoxymethane, BF3·Et2O, CH2Cl2. (iv) KOH, (CH2O)2. (v) 3-(triisopropylsilyl)-1-propynal,
BF3·Et2O, CH2Cl2. (vi) Zn(OAc)2·2H2O, CH2Cl2/ MeOH.
and meso-substituted porphyrin precursor 5 were prepared as
previously reported.[4c,18] The key intermediate pyrrole 10 was
synthesized via Suzuki coupling using bromopyrrole 9 and
boronic acid 8 which were prepared as reported procedures.[18a,19]
Results and Discussion
Most of efficient porphyrins are based on a 5,10,15,20-meso-
The reaction of pyrrole 10 with dimethoxymethane afforded
dipyrrylmethane 11, which was then decarboxylated under
alkaline conditions to give 12. Condensation of 12 with 3-
(triisopropylsilyl)-1-propynal[20] followed by oxidation and
metalation afforded -substituted zinc porphyrin 6. The mJS-
series and bJS-series dyes were synthesized via Sonogashira
coupling using the corresponding porphyrin precursors, which
substituted structure because structural modification at the meso-
positions of porphyrins is more accessible than at the -positions.
As mentioned, efficient porphyrin sensitizers have a commonly
used core, i.e., 5,15-bis(2,6-dialkoxyphenyl)porphyrin, and the
porphyrin ring is wrapped by four long alkoxyl chains. To improve
the light-harvesting capability of dyes, and thus the cell
performance of devices, the most widely used strategy is to
extend the -conjugation by insertion of acetylene and/or a variety
of aromatic units to the D--A architecture. However, such a
strategy induces more severe molecular aggregation so that
excess co-adsorbent such as CDCA has to be employed for
device fabrication in order to enhance the cell performance.[17a]
Obviously, introduction of more long alkoxyl chains to wrap the
porphyrin ring would outperform previously used system with only
four alkoxyl chains. Therefore, in this work judiciously tailored
double fence porphyrins were designed and synthesized with
aims of reducing dye aggregation and impeding charge
recombination. The synthetic route of novel -substituted zinc
porphyrin 6 is shown in Scheme 1 and the synthetic procedures
are detailed in Supporting Information. Anthracene-containing
triphenyl amine donor moiety 1, benzothiatrizole-containg
acceptor moiety 3, benzothiadizole-containg acceptor moiety 4,
were obtained by desilylation of
5 and 6 with tetra-n-
butylammonium fluoride (TBAF), respectively, with appropriate
donor/acceptor moieties.
The UV-vis spectra of mJS and bJS porphyrin sensitizers in
THF are displayed in Figure 3, and the absorption and emission
data are collected in Table 1. Energy levels of porphyrin frontier
orbitals can be perturbed by meso- and -substituents, thus the
excitation energy of porphyrin would be modulated according to
the Gouterman’s four-orbital model.[8] These new porphyrin dyes
exhibit more red-shifted and broadened absorptions with respect
to simple porphyrins such as 5,10,15,20-tetraarylporphyrin and
the Soret band is red-shifted (by 10-30 nm) and Q-bands are
slightly blue-shifted (by 6-7 nm) for the double fence bJS
porphyrins as compared to those of single fence mJS dyes.
Similar to LWP14 reported by Lin and coworkers,[17a] these
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. Absorption spectra of (a) mJS and (b) bJS dyes.
anthracene-containing porphyrin dyes show a strong and broad
Soret band in a range of 462 to 492 nm and a moderate Q-band
in a range of 673 to 698 nm, in which their Soret and Q bands are
both more red-shifted with respect to anthracene-free high
efficiency porphyrins SM315 and GY50 owning to more expended
-system.[5d,5e] Among these dyes, bJS2 and mJS2 have the
highest maximum molar absorption coefficient compared to their
corresponding series of porphyrins, respectively. Insertion of
bridge 2,1,3-benzothiadizole (BTD) or 1,2,3-benzotriazole (BTA)
between benzoic acid and acetylene moieties causes red-shift of
both Soret and Q band, thus increasing the light harvesting
capability. Such a phenomenon has also been observed for GY50
and SM315.
