Marked effects of indolyl vs. indolinyl substituent on solid-state structure, carrier mobility and photovoltaic efficiency of asymmetrical squaraine dyes

Article abstract of DOI:10.1039/c4ta03859b

Two solution-processed asymmetrical squaraines (ASQs) with cyclopenta[b]indolinyl (1a) and cyclopenta[b]indolyl (1b) as end cappers have been designed and synthesized. Although the internal molecular structure variations are minimal, the presence of the cyclopenta[b]indolinyl group endows 1a more planar molecular structure, which results in a much more compact solid-state structure (density is 1.317 g cm^-3 for 1a but is 1.187 g cm^-3 for 1b), dramatically affecting charge transport in the thin films. The hole mobility of 1a:PC₇₁BM blended film is about 7 times higher than that of 1b:PC₇₁BM. Consequently, the maximum power conversion efficiency (PCE) value of the organic photovoltaic cells (OPVs) based on 1a of up to 4.1%, approximately 80% higher than that of 1b, is one of the highest PCEs achieved for ASQ-based bulk-heterojunction (BHJ) OPVs. This journal is

Full text of DOI:10.1039/c4ta03859b

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Marked effects of indolyl vs. indolinyl substituent
on solid-state structure, carrier mobility and
photovoltaic efficiency of asymmetrical squaraine
dyes†
Cite this: J. Mater. Chem. A, 2014, 2,
18313
a
b
a
a
b
a
*
Lin Yang,‡ Qianqian Yang,‡ Daobin Yang, Qian Luo, Youqin Zhu, Yan Huang,
b
a
*
*
Suling Zhao and Zhiyun Lu
Two solution-processed asymmetrical squaraines (ASQs) with cyclopenta[b]indolinyl (1a) and cyclopenta[b]
indolyl (1b) as end cappers have been designed and synthesized. Although the internal molecular structure
variations are minimal, the presence of the cyclopenta[b]indolinyl group endows 1a more planar molecular
structure, which results in a much more compact solid-state structure (density is 1.317 g cmꢀ3 for 1a but is
1.187 g cmꢀ3 for 1b), dramatically affecting charge transport in the thin films. The hole mobility of 1a:PC71BM
blended film is about 7 times higher than that of 1b:PC71BM. Consequently, the maximum power conversion
efficiency (PCE) value of the organic photovoltaic cells (OPVs) based on 1a of up to 4.1%, approximately 80%
higher than that of 1b, is one of the highest PCEs achieved for ASQ-based bulk-heterojunction (BHJ) OPVs.
Received 26th July 2014
Accepted 12th September 2014
DOI: 10.1039/c4ta03859b
www.rsc.org/MaterialsA
SMOPVs is as high as 5.50%.10 It is noteworthy that the squar-
aine dye used in this report has a symmetrical molecular
1. Introduction
structure (SQ, D–A–D). Compared with SQs, the device perfor-
mance of BHJ-SMOPVs based on asymmetrical squaraines
(ASQs, D–A–D0) is much inferior to that of SQ ones, e.g., the
corresponding PCEs are only 0.20–2.05%.6,11–13 The low PCEs of
these devices should be ascribed to the relatively low open
circuit voltages (Voc, 0.24–0.69 V), ll factors (FF, 0.27–0.40) and
short circuit current density (Jsc, 1.40–9.05 mA cmꢀ2).6,11–13
However, asymmetrical squaraines (ASQs, D–A–D0) should be
more advantageous with regard to the ne-tuning of optoelec-
tronic properties14,15 and oen exhibit better solubility in
common organic solvents16 than SQs. In fact, ASQs are more
promising candidates relative to SQs in many optoelectronic
applications such as dye-sensitized solar cells, bioimaging
probes, and nonlinear optical materials.17–19
Recently, we have reported two asymmetrical squaraines and
rendered a very high Voc of 1.12 V based on its single BHJ
SMOPVs.20 However, the power conversion efficiency (PCE) was
limited to 2.82% due to its low FF of 0.38 and Jsc of 7.00 mA
cmꢀ2. Despite of the excellent absorption properties of squar-
aine dyes, most of the squaraine dyes based BHJ-OPVs suffer
from low FF and Jsc,6,11–13,20–23 which may be mostly ascribed to
the low hole mobility of the optical active layer.21–23 Therefore,
through rational molecular design, good planar squaraine
materials can be obtained to promote better intermolecular
packing for enhanced carrier mobility, essentially improving
the FF and Jsc of the corresponding BHJ-OPV devices to elevate
the PCE, which is greatly signicant in expanding the applica-
tion of squaraine materials in the OPV eld.23–25
Solution-processed organic photovoltaic (OPV) cell with a bulk-
heterojunction (BHJ) device structure is a quite promising
technique for renewable energy source applications due to its
low cost and potential for fabrication of large-area exible
devices.1–3 In addition, BHJ-OPVs based on small molecular
(SM) photoactive materials have attracted great interest recently
because of their unique advantages over their polymer coun-
terparts, such as well-dened molecular structure and molec-
ular weight, facile synthesis and high purity without batch-to-
batch variations.3–5
Among the SMOPV materials hitherto exploited, squaraine
derivatives have been considered to be excellent electron donor
candidates because of their intense absorption bands in the
visible and NIR regions, as well as excellent photochemical/
photophysical stability.6–9 So far, recorded power conversion
efficiency (PCE) of squaraine-based solution-processed BHJ-
aKey Laboratory of Green Chemistry and Technology (Ministry of Education), College of
Chemistry, Sichuan University, Chengdu 610064, P. R. China. E-mail: huangyan@scu.
edu.cn; luzhiyun@scu.edu.cn
bKey Laboratory of Luminescence and Optical Information (Ministry of Education),
Institute of Optoelectronics Technology, Beijing Jiaotong University, Beijing 100044,
P. R. China. E-mail: slzhao@bjtu.edu.cn
† Electronic supplementary information (ESI) available: Table of crystallographic
details for compounds 1a and 1b, ORTEP drawing of compounds 1a and 1b,
Comparison of calculated geometries of 1a and 1b with experimental data, AFM
images of the blended ASQ:PC71BM lms. CCDC 981604 and 981603. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c4ta03859b
‡ These authors contributed equally.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18313–18321 | 18313

View Article Online
Journal of Materials Chemistry A
Paper
2.3 X-ray crystallography
The crystallographic data for 1a and 1b reported here had been
deposited in the ESI.† Single crystal samples of 1a and 1b were
obtained from CH2Cl2/MeOH and CH2Cl2/acetone, respectively.
