Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-sensitized solar cells

Article abstract of DOI:10.1016/j.dyepig.2016.10.041

Four alkoxy-wrapped push-pull porphyrin dyes containing the phenothiazine derived donor and the ethynylbenzoic acid acceptor have been designed, synthesized and used as sensitizers for fabricating efficient dye-sensitized solar cells (DSSCs). Branched or

Full text of DOI:10.1016/j.dyepig.2016.10.041

Dyes and Pigments xxx (2016) 1e9
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Branched and linear alkoxy chains-wrapped push-pull porphyrins for
developing efficient dye-sensitized solar cells
,
*
Heli Song a, Xin Li b, Hans Ågren b, Yongshu Xie a
a Key Laboratory for Advanced Materials and Institute of Fine Chemicals, Shanghai Key Laboratory of Functional Materials Chemistry, School of Chemistry
and Molecular Engineering, East China University of Science and Technology, Meilong 130, Shanghai, 200237, China
b Division of Theoretical Chemistry and Biology, School of Biotechnology, KTH Royal Institute of Technology, SE-10691, Stockholm, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
Four alkoxy-wrapped push-pull porphyrin dyes containing the phenothiazine derived donor and the
ethynylbenzoic acid acceptor have been designed, synthesized and used as sensitizers for fabricating
efficient dye-sensitized solar cells (DSSCs). Branched or linear alkoxy chains were introduced to the
ortho-positions of the meso-phenyl moieties to suppress the dye aggregation and charge recombination.
The effect of alkoxy chains were investigated in the absence and presence of an additional electron-
withdrawing benzothiadiazole unit. In the former cases, almost identical photovoltaic efficiencies of
~8.3% were achieved for both the branched and the linear alkoxy chains, while in the latter cases, the
planar benzothiadiazole unit induces serious dye aggregation and charge recombination, resulting in
lower efficiencies of 6.46% and 7.50% for the linear and branched chains, respectively, even though
broader absorption was achieved. The relatively higher efficiency achieved for the dyes with branched
chains may be related to the better effect of suppressing the dye aggregation and charge recombination.
Furthermore, the coadsorption approach was employed, and a highest efficiency of 9.62% was achieved
for the dye that features branched chains and the benzothiadiazole unit. These results compose a novel
approach for developing efficient DSSCs by combining the coadsorbent with a porphyrin dye containing
both the additional benzothiadiazole acceptor and branched alkoxy chains.
Received 2 September 2016
Received in revised form
26 October 2016
Accepted 27 October 2016
Available online xxx
Keywords:
Dye-sensitized solar cells
Porphyrin
Sensitizer
Push-pull structure
Branched chains
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction
structural modification [13,14]. Thus, many porphyrin derivatives
have been synthesized and demonstrated to be good candidates as
Dye-sensitized solar cells (DSSCs), as one class of promising
techniques for solar-to-electricity conversion, have aroused inten-
sive research interest over the past two decades, showing the ad-
vantages of low manufacturing cost and decent photovoltaic
efficiency [1e3]. As the “light-harvester”, the sensitizer plays a
crucial role in DSSCs. Since the pioneering work on ruthenium-
DSSC sensitizers [15e21], and efficiencies up to ca. 13% have been
achieved using cobalt-based electrolytes [22,23].
It is noteworthy that most of the efficient examples of porphyrin
dyes have been designed using the “alkoxy-wrapped” push-pull
structures [24,25]. Specifically, the porphyrin cores are tethered
with an electron donor and an electron acceptor/anchoring group
at two opposite meso-positions, respectively, thus promoting the
electron injection process. Meanwhile, the other two meso-posi-
tions are substituted with phenyl moieties with long alkoxy chains
attached at the ortho positions with the purpose to improve the
€
complex-sensitized solar cells by Gratzel and co-workers [1],
great efforts have been devoted to seeking for efficient, low-cost,
stable and eco-friendly sensitizers [4e11].
As a class of natural dyes found in the photosynthesis systems,
porphyrins have inspired chemists to exploit their applications in
DSSC devices [12]. Featured with relatively strong absorption,
porphyrin dyes often exhibit excellent light harvesting. In addition,
solubility and reduce the
p-p aggregation of the dyes, as well as
preventing the unfavorable charge recombination between the
injected electrons in TiO2 and the oxidized redox mediator in the
electrolyte, and the photovoltaic performances of the dyes have
been demonstrated to be related to the lengths of alkoxy chains
[24]. Based on these backgrounds, it is envisioned that branched
alkoxy chains may be bulkier than linear ones and may be more
effective in preventing aggregation and in blocking the oxidized
up to four meso-positions and eight b-positions are available for
* Corresponding author.
E-mail address: yshxie@ecust.edu.cn (Y. Xie).
http://dx.doi.org/10.1016/j.dyepig.2016.10.041
0143-7208/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

2
H. Song et al. / Dyes and Pigments xxx (2016) 1e9
redox mediator from approaching TiO2 [26e28], and thus herein,
we will focus on investigating the effects of linear and branched
alkoxy chains on the photovoltaic behavior of DSSCs based on
porphyrin dyes.
absorption spectra were recorded on a Varian Cary 100 spectro-
photometer and fluorescence spectra were recorded on a Varian
Cray Eclipse fluorescence spectrophotometer. The cyclic voltam-
mograms of the dyes were obtained in acetonitrile with a Versastat
II electrochemical workstation (Princeton Applied Research) using
0.1 M TBAPF6 (Aldrich) as the supporting electrolyte, the sensitizer
attached to a nanocrystalline TiO2 film deposited on the conducting
FTO glass as the working electrode, a platinum wire as the counter
electrode, and a regular calomel electrode in saturated KCl solution
Four porphyrins XW18-XW21 (Fig. 1) were thus designed and
used as DSSC dyes. All of them possess the identical phenothiazine
derived donor, which has been demonstrated to be an efficient
donor by us and other groups [29e32]. XW18 and XW19 contain
linear octyloxy and branched 2-ethylhexyloxy chains, respectively.
On this basis, a benzothiadiazole (BTD) unit was further introduced
as an additional electron withdrawing group for broadening the
spectral response [33]. Thus, XW20 and XW21 were also designed
and synthesized. As a result, the cells based on XW18-XW21 exhibit
photovoltaic efficiencies of 8.32%, 8.28%, 6.46%, and 7.50%, respec-
tively. After coadsorption with CDCA, XW21 exhibits a highest PCE
of 9.62% among the four dyes. These results indicate that the
aggravated dye aggregation effect induced by the BTD unit may be
well suppressed by the coexistence of the branched alkoxy chains
and the coadsorbent, and thus high DSSC efficiencies may be
achieved.
as the reference electrode. The scan rate was fixed at 100 mV sꢀ1
.
Photovoltaic measurements were performed by employing an
AM 1.5 solar simulator equipped with a 300 W xenon lamp (model
no. 91160, Oriel). The power of the simulated light was calibrated to
100 mW cmꢀ2 using a Newport Oriel PV reference cell system
(model 91150 V). JeV curves were obtained by applying an external
bias to the cell and measuring the generated photocurrent with a
model 2400 source meter (Keithley Instruments, Inc. USA). The
voltage step and delay time of the photocurrent were 10 mV and
40 ms, respectively. Action spectra of the incident monochromatic
photon-to-electron conversion efficiency (IPCE) for the solar cells
were obtained with
a Newport-74125 system (Newport In-
struments). The intensity of monochromatic light was measured
with a Si detector (Newport-71640). The electrochemical imped-
ance spectroscopy (EIS) measurements of all the DSSCs were per-
formed using a Zahner IM6e Impedance Analyzer (ZAHNER-
Elektrik GmbH & CoKG, Kronach, Germany), with the frequency
range of 0.1 Hze100 kHz and the alternative signal of 10 mV. The
ZSimpWin software was used to fit the experimental EIS data of the
DSSCs.
2. Experimental section
2.1. Materials and instrumentation
All reagents and solvents were obtained from commercial
sources and used as received unless otherwise noted. DMF was
dried over 4 Å molecular sieves. Tetrabutylammonium hexa-
fluorophosphate (TBAPF6) was vacuum-dried for 48 h. The trans-
parent FTO conducting glass (fluorine-doped SnO2, transmission
2.2. Syntheses of the dyes
>90% in the visible range, sheet resistance 15U/square) and the TiO2
paste were purchased from Geao Science and Educational Co. Ltd.
