Tunable electrical memory characteristics of brush copolymers bearing electron donor and acceptor moieties

Article abstract of DOI:10.1039/c3tc30894d

A series of brush copolymers bearing N-phenylcarbazole (PK) and 2-biphenyl-5-(4-ethoxyphenyl)-1,3,4-oxadiazole (BEOXD) moieties in various compositions were studied in detail, in particular their electrical memory characteristics, optical and electrical properties, morphological structures, and interfaces. Nanoscale thin films of the brush copolymers in devices were found to exhibit excellent unipolar electrical memory versatility, which can easily be tuned by tailoring the chemical composition and by changing the film thickness. Moreover, the molecular orbitals and band gap can be tuned by changing the chemical composition. The novel memory characteristics of these copolymers originate primarily from the cooperative roles of the ambipolar PK and BEOXD moieties, which have different charge trapping and stabilization properties. The electrical memory behaviors were found to occur via a favorable hole injection from the electrode and to be governed by trap-limited space-charge limited conduction combined with ohmic conduction and local filament formation. Overall, the brush copolymers are very suitable active materials for the low-cost mass production of high performance, polarity-free digital memory devices that can be operated with very low power consumption, high ON/OFF current ratios, and high stability.

Full text of DOI:10.1039/c3tc30894d

Journal of
Materials Chemistry C
View Article Online
View Journal
PAPER
Tunable electrical memory characteristics of brush
copolymers bearing electron donor and acceptor
moieties†
Cite this: DOI: 10.1039/c3tc30894d
Kyungtae Kim,‡a Yi-Kai Fang,‡b Wonsang Kwon,a Seungmoon Pyo,c
b
a
*
*
Wen-Chang Chen and Moonhor Ree
A series of brush copolymers bearing N-phenylcarbazole (PK) and 2-biphenyl-5-(4-ethoxyphenyl)-1,3,4-
oxadiazole (BEOXD) moieties in various compositions were studied in detail, in particular their electrical
memory characteristics, optical and electrical properties, morphological structures, and interfaces.
Nanoscale thin films of the brush copolymers in devices were found to exhibit excellent unipolar electrical
memory versatility, which can easily be tuned by tailoring the chemical composition and by changing the
film thickness. Moreover, the molecular orbitals and band gap can be tuned by changing the chemical
composition. The novel memory characteristics of these copolymers originate primarily from the
cooperative roles of the ambipolar PK and BEOXD moieties, which have different charge trapping and
stabilization properties. The electrical memory behaviors were found to occur via a favorable hole injection
from the electrode and to be governed by trap-limited space-charge limited conduction combined with
ohmic conduction and local filament formation. Overall, the brush copolymers are very suitable active
materials for the low-cost mass production of high performance, polarity-free digital memory devices that
can be operated with very low power consumption, high ON/OFF current ratios, and high stability.
Received 13th May 2013
Accepted 6th June 2013
DOI: 10.1039/c3tc30894d
www.rsc.org/MaterialsC
In particular, poly(N-vinylcarbazole) has gained attention
because of its application in organic electronic devices as a hole
1 Introduction
transporting layer.19,20 Oxadiazole-containing polymers have
also been extensively investigated as electron-transport mate-
rials.21 Random carbazole and oxadiazole copolymers have
been fabricated, with applications in organic light emitting
diodes22–24 and as electrical memory materials.25 Random
copolymers with a high proportion of carbazole moieties exhibit
static random access memory (SRAM) behavior, whereas
random copolymers with a high proportion of oxadiazole
moieties exhibit no electrical memory behavior at all.25
Controversially, another polymer system containing only oxa-
diazole moieties has been reported to demonstrate electrical
memory characteristics.26 Nevertheless, the electrical memory
characteristics of polymer systems with carbazole and oxadi-
azole moieties have rarely been investigated. Questions remain
about the roles of carbazole and oxadiazole moieties in the
electrical memory characteristics of such polymers.
This paper reports the fabrication and electrical properties
of programmable memory devices based on a series of
brush copolymers containing both carbazole and oxadiazole
moieties: poly(9-(4-vinylphenyl)carbazole-ran-2-(4-vinylbiphenyl)-
5-(4-ethoxyphenyl)-1,3,4-oxadiazole) (PVPKmBEOXDn: m and n
are in mol%) (Fig. 1). The brush copolymers' optical and elec-
trochemical properties were measured. The copolymer lms'
electron densities and interfaces with metal electrodes were
Organic materials can be used to fabricate highly integrated
electronic devices with simplied manufacturing processes
yielding low-cost, exible, and lightweight devices with three-
dimensional stacking that have active device areas approaching
the nanoscale.1–4 Whereas the device fabrication processes and
the resistance to the external environment of small-molecule
organic materials have limitations,5–7 polymer materials exhibit
easy processability, exibility, high mechanical strength, and
good scalability. Moreover, polymer materials can be processed
at low cost and the multi-stack layer structures required for
highly dense memory devices can easily be fabricated from
them. Several polymers have been reported so far as memory
materials.8–18
aDepartment of Chemistry, Division of Advanced Materials Science, Center for
Electro-Photo Behaviors in Advanced Molecular Systems, Pohang Accelerator
Laboratory, Polymer Research Institute, and BK School of Molecular Science, Pohang
University of Science & Technology, Pohang 790-784, Republic of Korea. E-mail:
ree@postech.edu; Fax: +82-54-279-3399; Tel: +82-54-279-2120
bDepartment of Chemical Engineering and Institute of Polymer Science and
Engineering, National Taiwan University, Taipei 106, Taiwan. E-mail: chenwc@ntu.
edu.tw; Fax: +886-2-3366-5237; Tel: +886-2-3366-5236
cDepartment of Chemistry, Konkuk University, Seoul 143-701, Republic of Korea
† Electronic supplementary information (ESI) available: GIXS data, XR data and
analysis results, and I–V data. See DOI: 10.1039/c3tc30894d
‡ These authors equally contributed to this work.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 1 (a) Synthetic route to N-phenylcarbazole (PK) and 2-biphenyl-5-(4-ethoxyphenyl)-1,3,4-oxadiazole (BEOXD) containing brush homopolymers and their
copolymers (PVPKmBEOXDn). (b) A schematic diagram of the memory devices fabricated with nanoscale thin films of the PVPKmBEOXDn copolymers and aluminum (Al)
top and bottom electrodes.
examined by synchrotron X-ray reectivity (XR). Moreover, the (Fig. 2b), the oxidation half-wave potentials (E1/2) were deter-
brush copolymer lms' morphological structures were investi- mined to be 0.68 V for PVPK8BEOXD2, 0.82 V for PVPK5BEOXD5,
gated by synchrotron grazing incidence X-ray scattering (GIXS). and 0.70 V for PVPK2BEOXD8, with respect to the Ag/AgCl
These brush copolymers were found to exhibit excellent uni- standard electrode. In addition, the E1/2 values of PVPK and
polar electrical memory versatility, which can easily be tuned by PBEOXD were measured to be 0.51 V and 0.86 V, respectively.
