Three-Dimensionally Homoconjugated Carbon-Bridged Oligophenylenevinylene for Perovskite Solar Cells

Article abstract of DOI:10.1021/jacs.6b04002

Stabilization of the radical cationic state of a donor molecule by 3-D homoconjugation was probed using a substituted carbon-bridged oligophenylenevinylene backbone (COPV, or 5,5-diarylindeno[2,1-a]indenes). For molecules bearing electron-donating groups as the 5,5-aryl moieties, a one-electron oxidation of the COPV backbone results in delocalization of the cationic charge over the whole molecule with a small reorganization energy. The compounds forming a stable radical cation by 3-D homoconjugation produce a uniform amorphous film and show high short-circuit current, high fill factor, and hence high power-conversion efficiency when used as a hole-transporting layer of an organic-inorganic hybrid lead perovskite solar cell. This material thus shows a performance and stability in air comparable to those obtained with the benchmark material, spiro-MeOTAD.

Full text of DOI:10.1021/jacs.6b04002

Subscriber access provided by Northern Illinois University
Article
Three-Dimensionally Homoconjugated Carbon-Bridged
Oligophenylenevinylene for Perovskite Solar Cells
Qifan Yan, Yunlong Guo, Anna Ichimura, Hayato Tsuji, and Eiichi Nakamura
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b04002 • Publication Date (Web): 05 Aug 2016
Downloaded from http://pubs.acs.org on August 6, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Three-Dimensionally Homoconjugated Carbon-Bridged
Oligophenylenevinylene for Perovskite Solar Cells
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Qifan Yan, † Yunlong Guo,* Anna Ichimura, Hayato Tsuji,*,† Eiichi Nakamura*
Department of Chemistry, The University of Tokyo, 7ꢀ3ꢀ1 Hongo, Bunkyoꢀku, Tokyo 113ꢀ0033, Japan.
ABSTRACT: Stabilization of the radical cationic state of a donor molecule by 3ꢀD homoconjugation was probed using a
substituted carbonꢀbridged oligophenylenevinylene backbone (COPV or 5,5ꢀdiarylꢀindeno[2,1ꢀa]indenes). For molecules bearing
electronꢀdonating groups as the 5,5ꢀaryl moieties, a oneꢀelectron oxidation of the COPV backbone results in delocalization of the
cationic charge over the whole molecule with a small reorganization energy. The compounds forming a stable radical cation by 3ꢀD
homoconjugation produce a uniform amorphous film and show high shortꢀcircuit current, high fill factor, and hence high powerꢀ
conversion efficiency when used as a holeꢀtransporting layer of an organic–inorganic hybrid lead perovskite solar cell. This
material thus shows a performance and stability in air comparable to that obtained by the benchmark material, spiro-MeOTAD.
high Jsc but poor FF. Compounds 5 and 6, which form a
crystalline film and are incapable of radical cation
stabilization, showed poor performance for Jsc, openꢀcircuit
voltage (Voc), FF and PCE. A simple fluorene 7 lacking the
ability of stabilizing its oxidized form showed a low
performance and poor stability in air.
INTRODUCTION
Stabilization of negative charge on a 3ꢀD πꢀscaffold has
attracted considerable attention from chemists since the
discovery of the utility of fullerenes as electronꢀaccepting
material in organic and perovskite solar cells (SCs). Such 3ꢀD
molecules used as carrierꢀtransporting materials tend to have a
small reorganization energy and to give amorphous
morphology suitable for carrier transport in a film. On the
other hand, there has been a paucity of studies on the potential
of 3ꢀD stabilization of positive charge for hole transportation
where, chemically speaking, a radical cation of holeꢀ
transporting material plays the role. Having found 1 that a
carbonꢀbridged oligophenylenevinylene backbone (COPVꢀ1
or 5,5ꢀdiarylꢀindeno[2,1ꢀa]indenes, denoted as COPV
hereafter) provides a structurally robust 2ꢀD πꢀscaffold (cyan,
Figure 1a), we were intrigued by the possibility of 3ꢀD
homoconjugative stabilization of a radical cation through
installation of additional electronꢀdonating aminophenyl
groups orthogonal to the plane of the COPV πꢀsystem (Figure
1a). We report here the synthesis, properties and SC
application of the new electronꢀrich πꢀsystem (1–3; Figure 1b)
examined for the effects of the 3ꢀD homoconjugative
stabilization of a radical cationic state. The compounds form
an effective, amorphous holeꢀtransporting layer (HTL) for a
thinꢀfilm semitransparent2 lead perovskite SC device3, which
are already finding use in "solar window" applications.4 The
perovskite SCs using 1ꢀ3 show a level of powerꢀconversion
efficiency (PCE) and stability in air comparable to that using
a
homoconjugation
Ar2N
PCE 11-12%
NAr2
Ar2N
-2.7
NAr2
-3.9
-4.0
-4.4
FTO
N
-5.0
RO
NAr2: R = Me
OR
-4.9
Au
-5.4
PV
NAr'2: R = hexyl
Buffer
TiOx
b
Ar2N
Ar2N
NAr2
Ar2N
NAr2
5
Ar2N
8
NAr2
3
NAr2
Ph
Ph
Ph
Ph
1
2
C8H17
Ar2N
C8H17
Ar2N
NAr2
Ar2N
NAr2
NAr2
Ph
Ph
Ph
Ph
3
R1
4
R1
popularly employed spiro-MeOTAD (Figure 1b, bottom)–a
benchmark of holeꢀtransporting material.
5
Reference
Ar'2N
Ar2N
Ar2N
NAr2
NAr2
NAr'2
compounds 4–6, which have only the flat 2ꢀD conjugated
system, were found to perform poorly as an HTL in spite of
the similarity of their oxidation potentials to those of 1–3.
Compounds 1–3 form a stable radical cation and produce a
uniform amorphous film, and hence showed high shortꢀcircuit
current (Jsc), high fill factor (FF) and hence high PCE.
Compound 4, which forms an amorphous film but is incapable
of homoconjugative stabilization of a radical cation, showed
R1 R1
5: R1 = Ph, 6: R1 = Me
spiro-MeOTAD
C8H17
C8H17
Ar2N
NAr2
7
Figure 1. SC device configuration and molecular structures in
our study. a) Schematic illustration of threeꢀdimensional
ACS Paragon Plus Environment

Journal of the American Chemical Society
homoconjugation in COPV 1 and SC device configuration as
Page 2 of 9
a
1
2
3
4
5
6
7
well as energy diagram; b) structures of COPV 1–6
(numbering of carbon atoms on COPV are indicated in italic
red numbers) and reference molecules spiro-MeOTAD and 7.