Single crystal X-ray diffraction data were obtained on an Xcali-
bur E X-ray single crystal diffractometer equipped with a
Scheme 1 Molecular structures of 1a and 1b.
˚
graphite monochromator Mo-Ka (l ¼ 0.71073 A) radiation. The
data collection was executed using CrysAlisPro program. The
structures were determined using direct method and successive
Fourier difference syntheses (SHELXS-97) and rened using
full-matrix least squares procedure on F2 with anisotropic
thermal parameters for all non-hydrogen atoms (SHELXL-97).
Additional renement details and the resulting factors for 1a
and 1b are given in Supporting Information.
In this paper, we report the design and synthesis of two novel
ASQ compounds 1a and 1b (structures shown in Scheme 1),
which possess similar structures, but bearing cyclopenta[b]-
indolinyl (1a) and cyclopenta[b]indolyl-modied moieties (1b).
Herein, indolinyl and indolyl groups rather than alkyl(phenyl)
amino or 1,2,3,4-tetrahydroquinolinyl groups were selected as
the substituents due to their higher molecular rigidity and
better planarity,26–28 while a cyclopentyl segment was fused onto
the indoline or indole segments to endow the compounds with
good solution-processing capability. More importantly, the
different hybridization status of the C1 and C2 in 1a (sp3) and 1b
(sp2) and the different electron-donating capability of the N
atoms of the two compounds (marked in Scheme 1) would have
great effect on the steric and electronic congurations of the
substituents, thus the two objective compounds might show
different molecular geometries, hence different intermolecular
packing properties. This is quite propitious for us to gain
further insight into the correlations between molecular struc-
tures and photovoltaic properties. Preliminary results show that
compound 1a bearing cyclopenta[b]indolinyl unit displays a
much more planar geometry, and under AM 1.5G 1 sun illu-
mination at room temperature, its BHJ-OPV device renders a
rather high FF of 0.50 and Jsc of 10.49 mA cmꢀ2, leading to a
maximum PCE of 4.1%, which is one of the highest PCEs for
ASQ-based BHJ SMOPVs.6,11–13,20
2.4 Device preparation
Photovoltaic devices were fabricated on indium-tin oxide (ITO)
coated-glass substrate (sheet resistance ¼ 15 U sqꢀ1) with a
layered structure of glass/ITO/MoO3/ASQ:PC71BM/LiF/Al. The
ITO substrates were cleaned through sequential sonication in
detergent, deionized water, acetone and ethanol for 10 min
each, and nally blow-dried by high purity nitrogen. The
substrate was treated by UV-ozone for 5 min, then immediately
transferred into a high vacuum chamber for the depositionꢀo1f
ꢀ4
˚
˚
80 A MoO3 at a pressure of 3 ꢁ 10 Pa with a rate of 0.5 A s
.
Subsequently, the photoactive layer was fabricated by spin-
casting a blend of the target compounds and PC71BM in chlo-
roform solution with a total concentration of 20 mg mLꢀ1 (1500
rpm, 50 s) under a N2-lling glove box at 25 ꢂC. Then the
substrates were loaded into a vacuum chamber to nish the
ꢀ4
˚
˚
deposition of LiF (8 A) and Al (1000 A) at a pressure of 3 ꢁ 10
ꢀ1
Pa with a rate of 0.05 A s and 1.5 A sꢀ1, respectively. Depo-
sition rate and lm thickness were monitored in situ using a
quartz crystal oscillator mounted to the substrate holder. The
active area of OPV cells was 6 mm2. To obtain the average data
related to device performance, several batches of devices (8 cells
per batch) for each set of conditions were fabricated and tested.
The current–voltage curves under illumination were measured
using an Abet solar simulator with a Keithley 4200 source
measurement unit under AM 1.5G illumination (100 mW cmꢀ2),
aer spectral mismatch correction under an ambient atmo-
sphere at 25 ꢂC. Hole-only devices were fabricated with a
structure of ITO/MoO3 (8 nm)/ASQ:PC71BM (80 nm)/Au
(100 nm). External quantum efficiency (EQE) measurements
were performed in air using a QE/IPCE Measurements Solar Cell
Scan 100 (ZOLIX) system. Samples for atomic force microscopy
(AFM) measurements were prepared by casting ASQ:PC71BM
blended thin lms on ITO-coated glass substrates with pre-
deposited MoO3 layer (8 nm).
˚
˚
2. Experimental section
2.1 General
NMR spectra were recorded on a Bruker Avance AV II-400 MHz
instrument with tetramethylsilane as internal standard. High
resolution mass spectra were measured on a Shimadzu LCMS-
IT-TOF. FT-IR spectra were recorded on a Perkin-Elmer 2000
infrared spectrometer with KBr pellets under ambient atmo-
sphere. Absorption spectra of the target compounds in 5 ꢁ 10ꢀ6
mol Lꢀ1 chloroform solution and thin lm states were
measured with
a Hitachi U-4100 UV-vis-NIR scanning
spectrophotometer.
2.2 Electrochemistry
Cyclic voltammetry was performed in 0.10 mol Lꢀ1 Bu4NPF6/
anhydrous acetonitrile (2.5 ꢁ 10ꢀ4 mol Lꢀ1) with a LK 2005A
electrochemical work station, using a three-electrode cell with a
Pt working electrode, a Pt wire counter electrode, and a Ag/AgNO3
2.5 Synthesis
(0.1 mol Lꢀ1 in acetonitrile) reference electrode. Solutions were The synthetic routes of intermediates and target molecules are
deoxygenated by Ar bubbling for 30 min prior to each experi- outlined in Scheme 2. Compounds 2, 3a, and 3b were prepared
ment. At the end of each measurement, ferrocenium/ferrocene according to the procedures described in the literature.20,29
redox couple was used as an internal potential reference.
Butanol and toluene were freshly distilled from sodium prior to
18314 | J. Mater. Chem. A, 2014, 2, 18313–18321
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
Scheme 2 Synthetic routes to compounds 1a and 1b.
use. All the other chemicals, reagents, and solvents were used as
received from the suppliers.