tailoring the chemical composition and changing the lm E1/2 for the external ferrocene/ferrocenium (Fc/Fc+) system was
thickness. The memory mechanisms were also examined in measured to be 0.43 V vs. Ag/AgCl in acetonitrile. Assuming that
terms of the molecular orbitals, band gaps, and energy barriers the HOMO level for the Fc/Fc+ standard is ꢀ4.80 eV with respect
to the work function of the metal electrode of the copolymers. to the zero vacuum level, the HOMO levels were determined to
be ꢀ5.05 eV for PVPK8BEOXD2, ꢀ5.19 eV for PVPK5BEOXD5,
and ꢀ5.07 eV for PVPK2BEOXD8. Therefore, the LUMO levels
2 Results and discussion
were estimated to be ꢀ1.63 eV for PVPK8BEOXD2, ꢀ1.79 eV for
The optical and electrical properties of the brush copolymers PVPK5BEOXD5, and ꢀ1.70 eV for PVPK2BEOXD8. In compar-
were investigated by UV-vis spectroscopy and cyclic voltamme- ison, the HOMO levels of PVPK and PBEOXD were calculated to
try (CV) analysis. From the measured UV-vis spectra (Fig. 2a), be ꢀ4.88 eV and ꢀ5.23 eV, respectively. From these results, the
the band gaps (i.e., the differences between the highest occu- LUMO levels of PVPK and PBEOXD were determined to be ꢀ1.42
pied molecular orbital (HOMO) levels and the lowest unoccu- eV and ꢀ1.90 eV, respectively.
pied molecular orbital (LUMO) levels) were estimated to be
The morphological structures of nanoscale thin lms of the
3.42 eV for PVPK8BEOXD2, 3.40 eV for PVPK5BEOXD5, and brush copolymers were examined by synchrotron GIXS analysis.
3.37 eV for PVPK2BEOXD8. These band gaps are intermediate The PVPK8BEOXD2 copolymer produces featureless two-
between those (3.46 eV and 3.33 eV) of the PVPK and PBEOXD dimensional (2D) GISAXS and GIWAXS patterns (Fig. 3a and b).
homopolymers, respectively. From the measured CV data Similar GISAXS and GIWAXS patterns were observed for the
J. Mater. Chem. C
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 3 Synchrotron 2D GIXS patterns of 40 nm thick polymer films measured at
25 ꢁC: (a) 2D GISAXS of a PVPK8BEOXD2 film measured with ai ¼ 0.160ꢁ; (b) 2D
GISAXS of a PVPK8BEOXD2 film measured with ai ¼ 0.183ꢁ; (c) azimuthal scattering
profiles extracted from the 2D GIWAXS patterns along the amorphous halo ring.
Fig. 2 (a) UV-vis spectra and (b) CV responses of the PVPK and PBEOXD homo-
polymers and their copolymer films which were coated on quartz substrates and
fabricated with Au electrodes supported by silicon substrates, respectively. The
inset in (a) shows a magnification of the region around the onset of absorbances.
The CV measurements were carried out in aqueous 0.1 M tetrabutylammonium
tetrafluoroborate in acetonitrile using an electrochemical workstation (IM6ex synchrotron XR analysis. Fig. 4a shows the representative XR
impedance analyzer) with a platinum gauze counter electrode and an Ag/AgCl
(3.8 M KCl) reference electrode. A scan rate of 100 mV sꢀ1 was used.
proles obtained for three different samples prepared with the
PVPK8BEOXD2 copolymer. Similar XR proles were observed for
the other copolymers (Fig. S3†). As can be seen in Fig. 4a and
S3,† all the XR proles can be satisfactorily tted with the tting
other polymers (Fig. S1 and S2 in the ESI†) as well as for algorithm proposed by Parratt.27 The electron density proles
homopolymers. In particular, the GIWAXS patterns contain only across the thickness direction, as well as other structural
one broad, weak amorphous halo scattering ring with a parameters, were obtained from these XR analyses. The results
d-spacing of approximately 0.50 nm, which corresponds to the are summarized in Fig. 4b, S4–S6† and Table 1.
mean interdistance of the polymer chains. Note that the
For the PVPK8BEOXD2 lm coated onto the Si substrate, the
amorphous halo scattering ring is slightly anisotropic regard- XR analysis found that the polymer lm has an electron density
less of the copolymer composition. Thus, the azimuthal scat- re of 371.1 nmꢀ3, a thickness of 39.1 nm, and a surface
tering prole was extracted from the 2D GISWAXS pattern of roughness of 0.2 nm (Table 1). The re prole across the sample
each copolymer. The obtained azimuthal scattering proles are thickness was also determined (Fig. 4b). For the PVPK8BEOXD2
displayed in Fig. 3c as a function of the azimuthal angle m. All lm coated onto the Al bottom electrode, the XR analysis found
the proles have a maximum intensity at m ¼ 90ꢁ and a that a very thin aluminum oxide layer forms with a thickness of
minimum intensity at m ¼ 0ꢁ and 180ꢁ. These results indicate 2.3 nm and a surface roughness of 0.4 nm during the polymer
that for all the nanoscale thin lms of the polymers, the coating process (Table 1 and Fig. S4a†). The top coated polymer
polymer chains are oriented preferentially in the lm plane lm has an electron density of 375.2 nmꢀ3 (which is comparable
rather than randomly; however, their degree of in-plane to that of the lm prepared on the Si substrate), a thickness of
orientation is not particularly high.
39.8 nm, and a surface roughness of 0.6 nm. The lm surface is
In devices, the polymers are in physical contact with metal rougher than that (0.2 nm) of the lm prepared on the Si
electrodes. Thus, the interfaces in the devices were examined by substrate but not as rough as the Al bottom electrode (1.1 nm).
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C

View Article Online
Journal of Materials Chemistry C
Paper
of the polymer but lower than that of Al and furthermore much
lower than that of aluminum oxide. The polymer lm (bottom
layer) was determined to have an electron density of 373.7 nmꢀ3
and a roughness of 0.4 nm, which are the same as before the
deposition of the Al top layer. These results show that the
polymer lm surface undergoes no signicant damage during
the thermal deposition of the Al top electrode. Furthermore, no
aluminum oxide layer forms between the polymer lm and the
Al top electrode; the very thin interfacial layer might be due to
the mechanical mixing of the two layers that occurs on the
rough surface as well as the nature of Al layer formation via
thermal deposition.