397
402
0.6
418
RESULTS AND DISCUSSION
388
1. Synthesis of COPV 1–6
The flat 2ꢀD πꢀsystem of COPVs has been found to be useful
for dyeꢀsensitized SCs,6 solidꢀstate organic lasers,7 molecular
wires8 and energy and electron donors.9 COPV has a 2ꢀD
conjugated system flanked by geminal substituents (e.g., at
C5) that forms a 3ꢀD scaffold, preventing intermolecular
interactions. To study the feasibility of creating a 3ꢀD
homoconjugated donor system,9 we installed four (4ꢀ
arylamino)phenyl donor groups on both ends of the COPVꢀ
skeleton and on one of the geminal diphenyl groups (at C5) as
in 1–3, expecting that the COPV system and the geminal C5
groups would become homoconjugated in the radical cationic
state. Compound 3 has a spiroꢀlinkage mimicking spiro-
MeOTAD, 2 is an open chain analogue of 3, and 1 is its 4ꢀ
aminophenyl isomer. For comparison, we also synthesized
bis(4ꢀdiarylamino)phenylꢀCOPV 4–6 lacking the electronꢀ
donating geminal groups (Scheme 1).
8
9
0.3
0.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ε
400
500
λ
/nm
b
-0.03
Flu1+
0.26
Flu2+
0.07
-0.15
-0.05
-0.04
0.17
0.17
Scheme 1. Synthesis of COPV 1–5.
0.22
0.14
-0.03
-0.04
PV1+
0.33
0.32
0.21
Ar2N+
0.30
PV2+
0.18
-0.05
0.5
0.0
Potential vs Fc+/Fc (V)
Figure 2. Photoꢀ and electrochemical properties of 1–7. (a)
UV–vis absorption spectra of compounds 1–7 in dilute PhCl
solutions at 5 ꢀM. (b) Differential pulse voltammetry (bottom)
of compounds 1–7 in orthoꢀdichlorobenzene/CH3CN (v/v, 4/1)
at 0.5 mM. Oxidative peaks are labeled with the species
primarily involved in the oxidation. PV, bisdiarylamino COPV
moiety; Flu, bisdiarylaminofluorene moiety; Ar2N,
diarylamino moiety on the C5 positions in 1 and 2. Broken
lines show peak correlation.
2. Physical and optical properties of COPV
The electrochemical data in Figure 2b provide evidence for
strong interaction between the COPV framework and the
electronꢀrich geminal groups in the radical cation state of 1–3,
but not that of 4, 5. The fluorene reference compound 7 shows
the first and second oxidation peaks (Flu1+, Flu2+; Figure 2
top) at –0.03 and 0.26 V (vs. Fc+/Fc couple), respectively. The
amino COPV reference compounds 4 and 5 show the
corresponding peaks (PV1+, PV2+) at –0.04 V/0.17 V and –
0.05 V/0.17 V, respectively, and the tetramethyl analogue
COPV 6 (–0.15 V/0.07 V) is slightly easier to oxidize. It is
notable that the diaminofluoreneꢀCOPV 3 shows the Flu1+ and
Flu2+ peaks at 0.14 and 0.33 V, much higher by 0.17 and 0.07
V than those of the fluorene reference 7, indicating that the
oxidation of the fluorene moiety becomes much more difficult
In the neutral state of 1–3, there was little sign of interaction
between the amino COPV and the geminal 4ꢀaminophenyl
groups. For example, 1–3 (Figure 2a) show an absorption band
centered at 420 nm due to a π–π* electronic transition of the
COPV framework, and is identical to that of 4 and 5.
However, 3 shows an additional band at 397 nm due to a π–π*
electronic transition at the fluorene moiety. In addition, all 1–5
emit at 463–470 nm upon photoexcitation with almost
identical fluorescence quantum yield and lifetime (Table 1),
suggesting that photoexcitation of 1–5 involves almost solely
the COPV system.
Then, a strong interaction between the radical cation of the
COPV moiety and the electronꢀrich geminal groups was
identified by differential pulse voltammetry (DPV; Figure 2b).
ACS Paragon Plus Environment
2

Page 3 of 9
Journal of the American Chemical Society
once the COPV moiety is oxidized (PV1+). Likewise,
oxidation of the 4ꢀaminoaryl groups in 1 and 2 becomes so
a
1
7
0.4
0.2
2
3
4
5
6
7
8
difficult that they are oxidized only after the second oxidation
522 nm
of the COPV core (PV2+).
0.0 eq.
0.5 eq.
0.7 eq.
Neutral
Table 1. Photophysical and electrochemical dataa
1.0 eq. Monocation
ε
/104 M–
1 cm–1
compd
λ
abs/nm
λ
fl/nm
Φfl
τ/ns
Eox/V
–0.05, 0.18,
0.30 (2e)
9
1
2
417/435
418/435
397/431
4.3/4.3
4.7/4.6
6.4/4.2
465
467
463
0.87
0.91
0.87
2.03
2.06
2.02
0.0
0.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b
0.0 eq.
Neutral
–0.04, 0.21,
0.32 (2e)
0.5 eq.
0.7 eq.
4
1.0 eq. Monocation
–0.03, 0.14,
0.22, 0.33
3
4
665 nm
616 nm
417/433
420/435
402/416
388
4.4/4.4
4.7/4.7
4.1/4.0
3.5
466
470
447
417
0.83
0.80
0.84
0.50
1.86
1.93
1.53
1.27
–0.04, 0.17
–0.05, 0.17
–0.15, 0.07
–0.03, 0.26
5
6
0.2
0.0
7
a
Φfl (fluorescence quantum yield) was determined using an
absolute method; (fluorescence lifetime) was measured using a
timeꢀcorrelated singleꢀphoton counting method with a 405 nm
c
0.0 eq.
0.5 eq.
0.7 eq.
Neutral COPV/Flu
τ
3
COPV+/Flu+/Flu
1.0 eq.
NanoLED for 1–6 and a 365 nm NanoLED for 7, as excitation
source and fitted single exponentially with χ
2 from 0.89–1.21; Eox
0.4
0.2
0.0
was calculated against Fc+/Fc couple as external reference.
616 nm
665 nm
522 nm
3. Characteristics of neutral and radical cation
Absorption titration experiments on chemical oxidation
provided further support for the homoconjugation between the
radical cation of the COPV moiety and the 4ꢀaminophenyl
group in 3. Thus, a solution of compounds 3, 4 and 7 in
chlorobenzene was treated with aliquots of nitrosonium
400
600
λ / nm
hexafluorophosphate (NO+PF6 ),monitored with UV–vis
–
Figure 3. Titration experiments monitored by UV–vis
absorption spectra of a solution in PhCl at 5 ꢀM upon adding
absorption (Figure 3 and Figure S1). Upon addition of the
oxidant to a solution of the aminofluorene 7, a new peak at
522 nm rose at the expense of the band at 397 nm of the
neutral molecule (Figure 3a). The 522ꢀnm peak is due to the
SOMOꢀtoꢀLUMO transition of the fluorene radical cation.
Similarly to the previously reported parent COPVꢀ1,1a the
radical cation of 4 shows absorptions at 665 nm and 616 nm
due to 0–0 and 0–1 transitions, respectively (Figure 3b). In 3
(Figure 3c), where fluorene and indeno[2,1ꢀa]indene moieties
are homoconjugated, the three bands developed
simultaneously, indicating that both the COPV and the
geminal 4ꢀaminophenyl groups contribute to this spectral
change because of formation of the radical cation (see SI for
full range UV–vis–NIR spectra of 1–5 upon oxidation).
NO+PF6 : (a) 7, (b) 4 and (c) 3. Arrows indicate local maxima.