The FTO conducting glass was washed with a detergent solution,
deionized water, acetone and ethanol successively under ultra-
sonication for 20 min before use.
Compounds 1a-5a and 1b-2b were synthesized according to
reported methods [34e37].
1H NMR and 13C NMR spectra were obtained using a Bruker AM
400 spectrometer. HRMS measurements were performed using a
Waters LCT Premier XE spectrometer. Matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry (MALDI-
TOF-MS) was measured using a Shimadzu-Kratos model Axima
CFR þ mass spectrometer with dithranol as the matrix. UVeVis
2.2.1. Synthesis of compound 3b
To a degassed solution of compound 2b (3.63 g, 10.0 mmol) and
dipyrromethane (1.46 g, 10.0 mmol) in CH2Cl2 (1.5 L) was added
trifluoroacetic acid (0.75 mL, 10 mmol). After the solution was
stirred at room temperature under dinitrogen for 4 h in the dark,
DDQ (3.4 g, 15 mmol) was added and the mixture was stirred for an
Fig. 1. Molecular structures of dyes XW18-XW21.
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

H. Song et al. / Dyes and Pigments xxx (2016) 1e9
3
additional 1 h. The mixture was basified with Et3N (3 mL) and
filtered with silica. The solvent was removed under reduced pres-
sure and the residue was purified by column chromatography on
silica gel using petroleum ether/CH2Cl2 (2:1, v:v) as the eluent.
Recrystallization from CH2Cl2/MeOH gave compound 3b as purple
powders (1.3 g, yield: 27%). mp 70e72 ꢁC. 1H NMR (400 MHz,
2H, phenyl), 7.40e7.33 (m, 2H, phenyl), 7.01 (d, J ¼ 8.4 Hz, 5H,
phenyl), 6.96 (d, J ¼ 8.0 Hz, 1H, phenyl), 6.93 (d, J ¼ 8.8 Hz, 2H,
phenyl), 3.97 (t, J ¼ 6.8 Hz, 4H), 3.85 (t, J ¼ 6.4 Hz, 8H), 1.92 (s, 2H),
1.84e1.74 (m, 2H), 1.57e1.44 (m, 4H), 1.39e1.30 (m, 8H), 1.02e0.92
(m, 12H), 0.85e0.76 (m, 8H), 0.66e0.56 (m, 9H), 0.56e0.44 (m,
28H), 0.44e0.34 (m, 9H). 13C NMR (100 MHz, CDCl3)
d 159.91,
CDCl3)
d
10.11 (s, 2H, meso-H), 9.23 (d, J ¼ 4.8 Hz, 4H, pyrrolic), 8.96
158.43, 152.23, 151.34, 150.53, 149.03,145.08, 143.32, 135.55,132.54,
132.32, 132.21, 130.76, 130.03, 129.89, 127.41, 125.50, 125.44, 124.58,
124.44, 120.76, 118.17, 115.60, 115.15, 115.06, 114.77, 105.17, 104.77,
99.96, 95.03, 92.76, 68.61, 68.08, 47.75, 31.60, 31.51, 31.33, 31.26,
29.71, 29.22, 28.66, 28.61, 28.57, 26.86, 26.69, 25.72, 25.24, 22.65,
22.62, 22.25, 14.06, 14.04, 13.82. MS (MALDI-TOF, m/z): [MþH] calcd
for C96H119BrN5O5SZn, 1596.7407; found, 1596.202.
(d, J ¼ 4.8 Hz, 4H, pyrrolic), 7.71 (t, J ¼ 8.4 Hz, 2H, phenyl), 7.02 (d,
J ¼ 8.4 Hz, 4H, phenyl), 3.79e3.66 (m, 8H), 0.88e0.76 (m, 4H),
0.45e0.29 (m, 16H), 0.29e0.18 (m, 16H), 0.18e0.12 (m, 12H), 0.03
to ꢀ0.04 (m,12H), -2.99 (s, 2H). 13C NMR (100 MHz, CDCl3)
d 160.24,
147.62, 144.88, 130.58, 130.28, 129.92, 119.87, 111.40, 104.94, 103.76,
70.90, 38.44, 29.71, 29.49, 28.13, 22.77, 22.11, 13.24, 10.29. HRMS
(ESI, m/z): [MþH]þ calcd for C64H87N4O4, 975.6722; found,
975.6720.
6b. 125 mg, yield: 15%. 1H NMR (400 MHz, CDCl3)
d 9.65 (d,
J ¼ 4.8 Hz, 2H, pyrrolic), 9.58 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 8.89 (d,
J ¼ 4.4 Hz, 2H, pyrrolic), 8.85 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 7.81e7.74
(m, 2H, phenyl), 7.71 (t, J ¼ 8.4 Hz, 2H, phenyl), 7.48 (d, J ¼ 8.8 Hz,
2H, phenyl), 7.41e7.34 (m, 2H, phenyl), 7.06e6.91 (m, 8H, phenyl),
4.03e3.93 (m, 4H), 3.82e3.68 (m, 8H), 1.98e1.87 (m, 2H), 1.85e1.75
(m, 2H), 1.59e1.44 (m, 4H), 1.43e1.30 (m, 12H), 0.92 (t, J ¼ 6.8 Hz,
6H), 0.46e0.31 (m, 16H), 0.31e0.12 (m, 28H), 0.08e0.01 (m, 12H).
2.2.2. Synthesis of compound 4b
To a stirred solution of porphyrin 3b (2.0 g, 2.1 mmol) in CH2Cl2
(300 mL) was added a solution of NBS (730 mg, 4.1 mmol) in CH2Cl2
(30 mL). After the reaction was quenched with acetone (15 mL), the
solvent was removed under reduced pressure. The residue was
purified by column chromatography on silica gel using petroleum
ether/CH2Cl2 (3:1, v:v) as the eluent. Recrystallization from CH2Cl2/
MeOH gave 4b as purple powders (2.1 g, yield: 91%). mp 98e100 ꢁC.
13C NMR (100 MHz, CDCl3)
d 160.02, 158.48, 152.25, 151.45, 150.63,
149.02, 145.11, 143.31, 135.59, 132.39, 132.34, 132.22, 130.75, 130.61,
130.07, 129.91, 127.44, 125.53, 125.46, 124.62, 124.43, 120.50, 118.22,
115.62, 115.21, 114.78, 104.76, 99.84, 94.97, 92.76, 70.84, 70.79,
68.08, 47.74, 38.54, 31.60, 31.50, 29.60, 29.58, 29.24, 28.16, 26.87,
26.69, 25.73, 22.89, 22.64, 22.62, 22.21, 14.05, 14.03, 13.39, 10.44.
HRMS (ESI, m/z): [MþNa]þ calcd for C96H118BrN5NaO5SZn,
1618.7221; found, 1618.7203.
1H NMR (400 MHz, CDCl3)
d
9.48 (d, J ¼ 4.8 Hz, 4H, pyrrolic), 8.78 (d,
J ¼ 4.4 Hz, 4H, pyrrolic), 7.70 (t, J ¼ 8.4 Hz, 2H, phenyl), 6.98 (d,
J ¼ 8.4 Hz, 4H, phenyl), 3.74 (m, 8H), 0.97e0.73 (m, 4H), 0.48e0.30
(m, 16H), 0.30e0.13 (m, 28H), 0.06e0.00 (m, 12H), -2.58 (s, 2H). 13
C
NMR (100 MHz, CDCl3)
d 160.04, 130.23, 119.78, 114.10, 104.70,
102.13, 70.86, 38.53, 29.61, 28.15, 22.88, 22.16, 13.29, 10.41. HRMS
(ESI, m/z): [MþNa]þ calcd for C64H84Br2N4NaO4, 1153.4752; found,
1153.4748.
2.2.5. Synthesis of compounds 7a and 7b
General synthetic procedures for 7a and 7b.
To a degassed solution of compound 6a or 6b (100 mg,
0.063 mmol) and methyl 4-ethynylbenzoate (30 mg, 0.19 mmol) in
dry THF (30 mL) and Et3N (6 mL) were added Pd(PPh3)4 (36 mg,
0.031 mmol) and CuI (6 mg, 0.031 mmol) under a nitrogen atmo-
sphere. The mixture was degassed, re-filled with N2, and allowed to
stir at 60 ꢁC for 16 h, then the solvent was removed under reduced
pressure. The residue was purified by column chromatography on
silica gel using petroleum ether/CH2Cl2 (3:2, v:v) as eluent.