The XR analyses of the other polymers in equivalent struc-
tures yielded results that are similar to those for the
PVPK8BEOXD2 sample specimens, with minor differences in the
electron densities of the layers (Table 1, Fig. S5 and S6†). Finally,
we combined the XR analysis data for the Al/polymer and
polymer/Al samples to produce the whole electron density
proles for the Al/polymer/Al devices shown in Fig. 4c, S5d
and S6d.†
In view of these optical and electrochemical properties and
morphological and interfacial characteristics, devices with Al/
polymer/Al sandwich structures (Fig. 1b) were fabricated with
various homopolymer and copolymer lm thicknesses and
electrically evaluated.
Fig. 5 shows the current–voltage (I–V) characteristics of the
devices fabricated with 40 nm thick lms of the homopolymers
and their copolymers. Surprisingly, the PVPK homopolymer was
found to immediately reach a low resistance state (i.e., ON-state)
very early in the voltage sweep (Fig. 5a); the PVPK lm always
retains the ON-state in an electric eld. In contrast, the PBEOXD
homopolymer initially exhibits high resistance (i.e., OFF-state)
and no switching behavior during the voltage sweep, i.e. it
exhibits dielectric characteristics (Fig. 5e). The copolymers
also initially enter an OFF-state but exhibit various
switching behaviors depending on their chemical composition
(Fig. 5b–d).
For the PVPK8BEOXD2 copolymer, the current undergoes an
abrupt increase around +2.4 V in the rst positive voltage sweep
with a current compliance of 0.01 A, which corresponds to the
critical voltage Vc,ON to switch on the device (Fig. 5b). This OFF-
to-ON transition can function as a “writing” process. Once the
device reaches its ON-state, it remains there during the subse-
quent positive and negative voltage sweeps with a current
Fig. 4 Synchrotron XR analysis of PVPK8BEOXD2 films (ca. 40 nm thick) in contact
with the silicon (Si) substrate, Al bottom electrode, and Al top electrode (ca. 10 nm
thick): (a) representative XR profiles; (b) electron density profile of the polymer
film coated on the Si substrate; (c) electron density profile of the polymer in a
Si/Al/polymer/Al structure made by the combination of the electron density
profiles (Fig. S4†) determined from the analysis of the XR profiles for the sample
specimens of Si/Al/polymer and Si/Polymer/Al structures in (a). In (a), the symbols
are the measured data and the solid line represents the fit curve assuming a
homogeneous electron density distribution within the film. The inset shows a
magnification of the region around the two critical angles: ac,f and ac,s are the
critical angles of the film and the substrate (Si or Al electrode), respectively.
Thus, the slightly roughened surface might originate from the compliance of 0.01 A or higher and even aer the power is
rough surface of the Al bottom electrode. These results indicate turned off. These I–V characteristics were also observed in
that a very thin aluminum oxide layer forms on the Al bottom negative voltage sweeps (Fig. S7a†); Vc,ON was determined to be
electrode before and/or during the polymer coating process, but ꢀ2.9 V. The ON/OFF current ratio is in the 107 to 1012 range,
aluminum oxide ions and Al atoms do not diffuse from the Al depending on the turn-on compliance current and the reading
bottom electrode layer into the polymer top layer.
voltage; a higher turn-on compliance current and a lower
On the other hand, for the Al top electrode deposited onto reading voltage result in a higher ON/OFF current ratio. These
the polymer lm, the Al electrode layer was determined to have results indicate that the 40 nm thick PVPK8BEOXD2 layer in the
a thickness of 11.2 nm, an electron density of 856.3 nmꢀ3, and a device exhibits permanent nonvolatile memory behavior,
roughness of 1.1 nm (Table 1 and Fig. S4b†). An interfacial layer namely unipolar write-once-read-many-times (WORM) memory
was also found to form with a thickness of 1.3 nm, an electron characteristics.
density of 629.2 nmꢀ3, and a roughness of 0.1 nm. In particular,
this interfacial layer's electron density is much higher than that PVPK5BEOXD5 copolymer. In the positive voltage sweep with a
More interesting memory behaviors were found for the
J. Mater. Chem. C
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
Table 1 Structural parameters and electron density profiles of various bilayer samples prepared from the polymer film, Si substrate, and Al electrode
Bottom layer
Polymer layer
Top layer
Interlayerd
Sample
(top/bottom) da (nm) re (nmꢀ3
)
sc (nm) da (nm) re (nmꢀ3
)
sc (nm) da (nm) reb (nmꢀ3
)
sc (nm) da (nm) reb (nmꢀ3
)
sc (nm)
b
b
PVK8BEOXD2
Polymer/Si
Polymer/Al
Al/polymer
—
11.9
715.0
797.3
0.5
1.1
39.1
39.8
40.8
371.1
375.2
373.7
0.2
0.6
0.4
1.8
2.3
1.3
671.0
862.7
629.2
0.6
0.4
0.1
11.2
9.8
856.3
797.0
820.1
1.1
1.1
1.3
PVPK5BEOXD5
Polymer/Si
Polymer/Al
Al/polymer
—
11.4
718.2
764.0
0.9
0.9
38.1
41.8
40.1
378.2
380.5
378.3
0.2
0.6
0.2
1.9
0.6
1.0
637.0
886.1
695.1
0.5
0.3
0.9
PVPK2BEOXD8
Polymer/Si
Polymer/Al
Al/polymer
—
10.4
730.3
871.5
0.4
1.2
26.6
26.0
26.8
391.8
385.4
391.8
0.2
0.6
0.2
0.6
1.1
1.2
610.0
892.5
575.2
0.7
0.1
0.2
9.1
a
b
c
d
Layer thickness. Electron density of the layer. Roughness of the layer in contact with air, lower or upper layer. Silicon oxide layer for
Si(bottom)/polymer(top); aluminum oxide layer for Al/polymer; polymer–Al mixed layer (which is due to the roughness of the interface) for
polymer/Al systems.
The ON-state was, however, found to return to the OFF-state
during voltage sweeps with a higher current compliance (for
example, 0.1 A) than that (0.01 A) employed for creating the
WORM memory (see the third sweep in Fig. 5c). Similar memory
behaviors were observed in the negative voltage sweeps
(Fig. S7b†). The ON/OFF current ratio was estimated to be in the
105 to 108 range, depending on the reading voltage. Overall, the
40 nm thick PVPK5BEOXD5 lm in the device exhibits unipolar
WORM memory behavior as well as unipolar ash memory
characteristics controlled by adjusting the current compliance.
These memory behaviors are somewhat different from those of
the PVPK8BEOXD2 copolymer.