–
Density functional theory (DFT) calculations on both
neutral and radical cation states of COPV 1–6 were carried out
at the B3LYP/6ꢀ31G(d,p) theory level (details in Supporting
Information, Figure S2ꢀS7). The calculated total positive
charge on the two aminophenyl groups in the radical cations
of 1, 2 and 3 are 0.27, 0.20 and 0.34, respectively, whereas
that of 4 was only 0.08. Similarly, the spin density of the
radical cation of 3 delocalizes significantly on the fluorene
part, as shown in Figure S8. This is in stark contrast to
compound 4; the spin density of 4•+ resided mainly on the
COPV moiety (see SI for HOMOs of the neutral compounds).
Careful comparison of the charge distribution between
neutral and radical cation and the geometry parameters
supports the existence of homoconjugation in the radical
cation (Table 2). For example, the reorganization energies
upon formation of the radical cations for 3 were clearly
smaller than those for 4–6 (Table 2, second column). Upon
removal of one electron from COPV 1–5, the distance l shown
by a red arrow in Figure 4a shortens most significantly for 1
and 3. In agreement with this structural change, the radical
cations of 1 and 3 showed a larger positive charge on the
geminal aryl groups. The compounds 4 and 5 lack electronꢀ
donating diamine groups on the phenyl groups, and hence the
phenyl groups cannot take part in homoconjugation when the
compounds are oxidized (see Figure 2b, DPV). Smaller
ACS Paragon Plus Environment
3

Journal of the American Chemical Society
homoconjugation in 4–6 is seen in the smaller values of the
Page 4 of 9
4. SCs based on COPV as HTL
1
2
3
4
5
6
7
8
positive change and the distance reduction in Table 2.
Above studies indicate COPV's potential usefulness as HTL
materials, if the orbital energy levels match with the potentials
of PV and Au electrode. It was indeed the case. By
photoemission yield spectroscopy, we determined the HOMO
level of all of the compounds in their film state to be –4.9 eV
(–4.8 eV for 6). From the onset of the UV–vis absorption
spectra, we determined the band gap (Table S7 and Figure 2a).
These data suggest that they are suitable for blocking electrons
and transporting holes.
Table 2. DFT analysis of radical cations of 1–6 at the
B3LYP/6ꢀ31G(d,p) level of theory.a
Compound
Charge on C5
substituents
Reorganization
energy/eV
Distance
reduction/Å (%)
1
2
3
4
5
6
0.27
0.20
0.34
0.08
0.07
0.06
0.28
0.27
0.21
0.25
0.25
0.29
0.015 (0.60%)
0.003 (0.13%)
0.009 (0.36%)
0.004 (0.17%)
0.004 (0.14%)
0.001 (0.04%)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Measurements of hole mobility and conductivity for doped
and undoped compounds 1ꢀ7 as well as spiro-MeOTAD
indicated that doping of our compounds slightly increases the
mobility and the conductivity, and the doped data are
comparable to those of spiro-MeOTAD (in the order of 10ꢀ5
cm2 Vꢀ1sꢀ1 and 10ꢀ6 S/cm, respectively). These data obtained
for ITO/PEDOT:PSS/HTL(or HTLꢀdoped)/Au are shown in
Figure S9ꢀS10 and Table S8.
a
The charge on the C5 substituents refers to the difference
between the neutral and radical cationic states. Distance reduction
refers to the shortening of the distance between the ipsoꢀphenyl
carbon atom bonded to C5 and the vinyl carbon bonded to C5, as
indicated by a red arrow in Figure 4a.
Encouraged by the theoretical and experimental data
suggesting good holeꢀtransporting ability of a film of COPV
1–3, we examined their potential as a buffer layer in thinꢀfilm
perovskite SC devices, using 4ꢀ7 as references. For this
purpose we employed a planar TiOx as an ETL in our test
semitransparent SC devices, which we could reproducibly
fabricate in our hand. A solution of the COPVs and dopants,
LiꢀTFSI (lithium bis(trifluoromethane sulfonyl) imide) and
TBP (2,6ꢀdiꢀtertꢀbutylpyridine), were spinꢀcoated on a lead
perovskite layer prepared as reported previously.11 The whole
SC device consists of FTO/TiOx (45 nm)/CH3NH3PbI3–xClx
(230 nm nm)/buffer layer (200 nm)/Au (80 nm)/CYTOP and
tested under AM 1.5G illumination with a power density of
100 mW cm–2. This device shows an average transparency of
ca. 25% without metal electrode.
a
!
b
!1
1
2
!2
1000
1000
1000
0
0
0
&1000
-1000
-1000
315
315
320
325
330
330
320
325
330
315
320
325
Mag netic *field*/*mT
Magnetic field /mT
Magnetic field /mT
As shown in Figure 5a, the device consists of very uniform
layers of perovskite and doped compound 1 (1-doped) as
imaged with a low landing voltage scanning electron
microscopy (SEM). Such an SEM allows us to image directly
the surface of conducting and insulating materials without
metal coating that is usually required for conventional SEM
1000
1000
!3
3
4
!4
1000
1000
0
0
0
0
&1000
-1000
imaging (to avoid sample charging).
&1000
-1000
333300
320
320
325
325
331155
315
315
320
325
325
330
330
320
12.5 nm
Mag netic *field*/*mT
MagnMeatg niectic *ffieilde*/*lmdT /mT
Magnetic field / mT
b
a
1-doped
Au
Figure 4. The properties of radical cations. (a) Illustration of
the carbon–carbon distance and charge delocalization in the
radical cation. (b) ESR spectra of the radical cations of 1–4 in
PhCl.
1-doped
PV
TiOx
ITO
Glass
d
The ESR spectra of 1–4 in solution verified the radical
nature of the oxidized species (Figure 4b). The radical cation
of reference 4 showed an ESR signal with a g value of
2.00040, which is typical for organic radicals. Radical cations
of 1–3 also exhibited similar g values of 2.00039, 2.00038 and
2.00046, respectively. It is notable that the ESR spectrum of
the radical cation of 1 displayed pronounced hyperfine
structure caused by intervalence charge transfer with diaryl
amino groups on C5 implying effective interactions between
COPV moiety and substituents in 1•+.10a Broadened ESR signal
were observed for 2•+ and 3•+, suggesting delocalized radical.
In comparison, ESR spectrum of 4•+ displayed a much sharper
peak without hyperfine structure, indicative of a less
delocalized radical (Figure 4b).10b These results further
suggest that the homoconjugation is much more pronounced in
the radical cations of 1-3 than that of 4.
-25 nm
Rq=1.08 nm
c
spiro
Sample
Voltage/ V
Figure 5. Crossꢀsection image of SC, surface morphology of
doped compound 1 (1ꢀdoped) and JꢀV characteristics of SC
devices under AM 1.5G illumination. (a) Crossꢀsection SEM
image of device. Perovskite, TiOx and ITO layers containing
heavy metals are seen as gray layers because they emit more
ACS Paragon Plus Environment
4

Page 5 of 9
Journal of the American Chemical Society
a
backscattered electrons than organic materials, and hence the
organic HTL is seen to be darker. (b) AFM image of
b
1
2
3
4
compound 1(1ꢀdoped) (c) J–V curves and (d) relationships
between different SCs and shunt resistance, Rsh, as well as
series resistance, Rs.
