Anode interlayer in organic photovoltaics: Narrow bandgap small molecular materials as exciton-blocking layer

Article abstract of DOI:10.1002/jccs.201900123

Six materials were used as an interlayer at the anode side (anode interlayer [AIL]) of an archetypical planar heterojunction organic solar cell (OSC). In addition to two conventional wide bandgap hole transport materials (HTMs), tris(4-carbazol-9-ylphenyl

Full text of DOI:10.1002/jccs.201900123

Received: 1 April 2019
Revised: 10 April 2019
Accepted: 15 April 2019
DOI: 10.1002/jccs.201900123
A R T I C L E
Anode interlayer in organic photovoltaics: Narrow bandgap small
molecular materials as exciton-blocking layer
1
2
1,2
Jan Golder | Chiao-Wen Lin | Chin-Ti Chen
1
Department of Applied Chemistry, National
Abstract
Chiao Tung University, Hsinchu,
Taiwan, ROC
Six materials were used as an interlayer at the anode side (anode interlayer [AIL]) of an
archetypical planar heterojunction organic solar cell (OSC). In addition to two conven-
2
Institute of Chemistry, Academia Sinica,
Taipei, Taiwan, ROC
tional wide bandgap hole transport materials (HTMs), tris(4-carbazol-9-ylphenyl)amine
0
(
TCTA) and trans-4,4 -bis[N-(naphthalen-1-yl)-N-phenylamino]stilbene (NPAE), we
Correspondence
Chin-Ti Chen, Institute of Chemistry,
Academia Sinica, Taipei 11529, Taiwan,
ROC.
explore four narrow bandgap materials, bis(biphenylaminospiro)-fumaronitrile
PhSPFN), bis(N-(naphthalen-1-yl)-N-phenylamino)anthraquinone (NPAAnQ), bis-(di
(
Email: chintchen@gate.sinica.edu.tw
(2-fluorophenyl)aminospiro)-fumaronitrile (FPhSPFN), and bis[4-(N-(pyren-1-yl)-N-
phenylamino)phenyl]fumaronitrile (PyPAFN), the energy levels of which essentially
align with the ones of SubPc, the active light-absorbing material of the OSC study
herein. By using a narrow bandgap AIL, universally enhanced short-circuit current den-
sity and power conversion efficiencies (PCEs) have been achieved. In addition, one of
these materials, FPhSPFN, results in a PCE of 5.13%, which is the highest reported
value for SubPc solar cells with a similar architecture. This is ascribed to the formation
of an otherwise passive exciton-blocking interface. Furthermore, this demonstrates that
charge selectivity by way of a high-lying lowest unoccupied molecular orbital (LUMO)
energy level is not a prerequisite for successful AIL design. As such, in terms of energy
level alignment and bandgap energies, we establish a viable alternative approach toward
interface and interlayer material design.
K E Y W O R D S
AIL, EBL, exciton blocking, HTL, HTM, planar heterojunction, SubPc
1
| INTRODUCTION
anode interlayer (AIL)/cathode interlayer has led to a multi-
tude of design approaches. There are a range of AIL designs
For high-performance organic solar cell (OSC)s of either
planar heterojunction or bulk heterojunction configuration,
typically, the main light-absorbing materials are sandwiched
between charge-selective layers on the anode and cathode
sides, respectively. As such, a substantial effort has been
made regarding the study of electrode interfaces and the
development of materials generally termed electron transport
with various considerations, including hole extraction layer
(HEL), hole transport layer (HTL), and exciton dissociation
layer (EDL), leading to cascading heterojunction (CHJ)
[
6–13]
designs.
In addition, it is also well established that the
underlying interlayer may influence the active material layer
morphology and/or act as an exciton-blocking layer
[
14–16]
(ExBL).
Typically, small molecular-based AIL mate-
[1–5]
material and hole transport material (HTM).
In search of
rials consist of an aromatic or π-conjugated core substituted
higher performances, interfacial engineering of such an
with diarylamines, triarylamines, and/or carbazoles.
©
2019 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J Chin Chem Soc. 2019;1–11.
http://www.jccs.wiley-vch.de
1

2
GOLDER ET AL.
[
25]
Consequently, the highest unoccupied molecular orbital
spirobifluorene)-fumaronitriles, PhSPFN and FPhSPFN,
(
HOMO) energy levels of such materials can easily be
as well as newly synthesized bis[4-(N-(pyren-1-yl)-N-
phenylamino)phenyl]fumaronitrile (PyPAFN) and bis(N-
(naphthalen-1-yl)-N-phenylamino)anthraquinone
(NPAAnQ) (Figure 1a), are chosen as additional materials
in this study. Our results demonstrate that any AIL, acting as
an ExBL, is capable of increasing PCEs of OSCs studied
here. For the narrow bandgap materials, an increased short-
[
17–19]
modified.
With regard to the light-absorbing materials,
a high-lying LUMO energy level of the AIL is thought to be
an electron-blocking layer (EBL) and is hence charge
(i.e., hole) selective, although multiple functionalities are
usually ascribed to any given AIL. Recently, Lin et al.
obtained promising power conversion efficiencies (PCEs)
utilizing an AIL with a LUMO energy offset (with respect to
the donor material, boron subphthalocyanine chloride,
SubPc) of just ~0.1 eV.
series of energetically similar materials, which acted as
excellent ExBL, and high PCEs were achieved for this type
circuit current density (J ) is achieved primarily via exciton
SC
blocking, while the conventional HTL, HEL, and EDL also
serve their respective dual purpose. The highest reported
PCE of 5.13% among simple four-layer SubPc OSCs is
achieved by the inclusion of an FPhSPFN AIL. As the four
narrow bandgap AIL devices perform as well as, or out-
perform, the two wide bandgap ones in our study, the con-
ventional understanding of AIL having a high-lying LUMO
energy (and hence a reduced charge recombination at the
anode side) is not as necessary or less critical after all.
[20]
Shortly after, we also reported a
[21]
of SubPc-based OSCs.
exciton diffusion in organic materials,
In this context, and considering
[
22–24]
there remains a
question regarding whether an HTL or EDL, which pos-
sesses a high LUMO energy level (a large energy offset) and
aids charge extraction or forms a secondary exciton dissocia-
tion site, is necessary. Accordingly, we study two common
wide bandgap materials, as well as four unconventional nar-
row bandgap materials. For the wide bandgap AILs, we use
tris(4-carbazol-9-ylphenyl)amine (TCTA), a well-known
2
| RESULTS AND DISCUSSION
0
2.1 | Synthesis and characterization
and often used HTM, and a newly synthesized trans-4,4 -bis
[
(
N-(naphthalen-1-yl)-N-phenylamino]stilbene
Figure 1a). Moreover, relative to that of SubPc, both TCTA
(NPAE)
Narrow-bandgap AIL materials FPhSPFN and PhSPFN
[
25]
were synthesized according to previous reports
as both
and NPAE have substantially higher-lying LUMO energy
levels (Figure 1b). Therefore, both materials conform to the
principle of HTL, EBL, or EDL in conjunction with SubPc.