To evaluate energy levels of frontier orbitals, cyclic voltammetry
(CV) was employed to investigate the electrochemical properties
of mJS- and bJS- series of dyes using THF and 0.1 M
tetrabutylammonium hexafluorophosphate (TBAPF6) as solvent
and supporting electrolyte, respectively. The results are
presented in Figure 4 and Table 1. The energy levels of highest
occupied molecular orbital (EHOMO) were determined with redox
potential (E1/2(ox)) of the first oxidation for each sensitizer using
ferrocene/ferrocenium (Fc/Fc+, E1/2 = 0.63 V vs. normal hydrogen
electrode, NHE) as an internal standard. Energy levels of the first
excited states of sensitizers were estimated from EHOMO and
optical bandgap (E0-0), which was obtained from the cross section
of absorption and emission spectra, by the equation ES1 = EHOMO
– E0-0. The optical band gap of dye mJS2, mJS3, bJS2, and bJS3
are in a range of 1.74 to 1.79 eV, which is smaller than that of
GY50 (1.81 eV) ascribed to the extended -conjugation by
incorporating acetylene and anthracene unit to the donor moiety,
whereas dye mJS1(1.81 V vs. NHE) and bJS1(1.83 V vs. NHE)
built without electron-deficient BTD or BTA unit in acceptor moiety
have higher optical band gap. In comparison with mJS dyes, the
HOMO and S1 (the first excited state) of bJS dyes are destabilized
by only 0.03 eV and 0.05 eV, respectively, indicating the numbers
and positions of phenyl substitution only slightly disturb the
electronic structure of porphyrins because the dihedral angles
between porphyrin and phenyl rings are in the range 65-90
degrees. The HOMO levels of mJS- and bJS- series (0.91 to 0.82
V vs. NHE) are lower than those of GY50 and YD2-o-C8 (0.79
and 0.82 V vs. NHE), providing enough driving force for dye
regeneration. Furthermore, all the first excited levels are
considerably higher than the conduction band of TiO2. Thus,
efficient electron injection can be expected.
Table 1. Optical and Electrochemical Properties of Sensitizers mJS and bJS.
[a]
[b]
[c]
[d]
Absorption max, [a]
Emission max
(nm)
EHOMO
ES
1
E0-0
dye
(nm), [(105 M-1 cm-1)]
(V vs. NHE)
(V vs. NHE)
(eV)
1.81
1.77
1.74
1.83
1.79
1.76
mJS1
mJS2
mJS3
bJS1
bJS2
bJS3
462[1.91], 679[0.67]
471[2.20], 692[0.84]
472[1.36], 697[0.85]
471[1.59], 673[0.65]
484[2.12], 685[0.89]
492[1.21], 691[0.91]
700
709
734
685
699
723
0.91
-0.90
0.85
-0.92
0.86
-0.88
0.87
-0.96
0.82
-0.97
0.83
-0.93
[a ]Wavelength maxima for absorption (including Soret and Q- band) and emission are record in THF at 25 oC. [b] Determined by cyclic voltammetry (CV) in
THF containing 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as electrolyte and was calibrated with ferrocene/ferrocenium (Fc/Fc+) as internal
reference at 25 oC. [c] Energy level of first excited state (S1) was calculated by following equation: ES = EHOMO – E0-0. [d] Determined from the intersection of
1
normalized absorption and emission spectra in THF.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
porphyrin dyes was theoretically investigated via time-dependent
DFT (TDDFT) calculation at the same level, the result was
collected in Table S1 to Table S6. The TDDFT calculations shows
that the Soret and Q bands of all these porphyrin dyes are
dominated by HOMO-2 to LUMO+1 and HOMO to LUMO
transitions, respectively, regardless the type and position of
peripheral substituents. Four selected frontier orbitals including
HOMO-2, HOMO, LUMO, and LUMO+1 of mJS1 and bJS1 are
shown in Figure S33. The electron-donating -phenyl substituents
would raise the energy level of HOMO-2 and LUMO, which have
significant electron density on the -carbons, for bJS dyes.
Similarly, the HOMO is raised in mJS dyes owing to electron-
donating meso-phenyl groups as shown in Figure S33. Therefore,
bJS dyes would have red-shifted Soret band and blue-shifted Q
bands, consistent with what have been observed in UV-vis
spectra.
Figure 4. Schematic energy level diagram of mJS1, mJS2, mJS3, bJS1, bJS2,
and bJS3 based on their electrochemical and optical data.
