Synthesis and characterization of cyano-substituted pyridine derivatives for applications as exciton blockers in photovoltaic devices

Article abstract of DOI:10.1039/c2jm15287h

A new class of cyano-substituted pyridine derivatives have been synthesized and characterized. The relationship between their chemical structures, energy gap, energy levels, electron mobility and device performance was studied systematically by applying the new materials as EBL in standard CuPc/C ₆₀/EBL organic photovoltaic (OPV) devices. Their properties fulfill all the requirements for a good EBL material, such as a wide energy gap (E _g, 3.12-3.50 eV), a deep highest occupied molecular orbital (HOMO, 6.71-7.80 eV) energy level, a high electron mobility (in the order of 10 ^-5 cm² V^-1 s^-1) and a high glass transition temperature (T_g, 139-189°C), thus most of them show performances surpassing that of BCP. In particular, one device shows an efficiency of 2.12% and a half-efficiency lifetime of 340 h, which are, respectively, 44% and more than 5 times better than those (1.47%, 60 h) of the BCP-based device. Furthermore, we found that the lowest unoccupied molecular orbital (LUMO) energy level of the EBL can also influence the leakage current and thus the efficiency of the OPV devices. It is expected that these studies would provide useful guidance for designing high-performance EBL materials.

Full text of DOI:10.1039/c2jm15287h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 5107
www.rsc.org/materials
PAPER
Synthesis and characterization of cyano-substituted pyridine derivatives
for applications as exciton blockers in photovoltaic devices†
b
a
b
b
a
Juanjuan You,abc Ming-Fai Lo, Weimin Liu, Tsz-Wai Ng, Shiu-Lun Lai, Pengfei Wang
*
b
*
and Chun-Sing Lee
Received 18th October 2011, Accepted 5th January 2012
DOI: 10.1039/c2jm15287h
A new class of cyano-substituted pyridine derivatives have been synthesized and characterized. The
relationship between their chemical structures, energy gap, energy levels, electron mobility and device
performance was studied systematically by applying the new materials as EBL in standard CuPc/C60/
EBL organic photovoltaic (OPV) devices. Their properties fulfill all the requirements for a good EBL
material, such as a wide energy gap (Eg, 3.12–3.50 eV), a deep highest occupied molecular orbital
(HOMO, 6.71–7.80 eV) energy level, a high electron mobility (in the order of 10ꢀ5 cm2 Vꢀ1 sꢀ1) and
a high glass transition temperature (Tg, 139–189 ꢁC), thus most of them show performances surpassing
that of BCP. In particular, one device shows an efficiency of 2.12% and a half-efficiency lifetime of 340
h, which are, respectively, 44% and more than 5 times better than those (1.47%, 60 h) of the BCP-based
device. Furthermore, we found that the lowest unoccupied molecular orbital (LUMO) energy level of
the EBL can also influence the leakage current and thus the efficiency of the OPV devices. It is expected
that these studies would provide useful guidance for designing high-performance EBL materials.
However, the operational stabilities of such devices are still far
below requirements of most applications. While the stability
problem is clearly a pressing issue, so far not many systematic
studies have been carried out. Actually, the authors have shown
that oxygen species from the ITO is one factor that causes
degradation of devices using C60.7,14 Substantially improved
device stability has been achieved by inserting a fluorocarbon
layer between the ITO anode and the photoactive organic layers.7
On the other hand, crystallization of the bathocuproine (BCP) –
the most commonly used EBL under the cathode – has been
identified as another important cause of degradation.15 To
address this problem, several other organic materials, such as
Alq3 and its derivatives, have been used instead of BCP.15–21
While these materials do offer considerable stability enhance-
ment, their electrical performances are similar or only slightly
better than BCP. Up til now, there has been no report on a BCP
replacement which can simultaneously enhance both the stability
and efficiency considerably. It is thus highly desirable to develop
a high performance EBL for replacing BCP.
1. Introduction
Organic photovoltaic (OPV) devices have attracted much
attention recently for their various merits including low
manufacturing cost, slight weight, large area and potential flex-
ibility.1,2 In the past decade, most research efforts have been
focussed on enhancing power conversion efficiency. One of the
most important progresses is the introduction of an exciton-
blocking layer (EBL), which can prevent the active layer from
being damaged in the vacuum thermal-deposition process of the
metal electrode and reduce the loss of excitons via quenching by
keeping the excitons away from the metal cathode.3,4 In addition,
other approaches including the introduction of various anode
buffer layers,5–7 bulk heterojunction,3,8 and high performance
materials etc.9–12 have increased the power conversion efficiency
to a level where commercial applications become viable.13
aKey Laboratory of Photochemical Conversion and Optoelectronic
Materials Technical Institute of Physics and Chemistry, Chinese
Academy of Sciences, No. 29, Zhongguancun East Road, Haidian
District, Beijing, 100190, P. R. China. E-mail: wangpf@mail.ipc.ac.cn;
Fax: (+86) 10-8254-3512
bCenter of Super-Diamond and Advanced Films (COSDAF), Department
of Physics and Materials Science, City University of Hong Kong, Tat Chee
Avenue, Kowloon, Hong Kong SAR, P. R. China. E-mail: apcslee@cityu.