For the narrow bandgap AILs with a low-lying LUMO
energy level, previously known bis(biphenylamino-
molecules have slightly lower HOMO and very similar
LUMO energy levels with regard to SubPc. A further mate-
rial, PyPAFN, was designed utilizing a similar donor-accep-
tor-donor (D-A-D) architecture, in which an electron-
deficient fumaronitrile core, 2,3-diphenylfumaronitrile, is
chemically connected to N-pyren-1-yl-N-phenyl amine on
both ends of the molecule via facile Pd-catalyzed cross-cou-
pling. In addition, to further widen the scope of narrow-
bandgap AIL materials and to explore structure property
relationships upon insertion as AIL materials, NPAAnQ,
possessing an electron-accepting anthraquinone core, was
synthesized with the corresponding arylamine (N-naphtalen-
(
a)
F
CN
NC
O
CN
NC
N
N
N
F
N
F
N
N
O
F
N
N
NC
CN
N
N
N
N
1
-yl-N-phenyl amine) in a similar fashion. Finally, NPAE, a
N
N
naphthylphenylamine-substituted stilbene, was obtained via
a reductive McMurry coupling of the corresponding
aldehyde.
(
b)
The absorption and emission spectra in thin film of the
six AIL materials are shown in Figure 2. NPAE and TCTA
display an absorption maxima of 334 and 391 nm, both of
which can be mainly assigned to π–π* transition of trans-
stilbene and triphenylamine, respectively. The other four
materials, FPhSPFN, NPAAnQ, PhSPFN, and PyPAFN,
exhibit comparatively red-shifted absorption maxima
film
max
(
Abs. λ ) of 482, 447, 501, and 500 nm, respectively.
These are ascribed to intramolecular charge transfer, typical for
D-A-D type molecular structures. Their photoluminescence
(PL) spectra are shown in Figure 2b. Compared with those
FIGURE 1 (a) AIL molecular structures and (b) energy levels of
ITO/AIL/SubPc/C60/BCP/Al

GOLDER ET AL.
3
(
a)
1
1
1
1
0
0
0
0
.75
.50
.25
.00
.75
.50
.25
.00
PL spectra of arylamine-substituted trans-stilbenes. In order
to attain consistent and intercomparable optical bandgap
energies, all AIL materials were measured individually as
FPhSPFN
NPAANQ
PhSPFN
PyPAFN
NPAE
thin films to obtain the absorption onset wavelength (Abs.
film
onset
λ
) and are summarized in Table 1. Furthermore, the
TCTA
HOMO energy level of all involved AIL materials was
determined in the same fashion by photoelectron spectros-
copy in air (PESA). The results are shown in Figure 3, and
the data are summarized in Table 1. The HOMO energy
levels range from −5.51 to −5.73 eV, which is in close
alignment with that of SubPc (−5.6 eV), thus minimizing
hole extraction barriers. However, it is worthwhile to point
out that the determined HOMO energy level can vary
3
50 400 450 500 550 600 650 700
Wavelength (nm)
depending on the experimental method and is not always
(
b)
[26]
consistent with those reported in literature.
NPAE is con-
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
sidered to be a conventional AIL material, with its slightly
higher HOMO energy level with regard to SubPc and a wide
bandgap of 2.80 eV. This gives NPAE a threefold function-
ality: (a) blocking SubPc excitons from being quenched at
the indium tin oxide (ITO) interface, (b) forming a CHJ, and
TCTA
(c) blocking electrons from reaching the anode. A TCTA
AIL is expected to play a similar role, which, despite the ener-
getically lower HOMO of −5.71 eV compared with −5.6 eV of
[12]
SubPc, has been shown to enable exciton dissociation.
In
addition, the HTM-like TCTA may be able to extract direct
photo-generated free positives charges from the SubPc layer. In
contrast, FPhSPFN, NPAAnQ, PhSPFN, and PyPAFN
exhibit a rather close energy level alignment of their respective
LUMO (−3.31 eV to −3.64 eV) and HOMO (−5.58 eV to
3
50
450
550
650
750
850
950
Wavelength (nm)
FIGURE 2 (a) Absorption and (b) emission spectra of AIL
materials in thin film
−5.73 eV) energy levels to those (LUMO −3.6 eV and HOMO
−
5.6 eV) of SubPc. Thus, near-flat band conditions are created
of NPAE and TCTA, the four narrow-bandgap AIL mate-
at the AIL/SubPc interface, which intrinsically do not allow for
cascading energy transfer or facile electron blocking and also
limit eventual band bending/alignment. Furthermore, despite
their relatively low bandgap energies, there is quite limited
spectroscopic overlap of SubPc emission (Figure 5b) and the
absorption of AIL materials (Figure 2a), not to mention the
rials show a significantly longer wavelength emission max-
ima (Em. λfilm ) in the range of 600–650 nm. Notably, this
max
overlaps with SubPc absorption and could give rise to För-
ster resonance energy transfer (FRET) into the SubPc layer.
However, this was not observed experimentally and will be
discussed in a later section. While the PL maxima of TCTA
are in agreement with previous reports, NPAE shows typical
opt
optical bandgaps (Eg film) of AIL materials are all greater
than 2.0 eV (Table 1), the E og pt film of SubPc. This
TABLE 1 AIL material properties: Thin film absorption maxima (λfilm ), absorption onset (λ of i nl ms et), and optical bandgap (E go pt), which is
max
converted from λ fo i nl ms et. HOMO energy levels were obtained from PESA, and LUMO energy levels were calculated based on E og pt and HOMO
energy level
film
max
film
onset
opt
g
film
max
Abs. λ
482
(nm)
Abs. λ
580
(nm)
E
film (eV)
Em. λ
646
(nm)
HOMO (eV)
−5.73
LUMO (eV)
−3.59
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
2.14
2.41
2.04
2.13
2.80
3.31
447
514
611
−5.72
−3.31
501
608
685
−5.68
−3.64
500
582
644
−5.58
−3.45
391
442
452, 468
405
−5.51
−2.71
TCTA
334
375
−5.71
−2.40

4
GOLDER ET AL.