It is well known that porphyrin sensitizers wrapped by long
alkoxy groups exhibit better device performance with
[Co(bpy)3]3+/2+ than I-/I3 redox couple because of significantly
-
To gain further insight into the ground state geometry and
electron density distribution of the porphyrins, density functional
theory (DFT) calculations were performed on model porphyrins
and the results are displayed in Figure S32. Upon geometry
optimization, the porphyrin macrocycle remains planar for both
mJS and bJS series of porphyrins because the dialkoxylphenyl
groups at meso- or -positions are far away from coplanar to the
porphyrin ring (dihedral angles between phenyl ring and porphyrin
core: ~90o for mJS and 65o−70o for bJS) so that no steric
interaction between phenyl groups was observed. As expected,
the BTD and BTA unit adopt nearly coplanar conformation with
the porphyrin core, providing effective -conjugation of the donor-
-acceptor backbone. The dihedral angle between anthracene
and porphyrin plane are around 29o in all mJS and bJS dyes
ascribed to the steric interaction between 1,2-H atoms of
anthracene and the two closest -H atoms of porphyrin ring, which
is similar to that in Mg-TIPSTEP-anthryl (ca. 41o), a magnesium
porphyrin containing ethynyl-bridged anthracene unit reported by
Marsuo et al.[21] The optimized structures are consistent with the
red-shifted and broadened absorption spectra due to effective
electronic interaction among donor, porphyrin core, and acceptor
in all these new dyes via acetylene bridges. The calculated
HOMO−LUMO gaps obtained from DFT calculations are 2.27 eV
for YD2-o-C8, 1.92 eV for mJS1, 1.90 eV for mJS2, 1.84 eV for
mJS3, 1.96 eV for bJS1, 1.96 eV for bJS2, and 1.88 eV for bJS3,
respectively, and the trend of calculated HOMO−LUMO gaps are
consistent with experimental HOMO−LUMO gaps obtained from
optical data. The orbital distribution of HOMO is mainly populated
on the electron-donating group and slightly on the porphyrin
meso-carbons and nitrogens for all these new porphyrins. On the
other hand, the LUMO are mainly localized on porphyrin ring and
acceptor moiety in all porphyrins, in which BTD unit has more
contribution than BTA unit. Therefore, BTD-containing mJS3 and
bJS3 have lower LUMO energy level than BTA-incorporated
mJS2 and bJS2. It is worthy to note that the LUMO are also
populated on the acetylene bridge in mJS1 and bJS1 series,
which leads to lower LUMO energy level than YD2-o-C8, implying
that the acetylene bridge not only destabilizes HOMO but also
stabilizes LUMO of porphyrin sensitizers. As a result, a
decreasing HOMO−LUMO gap is observed for these new
porphyrins. Further, the excitation behavior of these new
suppressed charge recombination process between TiO2 electron
and [Co(bpy)3]3+. For most of efficient porphyrin sensitizers,
introduction of long alkoxy groups to the ortho-positions of the two
opposite meso-phenyls of porphyrin macrocycle provides an
efficient “fence” to reduce dye aggregation and impede mass
transport of [Co(bpy)3]3+/2+ to the TiO2 surface. In bJS-series,
phenyls with long alkoxy groups are introduced to the opposite -
carbons of porphyrin macrocycle to form a “double fence”
structure with expectation of better suppression of molecular
aggregation and charge recombination. Therefore, photovoltaic
performance of mJS- and bJS-based DSSCs were fabricated
using [Co(bpy)3]3+/2+ as redox shuttle to understand if double
fence porphyrin sensitizers have superior performance to the
single fence ones. As shown in Table 2 and Figure 5a, dye loading
amount of mJS-series are almost two times of bJS-series, this
difference can be partly ascribed to increased molecular sizes
from single to double fences porphyrin sensitizers. Although bJS
dyes have less uptake than mJS dyes, the power conversion
efficiency (PCE) of bJS dyes are way higher than mJS dyes. The
PCE of best-performing bJS- and mJS-based cells were in the
ranges of 8.03% to 10.69% and of 2.33% to 6.69%, respectively,
clearly demonstrating that double fence porphyrins have superior
performance than their single fence counterparts, particularly on
the photocurrent.