edu.hk
In general a good EBL material should have 1) a wide energy
gap (Eg) to block excitons diffusion to the cathode; 2) a deep
highest occupied molecular orbital (HOMO) energy level to
block hole leakage to the cathode; 3) a high electron mobility for
low electrical resistance; 4) a low optical absorption to minimize
photon loss and 5) a high glass transition temperature (Tg) for
thermal stability.22–26 We have previously used a pyridine deriv-
ative p-PPTNT as an electron-transporting and hole-blocking
layer in high-performance deep blue OLEDs.27,28 It appears that
the properties of p-PPTNT do fulfill all the requirements for
cGraduate University of Chinese Academy of Sciences, Beijing, 100190,
P. R. China
† Electronic supplementary information (ESI) available: thermal
properties measurements; electron mobility measurements; molecular
structure of p-MPyCN’s intermediate product. See DOI:
10.1039/c2jm15287h
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 5107–5113 | 5107

View Article Online
a good EBL material and similar pyridine derivatives might have
good potential for this application.
results would provide useful guidance for designing high-
performance EBL materials.
In this work, we designed and synthesized a series of cyano-
substituted pyridine derivatives with structures as shown in
Scheme 1. The design concept is based on the fact that by
introducing more cyano groups on the pyridine moiety, the
HOMO energy level can be drawn further downwards. The
added cyano groups would also increase the steric hindrance and
perturb the molecular planarity,29–32 leading to wider energy
gaps. Performances of the new compounds were studied by
applying them as EBL in standard CuPc/C60/EBL devices. Most
of the new compounds show performances surpassing that of
BCP. In particular, the device using m-MPyCN as EBL shows an
efficiency of 2.12% and a half-efficiency lifetime of 340 h, which
are, respectively, 44% and more than 5 times better than those
(1.47%, 60 h) of the BCP-based reference device. Furthermore, in
addition to the EBL requirements mentioned above, we also
found that the energy of the lowest unoccupied molecular orbital
(LUMO) of the EBL can influence the leakage current and thus
the efficiency of the OPV devices. Based on these results, we
analyzed the relationship between chemical structures, energy
gap, energy levels, electron mobility and performance of these
cyano-substituted pyridine derivatives. It is expected that these
2. Experimental section
2.1 Synthesis
The synthetic routes of the cyano-substituted pyridine deriva-
tives are shown in Scheme 1. A representative synthetic proce-
dure for m-MPyCN is as follows: 4-methylbenzoylacetonitrile
(0.95 g, 6 mmol) and isophthalaldehyde (0.20 g, 1.5 mmol) were
dissolved in 30 mL of glacial acetic acid containing ammonium
acetate (1.15 g, 15 mmol). The mixture was refluxed under an air
atmosphere at 120 ꢁC. After three hours, a yellow precipitate was
collected and washed with glacial acetic acid. The crude precip-
itate was then added to 30 mL of glacial acetic acid. Under rapid
stirring,
2,3-dichloro-5,6-dicyano-p-benzoquinone
(DDQ,
0.80 g, 3.5 mmol) was added to the glacial acetic acid solution.
The mixture was refluxed for one hour at 120 ꢁC. After cooling to
room temperature, the reaction mixture was filtered and washed
with methanol to obtain a white powder. The white powder was
further purified by column chromatography with dichloro-
methane as an eluting agent. Finally, 0.78 g white powder of
m-MPyCN was obtained (yield 75%).
Scheme 1 Molecular structures (p-PPTNT and BCP are included for reference) and synthetic routes of the new cyano-substituted pyridine derivatives.
5108 | J. Mater. Chem., 2012, 22, 5107–5113 This journal is ª The Royal Society of Chemistry 2012

View Article Online
4,40-(1,3-phenylene) bis(2,6-dip-tolylpyridine-3,5-dicarbonitrile)
a nitrogen atmosphere and encapsulated by ultraviolet light
curing resins.
1
(m-MPyCN). H NMR (400 MHz, CDCl3, d): 8.00 (d, 8H, J ¼
8.2), 7.90 (s, 4H), 7.35(d, 8H, J ¼ 8.1), 2.45 (s, 12H). 13C NMR
(400 MHz, CDCl3, d): 163.4, 158.8, 142.2, 134.8, 133.7, 131.5,
130.3, 129.7, 129.6, 115.9, 105.3, 21.7. HRMS (EI, m/z): [M+]
calcd for C48H32N6: 692.2688; found: 692.2700. Anal. calcd. for
C48H32N6: C 83.21, H 4.66, N 12.13; Found: C 83.22, H 4.71,
N 12.13.