The PVPK2BEOXD8 copolymer also exhibits interesting
electrical memory behavior. The copolymer switches on around
+2.8 V (¼Vc,ON) during a positive voltage sweep, but is reset to
the OFF-state by a reverse voltage sweep (Fig. 5d). This switching
behavior was found to be repeatable. Furthermore, the ON-state
returns to the OFF-state when the power is turned off. Similar
switching behaviors were observed with Vc,ON ¼ ꢀ2.4 V in the
negative voltage sweeps (Fig. S7c†). The ON/OFF current ratio
was estimated to be in the 106 to 1010 range, depending on the
reading voltage. These results show that the 40 nm thick
PVPK2BEOXD8 lm in the device exhibits unipolar DRAM
behavior, which is quite different from that of the other
copolymers.
Fig. 5 Typical I–V curves of the Al/polymer(40 nm thick)/Al devices, which were
Fig. 6 shows the I–V data for the devices fabricated with $70
measured by voltage sweeps with a compliance current set of 0.01 A: (a) PVPK; (b)
nm thick PVPK, PVPK8BEOXD2, and PVPK2BEOXD8 lms. The
PVPK8BEOXD2; (c) PVPK5BEOXD5; (d) PVPK2BEOXD8; (e) PBEOXD. The electrode
contact area was 0.5 ꢂ 0.5 mm2. The switching-OFF run in (c) was carried out with
100 nm thick PVPK lm still enters an ON-state under an elec-
tric eld, as is the case for the 40 nm thick PVPK lm discussed
above (Fig. 6a). However, the 130 nm thick lm exhibits WORM
memory behavior; Vc,ON is +2.5 V for the PVPK lm (Fig. 6b). In
a compliance current set of 0.1 A.
current compliance of 0.01 A, the copolymer switches on around contrast, the 70 nm and 100 nm thick PVPK8BEOXD2 lms
+3.0 V (¼Vc,ON), then remains in the ON-state during the exhibit DRAM behavior (Fig. 6c and d). Vc,ON increases from +6.0
subsequent voltage sweeps and even aer the power is turned V for the 70 nm thick lm to +9.1 V for the 100 nm thick lm;
off, and thus exhibits WORM memory characteristics (Fig. 5c). furthermore, the thicker lm has a lower ON/OFF current ratio.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 7 Typical I–V curves of the Al/polymer/Al devices, which were measured by
voltage sweeps with a compliance current set of 0.01 A: (a) 70 nm thick
PVPK5BEOXD5 film; (b) 100 nm thick PVPK5BEOXD5 film; (c) 25 nm thick
PVPK5BEOXD5 film; (d) 25 nm thick PBEOXD film; (e) 15 nm thick PBEOXD film.
The electrode contact area was 0.5 ꢂ 0.5 mm2.
Fig. 6 Typical I–V curves of the Al/polymer/Al devices, which were measured by
voltage sweeps with a compliance current set of 0.01 A: (a) 100 nm thick PVPK
film; (b) 130 nm thick PVPK film; (c) 70 nm thick PVPK8BEOXD2 film; (d) 100 nm
thick PVPK8BEOXD2 film; (e) 100 nm thick PVPK2BEOXD8 film; (f) 15 nm thick
PVPK2BEOXD8 film. The electrode contact area was 0.5 ꢂ 0.5 mm2.
These memory characteristics are quite different from the
WORM memory characteristics of the 40 nm thick lm. On
the other hand, a 100 nm thick lm of the PVPK2BEOXD8
copolymer exhibits insulator-like characteristics, with no
memory behavior (Fig. 6e). However, this copolymer does
surprisingly exhibit WORM memory behavior when its lm
thickness is 15 nm (Fig. 6f).
Other brush copolymers and the PBEOXD homopolymer
were also found to exhibit thickness-dependent I–V character-
istics. A 70 nm thick lm of the PVPK5BEOXD5 copolymer
exhibits DRAM behavior (Fig. 7a), but there is no memory
behavior in the 100 nm thick lm, which is most likely an
insulator (Fig. 7b). Interestingly, the 25 nm thick lm exhibits
nonvolatile memory behavior but the ON-state cannot be turned
off once switched on (i.e., WORM memory behavior) (Fig. 7c),
Fig. 8 Composition- and thickness-dependent memory characteristics of the
PVPKmBEOXDn copolymer system.
which is a little bit different from the behavior of the 40 nm copolymer means that a thicker lm layer is required in the
thick lm. In the case of the PBEOXD homopolymer, the 25 nm device to produce electrically bistable switching characteristics.
thick lm exhibits the DRAM behavior (Fig. 7d), whereas, For a copolymer with a given high VPK fraction, thicker lms
interestingly, the 15 nm thick lm exhibits WORM memory exhibit more stable permanent memory behavior (i.e., WORM
behavior (Fig. 7e).
memory) whereas lms of intermediate thickness exhibit tran-
Overall, the PVPKmBEOXDn copolymer system exhibits both siently stable memory characteristics (i.e., DRAM); otherwise,
composition- and thickness-dependent memory characteristics thinner lms are most likely to be in the ON-state under an
(Fig. 8), which are signicantly different from the I–V charac- electric eld, with no memory behavior. On the other hand, a
teristics reported previously for copolymers bearing carbazole higher BEOXD fraction in the copolymer means that a thinner
and oxadiazole derivatives.25,26 A higher VPK fraction in the lm layer is required in the device for electrically bistable
J. Mater. Chem. C
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
switching characteristics to arise. For a copolymer with a high the aid of its aromatic rings. The vinyl backbone can also be
BEOXD fraction, thinner lms exhibit more stable permanent classied as an electron donor because of its hydrocarbon
memory behavior (i.e., WORM memory) whereas lms of nature, although it might not have the capability to stabilize
intermediate thickness exhibit transiently stable memory trapped charges, with only a very weak power to trap holes.
characteristics (i.e., DRAM); otherwise, thicker lms are most However, due to its electron donor characteristics, the vinyl
likely to be in the OFF-state, with no memory behavior. In backbone could positively assist the hole trapping and stabili-
particular, the copolymer with equal proportions of VPK and zation of the PK moieties. On the other hand, for the PBEOXD
BEOXD moieties exhibits ash memory behavior in addition to homopolymer and copolymers, the oxadiazole moiety in the
the WORM memory characteristics for lms of a certain thick- side group is known to be an electron transporting group and
ness (40 nm thick).
can thus be classied as an electron acceptor, which can trap
The stabilities of the devices with 40 nm thick copolymer and stabilize electrons. The biphenyl and phenyl linkers in the
lm layers were tested with a reading voltage of +0.5 V at room side group are known to be electron decient, however, their
temperature in air ambient conditions. Their OFF-states were electron affinities are not high enough. Thus, they can be
found to be well retained without any degradation for a test classied as weak electron acceptors. However, they are
period of 10 h (Fig. S8†). Once the devices based on the aromatic and indeed can positively assist the electron trapping
PVPK8BEOXD2 and PVPK5BEOXD5 copolymers were switched and stabilization of the oxadiazole moiety. In contrast, the
on by applying a voltage of +3.5 V, the ON-states were also found electron donor characteristics of the vinyl backbone and the
to be well retained without any degradation over the test period ethoxy end of the side group could negatively affect the oxadi-
(Fig. S8a and b†). Moreover, when the ON state of the azole moiety's electron trapping and stabilization. Thus, the PK
PVPK5EBOXD5 device was switched back to the OFF-state by and BEOXD side groups can effectively act as charge-trapping
applying a voltage of 0.3 V with a compliance current of 0.1 A, sites. Therefore, the copolymers have a dual functionality and
the OFF-state was again conrmed to persist without any can trap holes and electrons and stabilize the trapped charges
degradation over the whole test period. In the case of the in a certain level, depending on their chemical composition.