5
6
7
8
Spiro-MeOTAD
Table 3. Performance parameters of thinꢀfilm perovskite SCs
using different buffer layers (see text for details).
Rsh/
kꢁ cm2
4.26
Rs/
ꢁ cm2
10.1
13.8
7.7
Jsc/
mA cm–2
19.9
Voc/
V
PCE/
%
Compound
FF
9
Figure 6. Stability of the solar cells using various HTL
materials. (a) Relative PCE of solar cells based on different
HTLs and (b) the appearance of various devices after 48 h in
air with humidity about 40 %.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
0.93
0.94
0.88
0.85
0.76
0.80
0.89
0.66
0.61
0.73
0.35
0.33
0.24
0.63
12.3
11.2
12.0
4.4
19.6
0.97
3
18.7
2.70
4
5a
14.9
0.20
40.5
106
4.5
1.1
0.29
6
8.3
1.6
0.09
98
In contrast to 1–4, 5 and 6 form a crystalline film, as
shown by the appearance of XRD diffraction peaks (5, 2
7
15.9
8.9
1.30
14.0
θ =
Spiro-
MeOTAD
3.47°, d = 2.54 nm; 6, 2 = 6.33°, d = 1.39 nm; see Figure
θ
18.4
0.98
0.68
12.2
4.76
10.6
S11), which would cause poor contact with the perovskite
layer and insufficient charge collection. Therefore, we found a
small Rsh, large Rs and very small Jsc of 4.5 and 8.3 mA cm–2
for 5 and 6, respectively (Table 3).
a
Doped with FK209. Rsh and Rs obtained from the J–V curve in
Figure 5.
As shown in the bottom of Table 3, the reference
perovskite SC system, studied here for a popular buffer, spiro-
MeOTAD, shows a PCE of 12.2% with Jsc of 18.4 mA cm–2,
Voc of 0.98 V and FF of 0.68, as shown in the bottom of Table
3, and the efficacies of the homoconjugated compounds 1–3
were found to be comparable. For example, a 1ꢀbased device
showed a PCE of 12.3% with Jsc of 19.9 mA cm–2, Voc of 0.93
V and FF of 0.66. The highest performance of 1 coincides well
with the highest degree of homoconjugation in the radical
cationic state as discussed above.
CONCLUSION
In summary, we have probed the origin of the high holeꢀ
transporting ability of popular spiro-MeOTAD and found that
stabilization of
a radical cationic state through 3ꢀD
homoconjugation can serve as a generally useful strategy for
the design of an amorphous and efficient HTL in a perovskite
SC. A few key features of the 3ꢀD COPV system have been
identified to be critical for good performance. First, the 3ꢀD
COPV 1–4 behave similarly in terms of the amorphous
nature of the films, as shown by the lack of XRD peaks and by
absorption spectra almost identical to those in solution (Figure
S11 and S12). Therefore, we expect a smooth interface with
perovskite and hence a high ability of the film to collect
charge at the interface with the perovskite film. In Figure 5b
and Figure S13 is shown the very smooth surface of HTL of
doped compound 1 (root mean square roughness (Rq) of 1.08
nm) as studied by atomicꢀforce microscopy (AFM). This is
supported by the high Jsc values of 14.9–19.9 mA cm–2 for 1 to
4 (Table 3). However, reference 4 differs markedly from 1–3
for its very small FF value of 0.35, large Rs of 40.5 ꢁ cm2 and
PCE of 4.4%, indicative of poor holeꢀconducting ability. A
device using the fluorene reference compound 7 showed PCE
of 8.9% (best data among multiple runs), which is 26% lower
than a spiroꢀMeOTAD device. Meanwhile, the stability of SC
using compound 7 was much worse than other HTLs: The
PCE value decreased to 20% of original value after 48 h in air
(Figure 6a) accompanying the color the PV film (Figure 6b).
Note that the stability of the devices using the compound 1
and Spiro-MeOTAD was found to be comparable. We can
ascribe these to the lack of homoconjugative stabilization of
the radical cation, which plays a key role in the increase of the
device stability and the hole transporting ability of HTL.
homoconjugated
system
has
intrinsically
smaller
reorganization energy than the flat 2ꢀD system and hence
contributes to higher charge transporting ability. Second, the
multiply substituted system allows tuning of the energy levels
of HOMO and LUMO of the film through structural changes
of the donor groups to obtain maximum hole conductivity. We
envisage that longer COPV systems will provide further
electronic tuning.7 Third, a 3ꢀD system tends to form an
amorphous film and hence produces a smoother interface with
neighboring layers to increase the current density. The high
performance of 1, comparable with that of spiro-MeOTAD,
supports the importance of 3ꢀD homoconjugation.12 Overall,
the homoconjugation strategy based on the COPV skeleton
improves charge carrier transportation without affecting the
energy levels of the HOMO and LUMO of the HTL and hence
is useful for the control of carrier injection. As combined with
recent information on the PV formation mechanism,13 we can
now have a prospect of designing PV devices through rational
chemical design of the active layer and interlayers.
EXPERIMENTAL SECTION
General information for synthetic procedures. All oxygenꢀ
and moistureꢀsensitive reactions were performed under a
nitrogen atmosphere using the standard Schlenk method,
lithium/naphthalene reagent was prepared under an argon
atmosphere. Analytical thinꢀlayer chromatography (TLC) was
performed using glass plates precoated with 0.25 mm, 230–
400 mesh silica gel impregnated with a fluorescent indicator
(254 nm). TLC plates were visualized by exposure to
ultraviolet light (UV). Flash column chromatography was
performed on Kanto silica gel 60 (spherical, neutral, 140–325
ACS Paragon Plus Environment
5

Journal of the American Chemical Society
mesh).14 Chemicals were purchased from Tokyo Kasei Co.
Page 6 of 9
δ 157.5, 156.3,
7.30 (m, 20H). 13C NMR (125 MHz, CDCl3)
1
2
3
4
5
6
7
8
and Aldrich Inc., and used as received unless otherwise
indicated. Reagent grade tetrahydrofuran (THF) and toluene
(Tol) were purified using a solventꢀpurification system. The
156.0, 154.2, 144.6, 142.5, 138.4, 137.8, 131.3, 130.6, 130.3,
128.7, 128.5, 127.8, 127.5, 127.2, 127.1, 126.4, 125.4, 122.9,
121.4, 120.7, 63.4, 62.6. HRMS (APCI–TOF): Calcd for
C40H26Br2, 664.0401 (M+), found, 664.0414.
following compounds were prepared following the literature:
15
bis(4ꢀoctylphenyl)methanone,
bromophenyl)methanone,
bis(4ꢀ
Compound 12 was precipitated from carbon tetrachloride
and washed with methanol to give the product with
satisfactory purity as a white solid (78% from 8). H NMR
16
17
bis(3ꢀbromophenyl)methanone
and CuBr2/Al2O3.18 1H and 13C NMR spectra were recorded on
a Bruker USꢀ500 (500 MHz) NMR spectrometer, using CDCl3
or C6D6 as the solvent unless otherwise noted. Chemical shifts
1
(500 MHz, CDCl3)
7.22–7.25 (m, 12H), 7.17–7.19 (m, 2H), 7.09–7.15 (m, 4H).