2-[(3-Butyl-1,1-dimethyl-1H-benzo[e]indol-2(3H)-ylidene)-
methyl]-4-[(2,6-dihydroxy-4-(1,3,3a,8b-tetrahydrocyclopenta[b]-
indol-4(2H)-yl)phenyl)-2,5-dien-1-ylidene]-3-oxocyclobut-1-en-1-
olate (1a). A mixture of compound 2 (1.10 g, 3.06 mmol) and 5a
(0.98 g, 3.67 mmol) in butanol (15 mL) and toluene (45 mL) was
reuxed under Ar for 36 h. Then the reaction mixture was
concentrated in vacuo, and the crude product was puried by
column chromatography (hexane/ethyl acetate ¼ 2/1, v/v) and
followed by recrystallization from CH2Cl2/MeOH ꢂto give golden
shiny crystals of 1a (1.30 g, 72%). m.p. 235–237 C. FT-IR: v ¼
1636 cmꢀ1 (squaraine carbonyl). 1H NMR (400 MHz, CDCl3, d):
12.36 (s, 2H, OH), 8.23 (d, J ¼ 8.8 Hz, 1H, ArH), 7.96 (d, J ¼ 8.0
Hz, 1H, ArH), 7.95 (d, J ¼ 8.8 Hz, 1H, ArH), 7.66 (t, J ¼ 7.2 Hz,
1H, ArH), 7.53 (t, J ¼ 7.6 Hz, 1H, ArH), 7.39 (d, J ¼ 8.8 Hz, 2H,
ArH), 7.18 (t, J ¼ 7.2 Hz, 2H, ArH), 6.95 (t, J ¼ 7.6 Hz, 1H, ArH),
6.36 (s, 2H, ArH), 5.99 (s, 1H, CH), 4.69 (br, 1H, CH), 4.24 (t, J ¼
7.6 Hz, 2H, CH2), 3.92 (t, J ¼ 7.6 Hz, 1H, CH), 2.04 (s, 6H, CH3),
2.01–1.93 (m, 2H, CH2), 1.92–1.84 (m, 2H, CH2), 1.71–1.64 (m,
2H, CH2), 1.56–1.46 (m, 2H, CH2), 1.45–1.37 (m, 2H, CH2), 1.04
(t, J ¼ 7.2 Hz, 3H, CH3). 13C NMR (100 MHz, CDCl3, d): 174.3,
168.6, 168.3, 151.3, 143.1, 137.7, 135.8, 135.0, 131.0, 129.2,
128.8, 127.3, 126.8, 126.4, 124.4, 123.8, 121.7, 121.1, 112.4,
109.4, 103.5, 95.8, 86.5, 67.6, 51.1, 44.5, 43.4, 33.6, 32.8, 28.7,
25.4, 23.3, 19.3, 12.8. HRMS (ESI)+ m/z: [M + H]+ calcd for
4-(3,5-Dimethoxyphenyl)-1,2,3,3a,4,8b-hexahydrocyclopenta
[b]indole (4a). A mixture of compound 3a (0.90 g, 5.65 mmol),
1-bromo-3,5-dimethoxybenzene (1.23 g, 5.65 mmol), NaOBu-t
[sodium tert-butoxide] (0.81 g, 8.48 mmol), Pd(OAc)2 [palla-
dium(II) acetate] (38 mg, 3%), and HP(t-Bu)3BF4 [tri(tert-
butyl)phosphine tetrauoroborate] (81 mg, 5%) in anhydrous
toluene (50 mL) was reuxed under Ar for 5 h. Aer cooling
down, the reaction mixture was ltered, and the ltrate was
concentrated in vacuo. The resulting crude product was
puried by silica gel column chromatography (hexane/ethyl
acetate ¼ 15/1, v/v) to give 4a (1.50 g, 90%) as a pale yellow oil.
1H NMR (400 MHz, CDCl3, d): 7.12 (d, J ¼ 7.2 Hz, 1H, ArH),
7.09–7.03 (m, 2H, ArH), 6.76–6.72 (m, 1H, ArH), 6.47 (d, J ¼
2.0 Hz, 2H, ArH), 6.12 (t, J ¼ 2.0 Hz, 1H, ArH), 4.72–4.67
(m, 1H, CH), 3.84–3.81 (m, 1H, CH), 3.79 (s, 6H, CH3), 2.07–
1.81 (m, 4H, CH2), 1.68–1.60 (m, 1H, CH2), 1.55–1.44
(m, 1H, CH2).
5-(1,3,3a,8b-Tetrahydrocyclopenta[b]indol-4(2H)-yl)benzene-
1,3-diol (5a). A solution of compound 4a (1.50 g, 5.07 mmol) in
anhydrous CH2Cl2 (60 mL) was added dropwise into BBr3
solution [boron tribromide] (50 mL, 1 mol Lꢀ1 in anhydrous
CH2Cl2, 50.7 mmol) slowly in an ice bath. Then the reaction
mixture was stirred at room temperature for 24 h and decanted
into ice water (200 mL). The organic phase was separated and
the aqueous phase was extracted with CH2Cl2 three times. The
combined organic phase was washed with saturated aqueous
NaHCO3 solution and water, dried over anhydrous Na2SO4, and
concentrated in vacuo. The resulting crude product was further
puried on silica gel column chromatography using dichloro-
methane (DCM) and methanol (100 : 1) as eluent to yield a
yellow solid compound 5a (0.98 g, 72%). 1H NMR (400 MHz,
CDCl3, d): 7.14–7.06 (m, 3H, ArH), 6.83–6.78 (m, 1H, ArH), 6.38
(s, 2H, ArH), 6.08 (s, 1H, ArH), 4.64–4.60 (m, 1H, CH), 3.87–3.78
(m, 1H, CH), 2.00–1.88 (m, 4H, CH2), 1.68–1.60 (br, 1H, CH2),
1.44–1.39 (br, 1H, CH2).
C
40H39N2O4, 611.2921; found, 611.2919.
4-(3,5-Dimethoxyphenyl)-1,2,3,4-tetrahydrocyclopenta[b]-
indole (4b). A mixture of compound 3b (4.20 g, 26.7 mmol),
1-bromo-3,5-dimethoxybenzene (5.80 g, 26.7 mmol), NaOBu-t
(5.13 g, 53.4 mmol), Pd(OAc)2 (240 mg, 1.07 mmol), and HP(t-
Bu)3BF4 (619 mg, 2.14 mmol) in anhydrous toluene (200 mL)
was reuxed under Ar for 5 h. The reaction mixture was
cooled down and ltered, then the ltrate was concentrated
in vacuo. The crude product was puried by silica gel column
chromatography (hexane/ethyl acetate ¼ 15/1, v/v) to give 4b
(5.78 g, 74%) as white solid. m.p. 90–91 ꢂC. 1H NMR (400
MHz, CDCl3, d): 7.51–7.46 (m, 2H, ArH), 7.13–7.11 (m, 2H,
ArH), 6.60 (d, J ¼ 2.0 Hz, 2H, ArH), 6.44 (t, J ¼ 2.4 Hz, 1H,
ArH), 3.83 (s, 6H, CH3), 2.94–2.88 (m, 4H, CH2), 2.58–2.51 (m,
2H, CH2).