2.3 Device and film characterization
Density–voltage (J–V) characteristics of the OPV devices were
measured with a programmable Keithley model 237 power
source. The current was measured in the dark and under illu-
mination with an intensity of 100 mW cmꢀ2 from an Oriel 150 W
solar simulator with AM1.5G filters. The light intensity was
calibrated with a reference Si diode with a KG-5 filter. Absorp-
tion spectra of solid-state films deposited on quartz substrates
were measured with a Perkin-Elmer Lambda 2S UV–visible
spectrophotometer. HOMO energy levels were measured with
thin films deposited on ITO glass substrates by ultra-violet
photoelectron spectroscopy (UPS) in a VG ESCA-LAB 220i-XL
surface analysis system. LUMO energy levels were estimated by
subtracting from HOMO with the optical band gap determined
from the absorption spectra of the solid-state films.
4,40-(1,3-phenylene) bis(2,6-diphenylpyridine-3,5-dicarbonitrile)
1
(m-PyCN). H NMR (400 MHz, CDCl3, d): 8.09 (m, 8H), 7.94
(s, 4H), 7.58(m, 12H). 13C NMR (400 MHz, CDCl3, d): 163.6,
158.7, 136.4, 134.8, 131.7, 130.5, 129.8, 129.6, 129.0, 115.6, 106.1.
HRMS (EI, m/z): [M+] calcd for C44H24N6: 636.2062; found:
636.2069. Anal. calcd. for C44H24N6: C 83.00, H 3.80, N 13.20;
Found: C 82.91, H 3.80, N 13.22.
4,40-(1,3-phenylene) bis(2,6-di(biphenyl-4-yl)pyridine-3,5-dicar-
bonitrile) (m-PhPyCN). 1H NMR (400 MHz, CDCl3, d): 8.24 (d,
8H, J ¼ 8.4), 7.98 (s, 4H), 7.80 (d, 8H, J ¼ 8.4), 7.69 (m, 8H), 7.49
(m, 8H), 7.42 (d, 4H, J ¼ 7.2). 13C NMR (400 MHz, CDCl3, d):
163.0, 159.0, 144.6, 140.0, 135.2, 134.8, 131.7, 130.5, 130.3, 129.6,
129.1, 128.3, 127.7, 127.4, 115.8, 105.7. MS (MALDI-TOF, m/z):
[M + Na]+ calcd for C68H40N6: 940.3314; found: 963.4. Anal.
calcd. for C68H40N6: C 86.79, H 4.28, N 8.93; Found: C 86.47, H
4.49, N 8.69.
3. Results and discussion
3.1 Thermal stability
The thermal properties of the new compounds measured with
thermogravimetric analyses (TGA) and differential scanning
calorimetry (DSC) are summarized in Table 1 and shown in
Fig. S1.† It can be seen that all the bilateral compounds possess
higher decomposition temperatures (Tds) than the unilateral
compound CNPyCN. Compounds with more cyano groups (m-
PyCN, p-MPyCN, m-MPyCN and m-PhPyCN) also have higher
Tds than that of p-PPTNT. The Tgs of these compounds have
a consistent trend with that of Tds. The Tgs of CNPyCN and
p-MPyCN were not observed. The former may have too low a Tg
to be detected, while the latter case is probably due to the more
bulky and rigid structure than that of the meta-substituted m-
MPyCN. A similar behavior has also been observed for similar
bulky structures such as para-substituting fluorene derivatives.33
The absence of Tg might be caused by hindering of molecular
segmental motion by the bulky and rigid structures. Compared
with the widely used BCP (with a Tg < 80 ꢁC)15 for EBL, most of
our cyano-substituted pyridine derivatives possess considerably
better thermal stability.
2,6-bis
(4-cyanophenyl)-4-phenylpyridine-3,5-dicarbonitrile
1
(CNPyCN). H NMR (400 MHz, CDCl3, d): 8.16 (d, 4H, J ¼
8.4), 7.88 (d, 4H, J ¼ 8.4), 7.66 (m, 5H). 13C NMR (400 MHz,
CDCl3, d): 161.7, 161.1, 140.0, 132.8, 131.8, 130.3, 129.6, 128.9,
118.0, 115.5, 115.0, 107.7. HRMS (EI, m/z): [M+] calcd for
C27H13N5: 407.1171; found: 407.1167. Anal. calcd. for
C27H13N5: C 79.59, H 3.22, N 17.19; Found: C 79.35, H 3.37,
N 16.94.