δ 7.42–7.43 (m, 2H), 7.29–7.31 (m, 8H),
9
in H and 13C NMR spectra were reported in parts per million
1
Compound 13 (73% from 8). Mp: >250 °C. 1H NMR (500
MHz, CDCl3) δ 7.73 (d, 2H, J = 8.5 Hz), 7.53 (dd, 2H, J = 8.5,
1.8 Hz), 7.44 (d, 1H, J = 7.5 Hz), 7.39–7.41 (m, 4H), 7.28–
7.36 (m, 7H), 7.20 (td, 1H, J = 7.5, 1.0 Hz), 7.08 (td, 1H, J =
7.8, 1.2 Hz), 7.00 (td, 1H, J = 7.5, 1.0 Hz), 6.96 (td, 1H, J =
7.2, 0.8 Hz), 6.91 (d, 2H, J = 1.5 Hz), 6.69 (d, 1H, J = 7.0 Hz),
6.27 (d, 1H, J = 7.5 Hz). HRMS (APCI–TOF): Calcd for
C40H24Br2, 662.0245 (M+), found, 662.0246.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(ppm,
δ
scale). Chemical shifts of H signals were referenced
with TMS (0 ppm) or residual C6D5H (7.16 ppm) as standards,
and those of 13C signals were referenced with CDCl3 (77.16
ppm) or C6D6 (128.06 ppm). 19 The data are presented as
follows: chemical shift, multiplicity (s = singlet, d = doublet, t
= triplet, q = quartet, m = multiplet and/or multiple
resonances, br = broad), coupling constant in hertz (Hz) and
integration. Melting points of solid materials were determined
on a MelꢀTemp II capillary meltingꢀpoint apparatus and are
uncorrected. Mass spectra were obtained on a Bruker
micrOTOF mass spectrometer. SEM imaging was conducted
on an FEI Magellan 400L at a landing voltage of 1 kV under a
reduced pressure of 5 ×10–5 Pa. AFM observations of PV thinꢀ
films were carried out on Bruker Multimode 8 instrument in
air.
General procedure for bromination with CuBr2/Al2O3. A
solution of compounds 9–13 in carbon tetrachloride (50 mL
per g) was added to CuBr2/Al2O3 (3 equiv per reaction site).
The reaction mixture was heated at reflux and monitored by
TLC. After cooling to room temperature, the reaction mixture
was filtered through a plug of silica gel. The solvents were
removed under vacuum and the residual was subjected to
column chromatography on silica gel eluted with
hexane/DCM to afford compounds 14–18 as white solids.
General procedure for lithium-induced reductive
cyclization and intramolecular Friedel–Crafts cyclization
reaction. A Schlenk tube charged with lithium (2.5 equiv to
compound 8) and naphthalene (2.5 equiv to compound 8) was
evacuated and backꢀfilled with argon three times, then dry
THF (0.25 mL per mg of lithium) was added. The tube was
sealed under an argon atmosphere and the reaction mixture
was stirred at room temperature for 5 h until the lithium was
depleted. A Schlenk tube charged with compound 8 was
evacuated and backꢀfilled with nitrogen three times, then dry
THF (the same for lithium) was added. The solution of 8 was
added to lithium/naphthalene dropwise, and stirred at room
temperature for 30 min before the corresponding ketone was
added under nitrogen flow. The reaction mixture was stirred at
room temperature for another 30 min before being quenched
with a few drops of NH4Cl (aq. saturated). The crude product
Compound 14 was prepared following the reported
procedure. 15
1
Compound 15 (99%). Mp: >250 °C. H NMR (500 MHz,
CDCl3)
δ 7.55 (d, 1H, J = 1.5 Hz), 7.45 (d, 1H, J = 1.5 Hz),
7.40 (d, 4H, J = 9.0 Hz), 7.20–7.30 (m, 12H), 7.09 (d, 4H, J =
8.5 Hz), 7.03 (d, 1H, J = 8.0 Hz), 6.94 (d, 1H, J = 8.0 Hz). 13C
NMR (125 MHz, CDCl3)
δ 159.2, 158.3, 155.4, 153.8, 141.5,
140.4, 137.0, 136.5, 132.1, 131.0, 130.7, 129.9, 128.9, 128.3,
127.7, 122.4, 121.9, 121.7, 120.8, 120.6, 63.5, 62.4. HRMS
(APCI–TOF): Calcd for C40H24Br4, 819.8612 (M+), found,
819.8599.
1
Compound 16 (96%). Mp: >250 °C. H NMR (500 MHz,
CDCl3)
7.13–7.43 (m, 20H), 7.03 (d, 1H, J = 8.0 Hz), 6.96 (d, 1H, J =
8.5 Hz). 13C NMR (125 MHz, CDCl3)
159.3, 157.8, 155.7,
δ 7.55 (d, 1H, J = 1.5 Hz), 7.46 (d, 1H, J = 1.5 Hz),
δ
was placed on
a short column of silica gel with
153.6, 143.4, 141.4, 136.9, 136.4, 131.2, 131.0, 130.8, 130.5,
129.0, 128.9, 127.7, 126.9, 123.2, 122.4, 121.7, 120.8, 120.7,
63.5, 62.6. HRMS (APCI–TOF): Calcd for C40H24Br4,
819.8612 (M+), found, 819.8608.
dichloromethane (DCM). After removal of solvents under
vacuum, the crude tertiary alcohol was dissolved in carbon
tetrachloride (0.4 mL per mg of 8) without further purification,
and boron trifluoride ether complex (2 ꢀL per mg of 8) was
added. The reaction mixture was stirred at room temperature
for 30 min (compounds 9–12) or at 70 °C for 3 h (compound
13) before being quenched with methanol. After removal of
solvents under vacuum, column chromatography over silica
gel eluted with hexane/DCM afforded the desired 9–13 as
white solids.
1
Compound 17 (95%). Mp: >250 °C. H NMR (500 MHz,
CDCl3)
δ 7.73 (d, 2H, J = 8.0 Hz), 7.54–7.56 (m, 3H), 7.31–
7.38 (m, 11H), 7.12 (d, 1H, J = 8.0 Hz), 7.09 (dd, 1H, J = 8.0,
1.5 Hz), 6.89 (d, 2H, J = 1.5 Hz), 6.80 (d, 1H, J = 2.0 Hz),
6.11 (d, 1H, J = 8.0 Hz). 13C NMR (125 MHz, CDCl3)
δ 158.7,
153.7, 150.3, 146.0, 141.6, 140.0, 138.4, 135.7, 132.0, 131.4,
130.8, 129.0, 128.7, 128.2, 127.8, 127.1, 126.9, 122.3, 122.2,
122.0, 120.8, 120.6, 120.5, 64.2, 62.2. HRMS (APCI–TOF):
Calcd for C40H22Br4, 817.8455 (M+), found, 817.8438.