Recrystallization from CH2Cl2/MeOH gave compound 7a or 7b as
dark green powders.
2.2.3. Synthesis of compound 5b
To a solution of 4b (1.8 g, 1.6 mmol) in CH2Cl2 (300 mL) was
added Zn(OAc)2$2H2O (7.0 g, 32 mmol) in MeOH (100 mL). The
mixture was refluxed for 12 h. Then the mixture was poured into
water, extracted with CH2Cl2, washed with water and dried over
Na2SO4. Evaporation of the solvents gave the target product 5b as
purple powders (1.6 g, yield: 87%). mp 145e147 ꢁC. 1H NMR
(400 MHz, CDCl3)
d
9.62 (d, J ¼ 4.8 Hz, 4H, pyrrolic), 8.89 (d,
J ¼ 4.8 Hz, 4H, pyrrolic), 7.70 (t, J ¼ 8.4 Hz, 2H, phenyl), 6.99 (d,
J ¼ 8.4 Hz, 4H, phenyl), 3.84e3.63 (m, 8H), 0.89e0.80 (m, 4H),
0.46e0.28 (m, 16H), 0.28e0.10 (m, 28H), 0.06e0.00 (m, 12H). 13C
7a. 43 mg, yield: 41%. mp 80e82 ꢁC. 1H NMR (400 MHz, CDCl3)
d
9.63 (d, J ¼ 4.4 Hz, 2H, pyrrolic), 9.62 (d, J ¼ 4.4 Hz, 2H, pyrrolic),
NMR (100 MHz, CDCl3)
d
160.03, 151.48, 149.69, 132.75, 132.63,
8.87 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 8.85 (d, J ¼ 4.4 Hz, 2H, pyrrolic),
8.14 (d, J ¼ 8.4 Hz, 2H, phenyl), 7.97 (d, J ¼ 8.4 Hz, 2H, phenyl),
7.81e7.73 (m, 2H, phenyl), 7.71 (t, J ¼ 8.4 Hz, 2H, phenyl), 7.47 (d,
J ¼ 8.8 Hz, 2H, phenyl), 7.41e7.32 (m, 2H, phenyl), 7.06e6.90 (m, 8H,
phenyl), 4.04e3.91 (m, 7H), 3.86 (t, J ¼ 6.4 Hz, 8H), 1.97e1.86 (m,
2H), 1.85e1.75 (m, 2H), 1.51e1.45 (m, 4H), 1.40e1.31 (m, 8H),
1.01e0.91 (m, 12H), 0.88e0.73 (m, 9H), 0.69e0.58 (m, 8H),
0.58e0.46 (m, 28H), 0.45e0.34 (m, 9H). 13C NMR (100 MHz, CDCl3)
130.03, 120.41, 115.06, 104.74, 103.84, 70.82, 53.39, 38.55, 29.60,
28.14, 22.90, 22.20,13.34,10.43. HRMS (ESI, m/z): [MþNa]þ calcd for
C
64H84Br2N4NaO4, 1153.4752; found, 1153.4748.
2.2.4. Syntheses of compounds 6a and 6b
General synthetic procedures for 6a and 6b.
To a degassed solution of compound 5a or 5b (600 mg,
0.50 mmol) and 3-ethynyl-10-hexyl-7-(4-(hexyloxy)phenyl)-10H-
phenothiazine (290 mg, 0.60 mmol) in THF (250 mL) and Et3N
(50 mL) were added Pd2(dba)3 (230 mg, 0.25 mmol) and AsPh3
(306 mg, 1.0 mmol). The mixture was degassed, re-filled with N2,
and allowed to stir at 60 ꢁC for 6 h, then the solvent was removed
under reduced pressure. The residue was purified by column
chromatography on silica gel using petroleum ether/CH2Cl2 (5:2,
v:v) as the eluent. Recrystallization from CH2Cl2/MeOH gave com-
pound 6a or 6b as dark green pastes.
d
166.54, 159.92, 158.40, 151.70, 151.41, 150.63, 150.51, 145.04,
143.26, 135.52, 132.18, 132.02, 131.73, 130.90, 130.74, 130.64, 130.33,
130.00, 129.84, 129.48, 129.02, 128.65, 127.38, 125.49, 125.40,
124.50, 124.38, 120.83, 118.10, 115.58, 115.43, 115.11, 114.76, 105.17,
101.15, 98.84, 96.65, 95.31, 94.61, 93.00, 68.63, 68.07, 52.17, 48.67,
47.74, 31.60, 31.50, 31.37, 29.71, 29.21, 28.71, 28.65, 28.61, 26.84,
26.68, 25.71, 25.26, 22.65, 22.62, 22.27, 14.05, 14.03, 13.83. HRMS
(ESI, m/z): [MþNa]þ calcd for C106H125N5NaO7SZn, 1698.8483;
found, 1698.8543.
6a. 104 mg, yield: 13%. 1H NMR (400 MHz, CDCl3)
d
9.65 (d,
7b. 38 mg, yield: 36%. mp 69e71 ꢁC. 1H NMR (400 MHz, CDCl3)
J ¼ 4.8 Hz, 2H, pyrrolic), 9.60 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 8.88 (d,
J ¼ 4.8 Hz, 2H, pyrrolic), 8.85 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 7.79e7.73
(m, 2H, phenyl), 7.70 (t, J ¼ 8.4 Hz, 2H, phenyl), 7.46 (d, J ¼ 8.8 Hz,
d
9.63 (d, J ¼ 4.4 Hz, 4H, pyrrolic), 8.88 (d, J ¼ 4.4 Hz, 2H, pyrrolic),
8.86 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 8.21 (d, J ¼ 8.4 Hz, 2H, phenyl), 8.04
(d, J ¼ 8.4 Hz, 2H, phenyl), 7.82e7.74 (m, 2H, phenyl), 7.71 (t,
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

4
H. Song et al. / Dyes and Pigments xxx (2016) 1e9
J ¼ 8.4 Hz, 2H, phenyl), 7.48 (d, J ¼ 8.0 Hz, 2H, phenyl), 7.41e7.33 (m,
2H, phenyl), 7.01 (d, J ¼ 8.8 Hz, 5H, phenyl), 6.98e6.91 (m, 3H,
phenyl), 4.04e3.87 (m, 7H), 3.86e3.66 (m, 8H), 1.97e1.87 (m, 2H),
1.86e1.74 (m, 2H), 1.58e1.44 (m, 4H), 1.39e1.31 (m, 8H), 0.96e0.80
(m, 10H), 0.49e0.32 (m, 16H), 0.31e0.12 (m, 28H), 0.10e0.01 (m,
Then water was added and the organic phase was extracted with
CH2Cl2. After evaporation of the solvents, the residue was purified
by column chromatography on silica gel using CH2Cl2/MeOH (20:1,
v:v) as eluent. Recrystallization from CH2Cl2/MeOH gave the target
products.
12H). 13C NMR (100 MHz, CDCl3)
d
166.68, 160.02, 158.49, 151.73,
XW18: dark green powders, 22 mg, yield: 83%. mp 220e222 ꢁC.
151.46, 150.79, 150.66, 145.17, 143.29, 135.61, 132.22, 132.14, 131.84,
131.21, 130.81, 130.54, 130.16, 130.10, 129.90, 129.77, 129.34, 128.99,
127.45, 125.54, 125.46, 124.61, 124.41, 120.52, 118.16, 115.70, 115.64,
115.21, 114.79, 104.79, 101.20, 98.79, 96.59, 95.35, 94.72, 92.87,
70.83, 68.09, 52.22, 47.75, 38.57, 31.60, 31.51, 29.71, 29.60, 29.24,
28.20, 26.87, 26.69, 25.73, 22.91, 22.65, 22.62, 22.25, 14.06, 14.04,
13.43, 10.44. HRMS (ESI, m/z): [MþH]þ calcd for C106H126N5O7SZn,
1676.8664; found, 1676.8691.