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18313–18321 | 18315

View Article Online
Journal of Materials Chemistry A
Paper
107.7, 103.7, 103.6, 90.0, 53.2, 45.1, 30.2, 28.4, 27.6, 26.0, 24.6,
20.4, 13.8. HRMS (ESI)+ m/z: [M + H]+ calcd for C40H37N2O4,
609.2764; found, 609.2763.
5-(2,3-Dihydrocyclopenta[b]indol-4(1H)-yl)benzene-1,3-diol
(5b). A solution of compound 4b (3.00 g, 10.2 mmol) in anhy-
drous CH2Cl2 (120 mL) was added dropwise into BBr3 (100 mL,
1 mol Lꢀ1 in anhydrous CH2Cl2, 102 mmol) slowly in an ice
bath. Then the reaction solution was stirred at room tempera-
ture for 24 h and decanted into ice water (400 mL). The organic
phase was separated, and the aqueous phase was extracted with
CH2Cl2 three times. The combined organic phase was washed
with saturated aqueous NaHCO3 and water, dried over anhy-
drous Na2SO4, the concentrated in vacuo. The resulting crude
product was further puried on silica gel column chromatog-
raphy using DCM–methanol (100 : 1) as eluent to yield a pale
yellow solid compound 5b (1.90 g, 70%). 1H NMR (400 MHz,
CDCl3, d): 7.51–7.45 (m, 2H, ArH), 7.14–7.09 (m, 2H, ArH), 6.51
(d, J ¼ 2.0 Hz, 2H, ArH), 6.31 (t, J ¼ 2.0 Hz, 1H, ArH), 2.91–2.87
(m, 4H, CH2), 2.57–2.50 (m, 2H, CH2).
3. Results and discussion
3.1 Synthesis
The synthetic routes to compounds 1a and 1b are outlined in
Scheme 2. Intermediates 2, 3a and 3b were prepared according
to literature reports.20,29 Intermediates 4a and 4b were synthe-
sized through the Buchwald–Hartwig cross-coupling of 3a/3b
with 1-bromo-3,5-dimethoxybenzene. Then, methoxy groups of
4a and 4b were converted into hydroxy groups with BBr3 to
render compounds 5a and 5b respectively. Finally, the two
objective molecules 1a and 1b were prepared by condensation
of 2 with 5a and 5b, respectively. The molecular structures of the
1a and 1b were characterized by 1H NMR, 13C NMR and HR-ESI-
MS, and conrmed by X-ray crystallographic analysis. It is
noteworthy that both the two compounds show good solubility
in common organic solvents like chloroform, chlorobenzene
and o-dichlorobenzene. Additionally, high quality lms could
be obtained through spin-coating, suggesting that they are very
suitable for solution processing.
2-[(3-Butyl-1,1-dimethyl-1H-benzo[e]indol-2(3H)-ylidene)
methyl]-4-[(4-(2,3-dihydrocyclopenta[b]indol-4(1H)-yl)-2,6-dihy-
droxyphenyl)-2,5-dien-1-ylidene]-3-oxocyclobut-1-en-1-olate
(1b). A mixture of compound 2 (1.35 g, 3.73 mmol) and 5b
(1.20 g, 4.52 mmol) in butanol (20 mL) and toluene (60 mL) was
reuxed under Ar for 36 h. Then the reaction mixture was
concentrated in vacuo, and the resulting crude product was
puried by silica gel column chromatography with hexane/ethyl
acetate as eluent (1/1, v/v), followed by recrystallization from
DCM/MeOH to give 1b (0.56 g, 25%). m.p. 257–259 ꢂC. FT-IR: v
3.2 Single crystal X-ray crystallography
1
¼ 1629 cmꢀ1 (squaraine carbonyl). H NMR (400 MHz, CDCl3,
To conrm the molecular structures and gain insight into the
intermolecular packing properties of 1a and 1b, their crystal
structures were determined via single crystal X-ray crystallog-
raphy. Both of 1a and 1b display monoclinic crystal system with
same space group of P21/n, and their ORTEP diagrams as well as
intermolecular packing patterns are shown in Fig. 1. The bond
d): 12.24 (s, 1H, OH), 12.16 (s, 1H, OH), 8.26 (d, J ¼ 8.8 Hz, 1H,
ArH), 8.03 (d, J ¼ 8.8 Hz, 1H, ArH), 8.02 (d, J ¼ 8.4 Hz, 1H, ArH),
7.72–7.69 (m, 2H, ArH), 7.61 (t, J ¼ 7.2 Hz, 1H, ArH), 7.48–7.43
(m, 2H, ArH), 7.16–7.14 (m, 2H, ArH), 6.56 (m, 2H, ArH), 6.20
(s, 1H, CH), 4.38 (t, J ¼ 7.6 Hz, 2H, CH2), 3.04 (t, J ¼ 6.4 Hz, 2H,
CH2), 2.89 (t, J ¼ 6.4 Hz, 2H, CH2), 2.58–2.51 (m, 2H, CH2), 2.07
(s, 6H, CH3), 1.98–1.90 (m, 2H, CH2), 1.57–1.51 (m, 2H, CH2),
1.07 (t, J ¼ 7.2 Hz, 3H, CH3). 13C NMR (100 MHz, o-dichloro-
benzene-d4, d): 184.0, 180.9, 178.3, 174.3, 167.3, 162.6, 162.2,
146.6, 144.8, 140.7, 138.1, 137.3, 133.3, 133.0, 132.8, 130.9,
128.3, 128.1, 126.4, 122.8, 121.7, 121.2, 119.3, 112.5, 110.7,
˚
˚
lengths of both C8–N1 (1.376 A) and C4–C5 (1.419 A) bonds in
1a are shorter than the corresponding ones in 1b (C27–N2:
˚
˚
1.411 A; C22–C24: 1.433 A), hence in comparison with cyclo-
penta[b]indolyl group, the cyclopenta[b]indolinyl substituent
should contribute more signicantly to the extension of
conjugation system in the objective compound, which might be
attributed to its more electron-donating nitrogen atom. This
Fig. 1 ORTEP and packing diagrams of 1a and 1b in their crystal structures.