4
3
3
2
2
1
1
0
5
0
5
0
5
0
5
0
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
5
.73 eV
5
.72 eV
TCTA
5
.71 eV
.68 eV
.58 eV
.51 eV
5
5
5
4
.8 4.9 5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 6.0
Energy (eV)
FIGURE 3 The HOMO energy level of AIL materials
determined by PESA
essentially eliminates the possibility of SubPc exciton trans-
fer into the AIL by way of FRET and also hinders electron
exchange via Dexter transfer, thus rendering the narrow-
bandgap AIL capable ExBLs. This applies to all AIL mate-
rials in this study and results in a higher exciton density at
the respective AIL–SubPc interface, which in turn facilitates
the exciton diffusion toward and subsequent dissociation at
the D/A (i.e., SubPc/C ) interface.
6
0
2
.2 | Thin film characterization of AIL/SubPc
To gain insight into the ITO/AIL and AIL/SubPc interface
and its influence on the SubPc morphology, atomic force
microscopy (AFM) was used to study the surface morphol-
ogies. Figure 4 shows the thin film surface morphologies of
each sample. On deposition of 2 nm AIL onto ITO glass, the
surface roughness is reduced from 3.03 to 2.54, 2.63, 2.51,
FIGURE 4 AFM images (5 μm × 5 μm) of (a) bare ITO glass,
(
(
b–g) 2 nm AIL material on ITO glass, (h) 11 nm SubPc on ITO, and
i–n) 11 nm SubPc on top of each AIL on ITO glass. The sequence of
AIL in (b–g) and (i–n) is FPhSPFN, NPAAnQ, PhSPFN, PyPAFN,
NPAE, and TCTA
2.48, 2.41, and 2.44 (±0.25 nm). This shows that whatever
AIL is used results in a smoother surface, and it also illus-
trates that all AIL thin films have a rather similar surface
morphology. Furthermore, the surface roughness of 11 nm
SubPc thin film remains nearly unchanged, from 1.36 to
absorption in the region between 350 and 550 nm is observed.
Considering the thin film absorption spectra of the AIL mate-
rials (Figure 2a), of which absorption λfilm is in the range of
max
1.35–1.48 nm, independent of the underlying AIL. Hence, a
~
425–550 nm, and checking the same wavelength region of
change of SubPc thin film morphology, if there is any
because of the AIL, can be neglected.
external quantum efficiency (EQE) spectra (Figure 6), there
is virtually no contribution to the photocurrent originating
from the AIL, Specifically, no signal unique to the absorp-
2
.3 | Characterization of AIL/SubPc thin films
film
tion λ
of each AIL material can be distinguished. Such a
max
signal in the EQE spectra would be necessary to account for
the gains in JSC. Thus, there is no additional exciton contri-
bution from the AIL. For the same reason, any FRET
between the AIL and the bulk SubPc layer seems unlikely.
Moreover, pronounced gains in EQE are all at a wavelength
of around 580 nm (absorption λmax of SubPc), a value at
which the AIL materials themselves show almost no absorp-
tion (Figure 2), not to mention the thinness (only 2 nm) of
To evaluate AIL material influences on the SubPc/C60 OSCs,
half-cells with the structure of ITO/AIL (2 nm)/SubPc
(11 nm) were fabricated. The layer thicknesses are the same
as those used for the complete cells. The absorption spectra of
such half-cells are shown in Figure 5a. Evidently, absorption
spectra of these half-cells are very similar with or without
AIL. Due to a relatively small thickness of the AIL and a
comparatively high absorbance of SubPc, only little increased

GOLDER ET AL.
5
(a)
with the ITO interface, and hence, photo-generated AIL
excitons and eventual emission are expected to be largely
quenched. Such considerations illustrate that the EQE
enhancement is mainly due to the elimination of direct con-
tact between ITO and SubPc. Therefore, such improvements
are due to factors such as the prevention of exciton
quenching, enhanced charge extraction, or additional exciton
dissociation inside the ITO/AIL/SubPc system. In light of
this, simple PL intensity measurements on half-cells of
ITO/SubPc with and without AIL in between have been car-
ried out. As the absorption intensities of each thin film sam-
ple are virtually the same at the excitation wavelength of
532 nm, the PL intensity shown in Figure 5b can be used as
a semiquantitative scale of exciton density.
0
0
0
0
0
0
0
0
.35
.30
.25
.20
.15
.10
.05
.00
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
TCTA
w/o AIL
3
00 350 400 450 500 550 600 650
Wavelength (nm)
Before going into details, we noticed that the primary
SubPc emission peak at 610 nm and a secondary peak at
(
b) 4.5
.0
.5
.0
.5
.0
.5
.0
.5
.0
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
TCTA
4
7
film.
12 nm, which are ascribed to defect structures within the
[27]
3
It is worth noting that the obtained PL spectra show
3
the same spectroscopic profile, indicating that the underlying
AIL thin film does not affect SubPc morphology. This sup-
ports the AFM measurement results. As is clear from
Figure 5b, the introduction of an AIL results in stark
increased PL intensities, indicating that the ITO/SubPc inter-
face is seriously exciton quenching. It can be reasonably
inferred that the exciton density of SubPc would be
increased once an AIL is inserted in-between ITO and
SubPc. Stopping or reducing exciton quenching at the ITO
anode increases the chance of SubPc excitons diffusing
toward, and being dissociated at, the SubPc/C60 interface.
However, such inference may not be the only reason for
wide-bandgap AIL. Figure 7 illustrates the correlation of PL
intensity of SubPc (obtained from the half-cells, Figure 5b)
and JSC of SubPc OSCs (see the following section for
Device characterization). A linear fit of narrow-bandgap
2
NPAE
w/o AIL
2
1
1
0
0
5
75 600 625 650 675 700 725 750 775
Wavelength (nm)
FIGURE 5 (a) Absorption and (b) emission spectra of ITO/AIL
2 nm)/SubPc (11 nm)
(
1
00
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
TCTA
SubPc
(w/o AIL)
80
60
40
20
0
4
4
3
3
2
2
1
1
0
.5
.0
.5
.0
.5
.0
.5
.0
.5
NPAAnQ
PyPAFN
FPhSPFN
PhSPFN
NPAE
TCTA
4
00 450 500 550 600 650 700 750
Wavelength (nm)
w/o AIL
FIGURE 6 Wavelength-dependent EQE of SubPc/BCP OSCs
with and without AIL
5
.80 6.00 6.20 6.40 6.60 6.80 7.00 7.20 7.40 7.60
JSC (mA/cm2)
the AIL. Finally, the additional AIL absorption in half-cells
is too limited to justify direct exciton contribution or energy
transfer. Moreover, the very thin AIL is in direct contact
FIGURE 7 PL intensity of ITO/AIL/SubPc half-cells versus
obtained JSC for each respective OSC

6
GOLDER ET AL.