1H NMR (400 MHz, CDCl3: DMSO-d6 ¼ 1:2)
d 12.90 (s, 1H, -COOH),
9.55 (d, J ¼ 4.8 Hz, 2H, pyrrolic), 9.54 (d, J ¼ 4.8 Hz, 2H, pyrrolic),
8.75 (d, J ¼ 4.4 Hz, 2H, phenyl), 8.73 (d, J ¼ 4.4 Hz, 2H, phenyl), 8.18
(d, J ¼ 8.4 Hz, 2H, phenyl), 8.14e8.09 (m, 2H, phenyl), 8.07 (d,
J ¼ 8.4 Hz, 2H, phenyl), 7.81 (d, J ¼ 8.4 Hz, 1H, phenyl), 7.72 (t,
J ¼ 8.4 Hz, 3H, phenyl), 7.50 (d, J ¼ 8.8 Hz, 2H, phenyl), 7.44e7.36 (m,
2H, phenyl), 7.12 (d, J ¼ 8.8 Hz, 1H, phenyl), 7.09e7.02 (m, 5H,
phenyl), 6.94 (d, J ¼ 8.8 Hz, 2H, phenyl), 4.02e3.94 (m, 4H), 3.88 (t,
J ¼ 6.1 Hz, 8H), 1.90e1.81 (m, 2H), 1.80e1.71 (m, 2H), 1.55e1.42 (m,
4H), 1.38e1.33 (m, 8H), 1.04e0.93 (m, 16H), 0.92e0.89 (m, 5H),
0.84e0.77 (m, 8H), 0.68e0.63 (m, 12H), 0.63e0.56 (m, 16H),
0.51e0.38 (m, 9H). HRMS (ESI, m/z): [MþNa]þ calcd for
2.2.6. Synthesis of compounds 7c and 7d
General synthetic procedures for 7c and 7d.
Compound 7c and 7d were prepared according to the similar
procedure as described for 7a and 7b by using methyl 4-(7-
C
105H123N5NaO7SZn, 1684.8327; found, 1684.8343.
ethynylbenzo[c]
[1,2,5]thiadiazol-4-yl)benzoate
(37
mg,
XW19: dark green powders, 21 mg, yield: 79%. mp 195e197 ꢁC.
0.13 mmol) instead of methyl 4-ethynylbenzoate, and dry THF/
piperidine (20 mL/1 mL) was used as the solvent.
1H NMR (400 MHz, CDCl3: DMSO-d6 ¼ 1: 2)
d 12.87 (s, 1H, -COOH),
9.60e9.45 (m, 4H, pyrrolic), 8.80e8.65 (m, 4H, pyrrolic), 8.18 (d,
J ¼ 8.4 Hz, 2H, phenyl), 8.13e8.09 (m, 1H, phenyl), 8.07 (d,
J ¼ 8.0 Hz, 2H, phenyl), 7.81 (d, J ¼ 8.4 Hz, 1H, phenyl), 7.72 (t,
J ¼ 8.4 Hz, 3H, phenyl), 7.51 (d, J ¼ 8.8 Hz, 2H, phenyl), 7.46e7.32 (m,
2H, phenyl), 7.14 (d, J ¼ 8.8 Hz, 1H, phenyl), 7.05 (d, J ¼ 8.8 Hz, 5H,
phenyl), 6.94 (d, J ¼ 8.8 Hz, 2H, phenyl), 3.97 (t, J ¼ 6.4 Hz, 4H),
3.87e3.62 (m, 8H), 1.92e1.81 (m, 2H), 1.80e1.70 (m, 2H), 1.57e1.42
(m, 4H), 1.41e1.29 (m, 8H), 0.97e0.83 (m, 10H), 0.68e0.38 (m, 24H),
0.38e0.13 (m, 32H). HRMS (ESI, m/z): [MþNa]þ calcd for
7c. 34 mg, yield: 30%. 1H NMR (400 MHz, CDCl3)
d 9.93 (d,
J ¼ 25.4 Hz, 2H, pyrrolic), 9.60 (d, J ¼ 12.4 Hz, 2H, pyrrolic), 8.94 (s,
2H, pyrrolic), 8.87 (s, 2H, pyrrolic), 8.25e7.82 (m, 5H, phenyl),
7.80e7.60 (m, 5H, phenyl), 7.48 (d, J ¼ 8.0 Hz, 2H, phenyl), 7.41e7.31
(m, 2H, phenyl), 7.13e6.85 (m, 8H, phenyl), 4.06e3.78 (m, 15H),
1.98e1.86 (m, 2H), 1.86e1.72 (m, 2H), 1.50e1.43 (m, 4H), 1.41e1.32
(m, 8H), 1.12e0.97 (m, 8H), 0.96e0.88 (m, 6H), 0.86e0.72 (m, 8H),
0.70e0.36 (m, 44H). 13C NMR (100 MHz, CDCl3)
d 166.62, 159.97,
158.45, 155.71, 152.51, 152.00, 151.29, 150.63, 150.52, 145.01, 143.26,
140.47, 135.52, 132.22, 131.64, 130.82, 130.75, 129.94, 129.84, 129.25,
129.01, 128.25, 127.84, 127.42, 125.49, 125.41, 124.45, 124.38, 120.84,
118.05, 117.90, 115.62, 115.57, 115.08, 114.79, 105.17, 101.55, 101.30,
98.88, 95.30, 92.99, 92.32, 68.70, 68.10, 53.42, 52.16, 47.77, 31.93,
31.61, 31.51, 31.38, 29.70, 29.32, 29.24, 28.74, 28.69, 28.66, 26.84,
26.69, 25.73, 25.31, 22.66, 22.62, 22.26, 14.13, 14.06, 14.04, 13.81. MS
(MALDI-TOF, m/z): [MþH] calcd for C112H128N7O7S2Zn, 1810.8608;
found, 1810.573.
C
105H123N5NaO7SZn, 1684.8327; found, 1684.8315.
XW20: brown powders, 23 mg, yield: 80%. mp 239e341 ꢁC. 1H
NMR (400 MHz, CDCl3: DMSO-d6 ¼ 1: 2)
d 12.93 (s,1H, -COOH), 9.91
(d, J ¼ 4.4 Hz, 2H, pyrrolic), 9.54 (d, J ¼ 4.0 Hz, 2H, pyrrolic), 8.79 (d,
J ¼ 2.8 Hz, 2H, pyrrolic), 8.71 (d, J ¼ 2.8 Hz, 2H, phenyl), 8.46e8.37
(m, 1H, phenyl), 8.25 (d, J ¼ 5.2 Hz, 2H, phenyl), 8.18 (d, J ¼ 8.0 Hz,
3H, phenyl), 7.82 (d, J ¼ 6.0 Hz, 1H, phenyl), 7.78e7.68 (m, 3H,
phenyl), 7.52 (t, J ¼ 8.4 Hz, 2H, phenyl), 7.47e7.36 (m, 2H, phenyl),
7.20e7.13 (m, 1H, phenyl), 7.09 (t, J ¼ 8.0 Hz, 5H, phenyl), 6.99e6.90
(m, 2H, phenyl), 4.07e3.94 (m, 4H), 3.89 (t, J ¼ 5.6 Hz, 8H),
1.90e1.78 (m, 2H), 1.79e1.68 (m, 2H), 1.55e1.40 (m, 4H), 1.40e1.29
(m, 8H), 1.06e0.83 (m, 22H), 0.83e0.71 (m, 8H), 0.71e0.49 (m,
28H), 0.48e0.35 (m, 8H). MS (MALDI-TOF, m/z): [MþH] calcd for
7d. 31 mg, yield: 27%. mp 85e87 ꢁC. 1H NMR (400 MHz, CDCl3)
d
9.99 (s, 2H, pyrrolic), 9.63 (s, 2H, pyrrolic), 8.94 (d, J ¼ 4.0 Hz, 2H,
pyrrolic), 8.86 (d, J ¼ 3.2 Hz, 2H, pyrrolic), 8.28 (d, J ¼ 6.8 Hz, 1H,
phenyl), 8.24 (d, J ¼ 8.0 Hz, 2H, phenyl), 8.16 (d, J ¼ 7.6 Hz, 2H,
phenyl), 7.94 (d, J ¼ 7.2 Hz, 1H, phenyl), 7.83e7.64 (m, 4H, phenyl),
7.48 (d, J ¼ 8.0 Hz, 2H, phenyl), 7.42e7.33 (m, 2H, phenyl),
7.06e6.92 (m, 8H, phenyl), 4.05e3.91 (m, 7H), 3.86e3.66 (m, 8H),
1.97e1.87 (m, 2H), 1.85e1.75 (m, 2H), 1.41e1.32 (m, 8H), 0.97e0.89
(m, 8H), 0.46e0.34 (m, 16H), 0.33e0.21 (m, 16H), 0.21e0.13 (m,
C
111H126N7O7S2Zn, 1796.8452; found, 1796.436.