18316 | J. Mater. Chem. A, 2014, 2, 18313–18321
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
deduction is further conrmed by 1H NMR characterization 36.67ꢂ for 1b and 9.87ꢂ for 1a; while the torsional angles
results, since the H1 signal of phenyl ring in 1a is found to show between the squarate ring and phenyl unit were calculated to be
more shielding effect than that in 1b (6.36 ppm vs. 6.56 ppm). As 4.93ꢂ for 1b and 0.84ꢂ for 1a. In addition, the hydrogen bond
˚
˚
expected, both indoline and indole subunits are found to show lengths (O/H–O) were measured to be 1.783 A and 1.794 A in
˚
˚
nearly coplanar geometries. Due to the different hybridization 1b, but 1.560 A and 1.625 A in 1a; and the corresponding
states of the C1 and C2 atoms (marked in Scheme 1) in 1a and phenolic protons of 1b (12.24 ppm, 12.16 ppm) show more
1b, the cyclopentane ring of 1a is found to align almost downeld-shied 1H NMR signals than those of 1a (12.36 ppm),
orthogonally to the indoline segment; yet that of 1b is nearly conrming that the intramolecular H-bonding interactions in
coplanar with the indole segment. Nevertheless, despite the 1b are weaker than those in 1a. Consequently, the molecular
more planar conformation of cyclopenta[b]indolyl relative to backbone of 1a possesses a more planar conguration relative
cyclopenta[b]indolinyl segment, the dihedral angles between to 1b, which might promote more effective intermolecular
phenyl and indolyl/indolinyl moieties were observed to be packing.
As illustrated in Fig. 1, both 1a and 1b were found to exhibit
relatively strong edge-to-p intermolecular stacking effect. But
for 1a, the distance between its two nearest neighboring mole-
˚
˚
cules is 3.354 A, which is shorter than that of 1b (3.465 A);
moreover, the density of 1a is much higher than that of 1b
(1.317 g cmꢀ3 vs. 1.187 g cmꢀ3) in single crystal states. Thus, the
intermolecular stacking of compound 1a was conrmed to be
more effective than that of 1b, which should be propitious to the
enhancement of charge carrier mobility.24,25,30–32
3.3 Photophysical characterization
The UV-vis absorption spectra of 1a and 1b in dilute chloroform
solution and thin solid lms samples spin-casted from chloro-
form solution are shown in Fig. 2, and the data are summarized
in Table 1. In solution, analogous to some ASQ dyes, 1a and 1b
both exhibit an intense absorption band with intramolecular-
charge-transfer (ICT) transition feature.12,33 The lmax of 1a and
1b are 681 nm and 638 nm, respectively, and the much bath-
ochromic-shied lmax of 1a relative to 1b suggests that 1a owns
a more extended molecular conjugation system. In addition to
its more red-shied ICT absorption band, 1a possesses a higher
absorption coefficient than that of 1b (2.4 ꢁ 105 Mꢀ1 cmꢀ1 vs.
1.2 ꢁ 105 Mꢀ1 cmꢀ1). This might be attributed to the more
efficient ICT effect within compound 1a due to its better
molecular planarity. In lm states, both the two compounds
display drastically broadened and red-shied absorption bands
(data shown in Table 1), which should arise from their intense
intermolecular interactions in condensed state. It is noteworthy
that although in dilute solution, the full width at half maxima
(FWHM) of the absorption band of 1a (39 nm) is much narrower
than that of 1b (78 nm), it becomes slightly broader than that of
1b in lm state (174 nm vs. 165 nm), which may be attributed to
the more condensed packing of 1a relative to 1b. Through their
Fig. 2 UV-vis absorption spectra of compounds 1a and 1b in chlo-
roform solution (top) and thin films (bottom). Insets are the photos of
solution and films.
Table 1 Photophysical and electrochemical properties of compounds 1a and 1b
FWHM [nm]
labs solutiona [nm]
Compound (3, 105 Mꢀ1 cmꢀ1
)
l
abs lm [nm] Solution Film DEoptb [eV] Eox [V] Ered [V] HOMOc [eV] LUMOc [eV] DECVd [eV]
1a
1b
681 (2.4)
638 (1.2)
654, 712
608, 658
39
78
174
165
1.54
1.64
0.37
0.52
ꢀ1.20
ꢀ1.15
ꢀ5.10
ꢀ5.22
ꢀ3.59
ꢀ3.63
1.51
1.59
Measured in CHCl3; c ¼ 5.0 ꢁ 10ꢀ6 mol Lꢀ1
.
On thin lms. c HOMO and LUMO were measured by cyclic voltammetry with Fc/Fc+ as internal
a
b
standard. DECV ¼ HOMO ꢀ LUMO.
d
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18313–18321 | 18317

View Article Online
Journal of Materials Chemistry A
Paper
absorption spectra in solid lm state, the optical band gaps of density functional theory (DFT) calculations were performed
1a and 1b were estimated to be 1.54 eV and 1.64 eV, both close using Gaussian 09 soware. For the singlet ground states (S0) of
to the ideal bandgap (1.50 eV) of photovoltaic donor materials.34 1a and 1b, the optimized molecular structures were calculated
at B3LYP/6-31G level. The optimized structures of the
compounds along with the numbering of important atoms, the
main geometrical parameters calculated, and some experi-
3.4 Electrochemical characterization
To estimate the energy levels of the frontier orbitals of the
objective compounds, cyclic voltammetry (CV) measurements
were carried out on 1a and 1b in acetonitrile solution with
Bu4NPF6 as the supporting electrolyte. The HOMO and LUMO
energy levels of the objective compounds were determined from
the peak of the rst oxidation and reduction waves by
comparison with the Fc/Fc+ redox couple whose energy level is
4.80 eV below vacuum.35 The cyclic voltammograms of 1a and
1b are shown in Fig. 3, and the data are summarized in Table 1.