AIL data points shows that PL intensity is directly propor-
tional to the obtained JSC values, whereas the wide-bandgap
AILs, TCTA and NPAE, do not follow this trend. In addi-
tion to the exciton-blocking function suggested for narrow-
bandgap AILs, hole extraction or formation of a secondary
exciton dissociation site is necessary to be included in the
explanation of a relatively higher JSC (or a relatively lower
PL intensity) found for the common wide-bandgap AILs.
Although there are fewer SubPc excitons (due to a lower PL
intensity) in TCTA or NPAE OSCs, the conversion ratio of
(a)
0
2
4
6
8
FPhSPAN
NPAAnQ
PhSPAN
PyPAFN
NPAE
TCTA
w/o AIL
-
-
-
-
exciton to current (i.e., J ) is a bit higher. As we have men-
SC
tioned earlier in Section 2.1, either TCTA or NPAE may
form a cascade heterojunction that facilities the hole extrac-
tion from SubPc, which reduces PL intensity of SubPc or
enhances JSC of OSC.
-
10
-
0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Voltage (V)
(b)
0
.01
In a comparative sense, we suggest that newly created
surface traps or pinholes are present at each AIL/SubPc
interface. Furthermore, the molecular orientation of, for
example, NPAAnQ may result in interactions with SubPc,
which hinder charge or exciton transport. Nevertheless, exci-
ton quenching in such systems is not well understood with
regard to common AIL materials and, for example,
0.001
1
E-4
E-5
E-6
E-7
E-8
E-9
1
1
1
1
1
[14,28–30]
quartz/organic interfaces.
current density range w/o AIL
current density range w/ AIL
2
.4 | Device characterization
The six materials were studied as AIL in OSCs having a general
structure of ITO/AIL(2 nm)/SubPc(11 nm)/C (45 nm)/2,9-
-
0.9 -0.6 -0.3 0.0
0.3
0.6
0.9
1.2
60
Voltage (V)
Dimethyl-4,7-diphenyl-1,10-phenanthroline (BCP)(7 nm)/Al
100 nm), which were fabricated by vacuum thermal deposition.
FIGURE 8 (a) J-V characteristics of ITO/AIL(2 nm)/SubPc(11 nm)/
60(45 nm)/BCP(7 nm)/Al(100 nm) and (b) same devices in the dark
(
C
The low work function, −5.7 eV, of ITO substrates was
maintained in all OSCs studied here in order to minimize hole
extraction barriers derived from various AIL/SubPc interfaces.
The thickness of all AIL was fixed to a very thin 2 nm in order
to reduce parasitic absorption and limit charge buildup at the
interface. This allows the study and facile comparison of OSC
performance among various AIL materials. As such, S-shaped
6
2.64% to 51.62–61.97%). Interestingly, devices using
narrow-bandgap AIL, FPhSPFN, NPAAnQ, PhSPFN, or
PyPAFN, achieve higher JSC than that of conventional
wide-bandgap materials, TCTA or NPAE. Such results sug-
gest that exciton blocking or exciton diffusion toward the
D/A interface is more beneficial than an additional interface
of exciton dissociation or an additional layer of hole extrac-
tion. Regarding open-circuit voltage (VOC), it broadly fol-
lows the expected energy level alignment between the ITO
Fermi level and the adjunct organic layer, that is, a lower- or
higher-lying AIL HOMO energy level leads to an increased
[31,32]
J–V characteristics can be mostly avoided.
In addition, a
thicker interlayer tends to weaken the built-in electric field, lead-
ing to a reduction of JSC and fill factor (FF). Although the AIL
thickness is not optimized for each material, the high EQE of
77–88% at 581 nm (Figure 6) for the AIL-containing OSCs
examined here also does not leave much room for further
improvement. Figure 8a shows the J–V characteristics of
OSCs under one sun illumination (AM 1.5 G,
[
33,34]
or decreased Voc, respectively.
containing OSCs exhibit one of the most pronounced S-
shaped J–V characteristics, and one of the highest V of
However, PyPAFN-
−2
1
00 mW cm ), and the averaged cell performances are sum-
O
marized in Table 2.
From the results, compared with SubPc/C60 OSCs with-
out AIL, the PCE of all AIL-containing OSCs increases
1.12 V is recorded. As PyPAFN does not have the lowest-
lying HOMO energy level, we attribute it to the charge
buildup at the ITO/PyPAFN/SubPc interfaces. Similarly, the
second most pronounced S-shaped J–V characteristics were
found for PhSPFN devices. Hence, for both materials
(PyPAFN and PhSPFN), the two highest series resistances
(from 4.08% to 4.23–5.13%) and can be mainly attributed to
−
2
the increase of JSC (from 5.97 to 6.56–7.44 mA cm ),
which largely compensates for a generally lower FF (from

GOLDER ET AL.
7
a
TABLE 2 Photovoltaic performance of OSCs with or without AIL under AM 1.5G irradiation
PCEavg (PCEmax
(%, %)
)
2
a
2
2
AIL
V
OC (V)
J
SC (mA/cm )
FF (%)
R
s
(Ω cm )
R
sh (k Ω cm )
w/o AIL
FPhSPFN
NPAAnQ
PhSPFN
PyPAFN
NPAE
1.09 ± 0.01
1.12 ± 0.01
1.11 ± 0.01
1.10 ± 0.01
1.12 ± 0.01
1.07 ± 0.02
1.11 ± 0.01
5.97 ± 0.09
7.16 ± 0.08
7.44 ± 0.11
6.95 ± 0.09
7.36 ± 0.09
6.87 ± 0.08
6.56 ± 0.10
62.64 ± 1.40
61.97 ± 1.55
58.01 ± 1.69
56.99 ± 1.53
51.62 ± 0.94
56.75 ± 2.25
57.34 ± 1.26
3.98 (4.08)
5.00 (5.13)
4.88 (5.02)
4.37 (4.46)
4.25 (4.30)
4.21 (4.35)
4.14 (4.23)
16.86 ± 0.51
20.11 ± 0.61
20.45 ± 0.59
43.85 ± 0.56
81.25 ± 2.29
28.73 ± 1.21
26.06 ± 0.63
0.95 ± 0.06
0.83 ± 0.09
0.93 ± 0.12
0.91 ± 0.10
0.77 ± 0.11
0.93 ± 0.21
0.69 ± 0.07
TCTA
a
Average values from more than 20 of each device.