XW21: brown powders, 23 mg, yield: 80%. mp 224e226 ꢁC. 1H
NMR (400 MHz, CDCl3: DMSO-d6 ¼ 1: 2)
d 12.85 (s, 1H, -COOH),
9.89 (d, J ¼ 4.4 Hz, 2H, pyrrolic), 9.52 (d, J ¼ 4.4 Hz, 2H, pyrrolic),
8.85e8.75 (m, 2H, pyrrolic), 8.75e8.67 (m, 2H, pyrrolic), 8.40 (d,
J ¼ 7.6 Hz, 1H, phenyl), 8.25 (d, J ¼ 7.6 Hz, 2H, phenyl), 8.18 (d,
J ¼ 8.0 Hz, 2H, phenyl), 8.15e8.14 (m, 1H, phenyl), 7.82 (d, J ¼ 8.0 Hz,
1H, phenyl), 7.73 (t, J ¼ 8.0 Hz, 3H, phenyl), 7.52 (d, J ¼ 8.4 Hz, 2H,
phenyl), 7.43 (d, J ¼ 8.0 Hz, 1H, phenyl), 7.40e7.36 (m, 1H, phenyl),
7.16 (d, J ¼ 8.4 Hz, 1H, phenyl), 7.07 (d, J ¼ 8.4 Hz, 5H, phenyl), 6.96
(d, J ¼ 8.0 Hz, 2H, phenyl), 4.04e3.95 (m, 4H), 3.88e3.66 (m, 8H),
1.92e1.81 (m, 2H), 1.81e1.71 (m, 2H), 1.57e1.43 (m, 4H), 1.42e1.30
(m, 8H), 0.98e0.83 (m, 10H), 0.69e0.38 (m, 24H), 0.38e0.11 (m,
32H). MS (MALDI-TOF, m/z): [MþH] calcd for C111H126N7O7S2Zn,
1796.8452; found, 1795.526.
12H), 0.10e0.01 (m, 12H). 13C NMR (100 MHz, CDCl3)
d 166.84,
160.05, 158.52, 156.12, 153.20, 152.13, 151.40, 150.85, 150.74, 145.19,
143.30, 141.55, 135.62, 132.39, 132.25, 131.77, 131.14, 130.83, 130.70,
130.55, 130.11, 129.90, 129.85, 129.71, 129.10, 128.54, 127.47, 125.55,
125.48, 124.62, 124.43, 120.58, 118.42, 118.16, 115.89, 115.64, 115.22,
114.81, 104.84, 101.72, 101.43, 98.86, 95.42, 92.90, 92.16, 70.88,
68.10, 52.23, 47.76, 38.57, 31.61, 31.51, 29.71, 29.62, 29.25, 28.19,
26.88, 26.69, 25.74, 22.93, 22.65, 22.62, 22.24, 14.06, 14.04, 13.44,
10.45. MS (MALDI-TOF, m/z): [MþH] calcd for C112H128N7O7S2Zn,
1810.8608; found, 1810.238.
2.2.7. Syntheses of XW18-XW21
2.3. Theoretical approach
General synthetic procedures for XW18-XW21.
To the respective solution of 7a-7d (27 mg for 7a or 7b and
29 mg for 7c or 7d, 0.016 mmol) in THF (20 mL) was added
LiOH$H2O (30 mg, 0.71 mmol) in H2O (2 mL). The mixture was
degassed, re-filled with N2, and allowed to stir at 60 ꢁC for 12 h.
The ground state geometries and frontier molecular orbitals of
the dyes were computed using the hybrid B3LYP functional and the
6-31G* basis set [38e40]. For zinc atoms the Los Alamos effective
core potential basis set (LANL2DZ) was used [41]. All calculations
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

H. Song et al. / Dyes and Pigments xxx (2016) 1e9
5
were carried out using the Gaussian09 program package [42].
The Zn(II) dibromoporphyrin intermediates 5a and 5b were pre-
pared according to reported procedures [34]. Then the donor and
acceptor moieties were successively introduced via Sonogashira
cross-coupling reactions. Finally, the target products were obtained
by hydrolysis of the esters. All key intermediates and target prod-
ucts were fully characterized by NMR and MS.
2.4. Fabrication of solar cells
The procedure for preparation of TiO2 electrodes and fabrication
of the sealed cells for photovoltaic measurements were adapted
€
from that reported by Gratzel and co-workers [43]. A screen-
printed double layer of TiO2 particles was used as the photo-
electrode. The detailed procedure was reported in our previous
work [44]. The films were then immersed into a 0.2 mM solution of
the dyes with or without CDCA in a mixture of toluene and ethanol
(1/4, v/v) for 10 h at 25 ꢁC. The counter electrode was also prepared
according to the procedure reported in our previous work [44].
Finally, the DSSCs were assembled, with the electrolyte solution
containing 0.1 M LiI, 0.05 M I2, 0.6 M 1-methyl-3-propyl-imidazo-
3.2. Optical and electrochemical properties
As shown in Fig. 2 and Table 1, all the four dyes feature typical
porphyrin absorption characteristics. XW18 and XW19 display an
intense Soret band around 457 nm and a clearly resolved Q band
around 663 nm. As for XW20 and XW21, incorporation of the BTD
group results in dramatically broadened absorption and more
intense and red-shifted Q-bands around 681 nm. The spectral
changes can be rationalized by the strong electron-deficient nature
of the BTD unit, which is believed to enhance the electronic
coupling between the porphyrin core and the acceptor part. These
changes are favorable for light-harvesting and are expected to
afford better cell performance.
lium iodide (PMII), and 0.5
acetonitrile.
M 4-tert-butylpyridine (TBP) in
3. Results and discussion
3.1. Syntheses
It should be noted that the absorptions of XW18 and XW19 are
almost identical, so are those of XW20 and XW21, indicating that
the different alkoxy chains only slightly influence the porphyrin
The synthetic routes for XW18-XW21 are depicted in Scheme 1.
RO
RO
OR
OR
RO
OR
OR
CHO
OR
OH
OH
NH
N
N
NH
N
N
(iv)
(v)
(iii)
(i)
(ii)
Br
Br
HN
HN
OR
OR
RO
OR
1a, 1b
2a, 2b
3a, 3b
4a, 4b
RO
RO
OR
OR
RO
OR
OR
RO
RO
OR
OR
RO
OR
OR
N
N
N
N
N
N
N
N
N
N
N
N
(vii)
(vi)
(viii)
Zn
Zn
Zn
Zn
Br
A1
A
Br
D
D
D
Br
N
N
N
N
RO
RO
6a, 6b
5a, 5b
7a-7d
XW18-XW21
C6H13
N
1a, 2a,...,7a, 7c, XW18, XW20
1b, 2b,...,7b, 7d, XW19, XW21
D =
S
R =
C6H13O
COOMe
7a, 7b
COOH
XW18, XW19
XW20, XW21
A1
=
A =
S
S
N
N
N
N
COOMe 7c, 7d
COOH
Scheme 1. Synthetic Routes for XW18-XW21. Reaction conditions: (i) 1-bromooctane or 2-ethylhexyl bromide, KOH, DMSO, rt, overnight; (ii) n-BuLi, TMEDA, THF, 0 ꢁC, 3 h; then
DMF, rt, 2 h; (iii) bis(1H-pyrrol-2-yl)methane, TFA, CH2Cl2, rt, 4 h; then DDQ, rt, 1 h; (iv) NBS, CH2Cl2, rt, 1 h; (v) Zn(OAc)2$2H2O, DCM/MeOH, reflux, 6 h; (vi) 3-ethynyl-10-hexyl-7-
(4-(hexyloxy)phenyl)-10H-phenothiazine, Pd2(dba)3, AsPh3, THF/Et3N, 55 ꢁC, 6 h; (vii) for 7a and 7b: methyl 4-ethynylbenzoate, Pd(PPh3)4, CuI, THF/Et3N, 55 ꢁC, overnight; for 7c
and 7d: methyl 4-(7-ethynylbenzo[c] [1,2,5]thiadiazol-4-yl)benzoate, Pd(PPh3)4, CuI, THF/piperidine, 60 ꢁC, overnight; (viii) LiOH$H2O, THF/H2O, 60 ꢁC, overnight.