4,40-(1,4-phenylene) bis(2,6-dip-tolylpyridine-3,5-dicarbonitrile)
(p-MPyCN). HRMS (EI, m/z): [M+] calcd for C48H32N6:
692.2688; found: 692.2698. Anal. calcd. for C48H32N6: C 83.21,
1
H 4.66, N 12.13; Found: C 82.21, H 4.70, N 11.86. Data for H
NMR and 13C NMR were not observed because of its insolubility
in commonly used reagents. Therefore, the molecular structure
and identification for its intermediate product are shown in the
ESI (see Scheme S1†).
3.2 Absorption properties
Optical absorption properties are important parameters
controlling the performance of EBLs in OPV devices. Absorp-
tion spectra of solid films of the pyridine derivates on quartz
substrates are shown in Fig. 1a. It can be seen that p-PPTNT has
a strong absorption band peaking at 290 nm and a shoulder at
about 340 nm. The two absorption bands can be assigned to the
p–p* transition of the pyridine moiety and intramolecular
charge transfer (ICT) state respectively. Comparing to
p-PPTNT, p-MPyCN and m-MPyCN have similar structures
except for more cyano groups on the pyridine moiety, shoulders
peaking at less than 250 nm and strong absorption bands
peaking around 313 nm are observed, indicating enhanced
intramolecular charge transfer. Moreover, steric hindrance of the
cyano groups in p-MPyCN and m-MPyCN can disturb
2.2 Device fabrication
OPV devices were fabricated on patterned ITO-coated glass
substrates with a sheet resistance of 30 U/,. The substrates were
cleaned with Decon 90, rinsed in de-ionized water, dried in an
oven, and finally exposed to UV-ozone for about 30 min. The
ITO substrates were then immediately transferred into a deposi-
tion chamber with a base pressure of 1 ꢂ 10ꢀ4 Pa. Organic layers
were sequentially deposited by conventional thermal evapora-
tion. The deposition rates were monitored using a quartz oscil-
lating crystal and controlled to be 0.1–0.2 nm sꢀ1. Then 150 nm of
Al cathode was deposited at a rate of 0.8–1.0 nm sꢀ1 through
a shadow mask to define an active device area of 0.12 cm2. The
devices were immediately transferred to the glove box with
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 5107–5113 | 5109

View Article Online
Table 1 Properties of the cyano-substituted pyridine derivatives and referenced BCP
Materials
Tg/Tm/Td [ꢁC]
lab.
a [nm]
HOMO [eV]
LUMO [eV]
Eg [eV]
me@6 Vb[cm2 Vꢀ1 sꢀ1
]
Reference
max.
BCP
<80/n.a./n.a.
189/342/520
139/363/388
149/341/402
n.a./338/426
141/339/408
n.a./n.a./325
n.a.
342
330
312
313
295
289
ꢀ6.50
ꢀ6.71
ꢀ6.81
ꢀ6.91
ꢀ6.93
ꢀ7.16
ꢀ7.80
ꢀ3.00
ꢀ3.59
ꢀ3.57
ꢀ3.55
ꢀ3.61
ꢀ3.70
ꢀ4.30
3.50
3.12
3.24
3.36
3.32
3.46
3.50
6 ꢂ 10ꢀ7
4.8 ꢂ 10ꢀ5
4.5 ꢂ 10ꢀ5
5.6 ꢂ 10ꢀ5
3.4 ꢂ 10ꢀ5
2.9 ꢂ 10ꢀ5
n.a.c
Refs 15, 36
This work
Refs 27, 28
This work
This work
This work
This work
m-PhPyCN
p-PPTNT
m-MPyCN
p-MPyCN
m-PyCN
CNPyCN
a
lab.max.: maximum wavelength of absorption (deposited on the quartz substrates). b me: electron mobility (measured by transient electroluminescence
method). c Due to the weak EL of CNPyCN, its transient EL signal cannot be detected.
250 nm and 340 nm are found in m-PhPyCN, which indicates
a higher degree of ICT.
Fig. 1b shows the absorption coefficients of the pyridine
derivatives and three other reference materials (BCP, CuPc and
C60). One key requirement of the EBL is low optical absorption
compared to the photo-active components in the OPV device.
The pyridine derivatives clearly satisfy this requirement as their
absorption coefficients are much lower than those of CuPc and
C
60 and are comparable to that of BCP.
3.3 Energy level and electron mobility
Energy level as well as electron mobility are also important
parameters for EBLs in OPV devices. The HOMO and LUMO
energies of the materials as determined by UPS and optical
absorption measurements were listed in Table 1. In agreement
with our original design concept, by introducing more cyano
groups on the pyridine moiety, p-MPyCN has a deeper HOMO
energy level and a wider Eg compared to p-PPTNT. Meanwhile,
the HOMO energy levels of cyano-substituted pyridine deriva-
tives can be tuned, by different substituents, from 6.71 to 7.80 eV
with an increased degree of electron-deficiency, indicating better
hole-blocking abilities than BCP (with a HOMO energy level of
6.5 eV). In addition, all of them have wide Egs larger than 3.0 eV,
explaining their abilities for blocking excitons.