2
(
R ) of 81.25 and 43.85 Ω cm were recorded. It is notewor-
utilizing the Shockley equation can lack physical meaning
when applied to typically low carrier-mobility organic mate-
s
thy that, despite these shortcomings, efficient exciton block-
ing still ensures an enhanced PCE, comparable with those of
conventional wide-bandgap materials, TCTA and NPAE.
[38,39]
rials.
Therefore, we suggest that, under illumination,
there is little need of using wide-bandgap AIL for SubPc/C -
60
In addition to VOC and J , it is important to note the dif-
based OSCs due to the efficient D/A interface for exciton dis-
sociation and the effective electron selectivity of BCP on the
cathode side. As long as the AIL is thin and its HOMO energy
level aligns well with that of SubPc, narrow-bandgap AILs
work well and outperform conventional wide-bandgap AILs.
SC
ferences in FF of each device. For example, the FPhSPAFN-
containing OSCs show the highest FF of 61.97%, only slightly
less than 62.64% of SubPc/C60 OSC without AIL. The
reduced FF occurs in parallel to the increased R of devices
S
with any additional AIL. In general, charge transport across
the interfaces is a key factor that influences the R of OSC. In
S
3
| CONCLUSIONS
order to gain further insight into FF and corresponding proper-
ties with the inclusion of AIL, the dark current characteristics
of the devices were recorded. As shown in Figure 8b, cells
with SubPc on bare ITO (i.e., without AIL) exhibit reverse
current densities, fluctuating by almost a factor of 10 at large
negative biases. A similar current density range has been
observed under the forward bias. Moreover, the shape of each
J–V curve at negative biases mostly remains the same. In com-
parison, upon introduction of any AIL materials, the reverse
current density is significantly suppressed, and the shape of
each J–V curve varies considerably. The suppressed reverse
current suggests an improved charge selectivity and enhanced
electron blocking at the anode interface. Considering the shape
of J–V curve, as well as the current densities appear arbitrary
irrelevant to the nature of AIL, we suggest that the very high-
lying LUMO energy level of NPAE and TCTA does not offer
many benefits over the LUMO energy level-matching mate-
rials FPhSPFN, NPAAnQ, PhSPFN, and PyPAFN. In addi-
In summary, we have investigated a series of materials as AIL
in a standardized SubPc/C60 OSC. Regarding their energy
bandgaps, two of these follow traditional HTM design criteria
and possess a relatively wide bandgap, while the four others
match the SubPc bandgap, that is, relatively narrow bandgap.
In both cases, photophysical measurements have demonstrated
that each material acts as a net exciton-blocking layer, thereby
increasing Jsc and PCE. In addition, AFM measurements do
not point to significant changes in the thin film morphology of
SubPc depending on the underlying materials, AIL or ITO. As
demonstrated in our study, the wide-bandgap, HTM-type AILs
serve a dual function, as increased JSC is also achieved by
enabling a secondary exciton dissociation site. On the other
hand, PL measurements have shown that, for bandgap energy-
matching materials, that is, narrow-bandgap AIL, the JSC of
OSC is proportional to the capability of exciton blocking. This
[35]
“
passive” interlayer results in enhanced device performance. In
tion, SubPc is known to be ambipolar, and thus, it stands to
reason that the reduction in dark current of the four narrow-
bandgap AIL materials originates from a reduced electron
transport through the AIL. We believe that the electron-
trapping nature of electron-withdrawing moieties, anthraqui-
none and fumaronitrile, and the thinness of the AIL are key
factors in this regard. It is noteworthy that, in this study, we do
not observe a clear correlation between VOC and dark current,
case of an FPhSPAFN AIL, SubPc/C OSCs yield the highest
60
reported PCE of 5.13% among these device types. Further dark
J–V measurements suggest that, for an efficient D/A hetero-
junction like SubPc/C60, electron-blocking by means of
inserting a high LUMO energy level AIL is not necessary.
These results show that mere exciton-blocking, while preserv-
ing a flat energy landscape with regard to ITO and the donor
material, is a viable approach. In terms of achieving higher
PCEs, this concept is competitive, if not superior, to established
[36,37]
which is sometimes proposed in the literature.
In addi-
tion, deriving parameters or extracting information by way of

8
GOLDER ET AL.
t
HTM design criteria. Therefore, this work encourages the pur-
suit of alternative approaches of interface engineering in OSC.
(100 mL), NaO Bu (10.89 g, 113 mmol), Pd(OAc)2
t
(0.231 g, 1 mmol), and P( Bu) (0.8 mL, 3.09 mmol) were
added. The mixture was then stirred at 120 C over a period
3
ꢀ
of 20 hr. The solvents were removed under reduced pressure
and the residue extracted thrice with dichloromethane. The
4
4
| EXPERIMENTAL
collected organic solvents were dried over MgSO , and then,
4
.1 | Materials and synthesis
the concentration was purified by column chromatography
1
4
-Bromobenzenaldehyde, TCTA, SubPc, BCP, and C60
to afford a yellow oil (17.22 g, yield 91%). H NMR
were purchased from Lumtec, Inc. A common synthetic pro-
(
(
400 MHz, CDCl ): δ (ppm) 7.88 (d, J = 8.4 Hz, 1H), 7.85
d, J = 8.4 Hz, 1H), 7.74 (d, J = 8.0 Hz, 1H), 7.44 (m, 2H),
3
cedure was used to prepare N-phenylnaphthalen-1-amine
[40]
and N-phenylpyren-1-amine.
2,6-Dibromoanthraquione,
7.32 (t, J = 8.0 Hz, 1H), 7.30 (d, J = 7.4 Hz, 1H), 7.28 (d,
J = 8.0 Hz, 2H), 7.17 (t, J = 8.0 Hz, 2H), 7.03 (d, J = 8 Hz,
2H), 6.98–6.91 (m, 3H).
bis(4-bromophenyl)fumaronitrile, PhSPFN, and FPhSPFN
[
25,41,42]
were prepared according to literature procedures
SubPc and the AIL materials were purified twice by vacuum
train sublimation before being used for device fabrication.
All reactions were carried out under nitrogen. Solvents were
carefully dried before use. The synthetic procedures and
structural characterizations are shown in Scheme 1.