During the anodic scan from 0 to 1.0 V, both the two
compounds show reversible two-electron oxidation processes,
with the peaks of the rst oxidation wave located at 0.37 V for 1a
and 0.52 V for 1b. Consequently, the HOMO levels of 1a and 1b
were estimated to be ꢀ5.10 eV and ꢀ5.22 eV, and the relatively
low HOMO levels of both compounds would benet much to
achieve relatively high Voc in BHJ-OPV devices.36 The lower
HOMO energy level of compound 1b relative to 1a can be
rationally ascribed to the presence of the less electron-rich
cyclopenta[b]indolyl substituent. Upon cathodic scan, 1a and 1b
were observed to show irreversible reduction signals with peak
waves of ꢀ1.20 V for 1a and ꢀ1.15 V for 1b; accordingly, their
LUMO energy levels were estimated to be ꢀ3.59 eV and ꢀ3.63
eV, respectively. Taking into account that the LUMO energy level
of PC71BM is as low as ꢀ4.00 eV,37 in the blending system of
PC71BM:1a (or 1b), efficient exciton dissociation and charge
separation could be realized.34 The electrochemical band gaps
(DECV) of compounds 1a and 1b were calculated to be 1.51 eV
and 1.59 eV, respectively, which is in line with their corre-
sponding optical band gaps.
mental data of 1a and 1b derived from X-ray crystallography can
be found in Supporting Information. The close similarity in
bond lengths and bond angles between calculated and experi-
mental values reveals the reliability of our computational
results. The electronic densities of the HOMO and LUMO
orbitals of the two objective compounds are depicted in Fig. 4.
For both 1a and 1b, the indolinyl and indolyl segments were
observed to contribute signicantly to the HOMOs, but little to
the LUMOs; therefore, the LUMO energy levels of 1a and 1b
should be similar, while the HOMO energy level of compound
1a should be higher than that of 1b due to the more electron-
donating indolinyl relative to the indolyl group.
3.6 Photovoltaic cell properties
Since both 1a and 1b show desirable solubility in common
solvents, intense absorption in red and near infrared region,
BHJ-OPV cells using 1a and 1b as electron donor materials and
[6,6]-phenylC71 butyric acid methyl ester (PC71BM) as electron
3.5 Theoretical calculation
To gain insight into the origin of the dissimilar electrochemical
properties of 1a and 1b induced by different substituents,
Fig. 4 HOMO and LUMO orbitals of compounds 1a (left) and 1b (right).
Fig. 5 J–V curves of the BHJ-OPVs based on 1a and 1b (illuminated
Fig. 3 Cyclic voltammograms of 1a and 1b in acetonitrile solution.
18318 | J. Mater. Chem. A, 2014, 2, 18313–18321
under AM 1.5G, 100 mW cmꢀ2).
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
model,39–41 and the structures were ITO/MoO3 (8 nm)/active
layer (80 nm)/Au (100 nm), where the active layers were 1a, 1b,
1a : PC71BM (1 : 8), or 1b : PC71BM (1 : 8) respectively. The hole
mobility of these pristine ASQ lms as well as ASQ:PC71BM
composite lms was calculated by tting the J–V characteristics
of the devices (shown in Fig. 6) using the Mott–Gurney law.42 As
expected, the results indicate that the hole mobility of 1a in
pristine and blending lm states are both much higher than
those of 1b. In pristine lm state, the hole mobility of 1a (1.63 ꢁ
10ꢀ5 cm2 Vꢀ1 sꢀ1) is about 2 times higher than that of 1b (4.88 ꢁ
10ꢀ6 cm2 Vꢀ1 sꢀ1); while in composition lm state, the hole
mobility of the 1a:PC71BM blending lm is calculated to be
9.18 ꢁ 10ꢀ6 cm2 Vꢀ1 sꢀ1, which is 7 times higher than that of
1b:PC71BM one (1.16 ꢁ 10ꢀ6 cm2 Vꢀ1 sꢀ1). The subtle structural
variation in the end-capping substituents of 1a and 1b result in
drastically different hole mobility in both pristine and blending
lms due to the different intermolecular packing of these two
ASQ compounds.
As the quality of interpenetrating network within the
blending lms would also have great effect on the carrier
mobility,43,44 atomic force microscopies (AFM) measurements
were conducted on the blending lms of 1a or 1b with PC71BM
(1 : 8). As shown in Fig. S2,† the two lm samples show obvi-
ously different morphology. The 1a:PC71BM blended lm shows
a smoother surface with a smaller surface roughness (0.25 nm
vs. 0.44 nm) than that of 1b:PC71BM, implying the inter-
penetrating network formed by 1a:PC71BM may be more bene-
cial to carrier transport.
Most importantly, the FF of 1a-based device is up to 0.47,
which is much higher than many devices using squaraine
derivatives as photovoltaic donor materials (0.27–0.40).6,11–13,20–23
Nevertheless, the FF of 1b-based device is only 0.32, which is
much lower than that of 1a, mostly due to its much lower hole
mobility of its blended lm. Besides hole mobility, the series
resistance (RS) of device is also a factor affecting FF and low
series resistance leads to high FF.30 The RS of devices were
dened from the slope of J–V curves in Fig. 5, which shows that
the RS of 1a-based device is much lower than that of 1b (10.87 U
cm2 vs. 61.24 U cm2). Consequently, the much higher FF of 1a-
based device relative to 1b-based one can be rationally attrib-
uted to the higher hole mobility as well as lower RS of the 1a-
device. Furthermore, in accordance with the literature report,8
Table
2 Photovoltaic performance of BHJ-OPVs based on
ASQs : PC71BM (1 : 8)
Active layer
(w/w)
Jsc [mA cmꢀ2
]
Voc [V]
FF
PCE [%]
1a
9.39 ꢃ 0.27
10.19 ꢃ 0.30
6.67 ꢃ 0.26
7.36 ꢃ 0.28
0.83 ꢃ 0.02 0.47 ꢃ 0.02 3.6 ꢃ 0.3
0.81 ꢃ 0.02 0.48 ꢃ 0.02 3.9 ꢃ 0.2
0.95 ꢃ 0.03 0.32 ꢃ 0.01 2.0 ꢃ 0.2
0.91 ꢃ 0.03 0.34 ꢃ 0.01 2.2 ꢃ 0.2
1aa
1b
1ba
a
Annealed at 90 ꢂC for 10 min.
acceptor material were fabricated in order to evaluate their
photovoltaic performance. The device structure was ITO/MoO3
(8 nm)/ASQ:PC71BM (80 nm)/LiF (0.8 nm)/Al (100 nm), where
MoO3 and LiF served as anode and cathode modication
layers,38 and the weight ratio of ASQ : PC71BM was selected as
1 : 8. The typical J–V curves of these BHJ-OPV cells are shown in
Fig. 5, and the representative characterization data are
summarized in Table 2.