The electron mobilities of the cyano-substituted pyridine
derivatives were measured using the transient electrolumines-
cence (EL) method.34,35 As shown in Table 1 and Table S1,† the
electron mobilities of all the cyano-substituted pyridine deriva-
tives are in the order of 10ꢀ5 cm2 Vꢀ1 sꢀ1, which are two orders of
magnitude higher than that of BCP36 and facilitate more efficient
electron transport through EBLs.
Fig. 1 a) Normalized absorption spectra of the cyano-substituted
pyridine derivatives on quartz substrates. b) Absorption coefficients of
the cyano-substituted pyridine derivatives, BCP, CuPc and C60 on quartz
substrates.
3.4 Photovoltaic performances
OPV devices with a structure of ITO/CuPc (35 nm)/C60 (45 nm)/
EBLs (5 nm)/Al (150 nm) were fabricated by using the six pyri-
dine derivatives and BCP as the EBL. For easy reference, the
energy levels of the used materials are shown in Fig. 2. Fig. 3
depicts the current density–voltage (J–V) characteristics of the
OPV devices with different EBLs under 1 Sun AM 1.5G (AM: air
mass, G: global) simulated solar illumination. Key device
performance parameters including short-circuit current (JSC),
open-circuit voltage (VOC), fill factor (FF), shunt resistance
(RSh), series resistance (RS), and power conversion efficiency
(PCE) are summarized in Table 2. For the first approximation,
molecular planarity leading to a shortened conjugation length
and blue-shift of the absorption peaks. On the other hand,
replacing the methyl substituent with weaker electron-donating
hydrogen or cyano substituents would weaken ICT, resulting in
blue-shifted absorption peaks in m-PyCN and CNPyCN. On the
contrary, by introducing a stronger electron-donating phenyl
substituent, two red-shifted absorption peaks centered around
5110 | J. Mater. Chem., 2012, 22, 5107–5113
This journal is ª The Royal Society of Chemistry 2012

View Article Online
BCP device. Of all the devices using the new compounds, only the
CNPyCN device shows a lower PCE (1.07%) than the reference
BCP device. Though CNPyCN has the deepest HOMO and the
widest Eg, the performance of its device is obviously limited by
the low FF (0.38).
As mentioned, a good EBL should effectively block excitons
while allowing passage of electrons to the cathode. A good
electron mobility is thus beneficial to the electron collection.
Fig. 4 shows a plot of the JSC against electron mobility of the
EBL. Except for the case of m-PhPyCN, the JSC shows an
obvious increasing trend with the EBL’s electron mobility, as
expected. The low JSC for the m-PhPyCN is likely to be related to
its relatively shallow HOMO level and small energy gap, which
weaken its exciton blocking ability.
It has been shown that the VOC of an OPV device is mainly
controlled by the energy offset at the donor/acceptor interfaces.39
However, this is not considered to be a key variant in the present
work, as all the devices have the same donor/acceptor interface
and electrodes. It has also been shown that the VOC is also
a function of the dark current as described by eqn (1).40
Fig. 2 The energy levels of the materials used.
ꢀ
ꢁ
nkT
q
JPhVOC
JS
VOC
VOC
¼
ln
þ 1 ꢀ
(1)
RShJS
where JS is the saturation dark current density (i.e. the saturation
leakage current when the diode is reversely biased in dark), n is
the ideality factor, RSh is the shunt resistance and JPh is
the photocurrent. When JPh/JS [ 1, VOC is proportional to
ln(JPh/JS), suggesting that a large JS would reduce the VOC. This
relation agrees well with our observation that the high VOC of
0.53 and 0.51 V from the m-MPyCN and the p-PPTNT
devices are accompanied by much smaller JS of 0.00058 and
Fig. 3 The current density–voltage (J–V) characteristics of the OPVs
with different EBLs under 1 Sun AM 1.5G simulated solar illumination.
RS and RSh were estimated from the inverse slope of the J–V
characteristic at zero J and V, respectively.37,38 It can be seen that
the devices using p-PPTNT (JSC ¼ 7.44 mA cmꢀ2, VOC ¼ 0.51 V,
FF ¼ 0.51; PCE ¼ 1.93%) and m-MPyCN (JSC ¼ 7.69 mA cmꢀ2
,
VOC ¼ 0.53 V, FF ¼ 0.52; PCE ¼ 2.12%) have considerably
better performance than the reference device using BCP (JSC
¼
6.33 mA cmꢀ2, VOC ¼ 0.47 V, FF ¼ 0.50, PCE ¼ 1.47%) as the
EBL. In particular, the PCE of the m-MPyCN device is about
44% higher than that of the BCP device. Furthermore, devices
using m-PhPyCN, p-MPyCN and m-PyCN also show higher
PCE (1.60%, 1.63% and 1.55%, respectively) than that of the
Fig. 4 The dependence of the JSC on the electron mobility of the EBL.