Compound
9 was prepared following the reported
procedure.15
Compound 10 (70% from 8). Mp: >250 °C. 1H NMR (500
7.43–7.45 (m, 1H), 7.33–7.38 (m, 5H), 7.10–
7.26 (m, 20H). 13C NMR (125 MHz, CDCl3)
157.4, 156.0,
1
Compound 18 (92%). Mp: >250 °C. H NMR (500 MHz,
MHz, CDCl3)
δ
CDCl3)
δ
7.53 (d, 2H, J = 1.5 Hz), 7.23–7.28 (m, 22H), 7.01
δ
(d, 2H, J = 8.0 Hz). 13C NMR (125 MHz, CDCl3)
δ
159.3,
142.6, 141.5, 138.4, 137.9, 128.6, 128.5, 127.7, 127.4, 126.4,
126.3, 125.4, 121.4, 120.6, 63.4, 62.3. HRMS (APCI–TOF):
Calcd for C40H26Br2, 664.0401 (M+), found, 664.0422.
154.9, 141.7, 137.2, 130.6, 128.8, 128.6, 128.4, 127.5, 122.1,
120.4, 63.4. HRMS (APCI–TOF): Calcd for C40H26Br2,
664.0401 (M+), found, 664.0381.
Compound 11 (62% from 8). Mp: >250 °C. 1H NMR (500
MHz, CDCl3)
δ
7.43–7.45 (m, 1H), 7.35–7.41 (m, 5H), 7.11–
ACS Paragon Plus Environment
6

Page 7 of 9
Journal of the American Chemical Society
1
Compound 19 was prepared from 2,7ꢀdibromoꢀ9ꢀHꢀ
Compound 5 (a yellow solid, 91%). Mp: 283–284 °C. H
NMR (500 MHz, C6D6): 7.62 (s, 2H), 7.48 (d, J = 7.0 Hz,
fluorene following the reported procedure.20
1
δ
2
3
4
5
6
7
8
9
8H), 7.18 (d, J = 8.0 Hz, 2H), 7.02 (d, J = 7.0 Hz, 12H), 6.96
(d, J = 8.5 Hz, 8H), 6.87 (d, J = 8.0 Hz, 2H), 6.71 (d, J = 8.5
Hz, 8H), 3.58 (t, 7.5 Hz, 8H), 1.58 (quint, J = 7.5 Hz, 8H),
1.31 (quint, J = 7.5 Hz, 8H), 1.15–1.25 (m, 16H), 0.86 (t, J =
General procedure for the Buchwald–Hartwig amination.
A Schlenk tube, charged with the corresponding substrate
bromide (14–19), bis(4ꢀmethoxyphenyl)amine (1.5 equiv to
C–Br bond), KOtꢀBu (1.2 equiv to amine), Pd(dba)2 or
Pd2(dba)3 (5% mol Pd per C–Br bond), and tꢀBu3P•HBF4 (2
equiv to Pd), was evacuated and backꢀfilled with nitrogen
three times and dry toluene (25 mL per g substrate bromide).
The reaction mixture was stirred at 40 °C for 30 min and then
heated at reflux for 48 h before cooling to room temperature.
After removal of solvent under vacuum, the residual was
flowed through a short column of silica gel with hexane/DCM
(plus 1% triethylamine). Preparative gel permeation
chromatography (Tol, refraction index detector) gave the
product.
7.0 Hz, 12H). 13C NMR (125 MHz, C6D6):
δ 158.9, 155.8,
153.7, 147.4, 144.0, 141.6, 132.5, 129.1, 128.7, 127.0, 126.5,
121.1, 120.5, 119.6, 115.5, 68.0, 63.7, 31.9, 29.6, 26.1, 23.0,
14.2. TOF MS (APCI+): 1243.7 [M]+; HRMS (APCI+) calcd
for C88H94N2O4 (M): 1242.7208; found: 1242.7226.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Compound 6 (yellow solid). Mp: 163–164 ºC. H NMR
(400 MHz, C6D6): δ 7.44 (s, 2H), 7.22 (d, J = 9.2 Hz, 8H),
7.15 (s, 4H), 6.83 (d, J = 9.2 Hz, 8H), 3.63 (t, J = 6.4 Hz, 8H),
1.41 (s, 12H), 1.60 (quint, J = 6.4 Hz, 8H), 1.34 (quint, J = 7.8
Hz, 8H), 1.15–1.29 (m, 16H), 0.86 (t, J = 6.9 Hz, 12H). 13C
NMR (125 MHz, C6D6): δ 161.2, 156.2, 155.0, 147.4, 142.9,
133.1, 126.8, 122.1, 120.3, 118.2, 116.3, 68.7, 45.7, 32.5, 30.3,
26.7, 25.3, 23.6,14.8. TOF MS (APCI+): 994.7 [M]+; HRMS
(APCI+) calcd for C68H86N2O4 (M): 994.6582; found:
994.6578.
1
Compound 1 (a yellow solid, 94%). Mp: 173–177 °C. H
NMR (500 MHz, C6D6) δ 7.68 (d, 1H, J = 2.0 Hz), 7.60 (d, 1H,
J = 2.0 Hz), 7.48–7.50 (m, 4H), 7.44 (d, 4H, J = 9.0 Hz), 7.34
(d, 1H, J = 8.0 Hz), 7.19 (d, 1H, J = 8.5 Hz), 6.86–7.02 (m,
28H), 6.63–6.69 (m, 16H), 3.31 (s, 12H), 3.28 (s, 6H), 3.27 (s,
6H). 13C NMR (125 MHz, C6D6)
δ 159.9, 159.0, 156.5, 156.2,
1
Compound 7 (a white solid, 57%). Mp: 91–93 °C. H
156.1, 154.6, 153.6, 148.0, 147.32, 147.25, 144.3, 141.8,
141.5, 135.6, 133.2, 132.7, 129.8, 129.1, 128.7, 127.3, 127.0,
126.6, 126.5, 121.4, 121.1, 120.8, 120.5, 120.3, 119.9, 119.8,
115.1, 115.0, 63.8, 62.5, 55.04, 55.00. HRMS (APCI–TOF):
Calcd for C96H81N4O8, 1417.6054 (M + H+), found, 1417.6069.
NMR (500 MHz, C6D6) δ 7.43 (d, 2H, J = 8.5 Hz), 7.32 (d, 2H,
J = 2.0 Hz), 7.21 (d, 8H, J = 9.0 Hz), 7.14–7.16 (m, 2H), 6.79
(d, 8H, J = 9.0 Hz), 3.33 (s, 12H), 1.72–1.75 (m, 4H), 0.99–
1.26 (m, 24H), 0.88 (t, 6H, J = 7.2 Hz). 13C NMR (125 MHz,
C6D6)
δ 156.2, 152.3, 148.0, 142.3, 135.5, 128.3, 126.4, 121.4,
1
Compound 2 (a yellow solid, 70%). Mp: 214–217 °C. H
120.0, 117.2, 115.1, 55.3, 55.0, 40.6, 32.3, 30.6, 29.85, 29.83,
24.6, 23.1, 14.4. HRMS (APCI–TOF): Calcd for C57H69N2O4,
845.5257 (M + H+), found, 845.5235.