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

6
H. Song et al. / Dyes and Pigments xxx (2016) 1e9
Fig. 3. Schematic energy levels of dyes XW18-XW21.
narrower energy gap of XW20 and XW21 relative to XW18 and
XW19 is in accordance with their broader and red-shifted ab-
sorption. The excited state oxidation potentials (E*ox) were obtained
to be in the range of -0.87 V ~ -1.00 V, negative enough for electron
injection as compared to that of the TiO2 conduction band (~-0.5 V)
(Fig. 3).
3.3. DFT calculations
We employed density functional theory (DFT) calculations to
optimize the ground state geometries of the dyes, and the calcu-
lated frontier molecular HOMO and LUMO orbitals are shown in
Fig. 4. The HOMO electrons of all four dyes are mainly distributed
over the donor part and the porphyrin core, while the LUMO
electrons are predominantly populated on the acceptor part and
the porphyrin core. Especially for XW20 and XW21, the LUMO
electrons are strongly pulled to the acceptor part by the electron-
withdrawing BTD unit. These observations can be rationalized by
enhanced charge-transfer from the donor to the acceptor, which
possibly favors the electron injection process.
Fig. 2. Absorption spectra of dyes XW18-XW21. (a) in THF solutions, (b) on TiO2 films
(3 mm in thickness).
absorption. Upon adsorption onto TiO2 films, both the Soret band
and Q-bands are broadened due to interactions among dye mole-
cules and/or between dyes and nanocrystalline TiO2. This obser-
vation is also favorable for light-harvesting.
3.4. Photovoltaic performances
Cyclic voltammetry was employed to investigate the oxidation
potentials of the dyes and evaluate the possibility of electron in-
jection and dye regeneration from the thermodynamic point of
view (Table 1 and Figure S37). The ground state oxidation potentials
(Eox) were estimated to be 0.90 Ve0.93 V vs NHE, which are more
positive than the potential of Iꢀ3 /Iꢀ (~0.4 V), ensuring sufficient
driving force for dye regeneration. E0-0 was calculated to be 1.90 eV
for XW18 and XW19 and 1.80 eV for XW20 and XW21. The
The photovoltaic performances of XW18-XW21 were tested
using iodine-based electrolyte. Fig. 5 shows the J-V curves and IPCE
action spectra of the devices and the corresponding photovoltaic
parameters are listed in Table 2.
Initially, the TiO2 films were sensitized with individual
porphyrin dyes. XW18 and XW19 show comparable performances,
with a PCE of ~8.3% and similar Jsc and Voc. On the other hand, cells
based on XW21 showed Voc and Jsc of 723 mV, and 15.66 mA cmꢀ2
,
Table 1
Absorption spectral and electrochemical data for XW18-XW21.
a
Dye
lmax, nm (ε, 103 Mꢀ1 cmꢀ1
)
lmax, nmb
E0-0, eVc
Eox, Vd (vs NHE)
E*ox, Ve (vs NHE)
XW18
XW19
XW20
XW21
457 (255.6), 664 (62.2)
457 (264.9), 663 (63.9)
465 (154.2), 682 (86.3)
465 (173.1), 681 (95.4)
461
460
466
466
1.90
1.90
1.80
1.80
0.92
0.90
0.93
0.91
ꢀ0.98
ꢀ1.00
ꢀ0.87
ꢀ0.89
a
Absorption maxima wavelengths (lmax) and molar extinction coefficients (ε) in THF solution.
Absorption maxima wavelengths on 3
b
c
mm transparent TiO2 films.
E0-0 was estimated from the wavelength at the intersection (linter) of normalized absorption and emission spectra using the equation E0-0 ¼ 1240/linter
.
d
Eox was obtained from the first oxidation potential measured in acetonitrile using 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as the electrolyte (working
electrode: FTO/TiO2/dye; reference electrode: SCE; counter electrode: Pt). The Eox values were calibrated with ferrocene/ferrocenium (Fc/Fcþ) as an external reference, and its
potential was estimated to be 0.45 V, vs SCE (Figure S37). By comparing this value with that of 0.63 V vs NHE, the Eox vs. SCE can be converted to that vs. NHE by adding a value
of 0.18 V.
e
E*ox was estimated from the equation Eo*x ¼ Eox e E0-0
.
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

H. Song et al. / Dyes and Pigments xxx (2016) 1e9
7
Fig. 4. Calculated frontier molecular HOMO and LUMO orbitals of XW18-XW21.
Fig. 5. J-V curves (a,b) and IPCE action spectra (c,d) of XW18-XW21-sensitized solar cells with or without CDCA as the coadsorbent.
respectively, both higher than the respective values of 716 mV and
13.36 mA cmꢀ2 achieved for XW20. As a result, a higher efficiency
of 7.50% was achieved for XW21 relative to 6.46% for XW20.
It is noteworthy that XW20 and XW21 exhibit dramatically red-
shifted IPCE onset wavelengths triggered by the strong “pulling”
effect of the BTD unit. However, their IPCE plateaus are relatively
low, especially for XW20, which exhibits IPCE values lower than
attenuated by the phenyl spacer inserted between the BTD unit and
the carboxyl anchoring group, it couldn't be fully prevented [23,45].
In other words, the electron-trapping effect of BTD may lead to
aggravated charge recombination, resulting in lower electron-
Table 2
Photovoltaic parameters of DSSCs based on XW18-XW21 under simulated AM 1.5 G
full sun illumination (Power 100 mW cmꢀ2) using iodine-based electrolytea.
50% at most of the wavelengths. Considering that IPCE
¼
LHE( hcoll hreg (where LHE( ) is the light-harvesting
l) ꢂ Finj
ꢂ
ꢂ
l
Dyes
CDCA [mM]
Voc [mV]
Jsc [mA$cmꢀ2
]
FF [%]
PCE [%]
efficiency, Finj is the electron injection yield from the photo-
excited dye into TiO2 conduction band, hcoll is the collection effi-
ciency of the photo-generated charge carriers and hreg is the dye
regeneration efficiency) [34], the lower IPCE of XW20 and XW21
may be ascribed to the following two main reasons: 1. The planar
BTD unit aggravates the dye-aggregation problem, causing self-
quenching of the excited dyes and in turn lowering the electron-
injection yield; 2. Although electron back-transfer may be
XW18
0
8
0
8
0
8
0
8
755
754
755
742
716
737
723
734
1
1
2
2
1
1
3
2
16.42 0.21
18.78 0.17
16.90 0.15
18.54 0.24
13.36 0.28
18.09 0.24
15.66 0.38
19.62 0.15
67.1 0.5
67.2 0.5
64.9 0.2
68.5 0.2
67.6 0.5
65.4 0.3
66.3 1.0
66.8 0.5
8.32 0.02
9.50 0.07
8.28 0.07
9.42 0.11
6.46 0.15
8.72 0.11
7.50 0.06
9.62 0.11
XW19
XW20
XW21
a
The parameters were obtained from the average value of three devices.
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

8
H. Song et al. / Dyes and Pigments xxx (2016) 1e9
Fig. 6. Plots of (a) chemical capacitance (Cm) and (b) electron lifetime (t) obtained through electrochemical impedance analysis as a function of bias voltages with or without CDCA
as the coadsorbent.
collection efficiency of XW20 and XW21. The above two reasons
may also account for the fact that the change of the linear chains to
the branched ones makes a difference only for the BTD-containing
dyes. When branched alkoxy chains were introduced into the BTD-
containing dye XW20 to afford XW21, the severe dye aggregation
may be suppressed, and thus an elevated efficiency was observed
for XW21 as compared with XW20.
Generally, the maximum attainable Voc of a DSSC device is
determined by the quasi Fermi level of electrons in TiO2 semi-
conductor and the potential of the redox species in the electrolyte.