On the basis of photovoltaic parameters, bJS-based cells have
both higher JSC and VOC than mJS ones, resulting in better PCE,
yet no obvious difference in FF. In mJS-based cells, mJS1
showed the best PCE of 6.69% with JSC of 10.551 mA cm-2, VOC
of 0.833 V, and FF of 0.762. Although the BTA- and BTD-
incorporated mJS2 and mJS3 have more red-shifted absorption
spectra than mJS1, they exhibited poor JSC of 5.468 and 3.792
mA cm-2, respectively; thus leading to low PCE. Because all three
mJS dyes have similar extinction coefficient in absorption spectra,
the decreased JSC of mJS2 and mJS3 is attributed to aggregation
of dye molecules. Introduction of anthracene and BTD (or BTA)
unit in porphyrin sensitizer extends -conjugation system but
enhances intermolecular - interaction at the same time so that
aggregation of porphyrin molecules becomes worse. As for bJS
dyes, they exhibited a trend similar to mJS dyes in the absorption
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 2. Photovoltaic Parameters of Devices Based on mJS dyes, bJS dyes, and YD2-o-C8 under AM1.5G illumination.[a]
Dye loading[b]
(10-8 mol cm-2)
dye
mJS1
mJS2
mJS3
bJS1
JSC (mA cm-2)
VOC (V)
FF
PCE (%)
10.551
(9.881 ± 0.595)
0.833
(0.836 ± 0.004)
0.762
(0.767 ± 0.005)
6.69
(6.33 ± 0.33)
7.59 ± 0.22
7.22 ± 0.63
7.74 ± 0.99
4.81 ± 0.52
4.93 ± 0.14
4.70 ± 0.31
-
5.468
(5.252 ± 0.224)
0.845
(0.843 ± 0.002)
0.752
(0.754 ± 0.002)
3.48
(3.34 ± 0.14)
3.729
(3.597 ± 0.128)
0.814
(0.811 ± 0.003)
0.768
(0.769 ± 0.001)
2.33
(2.24 ± 0.08)
12.525
(12.351 ± 0.158)
0.823
(0.813 ± 0.013)
0.779
(0.785 ± 0.006)
8.03
(7.89 ± 0.16)
16.586
(16.362 ± 0.376)
0.849
(0.851 ± 0.010)
0.759
(0.759 ± 0.006)
10.69
(10.56 ± 0.12)
bJS2
16.483
(16.603 ± 0.119)
0.836
(0.821 ± 0.014)
0.755
0.754 ± 0.008)
10.42
(10.28 ± 0.22)
bJS3
15.130
(14.665 ± 0.312)
0.849
(0.858 ± 0.008)
0.765
(0.750 ± 0.011)
9.83
(9.44 ± 0.27)
YD2-o-C8
[a]Photovoltaic parameters were obtained from the champion cells. Average values of four independent cells with standard error are presented in parentheses.
[b]Obtained by averaging of four independent cells.
Figure 5. (a) Photocurrent-voltage (J-V) curves under 1 sun illumination and (b) IPCE spectra of mJS-series, bJS-series, and YD2-o-C8.
spectra, therefore their better photovoltaic performance can be
attributed to the presence of second fence that helps alleviate dye
aggregation and reduce charge recombination. Thus, bJS2 and
bJS3 exhibit high JSC of 16.586 and 16.483 mA cm-1, respectively,
increased by ~60% with respect to the most efficient single fence
dye mJS1. It should be noted that VOC of bJS based cells (0.823
− 0.849 V) are slightly greater than mJS based cells (0.814 −
0.845 V). Considering that dye-loading on TiO2 for bJS dyes is
much lower than that for mJS counterparts, the effects of “double
fence” to mitigate charge recombination is pronounced.
Importantly, BTA-incorporated bJS2 and mJS2 have the highest
VOC values of 0.849 and 0.845, respectively, which may be due to
further decrease in charge recombination by the presence of a
long alkyl chain in BTA moiety. As a result, double fence porphyrin
bJS2 carrying long alkyl chain in acceptor moiety possesses the
best photovoltaic performance among all investigated dyes,
leading to an impressive PCE of 10.69%, which outperforms the
benchmark YD2-o-C8 dye (PCE = 9.83% in this work) by 8.7%
under same fabrication condition.
The photovoltaic properties of mJS and bJS dyes are also
investigated by incident photon-to-current conversion efficiency
(IPCE) spectra in this study. As shown in Figure 5b, mJS and bJS
dyes possess broad IPCE in the visible light region with a smaller
dip at around 600 nm when compared to that of YD2-o-C8 and
the onset wavelengths were observed at around 800 nm, which
was red-shifted by ca. 60 nm compared to YD2-o-C8. The
broadened characteristic in IPCE spectra with minor dip between
Soret and Q bands echoes the electronic absorption spectra for
these new dyes. The plateau heights of IPCE values are in the
order of bJS2 > bJS3 > bJS1 > mJS1 > mJS2 > mJS3, which
agree with the trend of JSC values in Table 2. bJS2 and bJS3
show much larger coverage and higher plateau than bJS1,
echoing the importance of BTD and BTA units for enhancing light-
harvesting in double fence porphyrin sensitizers as discussed
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 6. Logarithmic dependence of charge recombination resistance of (a) bJS and (b) mJS dyes at different applied bias in the dark
above. In contrast, mJS1 have the best light harvesting efficiency
among single fence mJS-series, implying that the use of a bigger
aromatic bridge would induces more aggregation. Thanks to the
double fences at the -positions in bJS series, the presence of
BTA and BTD bridges does not cause serious molecular
aggregation so that bJS2 and bJS3 exhibit excellent cell
performance.