PVPK2BEOXD8 device, the ON-state was well retained with a Under an electrical eld (i.e., an applied voltage), charges are
continuous voltage bias of +3.5 V (Fig. S8c†). Overall, all of the trapped in trapping sites. The charge-trapped sites are localized
devices with 40 nm thick copolymer lms exhibited excellent throughout the entire copolymer lm layers and are then
reliability even in ambient conditions.
stabilized in a certain level, and can act as stepping stones for
Representative I–V data for the devices were further analyzed charge ow, namely current ow. When the charge-trapped
in detail with various conduction models28–31 in order to inves- sites are localized with an interdistance of 1.0 nm or less as
tigate their electrical switching characteristics. Logarithmic stepping stones in the copolymer lm layers, current ow can
plots of the I–V data for the devices' OFF-states can be divided take place favorably between the electrodes in the device via the
into two regimes (Fig. S9†): below and above 1.0 V. For the lm layer. Thus, under an electric eld, local laments are
PVPK8BEOXD2 and PVPK2BEOXD8 copolymers, the currents formed from such charge-trapped sites that enable current ow
below 1.0 V can be satisfactorily tted with an ohmic model through the hopping process and result in an ON-state.
while those above 1.0 V are well tted with the trap-limited
For the devices with Al electrodes (F ¼ ꢀ4.28 eV), the
space-charge limited conduction (SCLC) model. For energy barriers to hole injection from the electrode to the active
PVPK5BEOXD5, the trap-limited SCLC model was found to polymer lm layer (HOMO level) were estimated to be 0.60 eV
satisfactorily t the I–V data in both regimes. These results for PVPK, 0.95 eV for PBEOXD, 0.77 eV for PVPK8BEOXD2,
indicate that all the devices in the OFF-state are governed by the 0.91 eV for PVPK5BEOXD5, and 0.79 eV for PVPK2BEOXD8. The
trap-limited SCLC mechanism combined with ohmic conduc- energy barriers to electron injection from the electrode to the
tion. On the other hand, for the devices in the ON-state, the I–V polymer lm layer (LUMO level) were estimated to be 2.86 eV for
data were found to be satisfactorily tted with ohmic conduc- PVPK, 2.38 eV for PBEOXD, 2.65 eV for PVPK8BEOXD2, 2.49 eV
tion (data not shown). Furthermore, their current levels in the for PVPK5BEOXD5, and 2.58 eV for PVPK2BEOXD8. The energy
ON-state were found to be independent of the device cell size, barrier to hole injection is lower than that for electron injection
which is indicative of the formation of heterogeneously local- for all the homopolymers and copolymers. Therefore, these
ized laments.
results suggest that the conduction processes in the devices are
The above results suggest that the memory behaviors of the dominated by hole injection.
brush copolymers are governed by trap-limited SCLC combined
The ordering of the energy barriers to hole injection in the
with ohmic conduction and local lament formation. The lms is PVPK < PVPK8BEOXD2 < PVPK2BEOXD8 < PVPK5BEOXD5
charge trapping sites required by these electrical switching < PBEOXD, as discussed above. One might, therefore, expect that
mechanisms might arise from the chemical components of the the ordering of the Vc,ON values would be PVPK < PVPK8BEOXD2 <
copolymers. For the PVPK homopolymer and the copolymers, PVPK2BEOXD8 < PVPK5BEOXD5 < PBEOXD. As discussed above,
the N-phenylcarbazole (PK) moiety in the side group mimics the copolymers' electrical switching behaviors were, however,
both conventional carbazole and triphenylamine, which are found to further depend on the VPK and BEOXD fractions (i.e.,
well known as hole transporting groups, and can thus be clas- chemical composition) as well as on the lm layer thickness.
sied as an electron donor. The PK moiety can trap holes on the Thus, the prediction of the Vc,ON values cannot be based only on
electron-rich nitrogen atom and stabilize the trapped holes with the energy barriers. Furthermore, it is very difficult to rationalize
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C

View Article Online
Journal of Materials Chemistry C
Paper
the relationships of the Vc,ON values with the energy barriers, the BEOXD-enriched copolymer, which can provide charge-
chemical compositions, lm thicknesses, and other associated trapping sites with lower capacity and stabilization power.
factors.
Thus, only thinner lms of the BEOXD-enriched copolymers
In view of the BEOXD unit's electron acceptor characteristics, exhibit electrical memory behaviors.
one might expect that the PBEOXD homopolymer and the
copolymers with a large BEOXD fraction can more favorably
trap electrons, leading to electron injection based conduction
3 Conclusions
processes in their devices. Considering the higher energy A series of brush copolymers bearing N-phenylcarbazole (PK)
barrier to electron injection from the electrode to the polymer and 2-biphenyl-5-(4-ethoxyphenyl)-1,3,4-oxadiazole (BEOXD)
lm layer discussed above, such conduction processes should moieties (PVPKmBEOXDn copolymers) were investigated in
require higher Vc,ON values. However, Vc,ON was found not to be detail, in particular their electrical memory characteristics,
simply proportional to the BEOXD fractions in the polymers. optical and electrical properties, morphological structures, and
This result suggests that such electron injection based interfaces. In nanoscale thin lms, the copolymers are amor-
conduction processes might not take place in the devices based phous but have a tendency to orient in the lm plane. In the
on the copolymers; if such a conduction process takes place at a devices, the brush copolymer lms were found to be in contact
certain electric eld, it may occur in a very minor portion rather with the Al top and bottom electrodes without diffusion of
than a major portion. Namely, hole injection based conduction the metal atoms and ions. The HOMO (ꢀ4.88 to ꢀ5.23 eV) and
processes were conrmed to be more favorable for the devices LUMO (ꢀ1.42 to ꢀ1.90 eV) levels and the band gaps (3.33 to
based on the BEOXD-rich copolymers.