4
.4 | 4-(N-(naphthalen-1-yl)-N-phenylamino)
benzaldehyde, 3
To a mixture of 2 (23.18 g, 63 mmol) in tetrahydrofuran
(THF) (150 mL), 4 N HCl (600 mL) was added and refluxed
for 12 h. After cooling to room temperature, the mixture was
extracted thrice with dichloromethane. The combined
4
.2 | 2-(4-Bromophenyl)-1,3 dioxolane, 1
A mixture of 4-bromobenzenaldehyde (11.10 g, 60 mmol),
ethylene glycol (36 mL), and p-toluenesulfonic acid (1.00 g)
in toluene (200 mL) was refluxed over 48 hr. After cooling,
the mixture was neutralized with saturated aqueous NaHCO3
and extracted thrice with ether. The combined organic sol-
organic phase was dried over MgSO and the solvents evap-
4
orated under reduced pressure. The product was obtained as
1
an orange oil (19.52 g, yield 96%). H NMR (400 MHz,
CDCl ): δ (ppm) 9.76 (s, 1H), 7.90 (d, J = 8.4 Hz, 1H), 7.85
3
(d, J = 7.8 Hz, 1H), 7.83 (d, J = 7.8 Hz, 1H), 7.63 (d,
J = 8.8 Hz, 2H), 7.50 (t, J = 7.4 Hz, 1H), 7.49 (t,
J = 7.4 Hz, 1H), 7.40 (d, J = 8.0 Hz, 1H), 7.38 (d,
J = 7.6 Hz, 1H), 7.31–7.21 (m 4H), 7.10 (t, J = 7.2 Hz, 1H),
6.86 (d, J = 8.8 Hz, 2H).
vents were dried over MgSO . After concentrating, the crude
4
product was purified by column chromatography using hex-
ane as the eluent to afford a yellow oil (13.09 g, yield 95%).
1
H NMR (400 MHz, CDCl ): δ (ppm) 7.49 (d, J = 8.4 Hz,
3
2H), 7.33 (d, J = 8.4 Hz, 2H), 5.75 (s, 1H), 4.05 (m, 4H).
0
4
.5 | Trans-4,4 -bis[N-(naphthalen-1-yl)-N-
4
.3 | (4-(1,3-Dioxolane)phenyl)-N-
phenylamino]stilbene, NPAE
phenylnaphthalen-1-amine, 2
A round-bottom flask was charged with Zn (0.47 g,
ꢀ
To a suspension of 1 (13.09 g, 57 mmol) and N-phenyl-
7.2 mmol) in THF (11 mL), and it was cooled to 0 C. TiCl4
1-naphthylamine (11.29 g, 52 mmol) in dry xylene
(0.4 mL, 3.6 mmol) was added drop wise, and the resulting
SCHEME 1 Synthetic scheme of NPAE, NPAAnQ, and PyPAFN

GOLDER ET AL.
9
mixture was refluxed for 30 min, upon which it was cooled
to 0 C. Compound 3 (0.58 g, 1.8 mmol), THF (4.4 mL), and
δ (ppm) 8.20 (d, J = 8.0 Hz, 4H), 8.15 (d, J = 7.6 Hz, 2H),
8.12–8.06 (m, 6H), 8.03–7.98 (m, 4H), 7.63 (d, J = 8.9 Hz,
4H), 7.31–7.24 (m, 10 H), 7.09 (t, J = 7.2 Hz, 2H), 6.97 (d,
ꢀ
dry pyridine (0.4 mL 5.40 mmol) were added into the
resulting dark suspension and refluxed for 4 hr. After
cooling to room temperature, the resulting mixture was
poured into CH Cl /H O (2:3, 100 mL) and filtered through
+
J = 8.9 Hz, 4H), EI-MS m/z = 812.3 (M ).
2
2
2
4.8 | Characterization
a celite pad. The filtrate was extracted thrice with dic-
hloromethane, and the collected organic phases were dried
1
H NMR spectra were measured on a Bruker AV-400 or
AV-500 MHz nuclear magnetic resonance spectrometer.
Fourier transform spectrometer at room temperature or
over MgSO . The solvents were removed, and the resulting
4
residue recrystallized from dichloromethane to afford the
1
ꢀ
product as a yellow solid (0.37 g, yield 66%). H NMR
50 C. EI-MS was performed by the Mass Spectroscopic
(
(
(
(
400 MHz, CDCl ): δ (ppm) 7.90 (d, J = 8.0 Hz, 4H), 7.86
d, J = 8.0 Hz, 4H), 7.75 (d, J = 8.0 Hz, 4H), 7.48–7.40
m, 4H), 7.35–7.23 (m, 8H), 7.04 (d, J = Hz, 4H), 6.94
Laboratory in-house service of the Institute of Chemistry.
Low-energy photoelectron spectrometer (Riken-Keiki AC-2)
or PESA was used to measure the ITO work function and
the HOMO energy level of the AIL materials, and the
absorption spectra of the materials were measured by a
ultraviolet-visible spectrophotometer (V-630, Jasco). The
3
d, J = 8 Hz, 4H), 6.93 (m, 2H), 6.86 (s, 2H), electron ioni-
+
zation mass spectroscopy (EI-MS) m/z = 614.2 (M ).
ITO substrate was pretreated with O plasma before use in
2
4
.6 | 2,6-Bis[N-(naphthalen-1-yl)-N-
order to increase the work function to 5.6–5.7 eV. Including
those of SubPc, optical energy gaps and LUMO level ener-
gies were obtained by estimating the absorption onset wave-
length. The surface morphology of the materials was
characterized by noncontact mode AFM (Park Systems, XE-
100). Steady-state PL experiments were performed with a
spectrometer system (Tau-3, Horiba). The excitation light
radiation, being selected at 532 nm with a 3-nm full-width-
at-half-maximum by an excitation monochromator from a
xenon lamp, was focused to the sample surface. The emitted
PL radiation was collected by a concave mirror in a non-
specular direction and sent to an emission monochromator
equipped with a multialkali photon-counting photomultiplier
tube (R2949, Hamamatsu). The PL spectra were recorded by
step-scanning the emission monochromator with a spectral
resolution of 3 nm and a spectral step of 0.5 nm with a 0.2 s
for signal integration time. The intensity spectral response of
the system was calibrated.