To further investigate charge transport and recombination
dynamics in the investigated devices, EIS were conducted by
applying different bias in the dark. Transmission line model (TLM)
was used to extract impedance elements from the spectra. In the
anthracene moiety.[23] Moreover, 1,2-H atoms of anthracene
induce steric congestion between anthracene moiety and nearby
aromatic rings, which may also lower intramolecular charge
transfer rate.[24] Similarly, the addition of BTA group increases RCT
of mJS2 compared to mJS1. Introduction of BTD in mJS3,
however, lowers its RCT to even below mJS1. RCT of YD2-o-C8 is
the lowest in Figure 6b, which can be explained by its significantly
higher JSC when compared with mJS dyes. To evaluate charge
transport in TiO2 film, RT at different applied bias is plotted in
Figure S35b. It can be seen that RT of mJS dyes are higher than
those in their analogous bJS dyes and YD2-o-C8, indicating that
charge injection from excited dyes in mJS based devices are far
from sufficient. Similar results were also observed by Wu et al., in
which aggregated dye molecules can form a dense layer on TiO2
surface, blocking charge transfer between oxidized dye and
electrolyte.[25]
medium applied bias region,
a 45 degree transmission
lineappears at high frequency region in Nyquist plot, which
represents the charge transport resistance (RT) in mesoporous
TiO2 film; followed by an arc in the middle frequency, which
reflects the charge recombination resistance (RCT
)
at
TiO2/dye/electrolyte interface (Figure S35a).[22] Charge transfer
resistance at Pt/electrolyte interface (RPt) and diffusion resistance
(ZD) in the electrolyte are too small to be observable. bJS dyes
exhibit comparable RCT to mJS dyes even though dye loading
amounts of bJS are only half of mJS dyes, showing superior
ability of bJS dyes in suppressing charge recombination. For
better comparison, RCT of bJS and mJS are separately
demonstrated in Figure 6 because JSC of two series dyes are
significantly different. The order of RCT of bJS-series devices is
bJS2 > bJS3 > YD2-o-C8 > bJS1 (Figure 6a), while the order of
RCT of mJS-series devices is mJS2 > mJS1 > mJS3 > YD2-o-C8
(Figure 6b). Among double fence bJS dyes, bJS1 exhibits lowest
RCT when compares with bJS2 and bJS3, indicating benzoic acid
acceptor cannot withdraw electrons efficiently from electron-rich
anthracene moiety. Thus, additional strong electron withdrawing
groups such as BTA in bJS2 and BTD in bJS3 are equipped to
enhance charge injection and thus RCT improves. RCT of bJS2 is
higher than that of bJS3 due to: (1) extra nitrogen atoms of BTA
lift the conduction band of TiO2, reducing electron recombination,
and (2) long alkyl group on nitrogen atom reduces dye
aggregation.[4c] Interestingly, the RCT of double fence bJS1 is
even lower than the one of single fence YD2-o-C8, this finding
suggests the electron-rich anthracene moiety hampers charge
injection to conduction band of TiO2, accelerating electron
recombination. Thomas et al. reported similar finding that without
strong withdrawing group, electron tends to accumulate on
Conclusion
Novel double fence porphyrin sensitizers bJS1−bJS3 along with
their single fence analogues mJS1−mJS3 have been designed
and synthesized. Introduction of acetylene and aromatic units to
the donor--acceptor backbone successfully achieve extended -
conjugation for both bJS and mJS sensitizers, leading to
decreased HOMO−LUMO gap and improved light harvesting
abilities of the sensitizers. However, employment of more
aromatic moieties to the molecules enhances intermolecular -
interaction of sensitizers, resulting in poor JSC and  value
(2.33%−6.69%) for mJS-based DSSCs. In contrast, bJS-based
DSSCs exhibited remarkable improvement in photovoltaic
performance ( = 8.03%−10.69%) of DSSCs under the similar
conditions. Photovoltaic parameters and EIS spectra show that
the major advantages of double fence structure in bJS dyes are
suppression of dye aggregation and of charge recombination,
leading to dramatically improved JSC. As a result, bJS2 and bJS3
exhibit high  values of 10.69% and 10.42%, respectively, which
are better than the benchmark dye YD2-o-C8 ( = 9.83%) under
the similar conditions. We have demonstrated a novel and
effective molecular engineering strategy to improve the cell
performance of porphyrin dyes. The double fence porphyrin core
provides a useful architecture, in which a number of donor, -
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
[9]
a) L. Schmidt-Mende, W. M. Campbell, Q. Wang, K. W. Jolley, D. L.
Officer, M. K. Nazeeruddin, M. Gratzel, Chemphyschem 2005, 6, 1253-
1258; b) Q. Wang, W. M. Carnpbell, E. E. Bonfantani, K. W. Jolley, D. L.