Table 2 The device characteristics of the OPVs with different EBLs
EBL
VOC [V]
JSC [mA cmꢀ2
]
FF
PCE [%]
RS [U cm2]
RSh [U cm2]
JS [mA cmꢀ2
]
BCP
0.47
0.46
0.51
0.53
0.46
0.46
0.45
6.33
6.64
7.44
7.69
6.91
6.38
6.21
0.50
0.53
0.51
0.52
0.51
0.52
0.38
1.47
1.60
1.93
2.12
1.63
1.55
1.07
52.29
11.11
15.88
16.57
12.67
11.93
25.61
289.16
370.37
327.42
500.00
300.38
472.44
156.25
0.0036
0.0035
0.0013
0.00058
0.0030
0.0047
0.0060
m-PhPyCN
p-PPTNT
m-MPyCN
p-MPyCN
m-PyCN
CNPyCN
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 5107–5113 | 5111

View Article Online
0.0013 mA cmꢀ2, respectively, than the other devices. On the
contrary, the CNPyCN device has the smallest VOC of 0.45 V,
corresponding to the highest JS of 0.0060 mA cmꢀ2. For other
materials, similar VOC are obtained as their JS show relatively
little variation.
p-PPTNT device showed little decrease within 360 h and decreases
by around 50% after 480 h. The m-MPyCN device shows a mild
degradation within the first 240 h and still maintains about 50% of
the initial PCE after 360 h. The poor stability of the BCP device
can be attributed to its crystallization, leading to pin-holes and
poor electrical conduction.16,20 Therefore, thermal stability of the
EBL is an important requirement to guarantee the device stability
and lifetime, which has been proved by replacing BCP
with materials of high Tgs. Thus, the stability of OPVs based on
p-PPTNT and m-MPyCN were dramatically improved, maybe
We proceed to analyze the relationship between the JS of the
devices and the energy levels of the EBLs (Fig. 2 and Table 1). BCP
has a HOMO energy of 6.5 eV, which is larger than that of C60
(6.2 eV) just by 0.3 eV. On the other hand, m-MPyCN and
p-PPTNT have HOMO energies of 6.91 and 6.81 eV, respectively,
which are much larger than that of BCP and are beneficial for
suppressing hole leakage. This in turn leads to lower JS and thus
higher VOC. However, this simple picture cannot explain the cases
for the p-MPyCN, m-PyCN and CNPyCN devices. Although
their HOMO energy levels are deep enough to block holes effec-
ꢁ
largely owing to their high Tgs of 139 and 149 C, respectively.
However, other factors that may affect the device stability are not
yet clear and require further investigations.
4. Conclusions
tively, their devices still have relatively high JS and lower VOC
indicating that there are other factors influencing the JS.
,
In summary, a novel class of cyano-substituted pyridine deriva-
tives has been designed and synthesized to meet the demand for
One possible cause for the large JS in the CNPyCN device is its
low-lying LUMO energy level of 4.3 eV, which is much closer to
the work function of Al (4.0 eV). This suggests that in reverse
bias, electrons can possibly be injected from Al into the LUMO
of CNPyCN. This is consistent with the observation that the
leakage current of the CNPyCN device is considerably larger
than those of the other devices. This also explains the smaller
VOC of the CNPyCN device. Early reports have demonstrated
that the LUMO position of the EBL seems not to influence the
photovoltaic performance due to the metal-deposition-induced
damage for charge transport.41,42 The present results suggest that
there are possible situations where the JS could be influenced
when the offset between the EBL’s LUMO and the cathode’s
work function is small. On the other hand, the causes of the
relatively large Js in the p-MPyCN and m-PyCN devices are not
yet clear and require further investigations.
ꢁ
EBLs. Most of them have high Tgs (139–189 C), high electron
mobilities (in the order of 10ꢀ5 cm2 Vꢀ1 sꢀ1) and deep HOMO
energy levels (6.71–7.80 eV). With these properties, their appli-
cations as EBLs in small molecular lamellar OPVs were investi-
gated. Devices with most of these cyano-substituted pyridine
derivatives achieved better performance than that of BCP. In
particular, a PCE up to 2.12% was obtained with m-MPyCN,
which is about 44% higher than the BCP-based reference device.
This performance enhancement is the best among all reported
BCP replacements. The high electron mobility and suitable
energy levels (including HOMO, LUMO and Eg) of the present
cyano-substituted pyridine derivatives account for the increases
in JSC, VOC and FF, and thus PCE in the corresponding devices.