The 1b-based OPV shows a higher Voc than that of 1a (0.95 vs.
0.83 V), which should stem from the lower HOMO energy level
of compound 1b relative to 1a. Nevertheless, BHJ-OPV device
with 1a:PC71BM composite system as active layer shows a
signicantly high FF of up to 0.47, together with a high Jsc of
9.39 mA cmꢀ2; while the 1b-based device has a moderate FF of
0.32 and Jsc of 6.67 mA cmꢀ2. Consequently, BHJ-OPV device
based on 1a shows an impressive PCE of 3.6%, which is much
higher than that of the 1b-based device (PCE: 2.0%).
The EQE curves of the two devices are shown in Fig. 6.
Obviously, squaraine dyes contribute more to the absorption in
the long wavelength region (550–800 nm), while PC71BM
contributes more to the absorption in the short wavelength
region (300–550 nm). Under the same blend ratio, the EQE data
in 300–550 nm region for 1a are much higher than that of 1b,
possibly due to the higher charge carrier mobility of 1a:PC71BM
lm, while the higher integral value in 550–800 nm for 1a may
be echoing the results of its better absorption and higher carrier
mobility.
To verify this hypothesis, four hole-only devices were
designed according to space-charge-limited current (SCLC)
Fig. 6 (a) External quantum efficiency curves of the two BHJ-OPVs and J–V characteristics of (b) pristine and (c) blend films single carrier devices
used for hole mobility measurement.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18313–18321 | 18319

View Article Online
Journal of Materials Chemistry A
Paper
our experimental results reveal that higher density of the
squaraine dye would result in higher FF of the OPV device.
Thereupon, for photovoltaic materials, density, the easily
measured physical quantity, might also be an important prop-
erty deserving of investigation.
Notes and references
¨
1 S. Gunes, H. Neugebauer and N. S. Saricici, Chem. Rev.,
2007, 107, 1324–1338.
2 Y. Liang and L. Yu, Acc. Chem. Res., 2010, 43, 1227–1236.
3 Y. Chen, X. Wan and G. Long, Acc. Chem. Res., 2013, 46, 2645–
2655.
It is noteworthy that the photovoltaic characterizations on 1a
and 1b are quite preliminary, since only the as-prepared
ASQs:PC71BM blending lms were used as the photoactive layer,
and no optimization on the device structure, layer thickness as
well as post-treatment conditions was conducted for achieving
higher PCEs. Therefore, much improved photovoltaic perfor-
mance should be expected when further optimizations like
thermo-annealing,38 solvent-annealing10 and incorporation of
additives45 are carried out. For example, aer the active layers
were thermo-annealed at 90 ꢂC for 10 min, the 1a and 1b based
devices both show increased Jsc and FF, resulting in a maximum
PCE of 4.1% and 2.4%, respectively (vide Table 2). Moreover, the
PCE of 4.1% is one of the highest PCEs for ASQ-based BHJ
SMOSCs,6,11–13,20 suggesting that ASQs are indeed quite prom-
ising photovoltaic materials.
4 Y. Lin, Y. Li and X. Zhan, Chem. Soc. Rev., 2012, 41, 4245–4272.
¨
5 A. Mishra and P. Bauerle, Angew. Chem., Int. Ed., 2012, 51,
2020–2067.
¨
6 U. Mayerhoffer, K. Deing, K. Grub, H. Braunschweig,
¨
K. Meerholz and F. Wurthner, Angew. Chem., Int. Ed., 2009,
48, 8776–8779.
7 S. Y. Wang, L. Hall, V. V. Diev, R. Haiges, G. D. Wei, X. Xiao,
P. I. Djurovich, S. R. Forrest and M. E. Thompson, Chem.
Mater., 2011, 23, 4789–4798.
8 G. Wei, X. Xiao, S. Wang, K. Sun, K. J. Bergemann,
M. E. Thompson and S. R. Forrest, ACS Nano, 2012, 6, 972–
978.
9 J. S. Huang, T. Goh, X. K. Li, M. Y. Sfeir, E. A. Bielinski,
S. Tomasulo, M. L. Lee, N. Hazari and A. D. Taylor, Nat.
Photonics, 2013, 7, 480–486.
10 G. D. Wei, S. Y. Wang, K. Sun, M. E. Thompson and
S. R. Forrest, Adv. Energy Mater., 2011, 1, 184–187.
4. Conclusion
Two ASQ derivatives bearing similar molecular skeleton but
different electron-donating end-cappers, namely 1a and 1b, have
been designed and synthesized. The small variations in their
molecular structures result in dramatically different electronic
and steric structures, and hence different photovoltaic perfor-
mances. In comparison with compound 1b bearing cyclopenta[b]-
indolyl substituent, compound 1a with cyclopenta[b]indolinyl
as end-capper shows more planar molecular geometry, more
intense intermolecular interactions, higher density in crystal-
line state, bathochromic-shied absorption spectra, higher
absorption coefficient, and higher hole mobility. Consequently,
BHJ-SMOPV device based on 1a displays drastically higher FF
and Jsc than those of 1b-device, and a high PCE of 4.1% is
achieved. All these preliminary results demonstrate that ASQs
are quite promising candidates for SMOPV application, and
their optoelectronic properties could be easily ne-tuned by
manipulating the character of D and D0 subunits. Moreover, for
1a and 1b, the higher density of the squaraine dye would result
in higher FF of the OPV device, which is in accordance with the
literature report. Therefore, density might also be an important
physical quantity deserving of investigation for the evaluation of
the photovoltaic performance of organic materials.
¨
¨
11 K. C. Deing, U. Mayerhoffer, F. Wurthner and K. Meerholz,
Phys. Chem. Chem. Phys., 2012, 14, 8328–8334.
12 S. So, H. Choi, H. M. Ko, C. Kim, S. Paek, N. Cho, K. Song,
J. K. Lee and J. Ko, Sol. Energy Mater. Sol. Cells, 2012, 98,
224–232.
13 L. Beverina, R. Ruffo, M. M. Salamone, E. Ronchi, M. Binda,
D. Natali and M. Sampietro, J. Mater. Chem., 2012, 22, 6704–
6710.
14 S. S. Pandey, R. Watanabe, N. Fujikawa, Y. Ogomi,
Y. Yamaguchi and S. Hayase, Proc. SPIE, 2011, 8111.
15 J. Fabian, H. Nakazumi and M. Matsuoka, Chem. Rev., 1992,
92, 1197–1226.