Given the same electrolyte used in the test, the Voc should be only
limited by the quasi Fermi level in TiO2, which is governed by the
TiO2 conduction band position and the electron injection/charge
recombination process [47]. The attained Cm and
t values provide
To further improve the DSSC efficiencies, especially for those
based on XW20 and XW21, which exhibit quite broad absorption,
CDCA was used as a coadsorbent to suppress dye aggregation and
charge recombination [46]. As expected, all the dyes display
improved performances under an optimized CDCA concentration of
8 mM (see Table S4 for the data of the optimization of CDCA con-
centrations). Specifically, for XW20 and XW21, the Jsc and Voc were
simultaneously elevated, leading to a drastic improvement in their
PCEs; while for XW18 and XW19, an increase of Jsc was accompa-
nied by a slight sacrifice of Voc. Therefore, their improvement in PCE
was relatively small. As a result, benefiting from both BTD and
branched chains, the combination of XW21 with CDCA gives a
highest PCE. The final PCEs lie in the order of XW21 (9.62%) > XW18
(9.50%) z XW19 (9.42%) > XW20 (8.72%).
The different coadsorption behaviors between the dyes in the
cases with and without the BTD unit indicate that the dye aggre-
gation for XW18 is not very serious due to its relatively small
conjugation framework, and thus when branched alkoxy units
were used to afford XW19, the aggregation effect were not obvi-
ously affected, and thus similar photovoltaic characteristics were
observed for XW18 and XW19. In contrast, XW20 contains a larger
conjugation framework due to the presence of the planar BTD unit,
which may induce serious dye aggregation effect, and thus it ex-
hibits a lower efficiency relative to XW18 and XW19. When
branched alkoxy chains were introduced to afford XW21, the dye
aggregation effect may be suppressed, and thus an elevated effi-
ciency was observed for XW21 as compared with XW20. However,
its efficiency is still lower than XW18 and XW19 in spite of its
broader absorption, indicating that additional approaches are
required for further suppressing the dye aggregation effect. Hence,
when CDCA was further used as the coadsorbent, a highest effi-
ciency of 9.62% was observed for XW21.
some perception into these factors, respectively.
The similar Cm values at a fixed bias voltage (Fig. 6a) signify
negligible differences in the conduction band positions for the four
dyes, while the electron lifetimes of XW18 and XW19-based cells
are obviously longer than those based on XW20 and XW21
(Fig. 6b), indicating that XW20 and XW21 suffer from more serious
dye-aggregation/charge recombination, which leads to their lower
observed Voc values (vide supra). On the other hand, the electron
lifetime of XW21 is slightly longer than that of XW20, indicative of
the effect of branched chains on suppressing the dye-aggregation/
charge recombination.
After coadsorption with CDCA, the Cm values of the cells based on
XW18 and XW19 were observed to be slightly decreased (Fig. 6a),
indicating an upward shift of TiO2 conduction band edge, which is
favorable for improving Voc. But their electron lifetimes were also
decreased (Fig. 6b), resulting in the net effect of slightly decreased
Voc values. On the other hand, cells based on XW20 and XW21
showed obviously decreased Cm and slightly elevated or almost
unchanged electron lifetimes after coadsorption, resulting in their
improved Voc values (vide supra).
4. Conclusions
In summary, four push-pull porphyrin dyes XW18-XW21
were designed, synthesized and used as DSSC sensitizers. Linear
and branched alkoxy chains were used to wrap the porphyrin
framework. For the dyes containing the ethynylbenzoic acid
group as the acceptor, the changes in the alkoxy chains make
little difference in the photovoltaic performance, while for dyes
containing the BTD unit as the auxiliary acceptor, branched
chains result in obviously better photovoltaic behavior than the
linear ones. Finally, by employing CDCA as the coadsorbent, the
porphyrin dye containing both BTD and branched chains (XW21)
exhibits a highest PCE of 9.62%. This work clearly demonstrates
the inherent advantages of the additional BTD acceptor and the
possibility of achieving high DSSC efficiencies by combining the
coadsorbent CDCA with a porphyrin dye containing both the BTD
acceptor and branched alkoxy chains. However, a large amount of
CDCA is required for suppressing the dye aggregation/charge
recombination for XW21, indicating that the effect of branched
chains is still limited. Further structural modification of the
3.5. Electrochemical impedance analysis
To further understand the difference in Voc for the devices based
on XW18-XW21, electrochemical impedance spectroscopy (EIS)
was performed in the dark at a series of bias voltages near their Voc
values. Data were fitted with an equivalent circuit to obtain infor-
mation about the chemical capacitance (Cm) and electron lifetime
(t).
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041

H. Song et al. / Dyes and Pigments xxx (2016) 1e9
9
sensitized solar cells. Chem Eur J 2013;19:17075e81.
[21] Li Z, Li Q. Molecular engineering and cosensitization for developing efficient
porphyrin dyes is under way in our laboratory.
solar cells based on porphyrin dyes with an extended
Chem 2014;57:1491.
p framework. Sci China
Acknowledgements
[22] Yella A, Mai CL, Zakeeruddin SM, Chang SN, Hsieh CH, Yeh CY, et al. Molecular
engineering of pushepull porphyrin dyes for highly efficient dye-sensitized
solar cells: the role of benzene spacers. Angew Chem Int Ed 2014;53:2973e7.
[23] Mathew S, Yella A, Gao P, Humphry-Baker R, CurchodBasile FE, Ashari-
Astani N, et al. Dye-sensitized solar cells with 13% efficiency achieved through
the molecular engineering of porphyrin sensitizers. Nat Chem 2014;6:242e7.
[24] Wang CL, Lan CM, Hong SH, Wang YF, Pan TY, Chang CW, et al. Enveloping
porphyrins for efficient dye-sensitized solar cells. Energy Environ Sci 2012;5:
6933e40.
This work was financially supported by NSFC (21472047,
91227201), the Program for Professor of Special Appointment
(Eastern Scholar, GZ2016006) at Shanghai Institutions of Higher
Learning, and the Fundamental Research Funds for the Central
Universities (WK1616004).
[25] Ripolles-Sanchis T, Guo BC, Wu HP, Pan TY, Lee HW, Raga SR, et al. Design and
characterization of alkoxy-wrapped push-pull porphyrins for dye-sensitized
solar cells. Chem Commun 2012;48:4368e70.
[26] Hung WI, Liao YY, Lee TH, Ting YC, Ni JS, Kao WS, et al. Eugenic metal-free
sensitizers with double anchors for high performance dye-sensitized solar
cells. Chem Commun 2015;51:2152e5.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.dyepig.2016.10.041.
[27] Jia HL, Zhang MD, Yan W, Ju XH, Zheng HG. Effects of structural optimization
on the performance of dye-sensitized solar cells: spirobifluorene as a prom-
ising building block to enhance Voc. J Mater Chem A 2016;4:11782e8.
[28] adi Jr FM, Kang X, O'Neil D, Pajares G, Getmanenko YA, Szymanski P, et al.
Near-infrared asymmetrical squaraine sensitizers for highly efficient dye
References
[1] O'Regan B, Gratzel M. A low-cost, high-efficiency solar cell based on dye-
sensitized colloidal TiO2 films. Nature 1991;353:737e40.
[2] Wang SW, Chou CC, Hu FC, Wu KL, Chi Y, Clifford JN, et al. Panchromatic Ru(II)
sensitizers bearing single thiocyanate for high efficiency dye sensitized solar
cells. J Mater Chem A 2014;2:17618e27.
[3] Zhang S, Yang X, Numata Y, Han L. Highly efficient dye-sensitized solar cells:
progress and future challenges. Energy Environ Sci 2013;6:1443e64.
[4] Bessho T, Zakeeruddin SM, Yeh CY, Diau EWG, Gratzel M. Highly efficient
sensitized solar cells: the effect of
p-bridges and anchoring groups on solar
cell performance. Chem Mater 2015;27:2480e7.
[29] Hua Y, Chang S, Huang D, Zhou X, Zhu X, Zhao J, et al. Significant improvement
of dye-sensitized solar cell performance using simple phenothiazine-based
dyes. Chem Mater 2013;25:2146e53.
[30] Lin RYY, Wu FL, Li CT, Chen PY, Ho KC, Lin JT. High-performance aqueous/
organic dye-sensitized solar cells based on sensitizers containing triethylene
oxide methyl ether. ChemSusChem 2015;8:2503e13.