Officer, P. J. Walsh, K. Gordon, R. Humphry-Baker, M. K. Nazeeruddin,
M. Gratzel, J. Phys. Chem. B 2005, 109, 15397-15409; c) W. M.
Campbell, K. W. Jolley, P. Wagner, K. Wagner, P. J. Walsh, K. C. Gordon,
L. Schmidt-Mende, M. K. Nazeeruddin, Q. Wang, M. Gratzel, D. L. Officer,
J. Phys. Chem. C 2007, 111, 11760-11762; d) S. L. Wu, H. P. Lu, H. T.
Yu, S. H. Chuang, C. L. Chiu, C. W. Lee, E. W. G. Diau, C. Y. Yeh, Energ.
Environ. Sci. 2010, 3, 949-955.
bridge, and acceptor can be easily introduced to the molecules for
use in high-performance DSSCs.
Acknowledgements
This work was supported by the Ministry of Science and
Technology, Taiwan (MOST107-2113-M-005-010-MY3 and
MOST108-2221-E-007-102-MY3). We thank the Instrument
Center at National Chung Hsing University for providing valuable
assistance on ESI-MS and FT-IR measurements. We thank to
National Center for High-performance Computing (NCHC) for
providing computational and storage resources. C.-Y.Y. is also
grateful for financial support from the “Innovation and
Development Center of Sustainable Agriculture” of The Featured
Areas Research Center Program within the framework of the
Higher Education Sprout Project by the Ministry of Education
(MOE) in Taiwan.
[10] a) T. Bessho, S. M. Zakeeruddin, C. Y. Yeh, E. W. G. Diau, M. Gratzel,
Angew. Chem. Int. Edit. 2010, 49, 6646-6649; b) M. K. Nazeeruddin, A.
Kay, I. Rodicio, R. Humphrybaker, E. Muller, P. Liska, N. Vlachopoulos,
M. Gratzel, J. Am. Chem. Soc. 1993, 115, 6382-6390; c) C. J. Barbe, F.
Arendse, P. Comte, M. Jirousek, F. Lenzmann, V. Shklover, M. Gratzel,
J. Am. Ceram. Soc. 1997, 80, 3157-3171; d) M. K. Nazeeruddin, F. De
Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito, T. Bessho,
M. Gratzel, J. Am. Chem. Soc. 2005, 127, 16835-16847; e) Q. J. Yu, Y.
H. Wang, Z. H. Yi, N. N. Zu, J. Zhang, M. Zhang, P. Wang, ACS Nano
2010, 4, 6032-6038.
[11] a) S. Chang, H. Wang, L. T. Lin Lee, S. Zheng, Q. Li, K. Y. Wong, W.-K.
Wong, X. Zhu, W.-Y. Wong, X. Xiao, T. Chen, J. Mater. Chem. C 2014,
2, 3521-3526; b) J.-W. Shiu, Y.-C. Chang, C.-Y. Chan, H.-P. Wu, H.-Y.
Hsu, C.-L. Wang, C.-Y. Lin, E. W.-G. Diau, J. Mater. Chem. A 2015, 3,
1417-1420; c) K. Zeng, Y. Chen, W.-H. Zhu, H. Tian, Y. Xie, J. Am. Chem.
Soc. 2020, 142, 5154-5161; d) J. M. Ji, H. R. Zhou, Y. K. Eom, C. H. Kim,
H. K. Kim, Adv. Energy Mater. 2020, 10, 2000124.
Conflict of interest
The authors declare no conflict of interest.
[12] a) C.-Y. Lin, Y.-C. Wang, S.-J. Hsu, C.-F. Lo, E. W.-G. Diau, J. Phys.
Chem. C 2010, 114, 687-693; b) M. Tanaka, S. Hayashi, S. Eu, T.
Umeyama, Y. Matano, H. Imahori, Chem. Commun. 2007, 2069-2071; c)
C. L. Mai, W. K. Huang, H. P. Lu, C. W. Lee, C. L. Chiu, Y. R. Liang, E.
W. G. Diau, C. Y. Yeh, Chem. Commun. 2010, 46, 809-811; d) Y.
Kurumisawa, T. Higashino, S. Nimura, Y. Tsuji, H. Iiyama, H. Imahori, J.
Am. Chem. Soc. 2019, 141, 9910-9919; e) T. Higashino, Y. Kurumisawa,
S. Nimura, H. Iiyama, H. Imahori, Eur. J. Org. Chem. 2018, 2018, 2537-
2547.
Keywords: electrochemistry, energy conversion, porphyrinoids,
renewable resources
[1]
C. Battaglia, A. Cuevas, S. De Wolf, Energ. Environ. Sci. 2016, 9, 1552-
1576.