Considerably prolonged OPV lifetimes have also been demon-
strated for p-PPTNT and m-MPyCN based devices, which may
be mainly attributed to their high Tgs. The present results show
that cyano-substituted pyridine would be potentially an impor-
tant class of material for high-performance EBLs. The systematic
studies here are also believed to provide useful guidance for the
design of EBL materials.
The FF is closely related to the series resistance (RS) and shunt
resistance (RSh) of OPVs. From Table 2, it can be seen that all
devices using the new compounds show smaller RS values than
that of the reference device using BCP, indicating the higher
electron mobilities of the cyano-substituted pyridine than that of
BCP. With smaller RS and larger RSh values than other devices,
the FF of the devices using m-PhPyCN and m-PyCN reach 0.53
and 0.52, respectively. A FF of 0.52 was also achieved for the m-
MPyCN device owing to its large RSh of 500 U cm2. For the
CNPyCN device, it has a low FF of 0.38 and a relatively small
RSh of 156.25 U cm2. The small RSh suggests that more carriers
would be lost due to recombination and thus resulting in a small
FF. Causes for the small RSh are not entirely clear yet. Never-
theless, this might be caused by more severe damage by Al
cathode deposition on CNPyCN than on other materials.
Further investigation is still needed to fully understand this.
Acknowledgements
This work is financially supported by the National High Tech-
nology Research and Development Program of China
3.5 Stability
As mentioned, the most important hurdle for commercialization
of OPV devices is their poor stability. We have tested the device
stability of the two compounds (m-MPyCN and p-PPTNT) with
the highest electrical performance. Encapsulated devices using
BCP, m-MPyCN and p-PPTNT as EBL were tested after different
storage periods. As shown in Fig. 5, within 120 h the PCE of the
BCP device has decreased to 0. On the other hand, the PCE of the
Fig. 5 PCE degradation of OPVs with p-PPTNT, m-MPyCN and BCP.
This journal is ª The Royal Society of Chemistry 2012
5112 | J. Mater. Chem., 2012, 22, 5107–5113

View Article Online
20 S. W. Liu, C. C. Lee, C. F. Lin, J. C. Huang, C. T. Chen and J. H. Lee,
J. Mater. Chem., 2010, 20, 7800.
21 N. N. Wang, J. S. Yu, Y. Zang, J. Huang and Y. D. Jiang, Sol. Energy
Mater. Sol. Cells, 2010, 94, 263.
22 D. F. O’Brien, P. Burrows, S. R. Forrest, B. E. Koene, D. E. Loy and
M. E. Thompson, Adv. Mater., 1998, 10, 1108.
23 S. J. Liu, F. He, H. Wang, H. Xu, C. Y. Wang, F. Li and Y. G. Ma, J.
Mater. Chem., 2008, 18, 4802.
(863 Program) (Grant No. 2011AA03A110), the Beijing Natural
Science Foundation (Grant No. 2111002), the National Natural
Science Foundation of China (Grant No. 61178061), the
Research Grants Council of Hong Kong (Grant No: CityU
101707).
24 S. Bertho, G. Janssen, T. J. Cleij, B. Conings, W. Moons, A. Gadisa,
J. D’Haen, E. Goovaerts, L. Lutsen, J. Manca and D. Vanderzande,
Sol. Energy Mater. Sol. Cells, 2008, 92, 753.
25 F. M. Hsu, C. H. Chien, Y. J. Hsieh, C. H. Wu, C. F. Shu, S. W. Liu
and C. T. Chen, J. Mater. Chem., 2009, 19, 8002.
26 A. R. G. Smith, J. L. Ruggles, H. Cavaye, P. E. Shaw, T. A. Darwish,
M. James, I. R. Gentle and P. L. Burn, Adv. Funct. Mater., 2011, 21,
2225.
27 N. Li, P. F. Wang, S. L. Lai, W. M. Liu, C. S. Lee, S. T. Lee and
Z. T. Liu, Adv. Mater., 2010, 22, 527.
28 N. Li, S. L. Lai, W. M. Liu, P. F. Wang, J. J. You, C. S. Lee and
Z. T. Liu, J. Mater. Chem., 2011, 21, 12977.
29 C. L. Li, S. J. Shieh, S. C. Lin and R. S. Liu, Org. Lett., 2003, 5,
1131.
30 G. Bringmann, A. J. P. Mortimer, P. A. Keller, M. J. Gresser,
J. Garner and M. Breuning, Angew. Chem., Int. Ed., 2005, 44,
5384.
Notes and references
1 C. W. Tang, Appl. Phys. Lett., 1986, 48, 183.
2 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789.
3 P. Peumans, V. Bulovic and S. R. Forrest, Appl. Phys. Lett., 2000, 76,
2650.