€
mesoscopic dye-sensitized solar cells based on donoreacceptor-substituted
porphyrins. Angew Chem Int Ed 2010;49:6646e9.
[31] Sun X, Wang Y, Li X, Agren H, Zhu W, Tian H, et al. Cosensitizers for simul-
taneous filling up of both absorption valleys of porphyrins: a novel approach
for developing efficient panchromatic dye-sensitized solar cells. Chem Com-
mun 2014;50:15609e12.
[32] Xie Y, Tang Y, Wu W, Wang Y, Liu J, Li X, et al. Porphyrin cosensitization for a
photovoltaic efficiency of 11.5%: a record for non-ruthenium solar cells based
on iodine electrolyte. J Am Chem Soc 2015;137:14055e8.
[33] Tang Y, Wang Y, Li X, Ågren H, Zhu WH, Xie Y. Porphyrins containing a tri-
phenylamine donor and up to eight alkoxy chains for dye-sensitized solar
cells: a high efficiency of 10.9%. ACS Appl Mater Interfaces 2015;7:27976e85.
[34] Yella A, Lee HW, Tsao HN, Yi C, Chandiran AK, Nazeeruddin MK, et al.
Porphyrin-sensitized solar cells with cobalt (II/III)eBased redox electrolyte
exceed 12 percent efficiency. Science 2011;334:629e34.
[5] Lu X, Feng Q, Lan T, Zhou G, Wang ZS. Molecular engineering of quinoxaline-
based organic sensitizers for highly efficient and stable dye-sensitized solar
cells. Chem Mater 2012;24:3179e87.
[6] Yao Z, Wu H, Li Y, Wang J, Zhang J, Zhang M, et al. Dithienopicenocarbazole as
the kernel module of low-energy-gap organic dyes for efficient conversion of
sunlight to electricity. Energy Environ Sci 2015;8:3192e7.
[7] Kakiage K, Aoyama Y, Yano T, Oya K, Fujisawa Ji, Hanaya M. Highly-efficient
dye-sensitized solar cells with collaborative sensitization by silyl-anchor and
carboxy-anchor dyes. Chem Commun 2015;51:15894e7.
[8] Liu YC, Chou HH, Ho FY, Wei HJ, Wei TC, Yeh CY. A feasible scalable porphyrin
dye for dye-sensitized solar cells under one sun and dim light environments.
J Mater Chem A 2016;4:11878e87.
[9] Higashino T, Fujimori Y, Sugiura K, Tsuji Y, Ito S, Imahori H. Tropolone as a
high-performance robust anchoring group for dye-sensitized solar cells.
Angew Chem Int Ed 2015;54:9052e6.
[35] Lee CY, Hupp JT. Dye sensitized solar cells: TiO2 sensitization with a bodipy-
porphyrin antenna system. Langmuir 2010;26:3760e5.
€
[36] Chen ZK, Lee NHS, Huang W, Xu YS, Cao Y. New phenyl-substituted PPV de-
rivatives for polymer light-emitting DiodesꢀSynthesis, characterization and
StructureꢀProperty relationship study. Macromolecules 2003;36:1009e20.
[37] Sasaki K, Nomura K. Photoelectric conversion element and dye-sensitized
solar cell. PCT Int Appl 2013; WO 2013089087 A1 20130620.
[10] Wu Y, Zhu WH, Zakeeruddin SM, Gratzel M. Insight into DeAꢀ
peA structured
sensitizers: a promising route to highly efficient and stable dye-sensitized
solar cells. ACS Appl Mater Interfaces 2015;7:9307e18.
[11] Guo FL, Li ZQ, Liu XP, Zhou L, Kong FT, Chen WC, et al. Metal-free sensitizers
containing hydantoin acceptor as high performance anchoring group for dye-
sensitized solar cells. Adv Funct Mater 2016. http://dx.doi.org/10.1002/
adfm.201601305.
[38] Becke AD. Densityefunctional thermochemistry. III. The role of exact ex-
change. J Chem Phys 1993;98:5648e52.
[39] Lee C, Yang W, Parr RG. Development of the Colle-Salvetti correlation-energy
formula into a functional of the electron density. Phys Rev B 1988;37:785e9.
[40] Hehre WJ, Ditchfield R, Pople JA. Selfeconsistent molecular orbital methods.
XII. further extensions of gaussianetype basis sets for use in molecular orbital
studies of organic molecules. J Chem Phys 1972;56:2257e61.
[12] Gobeze HB, Das SK, D'Souza F. Femtosecond transient absorption study of
supramolecularly assembled metal tetrapyrroleeTiO2 thin films. J Phys Chem
C 2014;118:16660e71.
[13] Li LL, Diau EWG. Porphyrin-sensitized solar cells. Chem Soc Rev 2013;42:
291e304.
[41] Hay PJ, Wadt WR. Ab initio effective core potentials for molecular calculations.
Potentials for the transition metal atoms Sc to Hg. J Chem Phys 1985;82:
270e83.
[14] Chen B, Sun L, Xie YS. Modulation of photovoltaic behavior of dye-sensitized
solar cells by electron donors of porphyrin dyes and cosensitization. Chin
Chem Lett 2015;26:899e904.
€
[42] Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al.
Gaussian 09 revision A. 2. Wallingford, CT: Gaussian, Inc; 2009.
[15] Urbani M, Gratzel M, Nazeeruddin MK, Torres T. Meso-substituted porphyrins
for dye-sensitized solar cells. Chem Rev 2014;114:12330e96.
[16] Cabau L, Vijay Kumar C, Moncho A, Clifford JN, Lopez N, Palomares E. A single
atom change "switches-on" the solar-to-energy conversion efficiency of Zn-
porphyrin based dye sensitized solar cells to 10.5%. Energy Environ Sci
2015;8:1368e75.
[17] Lu J, Li H, Liu S, Chang YC, Wu HP, Cheng Y, et al. Novel porphyrin-preparation,
characterization, and applications in solar energy conversion. Phys Chem
Chem Phys 2016;18:6885e92.
€
[43] Ito S, Murakami TN, Comte P, Liska P, Gratzel C, Nazeeruddin MK, et al.
Fabrication of thin film dye sensitized solar cells with solar to electric power
conversion efficiency over 10%. Thin Solid Films 2008;516:4613e9.
[44] Wang Y, Chen B, Wu W, Li X, Zhu W, Tian H, et al. Efficient solar cells
sensitized by porphyrins with an extended conjugation framework and a
carbazole donor: from molecular design to cosensitization. Angew Chem Int
Ed 2014;53:10779e83.
[45] Chai Z, Wu M, Fang M, Wan S, Xu T, Tang R, et al. Similar or totally different:
the adjustment of the twist conformation through minor structural modifi-
cation, and dramatically improved performance for dye-sensitized solar cell.
Adv Energy Mater 2015:5. http://dx.doi.org/10.1002/aenm.201500846.
[46] Shi J, Chai Z, Tang R, Hua J, Li Q, Li Z. New triphenylamine-based sensitizers
bearing double anchoring units for dye-sensitized solar cells. Sci China Chem
2015;58:1144.
[18] Wang CL, Zhang M, Hsiao YH, Tseng CK, Liu CL, Xu M, et al. Porphyrins bearing
a consolidated anthryl donor with dual functions for efficient dye-sensitized
solar cells. Energy Environ Sci 2016;9:200e6.
[19] Li C, Luo L, Wu D, Jiang R, Lan J, Wang R, et al. Porphyrins with intense ab-
sorptivity: highly efficient sensitizers with a photovoltaic efficiency of up to
10.7% without a cosensitizer and a coabsorbate. J Mater Chem A 2016;4:
11829e34.
[47] Ning Z, Fu Y, Tian H. Improvement of dye-sensitized solar cells: what we
know and what we need to know. Energy Environ Sci 2010;3:1170e81.
[20] Kurotobi K, Toude Y, Kawamoto K, Fujimori Y, Ito S, Chabera P, et al. Highly
asymmetrical porphyrins with enhanced pushepull character for dye-
Please cite this article in press as: Song H, et al., Branched and linear alkoxy chains-wrapped push-pull porphyrins for developing efficient dye-
sensitized solar cells, Dyes and Pigments (2016), http://dx.doi.org/10.1016/j.dyepig.2016.10.041