NMR (500 MHz, C6D6) δ 7.60 (d, 1H, J = 2.5 Hz), 7.50 (d, 1H,
J = 2.0 Hz), 7.42–7.43 (m, 2H), 7.34–7.36 (m, 4H), 7.13–7.17
(m, 2H), 7.06–7.08 (m, 4H), 6.86–7.01 (m, 25H), 6.76 (dd, 1H,
J = 8.2, 2.2 Hz), 6.65–6.72 (m, 16H), 3.31 (s, 12H), 3.284 (s,
UV–vis and PL spectra measurement. The UV–vis
spectrum of the perovskite thin films on a PEDOT:PSS
surface was recorded using a Jasco Vꢀ670 spectrophotometer.
Photoluminescence of the PV film on glass was performed
using a fluorescence spectrophotometer (HITACHI, Fꢀ4500).
6H), 3.278 (s, 6H). 13C NMR (125 MHz, C6D6)
δ 158.9, 158.5,
156.3, 156.2, 156.1, 154.0, 153.9, 149.1, 147.1, 144.8, 144.1,
141.9, 141.8, 141.4, 132.6, 132.4, 129.3, 129.0, 128.7, 126.9,
126.8, 126.4, 121.8, 121.5, 121.3, 121.0, 120.9, 120.1, 120.0,
119.5, 119.0, 115.0, 63.9, 63.6, 55.1, 55.0. HRMS (APCI–
TOF): Calcd for C96H81N4O8, 1417.6054 (M + H+), found,
1417.6101.
Materials. Methylammonium iodide (MAI) was prepared
following previous reports.11 In a glove box (N2 atmosphere),
MAI, PbI2 (TCI, 99.999%) and PbCl2 (Aldrich, 99.999%)
were dissolved in N,Nꢀdimethylformamide (DMF, Tokyo
Chemical Industry Co., 99.5%).
1
Compound 3 (a yellow solid, 95%). Mp: 162–167 °C. H
NMR (500 MHz, C6D6)
δ 7.54 (d, 1H, J = 2.0 Hz), 7.37–7.41
XRD diffraction measurements. The XRD experiment was
performed on a Rigaku SmartLab Xꢀray diffractometer
equipped with a scintillation counter. The measurement
employed the Cu Kα line, focused radiation at 9 kW (45 kV,
(m, 6H), 6.99–7.12 (m, 20H), 6.94 (d, 4H, J = 8.5 Hz), 6.83 (d,
4H, J = 9.0 Hz), 6.67–6.74 (m, 13H), 6.58–6.61 (m, 5H), 6.53
(d, 1H, J = 8.5 Hz). 13C NMR (125 MHz, C6D6)
δ 159.3, 157.2,
156.2, 156.0, 155.9, 155.7, 151.4, 148.3, 147.4, 147.3, 143.9,
141.9, 141.7, 136.0, 134.2, 132.0, 128.8, 127.0, 126.4, 126.1,
122.2, 121.7, 121.0, 120.6, 120.5, 119.8, 119.6, 118.5, 117.7,
115.01, 114.97, 144.8, 64.1, 63.1, 55.1, 55.0, 54.9. HRMS
(APCI–TOF): Calcd for C96H79N4O8, 1415.5898 (M + H+),
found, 1415.5902.
200 mA) power using a 0.02° 2
θ step scan from 3.0–35° with
a scanning speed of 3° min−1.
Evaluation of PV devices: Current–voltage sweeps were
taken on a Keithley 2400 source measurement unit controlled
by a computer. The light source used to determine the PCE
was an AM 1.5G solar simulator system (Sumitomo Heavy
Industries Advanced Machinery) with an intensity of 100
mW/cm2. The SCs were masked with a metal aperture to
define the active area of 4 mm2.
Preparation of SCs. A fluorineꢀdoped tin oxide (FTO) layer
on a glass substrate was used for this study. Prior to the
formation of the buffer layer, the patterned FTO glass was
ultrasonically cleaned using a surfactant, rinsed with water and
then finally given 3 min UV–ozone treatment. A 45ꢀnm thick
electronꢀtransporting layer TiOx was deposited by spinꢀcoating
(3000 rpm for 30 s) of the precursor solution (see materials
section) and annealed at 500 °C for 30 min in air
atmosphere.11 To form the CH3NH3PbI3–xClx layer, a 40 wt %
1
Compound 4 (a yellow solid, 71%). Mp: 183–187 °C. H
NMR (500 MHz, C6D6) δ 7.65 (d, 1H, J = 2.0 Hz), 7.58 (d, 1H,
J = 2.0 Hz), 7.50–7.54 (m, 8H), 7.30 (d, 1H, J = 8.5 Hz), 7.20
(d, 1H, J = 8.5 Hz), 6.64–7.06 (m, 20H), 6.61–6.65 (m, 8H),
3.28 (s, 6H), 3.26 (s, 6H), 2.44 (t, 4H, J = 7.8 Hz), 1.46–1.49
(m, 4H), 1.20–1.30 (m, 20H), 0.90 (t, 6H, J = 7.2 Hz). 13C
NMR (125 MHz, C6D6)
δ 159.4, 159.0, 156.2, 154.4, 153.7,
147.37, 147.33, 144.2, 141.80, 141.77, 141.55, 141.48, 132.9,
132.7, 129.14, 129.10, 128.8, 128.7, 127.0, 126.5, 121.3,
121.1, 120.7, 120.6, 119.8, 115.0, 63.8, 63.3, 55.0, 36.0, 32.3,
31.8, 29.91, 29.89, 29.7, 23.1, 14.4. HRMS (APCI–TOF):
Calcd for C84H87N2O4, 1187.6666 (M + H+), found, 1187.6656.
ACS Paragon Plus Environment
7

Journal of the American Chemical Society
precursor solution (4:1:1 mole ratio of CH3NH3I:PbI2:PbCl2)
Page 8 of 9
1
2
3
4
5
6
7
8
in DMF was spinꢀcoated onto the TiOx layer at 500 rpm for 3 s
and 4000 rpm for 30 s. Further, it was annealed at 100 °C for
35 min in a nitrogenꢀfilled glove box. The HTL (see materials
section) 60 mg/mL in chlorobenzene with dopants (15 ꢀL, 520
mg/mL LiꢀTFSI in CH3CN and 22.5 ꢀL TBP) was then
deposited by spin coating (2200 rpm for 30 s). The top
electrode (Au, 80 nm) was deposited through a metal shadow
mask, which defined a 2 mm stripe pattern perpendicular to
the ITO stripe.
(5) (a) Burschka, J.; Pellet, N.; Moon, S. J.; Baker, R. H.; Gao, P.;
Nazeeruddin, M. K.; Grätzel, M. Nature 2013, 499, 316. (b) Liu, M.;
Johnston, M. B.; Snaith, H. J. Nature 2013, 501, 395.
(6) Zhu, X.; Tsuji, H.; Yella, A.; Chauvin, A.ꢀS.; Grätzel, M.;
Nakamura, E. Chem. Commun. 2013, 49, 582.
(7) MoralesꢀVidal, M.; Boj, P. G.; Villalvilla, J. M.; Quintana, J. A.;
Yan, Q.; Lin, N.ꢀT.; Zhu, X.; Ruangsupapichat, N.; Casado, J.; Tsuji,
H.; Nakamura, E.; DíazꢀGarcía, M. A. Nat. Commun. 2015, 6, 8458.