3.46 eV) of the copolymers can be tuned by tailoring their
In fact, the oxadiazole moiety's electron affinity originates compositions.
from the electron deciency generated by its heterocyclic
The excellent electrical memory versatility of the nanoscale
conjugation ring and further from the ability to stabilize trap- copolymer layers in the devices with the Al top and bottom elec-
ped electrons by delocalization into the heterocyclic conjuga- trodes can be tuned by tailoring the chemical composition and
tion ring. In the case of the BEOXD unit, electron deciency and the lm layer thickness: unipolar DRAM, ash memory, and
stabilization power are enhanced somewhat by the electron WORM memory characteristics were all demonstrated, including
decient characteristics of the additional phenyl and biphenyl conductor-like and insulator-like behaviors. These novel memory
linkers, as discussed above. However, the oxadiazole moiety characteristics originate from the cooperative roles of the ambi-
contains lone pairs of electrons on the oxygen and nitrogen polar PK and BEOXD moieties, which have signicantly different
atom components. In addition, the ethoxy end group is an charge trapping and stabilization powers; despite their ambipo-
electron donor with a further two lone pair electrons. Taking larities, the PK moiety has a higher affinity to holes whereas the
these facts into account, the BEOXD unit might also have an BEOXD moiety has a higher affinity to electrons. The electrical
affinity for holes and be able to stabilize them. Therefore, the memory behaviors were found to occur via a favorable hole
BEOXD unit can act as an electron trapping site as well as a hole injection from the electrode and further to be governed by trap-
trapping site, depending on circumstances. Of course, its limited space-charge limited conduction (SCLC) combined with
capacity for hole trapping and stabilization is weaker than its ohmic conduction and local lament formation.
capacity for electron trapping and stabilization and, further-
Overall, this study has demonstrated that PVPKmBEOXDn
more, weaker than the PK unit's capacity for hole trapping and copolymers are suitable active materials for the low-cost mass
stabilization. Nevertheless, thinner lms of the PBEOXD production of high performance, polarity-free digital memory
homopolymer and the BEOXD-enriched copolymers exhibit devices that can be operated with very low power consumption,
excellent DRAM and WORM memory behaviors in devices. high ON/OFF current ratios, and high stability. Moreover, the
These results conrm that the BEOXD moieties in the homo- memory modes of these copolymers can easily be tuned by
polymer and copolymers are ambipolar and thus can trap holes, changing the composition and lm thickness.
which results in current ows under the appropriate conditions.
The various observed I–V characteristics might indeed be
primarily due to differences between the charge trapping and
4 Experimental
stabilization powers of the PK and BEOXD units in the 2,2,6,6-Tetramethylpiperidine-1-oxyl (TEMPO) and benzoyl
PVPKmBEOXDn copolymers. Taking these factors into account, peroxide (BPO) were obtained from Acros (Geel, Belgium).
the observed lm thickness dependent memory behaviors Pd(PPh3)4 and sodium carbonate were purchased from Strem
suggest that the efficiency of the local lament based conduc- Chemicals (Newburyport, MA, USA) and Showa (Tokyo, Japan),
tion process is a function of the travel path distance (i.e., the respectively. Anhydrous toluene and dimethylformamide (DMF)
conduction path length) in the copolymer layer between the top were purchased from TEDIA (Faireld, CT, USA). Tetrabutyl-
and bottom electrodes. The local lament based conduction ammonium perchlorate (TBAP) was obtained from TCI (Tokyo,
path length is longer in the PK-enriched copolymers and so can Japan). Other chemicals were purchased from Aldrich (Mil-
provide charge-trapping sites with higher capacity and stabili- waukee, WI, USA). All the chemicals were used as received
zation power. In other words, thicker lms of the PK-enriched unless otherwise noted. TBAP was puried by recrystallization
copolymer can exhibit electrical memory behaviors. In contrast, twice from ethyl acetate and then dried under vacuum prior
the local lament based conduction path length is shorter in to use.
J. Mater. Chem. C
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
2-(4-Bromophenyl)-5-(4-ethoxyphenyl)-1,3,4-oxadiazole was TGA (Waltham, MA, USA). All the polymers were found to be
synthesized (Fig. 1a), according to the method reported in the soluble in common solvents such as chloroform, THF, toluene,
literature.31 The obtained compound (10.36 g, 30.0 mmol) and chlorobenzene, dichlorobenzene, and dimethylacetamide.
4-vinylphenylboronic acid (5.33 g, 36.0 mmol) were reacted
Devices were fabricated as follows. For each polymer, homo-
together with the aid of Pd(PPh3)4 (1.23 g, 1.05 mmol) and geneous solutions with two different concentrations (0.7 and
Na2CO3 (3.81 g, 30.0 mmol) in a mixture of toluene (210 mL), 1.0 wt%) were prepared in chloroform, followed by ltrations
ꢁ
ethanol (6 mL) and deionized distilled water (30 mL) at 90 C through PTFE-membrane micro lters with a pore size of 0.2 mm.
under a nitrogen atmosphere for 16 h. Aer the completion of Aluminum (Al) bottom electrodes with a thickness of 300 nm
the reaction, the mixture was cooled down to room temperature were deposited on precleaned silicon (Si) substrates by electron-
and ltered. The solvent was removed by evaporation under a beam evaporation. Each polymer solution was spin-coated onto
reduced pressure to obtain a crude product. The product was the Al bottom electrodes at 2000 rpm for 40 s, followed by drying
puried by column chromatography on a silica gel column at 80 ꢁC under vacuum for 1 day. The resulting lms' thicknesses
using dichloromethane as an eluent, washed with methanol were determined using a spectroscopic ellipsometer (model
(twice), and dried under vacuum at 70 ꢁC to afford 7.44 g (67%) M2000, Woollam, Lincoln, NE, USA). Then, Al top electrodes
of the target monomer, 2-(4-vinylbiphenyl)-5-(4-ethoxyphenyl)- were thermally evaporated onto the polymer lms at a pressure
1,3,4-oxadiazole (BEOXD), as a green powder. The obtained of ca. 10ꢀ6 torr through a shadow mask. The resulting top
BEOXD monomer was characterized by proton nuclear electrodes had a thickness of 300 nm and the area varied from
magnetic resonance (1H NMR) spectroscopy using a Bruker 2.0 mm ꢂ 2.0 mm to 0.5 mm ꢂ 0.5 mm.
NMR spectrometer (model Avance 300 MHz FT-NMR,
For XR experiments, three kinds of sample specimens were
1
¨
Fallanden, Switzerland). H NMR (300 MHz, CDCl3, d (ppm)): prepared for each brush copolymer: (i) polymer lms (40 nm
1.46 (t, 3H), 4.13 (q, 2H), 5.31 (d, 1H, ArCH]CH cis), 5.83 (d, 1H, thick) on the Si substrates with a native oxide layer, (ii) polymer
ArCH]CH trans), 6.79 (dd, 1H, ArCH]CH), 7.01 (d, 2H), 7.51 lms (30 nm thick) on 10 nm thick Al-deposited Si substrates
(d, 2H), 7.62 (d, 2H), 7.74 (d, 2H), 8.06 (d, 2H), 8.17 (d, 2H).