phenylamino]anthraquinone, NPAAnQ
2,6-Dibromoanthraquione (0.70 g, 1.92 mmol), N-phenyl-
t
naphthalen-1-amine (0.84 g, 3.84 mmol), NaO Bu (1.27 g,
t
13.2 mmol), Pd(OAc)2 (0.03 g, 0.12 mmol), P( Bu)3
(0.65 mL, 0.49 M solution in toluene), and toluene (50 mL)
were mixed together in a Schlenk flask. The reaction was
refluxed for 24 hr. Subsequently, the solvent was removed
under reduced pressure and the crude product extracted
thrice with dichloromethane. The organic phases were col-
lected, dried over MgSO and suction-filtered through a
short plug of silica. After the solvents were removed under
reduced pressure, the crude product was subjected to a train
sublimation to give an orange solid (0.71 g, yield 58%). H
NMR (500 MHz, CDCl , 50 C): δ (ppm) 7.98 (d,
J = 8.5 Hz, 2H), 7.94 (d, J = 8.0 Hz, 2H), 7.89 (d,
J = 8.0 Hz, 2H), 7.87 (d, J = 8.5 Hz, 2H), 7.71 (s, 2H), 7.54
4
1
ꢀ
3
(
t, J = 7.5 Hz, 2H), 7.50 (t, J = 7.5 Hz, 2H), 7.42 (d,
J = 7.5 Hz, 2H), 7.41 (t, J = 8.5 Hz, 2H), 7.33 (t, J = 7.5 Hz,
H), 7.27 (d, J = 7.5 Hz, 4H), 7.15 (t, J = 7.5 Hz, 2H), 7.02
4
4.9 | Solar cell fabrication and testing
+
(d, J = 8.0 Hz, 2H), EI-MS m/z = 643.2 (M + H) .
All devices were fabricated with a conventional structure
of ITO/AIL (2 nm)/SubPc (11 nm)/C60 (45 nm)/BCP
4
.7 | Bis[4-(N-(pyren-1-yl)-N-phenylamino)
(7 nm)/Al (100 nm) by vacuum-thermal evaporation. The
phenyl]fumaronitrile, PyPAFN
organic layers were sequentially deposited on patterned
ITO glass purchased from Lumtec with a sheet resistance
of 15 Ω/sq. Prior to evaporation, the substrates were
cleaned by successive sonication in detergent, deionized
water, acetone, and isopropanol. Substrate treatment with
A two-neck flask was loaded with bis(4-bromophenyl)fum-
aronitrile (1.00 g, 2.57 mmol), N-phenylpyren-1-amine
(1.51 g 5.15 mmol), Cs CO (2.52 g, 7.73 mmol), and
2 3
Pd(OAc) (17.4 mg, 0.08 mmol). Dried toluene (50 mL)
followed by 0.49 mL P( Bu) (0.49 M in tolune) were added
2
t
O plasma (PCD 150, All Real) for 5 min, with a power
2
3
by syringe, and the resulting mixture was refluxed for 24 hr.
The solvents were removed under reduced pressure and the
of 40 W and an initial oxygen pressure of 180 mTorr, was
utilized to modify the work function of the ITO. Using a
photoelectron spectrometer, the ITO work function was
measured within 5.70 ± 0.05 eV after such O2 plasma
crude product subjected to a train sublimation to give a pur-
1
ple solid (1.01 g, yield 48%). H NMR (400 MHz, CDCl ):
3

1
0
GOLDER ET AL.
treatment. Samples were then rapidly transferred into the
[5] H. Gommans, B. Verreet, B. P. Rand, R. Muller, J. Poortmans,
P. Heremans, J. Genoe, Adv. Funct. Mater. 2008, 18, 3686.
evaporation chamber via a connected N -filled glove box.
2
[
6] Z. Zheng, Q. Hu, S. Zhang, D. Zhang, J. Wang, S. Xie, R. Wang,
Y. Qin, W. Li, L. Hong, N. Liang, F. Liu, Y. Zhang, Z. Wei,
Z. Tang, T. P. Russell, J. Hou, H. Zhou, Adv. Mater. 2018, 30,
The evaporation rate was controlled by a quartz-crystal
monitor that was calibrated through a series of thickness
measurements of each layer with a surface profiler (Veeco
Dektak 150). The evaporation rate for each layer and cath-
ode was maintained as follows: AIL 1 ± 0.1 Å/s, SubPc
1
801801.
[7] C. Kulshreshtha, G. W. Kim, R. Lampande, D. H. Huh, M. Chae,
J. H. Kwon, J. Mater. Chem. A 2013, 1, 4077.
[
8] Z. El Jouad, M. Morsli, G. Louarn, L. Cattin, M. Addou,
J. C. Bernède, Sol. Energy Mater. Sol. Cells 2015, 141, 429.
9] A. Barito, M. E. Sykes, D. Bilby, J. Amonoo, Y. Jin, S. E. Morris,
P. F. Green, J. Kim, M. Shtein, J. Appl. Phys. 2013, 113, 203110.
1.7 ± 0.3 Å/s, C60 1.0 ± 0.1 Å/s, BCP 1.3 ± 0.3 Å/s, and
Al 10 ± 3 Å/s. During evaporation, the chamber pressure
was maintained below 9 × 10− Torr. After the deposition
of the interlayer, the vacuum was broken in order to load
further materials, and the sample was transferred tempo-
[
6
[
10] K. Nakano, K. Tajima, Adv. Mater. 2017, 29, 1603269.
[11] F. Jin, Z. Su, B. Chu, P. Cheng, J. Wang, H. Zhao, Y. Gao,
X. Yan, W. Li, Sci. Rep. 2016, 6, 26262.
rally back into the N -filled glove box. For each
2
[
12] A. Barito, M. E. Sykes, B. Huang, D. Bilby, B. Frieberg, J. Kim,
P. F. Green, M. Shtein, Adv. Energy Mater. 2015, 4, 1400216.
13] J. C. Wang, S. Q. Shi, C. W. Leung, S. P. Lau, K. Y. Wong, Appl.
Phys. Lett. 2012, 100, 053301.
interlayer-containing device, a reference cell (SubPc/BCP
OSC w/o AIL) was fabricated simultaneously to ensure
comparability. After the evaporation process, the devices
[
were transferred back into a N -filled glove box and
[14] M. Hirade, C. Adachi, Appl. Phys. Lett. 2011, 100, 153302.
[15] L. Caliò, P. R. Bhushan, J. Benduhn, K. Vandewal, H.-
G. Rubahn, M. Madsen, S. Kazim, S. Ahmad, Sustainable Energy
Fuels 2018, 2, 2296.