[2]
[3]
[4]
T. L. Jester, Prog. Photovoltaics 2002, 10, 99-106.
B. Oregan, M. Gratzel, Nature 1991, 353, 737-740.
[13] a) T. Higashino, Y. Fujimori, K. Sugiura, Y. Tsuji, S. Ito, H. Imahori,
Angew. Chem. Int. Edit. 2015, 54, 9052-9056; b) T. Higashino, Y.
Kurumisawa, N. Cai, Y. Fujimori, Y. Tsuji, S. Nimura, D. M. Packwood, J.
Park, H. Imahori, Chemsuschem 2017, 10, 3347-3351; c) T. Higashino,
S. Nimura, K. Sugiura, Y. Kurumisawa, Y. Tsuji, H. Imahori, ACS Omega
2017, 2, 6958-6967.
a) M. Gratzel, Accounts Chem. Res. 2009, 42, 1788-1798; b) A. Hagfeldt,
G. Boschloo, L. C. Sun, L. Kloo, H. Pettersson, Chem. Rev. 2010, 110,
6595-6663; c) Y. S. Tingare, N. S. Vinh, H. H. Chou, Y. C. Liu, Y. S. Long,
T. C. Wu, T. C. Wei, C. Y. Yeh, Adv. Energy Mater. 2017, 7, 1700032.
a) M. S. Ahmad, A. K. Pandey, N. Abd Rahima, Renew. Sust. Energ. Rev.
2017, 77, 89-108; b) Y. J. Son, J. S. Kang, J. Yoon, J. Kim, J. Jeong, J.
Kang, M. J. Lee, H. S. Park, Y. E. Sung, J. Phys. Chem. C 2018, 122,
7051-7060; c) K. Zhu, N. R. Neale, A. Miedaner, A. J. Frank, Nano Lett.
2007, 7, 69-74; d) S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F.
E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K.
Nazeeruddin, M. Gratzel, Nat. Chem. 2014, 6, 242-247; e) A. Yella, C. L.
Mai, S. M. Zakeeruddin, S. N. Chang, C. H. Hsieh, C. Y. Yeh, M. Gratzel,
Angew. Chem. Int. Edit. 2014, 53, 2973-2977; f) Y. Saygili, M. Soderberg,
N. Pellet, F. Giordano, Y. M. Cao, A. B. Munoz-Garcia, S. M.
Zakeeruddin, N. Vlachopoulos, M. Pavone, G. Boschloo, L. Kavan, J. E.
Moser, M. Gratzel, A. Hagfeldt, M. Freitag, J. Am. Chem. Soc. 2016, 138,
15087-15096; g) Y. J. Wang, T. W. Hamann, Chem. Commun. 2018, 54,
12361-12364; h) T. Higashino, H. Iiyama, S. Nimura, Y. Kurumisawa, H.
Imahori, Inorg. Chem. 2020, 59, 452-459; i) K. Kannankutty, C.-C. Chen,
V. S. Nguyen, Y.-C. Lin, H.-H. Chou, C.-Y. Yeh, T.-C. Wei, ACS Appl.
Mater. Interfaces 2020, 12, 5812-5819; j) A. Yella, H. W. Lee, H. N. Tsao,
C. Y. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W. G. Diau, C. Y. Yeh,
S. M. Zakeeruddin, M. Gratzel, Science 2011, 334, 629-634.
a) A. Carella, F. Borbone, R. Centore, Front. Chem. 2018, 6, 481; b) H.
A. Maddah, V. Berry, S. K. Behura, Renew. Sust. Energ. Rev. 2020, 121,
109678.
[5]
[14] a) S. H. Kang, S. Y. Jung, Y. W. Kim, Y. K. Eom, H. K. Kim, Dyes
Pigments 2018, 149, 341-347; b) H. Zhou, J.-M. Ji, S. H. Kang, M. S.
Kim, H. S. Lee, C. H. Kim, H. K. Kim, J. Mater. Chem. C 2019, 7, 2843-
2852; c) J.-M. Ji, S. H. Kim, H. Zhou, C. H. Kim, H. K. Kim, ACS Appl.
Mater. Interfaces 2019, 11, 24067-24077.
[15] a) C. W. Lee, H. P. Lu, C. M. Lan, Y. L. Huang, Y. R. Liang, W. N. Yen,
Y. C. Liu, Y. S. Lin, E. W. G. Diau, C. Y. Yeh, Chem.-Eur. J. 2009, 15,
1403-1412; b) G. Yang, Y. Tang, X. Li, H. Å gren, Y. Xie, ACS Appl. Mater.
Interfaces 2017, 9, 36875-36885; c) S. Mathew, N. A. Astani, B. F. E.
Curchod, J. H. Delcamp, M. Marszalek, J. Frey, U. Rothlisberger, M. K.