4 H. Gommans, B. Verreet, B. P. Rand, R. Muller, J. Poortmans,
P. Heremans and J. Genoe, Adv. Funct. Mater., 2008, 18, 3686.
5 V. Shrotriya, G. Li, Y. Yao, C. W. Chu and Y. Yang, Appl. Phys.
Lett., 2006, 88, 073508.
6 Y. M. Hwang, S. J. Moon, W. W. So, S. C. Yoon, C. J. Lee and
W. S. Shin, Mol. Cryst. Liq. Cryst., 2008, 491, 324.
7 M. F. Lo, T. W. Ng, S. L. Lai, M. K. Fung, S. T. Lee and C. S. Lee,
Appl. Phys. Lett., 2011, 99, 033302.
8 G. Li, V. Shrotriya, J. S. Huang, Y. Yao, T. Moriarty, K. Emery and
Y. Yang, Nat. Mater., 2005, 4, 864.
31 A. Mazzanti, L. Lunazzi, M. Minzoni and J. E. Anderson, J. Org.
Chem., 2006, 71, 5474.
9 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger
and G. C. Bazan, Nat. Mater., 2007, 6, 497.
32 Y. R. Lai, L. Xing, G. I. Poda and Y. D. Hu, Drug Metab. Dispos.,
2007, 35, 937.
33 S. H. Ye, Y. Q. Liu, C. A. Di, H. X. Xi, W. P. Wu, Y. G. Wen, K. Lu,
C. Y. Du, Y. Liu and G. Yu, Chem. Mater., 2009, 21, 1333.
34 S. Barth, P. Muller, H. Riel, P. F. Seidler, W. Riess, H. Vestweber and
H. Bassler, J. Appl. Phys., 2001, 89, 3711.
10 S. H. Park, A. Roy, S. Beaupre, S. Cho, N. Coates, J. S. Moon,
D. Moses, M. Leclerc, K. Lee and A. J. Heeger, Nat. Photonics,
2009, 3, 297.
11 N. Blouin, A. Michaud and M. Leclerc, Adv. Mater., 2007, 19, 2295.
12 Y. Y. Liang, Y. Wu, D. Q. Feng, S. T. Tsai, H. J. Son, G. Li and
L. P. Yu, J. Am. Chem. Soc., 2009, 131, 56.
35 T. C. Wong, J. Kovac, C. S. Lee, L. S. Hung and S. T. Lee, Chem.
Phys. Lett., 2001, 334, 61.
13 Y. Y. Liang, Z. Xu, J. B. Xia, S. T. Tsai, Y. Wu, G. Li, C. Ray and
L. P. Yu, Adv. Mater., 2010, 22, E135.
36 Z. Y. Xie, T. C. Wong, L. S. Hung and S. T. Lee, Appl. Phys. Lett.,
2002, 80, 1477.
14 T. W. Ng, M. F. Lo, Y. C. Zhou, Z. T. Liu, C. S. Lee, O. Kwon and
S. T. Lee, Appl. Phys. Lett., 2009, 94, 193304.
37 A. Moliton and J. M. Nunzi, Polym. Int., 2006, 55, 583.
38 Yuan Li, Organic Photovoltaics Analysis Platform (OPVAP),
www.OPVAP.inwake.com.
39 B. P. Rand, D. P. Burk and S. R. Forrest, Phys. Rev. B: Condens.
Matter Mater. Phys., 2007, 75, 115327.
15 K. S. Yook, S. O. Jeon, O. Y. Kim and J. Y. Lee, Electrochem. Solid-
State Lett., 2011, 14, B59.
16 M. F. Lo, T. W. Ng, S. L. Lai, F. L. Wong, M. K. Fung, S. T. Lee and
C. S. Lee, Appl. Phys. Lett., 2010, 97, 143304.
17 Q. L. Song, F. Y. Li, H. Yang, H. R. Wu, X. Z. Wang, W. Zhou,
J. M. Zhao, X. M. Ding, C. H. Huang and X. Y. Hou, Chem. Phys.
Lett., 2005, 416, 42.
40 N. Li, B. E. Lassiter, R. R. Lunt, G. Wei and S. R. Forrest, Appl.
Phys. Lett., 2009, 94, 023307.
41 Z. R. Hong, Z. H. Huang and X. T. Zeng, Chem. Phys. Lett., 2006,
425, 62.
42 B. P. Rand, J. Li, J. G. Xue, R. J. Holmes, M. E. Thompson and
S. R. Forrest, Adv. Mater., 2005, 17, 2714.
18 Q. L. Song, M. L. Wang, E. G. Obbard, X. Y. Sun, X. M. Ding,
X. Y. Hou and C. M. Li, Appl. Phys. Lett., 2006, 89, 251118.
19 Q. L. Song, C. M. Li, M. L. Wang, X. Y. Sun and X. Y. Hou, Appl.
Phys. Lett., 2007, 90, 071109.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 5107–5113 | 5113