2
encapsulated to prevent damage from the oxygen and
moisture. A total of 82 devices were prepared and ana-
lyzed in the dark and under simulated AM 1.5G solar illu-
mination from a class A solar simulator (Newport 300 W)
[16] W. Chen, D.-C. Qi, H. Huang, X. Gao, A. T. S. Wee, Adv. Funct.
Mater. 2011, 21, 410.
2
with an illumination intensity of 100 mW/cm , which was
[17] P. Agarwala, D. Kabra, J. Mater. Chem. A 2017, 5, 1348.
controlled by a programmable source meter (Keithley
[18] S. S. Reddy, K. Gunasekar, J. H. Heo, S. H. Im, C. S. Kim, D.-
2
400). The light intensity was calibrated using a VLSI
H. Kim, J. H. Moon, J. Y. Lee, M. Song, S.-H. Jin, Adv. Mater.
2
016, 28, 686.
StSRC-1000-TC-QZ reference solar cell, while series resis-
tance (R ) and shunt resistance (R ) were estimated by
[
[
19] D. E. M. Rojas, K. T. Cho, Y. Zhang, M. Urbani, N. Tabet,
S
sh
G. Torre, M. K. Nazeeruddin, T. Torres, Adv. Energy Mater.
taking the reciprocal from the slope of a tangent to the J–
V curve at the open- and short-circuit conditions, respec-
tively. For monochromatic incident photon-to-current effi-
ciency or EQE measurements, a 100 W Xe lamp was used
to generate the bias light.
2
018, 8, 1800681.
20] C.-F. Lin, S.-H. Li, C.-C. Lee, T. Sakurai, M. Kubota, W. C. Su,
J.-C. Huang, T.-L. Chiu, H.-C. Han, L.-C. Chen, C.-T. Chen, J.-
H. Lee, Sol. Energy Mater. Sol. Cells 2015, 137, 138145.
21] J. Golder, C.-W. Lin, H.-W. Chen, Y.-B. Lan, C.-T. Chen, J.-
K. Wang, J. Phys. D Appl. Phys. 2018, 51, 314002.
[
[22] Y. Nakayama, T. L. Nguyen, Y. Ozawa, S. Machida, T. Sato,
H. Tokairin, Y. Noguchi, H. Ishii, Adv. Energy Mater. 2014, 4,
ACKNOWLEDGMENTS
1
301354.
[23] S. M. Menke, R. J. Holmes, Energy Environ. Sci. 2014,
, 499.
[
[
This research was supported in part by the Ministry of Sci-
ence and Technology of Taiwan, iMATE Program of Acade-
mia Sinica, and National Chiao Tung University.
7
24] T. M. Clarke, J. R. Durrant, Chem. Rev. 2010, 110, 6736.
25] Y.-T. Lee, C.-L. Chiang, C.-T. Chen, Chem. Commun. 2008,
2
, 217.
[
[
26] J.-L. Bredas, Mater. Horiz. 2014, 1, 17.
27] H. Gommans, S. Schols, A. Kadashchuk, P. Heremans, J. Phys.
Chem. C 2009, 113, 2974.
ORCID
Chin-Ti Chen https://orcid.org/0000-0002-1493-2533
[
28] X. Xiao, J. D. Zimmerman, B. E. Lassiter, K. J. Bergemann,
S. R. Forrest, Appl. Phys. Lett. 2013, 102, 073302.
29] J. Wang, T. Wang, D. Cao, X. Zhao, J. Liu, M. Zhuo, B. Mi,
Z. Gao, Org. Electron. 2015, 23, 11.
[
REFERENCES
[
[
1] Z. Yin, J. Wei, Q. Zheng, Adv. Sci. 2016, 3, 1500362.
2] J. D. Servaites, M. A. Ratner, T. J. Marks, Energy Environ. Sci.
[30] A. van Dijken, A. Perro, E. A. Meulenkamp, K. Brunner, Org.
Electron. 2003, 4, 131.
2
011, 11, 4410.
[31] I. Irfan, S. Graber, F. So, Y. Gaoa, Org. Electron. 2012, 13,
2028.
[32] W. Tress, K. Leo, M. Riede, Adv. Funct. Mater. 2011, 21, 2140.
[
3] C. Sun, Z. Wu, Z. Hu, J. Xiao, W. Zhao, H.-W. Li, Q.-Y. Li, S.-
W. Tsang, Y.-X. Xu, K. Zhang, H.-L. Yip, J. Hou, F. Huang,
Y. Cao, Energy Environ. Sci. 2017, 10, 1784.
[4] P. Peumans, S. R. Forrest, Appl. Phys. Lett. 2001, 79, 126.

GOLDER ET AL.
11
[
33] S. Braun, W. R. Salaneck, M. Fahlman, Adv. Mater. 2009, 21,
[41] A. Y.-Z. Lin, C. H. Huang, Y. J. Chang, C.-W. Yeh, T.-M. Chin,
K.-M. Chi, P.-T. Chou, M. Watanabe, T. J. Chow, Tetrahedron
2014, 70, 262.
[42] B. H.-C. Yeh, W.-C. Wu, Y.-S. Wen, D.-C. Dai, J.-K. Wang, C.-
T. Chen, J. Org. Chem. 2004, 69, 6455.
1
450.
[
[
34] K. Akaike, Y. Kubozono, Org. Electron. 2013, 14, 1.
35] H. T. Chandran, T.-W. Ng, Y. Foo, H.-W. Li, J. Qing, X.-K. Liu,
C.-Y. Chan, F.-L. Wong, J. A. Zapien, S.-W. Tsang, M.-F. Lo,
C.-S. Lee, Adv. Mater. 2017, 29, 1606909.
[
[
36] N. K. Elumalai, A. Uddin, Energy Environ. Sci. 2016, 9, 391.
37] M. D. Perez, C. Borek, S. R. Forrest, M. E. Thompson, J. Am.
Chem. Soc. 2009, 131, 9281.
How to cite this article: Golder J, Lin C-W,
Chen C-T. Anode interlayer in organic photovoltaics:
Narrow bandgap small molecular materials as
exciton-blocking layer. J Chin Chem Soc. 2019;1–11.
https://doi.org/10.1002/jccs.201900123
[
[
[
38] U. Würfel, D. Neher, A. Spies, S. Albrecht, Nat. Commun. 2015,
6
, 6951.
39] I. Ramirez, M. Causa, Y. Zhong, N. Banerji, M. Riede, Adv.
Energy Mater. 2018, 8, 1703551.
40] I. Neogi, S. Jhulki, A. Ghosh, T. J. Chow, J. N. Moorthy, Org.
Electron. 2014, 8, 3766.