16 K. Y. Law and F. C. Bailey, J. Org. Chem., 1992, 57, 3278–3286.
17 C. Qin, W. Y. Wong and L. Han, Chem.–Asian J., 2013, 8,
1706–1719.
18 W. D. Wang, A. Fu, J. S. You, G. Gao, J. B. Lan and L. J. Chen,
Tetrahedron, 2010, 66, 3695–3701.
19 L. Beverina and P. Salice, Eur. J. Org. Chem., 2010, 7, 1207–
1225.
20 D. Yang, Q. Yang, L. Yang, Q. Luo, Y. Huang, Z. Lu and
S. Zhao, Chem. Commun., 2013, 49, 10465–10467.
21 F. Silvestri, M. D. Irwin, L. Beverina, A. Facchetti,
G. A. Pagani and T. J. Marks, J. Am. Chem. Soc., 2008, 130,
17640–17641.
Acknowledgements
The nancial support for this work of the National Natural
Science Foundation of China (project no. 21190031 and
21372168) and New Century Excellent Talents in University
(grant no. NCET-10-0220) is acknowledged. Thanks are due to
the Analytical & Testing Center of Sichuan University and
Comprehensive Training Platform of Specialized Laboratory,
College of Chemistry, Sichuan University for providing NMR,
HRMS and single crystal X-ray diffraction data for the inter-
mediates and objective molecules.
¨
22 S. Bruck, C. Krause, R. Turrisi, L. Beverina, S. Wilken,
¨
W. Saak, A. Lutzen, H. Borchert, M. Schiek and J. Parisi,
Phys. Chem. Chem. Phys., 2014, 16, 1067–1077.
23 J. Roncali, P. Leriche and P. Blanchard, Adv. Mater., 2014, 26,
3821–3838.
24 O. P. Lee, A. T. Yiu, P. M. Beaujuge, C. H. Woo,
T. W. Holcombe, J. E. Millstone, J. D. Douglas, M. S. Chen
´
and J. M. Frechet, Adv. Mater., 2011, 23, 5359–5363.
18320 | J. Mater. Chem. A, 2014, 2, 18313–18321
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
25 D. Bagnis, L. Beverina, H. Huang, F. Silvestri, Y. Yao, H. Yan, 35 J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt,
¨
G. A. Pagani, T. J. Marks and A. Facchetti, J. Am. Chem. Soc.,
2010, 132, 4074–4075.
H. Bassler, M. Porsch and J. Daub, Adv. Mater., 1995, 7,
551–554.
26 D. Demeter, V. Jeux, P. Leriche, P. Blanchard, Y. Olivier, 36 L. Huo, J. Hou, S. Zhang, H. Y. Chen and Y. Yang, Angew.
J. Cornil, R. Po and J. Roncali, Adv. Funct. Mater., 2013, 23,
4854–4861.
Chem., Int. Ed., 2010, 49, 1500–1503.
37 G. Chen, H. Sasabe, Z. Wang, X. Wang, Z. Hong, J. Kido and
Y. Yang, Phys. Chem. Chem. Phys., 2012, 14, 14661–14666.
¨
¨
¨
27 H. Burckstummer, N. M. Kronenberg, M. Gsanger, M. Stolte,
¨
K. Meerholz and F. Wurthner, J. Mater. Chem., 2010, 20, 240– 38 Q. Q. Yang, D. B. Yang, S. L. Zhao, Y. Huang, Z. Xu, X. D. Liu,
243.
W. Gong, X. Fan, Q. Y. Huang and X. R. Xu, Appl. Surf. Sci.,
2013, 284, 849–854.
28 A. S. Hashmi, M. Rudolph, J. W. Bats, W. Frey, F. Rominger
and T. Oeser, Chem.–Eur. J., 2008, 14, 6672–6678.
29 G. S. Welmaker and J. E. Sabalski, Tetrahedron Lett., 2004, 45,
4851–4854.
30 B. Qi and J. Wang, Phys. Chem. Chem. Phys., 2013, 15, 8972–
8982.
31 G. Chen, H. Sasabe, Y. Sasaki, H. Katagiri, X. F. Wang,
T. Sano, Z. R. Hong, Y. Yang and J. Kido, Chem. Mater.,
2014, 26, 1356–1364.
39 H. Azimi, A. Senes, M. C. Scharber, K. Hingerl and
C. J. Brabec, Adv. Energy Mater., 2011, 1, 1162–1168.
40 Z. He, C. Zhong, X. Huang, W. Y. Wong, H. Wu, L. Chen,
S. Su and Y. Cao, Adv. Mater., 2011, 23, 4636–4643.
41 V. D. Mihailetchi, J. K. J. van Duren, P. W. M. Blom,
J. C. Hummelen, R. A. J. Janssen, J. M. Kroon,
M. T. Rispens, W. J. H. Verhees and M. M. Wienk, Adv.
Funct. Mater., 2003, 13, 43–46.
32 L. S. Li, Y. Y. Huang, J. B. Peng, Y. Cao and X. B. Peng, J. 42 N. F. Mott and R. W. Gurney, Electronic Process in Ionic
Mater. Chem. A, 2013, 1, 2144–2150. Crystals, Dover, New York, 1964.
33 J. H. Yum, P. Walter, S. Huber, D. Rentsch, T. Geiger, 43 A. M. Ballantyne, L. Chen, J. Dane, T. Hammant, F. M. Braun,
¨
¨
F.
Nuesch,
F.
De
Angelis,
M.
Gratzel
and
M. Heeney, W. Duffy, I. McCulloch, D. D. C. Bradley and
J. Nelson, Adv. Funct. Mater., 2008, 18, 2373–2380.
M. K. Nazeeruddin, J. Am. Chem. Soc., 2007, 129, 10320–
10321.
44 J. F. Chang, B. Q. Sun, D. W. Breiby, M. M. Nielsen,
¨
¨
34 M. C. Scharber, D. Wuhlbacher, M. Koppe, P. Denk,
T. I. Solling, M. Giles, I. McCulloch and H. Sirringhaus,
C. Waldauf, A. J. Heeger and C. L. Brabec, Adv. Mater.,
2006, 18, 789–794.
Chem. Mater., 2004, 16, 4772–4776.
45 N. Cho, S. Paek, J. Jeon, K. Song, G. D. Sharma and J. Ko, J.
Mater. Chem. A, 2014, 2, 12368–12379.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A, 2014, 2, 18313–18321 | 18321