(here, the Al layer was deposited by electron-beam evaporation),
On the other hand, 9-(4-vinylphenyl)carbazole (VPK) was and (iii) thermally evaporated Al lms (10 nm thick) on 30 nm
synthesized (Fig. 1b), according to the method in the literature thick polymer lms coated onto the Si substrates. Synchrotron
report.32
XR measurements were performed at the 3D and 8D beamlines
From the above BEOXD and VPK monomers, a series of of the Pohang Accelerator Laboratory (PAL).33 A Si(111) double-
brush copolymers were synthesized in various compositions via crystal monochromator was used to select a wavelength l of
nitroxide-mediated free radical polymerization (NMFRP) 0.1541 nm within an energy resolution of Dl/l ¼ 5 ꢂ 10ꢀ4, and a
(Fig. 1a), and further puried by Soxhlet extraction in acetone. Sagittal bender for the second crystal was used to focus the X-ray
The obtained brush copolymers were nally dried under beam in the horizontal direction. The primary beam was
vacuum at 60 ꢁC for 1 day: PVPK (Mw ¼ 12 000 and PDI ¼ 1.30; dened by four slits before the sample, and another two slits
ꢁ
ꢁ
Tg ¼ 210 C and Td ¼ 405 C), PVPK8BEOXD2 (77.5 mol% VPK were used as receiving slits aer the sample. The beam was
and 22.5 mol% BEOXD; Mw ¼ 9900 and PDI ¼ 1.23; Tg ¼ 202 ꢁC collimated at the sample position to 2 mm (horizontal) by 0.1
and Td ¼ 368 ꢁC), PVPK5BEOXD5 (49.6 mol% VPK and 50.4 mol mm (vertical). The measured reected intensity was normalized
% BEOXD; Mw ¼ 10 000 and PDI ¼ 1.31; Tg ¼ 197 ꢁC and Td ¼ to the intensity of the primary beam, which was monitored with
377 ꢁC), PVPK2BEOXD8 (14.0 mol% VPK and 86.0 mol% BEOXD; an ionization chamber. Specular reection was measured in the
Mw ¼ 10 800 and PDI ¼ 1.27; Tg ¼ 193 ꢁC and Td ¼ 389 ꢁC), and q–2q scanning mode. The reectivity R, i.e. the ratio of the
ꢁ
PBEOXD (Mw ¼ 14 200 and PDI ¼ 1.38; Tg ¼ 176 C and Td ¼ reected beam intensity to the primary beam intensity, was
measured down to just above 10ꢀ8. To obtain accurate deter-
ꢁ
406 C) (Fig. 1).
All the obtained polymers and their compositions were minations of the critical angles of the silicon substrate and of
characterized in CDCl3 by 1H NMR spectroscopy (300 MHz). the thin lm, 2q was scanned at small increments of 0.01ꢁ at
Weight-average weights (Mw ) and polydispersity indices (PDI) angles smaller than 1.0ꢁ. At higher angles, the step width was
were determined by gel permeation chromatography (GPC) increased to 0.02–0.1ꢁ. The obtained data underwent data
using a Lab Alliance RI2000 instrument (Shah Alam, Malaysia; binning, geometrical correction, and the background subtrac-
one column, MIXED-D from Polymer Laboratories, Amherst, tion procedure described in the literature.33,34
MA, USA) connected with a refractive index detector (Schambeck
GIXS measurement was performed at the PAL 3C beam-
SFD Gmbh, Bad Honnef, Germany) which was calibrated with line.35,36 Measurements were performed at a sample-to-detector
polystyrene standards. All GPC analyses were performed using distance (SDD) of 125 mm for grazing incidence wide-angle
tetrahydrofuran (THF) eluent at a ow rate of 1 mL minꢀ1 at X-ray scattering (GIWAXS) and 2220 mm for grazing incidence
40 ꢁC. Glass transition and degradation temperatures (Tg and Td) small-angle X-ray scattering (GISAXS). Scattering data were
were measured by differential scanning calorimetry (DSC) and typically collected for 30 s using an X-ray radiation source of
thermogravimetry (TGA), respectively. DSC measurements were l ¼ 0.138 nm with a two-dimensional charge-coupled detector
performed under a nitrogen atmosphere at a heating rate of (Roper Scientic, Trenton, NJ, USA). The incidence angle ai of
20 ꢁC minꢀ1 using a TA instrument DSC-Q100 (New Castle, DE, the X-ray beam was set at 0.183ꢁ for wide angle scattering and
USA) while TGA analysis was carried out at a heating rate of 20 ꢁC 0.160ꢁ for small-angle scattering, which is between the critical
minꢀ1 under a nitrogen atmosphere using a PerkinElmer Pyris 1 angle of the polymer lms and the silicon substrate
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C

View Article Online
Journal of Materials Chemistry C
Paper
(ac,f and ac,s). Scattering angles were corrected according to the
positions of the X-ray beams reected from the silicon substrate
with respect to a pre-calibrated silver behenate (TCI, Tokyo,
Japan) powder. Aluminum foil pieces were applied as a semi-
transparent beam stop because the intensity of the specular
reection from the substrate was much stronger than the
scattering intensity of the polymer lms near the critical angle.
I–V characteristics of the devices were tested under ambient
air conditions using a Keithley 4200 semiconductor analyzer
(Cleveland, OH, USA). I–V curves were recorded by performing
forward and reverse voltage scans between ꢀ4.0 V and +4.0 V at
a scan rate of 500 mV sꢀ1. Optical properties were measured
7 D. Kolosov, D. S. English, V. Bulovic, P. F. Barbara,
S. R. Forrest and M. E. Thompson, J. Appl. Phys., 2001, 90,
3242.
¨
8 D. Ma, M. Aguiar, J. A. Freire and I. A. Hummelgen, Adv.
Mater., 2000, 12, 1063.