(8) Sukegawa, J.; Schubert, C.; Zhu, X.; Tsuji, H.; Guldi, D. M.;
Nakamura, E. Nat. Chem. 2014, 6, 899.
(9) Sukegawa, J.; Tsuji, H.; Nakamura, E. Chem. Lett. 2014, 43, 699.
(10) (a) Nelsen, S. F.; Ismagilov, R. F.; Powell, D. R. J. Am. Chem.
Soc. 1996, 118, 6313. (b) We thank one of a reviewer for this
interpretation of the ESR spectra.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information. DFT calculations and UV–vis–NIR
titration experiments, this material is available free of charge
via the Internet at http://pubs.acs.org.
( 11 ) (a) Guo, Y.; Liu, C.; Inoue, K.; Harano, H.; Tanaka, H.;
Nakamura, E. J. Mater. Chem. A 2014, 2, 13827. (b) Guo, Y.; Liu, C.;
Tanaka, H.; Nakamura, E. J. Phys. Chem. Lett. 2015, 6, 535. (c)
Krishna, A.; Sabba, D.; Li, H.; Yin, J.; Boix, P. P.; Soci, C.;
Mhaisalkar, S. G.; Grimsdale, A. C. Chem. Sci. 2014, 5, 2702. (d)
Gratia, P.; Magomedov, A.; Malinauskas, T.; Daskeviciene, M.;
Abate, A.; Ahmad, S.; Grätzel, M.; Getautis, V.; Nazeeruddin, M. K.
Angew. Chem. Int. Ed. 2015, 54, 11409. (e) Li, M. H.; Hsu, C. W.;
Shen, P. S.; Cheng, H. M.; Chi, Y.; Chen, P.; Guo, T. F. Chem.
Commun. 2015, 51, 15518. (f) Park, S.; Heo, J. H.; Cheon, C. H.; Kim,
H.; Im, S. H.; Son, H. J. J. Mater. Chem. A 2015, 3, 24215. (g) Liu,
Y.; Hong, Z.; Chen, Q.; Chen, H.; Chang, W. H.; Yang, Y.; Song, T.
B.; Yang, Y. Adv. Mater. 2016, 28, 440. (h) Rakstys, K. Saliba, M.;
Gao, P.; Gratia, P.; Kamarauskas, E.; Paek, S.; Jankauskas, V.;
Nazeeruddin, M. K. Angew. Chem. Int. Ed. 2016, 55, 7464.
(12) Saragi, T. P. I.; Spehr, T.; Siebert, A.; FuhrmannꢀLieker, T.;
Salbeck, J. Chem. Rev. 2007, 107, 1011.
AUTHOR INFORMATION
Corresponding Author
*guoyunlong@chem.s.uꢀtokyo.ac.jp
*tsujiha@kanagawaꢀu.ac.jp
*nakamura@chem.s.uꢀtokyo.ac.jp
Present Addresses
†Prof. Hayato Tsuji, Faculty of Science, Kanagawa University,
2946 Tsuchiya, Hiratsuka, Kanagawa 259ꢀ1293, Japan.
†Dr. Qifan Yan, School of Chemistry and Molecular Engineering,
East China University of Science and Technology, Meilong road
130, Shanghai 200237, China.
(13) Guo, Y.; Shoyama, K.; Sato, W.; Matsuo, Y.; Inoue, K.; Harano,
K.; Liu, C.; Tanaka, H.; Nakamura, E. J. Am. Chem. Soc. 2015, 137,
15907–15914. Guo, Y.; Sato, W.; Shoyama, K.; Nakamura, E. J. Am.
Chem. Soc., 2016, 138, 5410ꢀ5416.
(14) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923.
(15) Zhu, X.; Tsuji, H.; Nakabayashi, K.; Ohkoshi, S.ꢀi.; Nakamura, E.
J. Am. Chem. Soc. 2011, 133, 16342.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENTS
(16)Zhang, Y.; Lu, Z.; Desai, A.; Wulff, D. W. Org. Lett. 2008, 10,
5429.
We thank MEXT for financial support [KAKENHI 15H05754
to E.N., the Strategic Promotion of Innovative Research, JST
to Y.G.; Q.Y. is grateful to the JSPS Postdoctoral Fellowship
for Foreign Researchers (no. P13334)]. We thank Dr. NaiꢀTi
Lin for synthesis of compound 6.
(17) Mills, N. S.; Tirla, C.; Benish, M. A.; Rakowitz, A. J.; Bebell, L.
M.; Hurd, C. M. M.; Bria, A. L. M. J. Org. Chem. 2005, 70, 10709.
(18) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.
Organometallics 2010, 29, 2176.
(19) Kodomari, M.; Satoh, H.; Yoshitomi, S. J. Org. Chem. 1988, 13,
2093.
(20) Hung, M.ꢀC.; Liao, J.ꢀL.; Chen, S.ꢀA.; Chen, S.ꢀH.; Su, A.ꢀC. J.
Am. Chem. Soc. 2005, 127, 14576.
REFERENCES
(1) (a) Mayorga Burrezo, P.; Zhu, X.; Zhu, S.ꢀF.; Yan, Q.; Lopez
Navarrete, J. T.; Tsuji, H.; Nakamura, E.; Casado, J. J. Am. Chem.
Soc. 2015, 137, 3834–3843. (b) Zhu, X.; Tsuji, H.; Lopez Navarrete, J.
T.; Casado, J.; Nakamura, E. J. Am. Chem. Soc. 2012, 134, 19254.
(2) (a) RoldánꢀCarmona, C.; Malinkiewicz, O.; Betancur, R.; Longo,
G.; Momblona, C.; Jaramillo, F.; Camacho, L.; Bolink, H. J. Energy
Environ. Sci. 2014, 7, 2968. (b) Guo, Y.; Shoyama, K.; Sato, W.;
Nakamura, E. Adv. Energy Mater. 2016, 6, 1502317.
(3) (a) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am.
Chem. Soc. 2009, 131, 6050. (b) Nishimura, H.; Ishida, N.; Shimazaki,
A.; Wakamiya, A.; Saeki, A.; Scott, L. T.; Murata, Y. J. Am. Chem.
Soc. 2015, 137, 15656–15659. (c) Grätzel, M. Nat. Mater. 2014, 13,
838. (d) Green, M. A.; HoꢀBaillie, A.; Snaith, H. J. Nat. Photon 2014,
8, 506. (e) Yin, W.ꢀJ.; Yang, J.ꢀH.; Kang, J.; Yan, Y.; Wei, S.ꢀH. J.
Mater. Chem. A 2015, 3, 8926.
(4) (a) Matsuo, Y.; Sato, Y.; Niinomi, T.; Soga, I.; Tanaka, H.;
Nakamura, E. J. Am. Chem. Soc., 2009, 131, 16048ꢀ16050. (b) Cf.
http://www.japantimes.co.jp/news/2016/02/07/business/tech/technolo
gyꢀimprovesꢀfutureꢀorganicꢀsolarꢀcellsꢀlooksꢀbright/#.V5rudo787yc
ACS Paragon Plus Environment
8

Page 9 of 9
Journal of the American Chemical Society
1
2
3
TOC
4
5
6
7
homoconjugation
Ar2N
NAr2
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ar2N
NAr2
ACS Paragon Plus Environment
9