Nazeeruddin, M. Grätzel, J. Mater. Chem. A 2016, 4, 2332-2339; d) K.
Zeng, Z. Tong, L. Ma, W.-H. Zhu, W. Wu, Y. Xie, Energ. Environ. Sci.
2020, Advance Article.
[16] Y. Lu, H. Song, X. Li, H. Å gren, Q. Liu, J. Zhang, X. Zhang, Y. Xie, ACS
Appl. Mater. Interfaces 2019, 11, 5046-5054.
[17] a) C. L. Wang, M. Zhang, Y. H. Hsiao, C. K. Tseng, C. L. Liu, M. F. Xu,
P. Wang, C. Y. Lin, Energ. Environ. Sci. 2016, 9, 200-206; b) J. Pan, H.
Song, C. Lian, H. Liu, Y. Xie, Dyes Pigments 2017, 140, 36-46; c) J. V.
S. Krishna, D. Koteshwar, T. H. Chowdhury, S. P. Singh, I. Bedja, A.
Islam, L. Giribabu, J. Mater. Chem. C 2019, 7, 13594-13605.
[6]
[7]
[18] a) C. Y. Lee, J. T. Hupp, Langmuir 2010, 26, 3760-3765; b) H.-H. Chou,
K. S. K. Reddy, H.-P. Wu, B.-C. Guo, H.-W. Lee, E. W.-G. Diau, C.-P.
Hsu, C.-Y. Yeh, ACS Appl. Mater. Interfaces 2016, 8, 3418-3427.
[19] P. Devasthale, W. Wang, J. Mignone, K. Renduchintala, S.
Radhakrishnan, J. Dhanapal, J. Selvaraj, R. Kuppusamy, M. A.
Pelleymounter, D. Longhi, N. Huang, N. Flynn, A. V. Azzara, K. Rohrbach,
J. Devenny, S. Rooney, M. Thomas, S. Glick, H. Godonis, S. Harvey, M.
a) S. H. Kang, M. J. Jeong, Y. K. Eom, I. T. Choi, S. M. Kwon, Y. Yoo, J.
Kim, J. Kwon, J. H. Park, H. K. Kim, Adv. Energy Mater. 2017, 7,
1602117; b) J.-M. Ji, H. Zhou, H. K. Kim, J. Mater. Chem. A 2018, 6,
14518-14545.
[8]
M. Gouterman, J. Mol. Spectrosc. 1961, 6, 138-163.
8
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
J. Cullen, H. Zhang, C. Caporuscio, P. Stetsko, M. Grubb, C. Huang, L.
Zhang, C. Freeden, B. J. Murphy, J. A. Robl, W. N. Washburn, Bioorg.
Med. Chem. Lett. 2015, 25, 4412-4418.
[20] O. Robles, F. E. McDonald, Org. Lett. 2008, 10, 1811-1814.
[21] T. Nakagawa, J. Hatano, Y. Matsuo, J. Porphyr. Phthalocya. 2014, 18,
735-740.
[22] a) F. Fabregat-Santiago, J. Bisquert, G. Garcia-Belmonte, G. Boschloo,
A. Hagfeldt, Sol. Energy Mater. Sol. Cels 2005, 87, 117-131; b) J.
Bisquert, Phys. Chem. Chem. Phys. 2000, 2, 4185-4192.
[23] K. R. J. Thomas, P. Singh, A. Baheti, Y. C. Hsu, K. C. Ho, J. T. Lin, Dyes
Pigments 2011, 91, 33-43.
[24] a) C. Teng, X. C. Yang, C. Yang, S. F. Li, M. Cheng, A. Hagfeldt, L. C.
Sun, J. Phys. Chem. C 2010, 114, 9101-9110; b) R. Y. Y. Lin, H. W. Lin,
Y. S. Yen, C. H. Chang, H. H. Chou, P. W. Chen, C. Y. Hsu, Y. C. Chen,
J. T. Lin, K. C. Ho, Energ. Environ. Sci. 2013, 6, 2477-2486.
[25] C. H. Wu, M. C. Chen, P. C. Su, H. H. Kuo, C. L. Wang, C. Y. Lu, C. H.
Tsai, C. C. Wu, C. Y. Lin, J. Mater. Chem. A 2014, 2, 991-999.
9
This article is protected by copyright. All rights reserved.

10.1002/anie.202013964
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Double fence porphyrin dyes bJS1−bJS3 were designed and synthesized for high efficiency dye-sensitized solar cells (DSSCs).
Compared to typical single fence porphyrin dyes, double fence porphyrins show reduced dye aggregation and charge recombination,
which are evidenced by photovoltaic data, thus giving much better device performance than their single fence counterparts.
10
This article is protected by copyright. All rights reserved.