9 S. Baek, D. Lee, J. Kim, S.-H. Hong, O. Kim and M. Ree, Adv.
Funct. Mater., 2007, 17, 2637; D. Lee, S. Baek, M. Ree and
O. Kim, Jpn. J. Appl. Phys., 2008, 47, 5665; D. Lee, S. Baek,
M. Ree and O. Kim, Electron. Lett., 2008, 44, 596; Y.-S. Lai,
C.-H. Tu, D.-L. Kwong and J. S. Chen, Appl. Phys. Lett.,
2005, 87, 122101; Y.-S. Lai, C.-H. Tu, D.-L. Kwong and
J. S. Chen, IEEE Electron Device Lett., 2006, 27, 451.
using a Scinco ultraviolet-visible (UV-vis) spectrometer (model 10 T. J. Lee, S. Park, S. G. Hahm, D. M. Kim, K. Kim, J. Kim,
S-3100, Seoul, Korea). Cyclic voltammetry (CV) measurements
were carried out in a 0.1 M solution of tetrabutylammonium
W. Kwon, Y. Kim, T. Chang and M. Ree, J. Phys. Chem. C,
2009, 113, 3855.
tetrauoroborate in acetonitrile by using an electrochemical 11 H. S. Majumdar, A. Bandyopadhyay, A. Bolognesi and
workstation (IM6ex impedance analyzer, Zahner, Kronach,
Germany) with a platinum gauze counter electrode and an
Ag/AgCl (3.8 M KCl) reference electrode, and the polymers were
A. J. Pal, J. Appl. Phys., 2002, 91, 2433; E. Y. H. Teo,
Q. D. Ling, Y. Song, Y. P. Tan, W. Wang, E. T. Kang,
D. S. H. Chan and C. Zhu, Org. Electron., 2006, 7, 173.
coated onto the Au bottom electrode, which was deposited on a 12 S. L. Lim, Q. Ling, E. Y. H. Teo, C. X. Zhu, D. S. H. Chan,
silicon wafer. A scan rate of 100 mV sꢀ1 was used.
E.-T. Kang and K. G. Neoh, Chem. Mater., 2007, 19, 5148.
13 Q. Ling, Y. Song, S. J. Ding, C. Zhu, D. S. H. Chan,
D.-L. Kwong, E.-T. Kang and K. G. Neoh, Adv. Mater., 2005,
17, 455; L.-H. Xie, Q.-D. Ling, X.-Y. Hou and W. Huang,
J. Am. Chem. Soc., 2008, 130, 2120; M. Karakawa,
M. Chikamatsu, Y. Yoshida, R. Azumi, K. Yase and
C. Nakamoto, Macromol. Rapid Commun., 2007, 28, 1479;
S. K. Majee, H. S. Majumdara, A. Bolognesi and A. J. Pal,
Synth. Met., 2006, 156, 828; T.-W. Kim, K. Lee, S.-H. Oh,
G. Wang, D.-Y. Kim, G.-Y. Jung and T. Lee, Nanotechnology,
2008, 19, 405201.
Acknowledgements
This study was supported by the National Research Foundation
(NRF) of Korea (Doyak Program 2011-0028678 and the Center
for Electro-Photo Behaviors in Advanced Molecular Systems
(2010-0001784)) and the Ministry of Science, ICT & Future
Planning (MSIP) (World Class University Program (R31-2008-
000-10059-0) and BK21 Program). Synchrotron GIXS and XR
measurements were supported by MSIP, POSCO and POSTECH
Foundation. The work at National Taiwan University was sup- 14 Q.-D. Ling, D.-J. Liaw, E. Y.-H. Teo, C. Zhu, D. S.-H. Chan,
ported by the National Science Council under the contract
number, NSC 98-2221-E-002-006-MY3.
E.-T. Kang and K.-G. Neoh, Polymer, 2007, 48, 5182.
15 S. G. Hahm, S. Choi, S.-H. Hong, T. J. Lee, S. Park, D. M. Kim,
W.-S. Kwon, K. Kim, O. Kim and M. Ree, Adv. Funct. Mater.,
2008, 18, 3276.
16 D. M. Kim, S. Park, T. J. Lee, S. G. Hahm, K. Kim, J. C. Kim,
W. Kwon and M. Ree, Langmuir, 2009, 25, 11713; K. Kim,
S. Park, S. G. Hahm, T. J. Lee, D. M. Kim, J. C. Kim,
W. Kwon, Y.-G. Ko and M. Ree, J. Phys. Chem. B, 2009, 113,
9143; T. J. Lee, C.-W. Chang, S. G. Hahm, K. Kim, S. Park,
D. M. Kim, J. Kim, W.-S. Kwon, G.-S. Liou and M. Ree,
Nanotechnology, 2009, 20, 135204.
References
1 Q. D. Ling, D.-J. Liaw, C. Zhu, D. S.-H. Chan, E.-T. Kang and
K.-G. Neoh, Prog. Polym. Sci., 2008, 33, 917.
2 K. S. Kwok and J. C. Ellenbogen, Mater. Today, 2002, 5, 28;
J. C. Scott, Science, 2004, 304, 62.
3 Z. Bao, A. J. Lovinger and A. Dodabalapur, Appl. Phys. Lett.,
1996, 69, 3066; S. Choi, S. Hong, S. H. Cho, S. Park,
S. M. Park, O. Kim and M. Ree, Adv. Mater., 2008, 20, 17 S. Park, K. Kim, D. M. Kim, W. Kwon, J. Choi and M. Ree, ACS
1766.
Appl. Mater. Interfaces, 2011, 3, 765; S. Park, K. Kim,
J. C. Kim, W. Kwon, D. M. Kim and M. Ree, Polymer, 2011,
52, 2170; S. G. Hahm, T. J. Lee, D. M. Kim, W. Kwon,
Y.-G. Ko, T. Michinobu and M. Ree, J. Phys. Chem. C, 2011,
115, 21954.
4 Y. Yang, J. Ouyang, L. Ma, R. J. Tseng and C. Chu, Adv. Funct.
Mater., 2006, 16, 1001.
5 A. R. Elsharkawi and C. Kao, J. Phys. Chem. Solids, 1977, 38,
95; C. Chiang, Solid State Commun., 1981, 39, 111.
6 H. J. Gao, K. Sohlberg, Z. Q. Xue, H. Y. Chen, S. M. Hou, 18 S. G. Hahm, N.-G. Kang, W. Kwon, K. Kim, Y.-G. Ko, S. Ahn,
L. P. Ma, X. W. Fang, S. J. Pang and S. J. Pennycook, Phys.
Rev. Lett., 2000, 84, 1780; Z. J. Donhauser, B. A. Mantooth,
K. F. Kelly, L. A. Bumm, J. D. Monnell, J. J. Stapleton,
D. W. Price Jr, A. M. Rawlett, D. L. Allara, J. M. Tour and
P. S. Weiss, Science, 2001, 292, 2303; J. Chen and D. Ma,
Appl. Phys. Lett., 2005, 87, 23505; C.-H. Tu, Y.-S. Lai and
D.-L. Kwong, Appl. Phys. Lett., 2006, 89, 062105.
B.-G. Kang, T. Chang, J.-S. Lee and M. Ree, Adv. Mater., 2012,
24, 1062; D. M. Kim, Y.-G. Ko, J. K. Choi, K. Kim, W. Kwon,
J. Jung, T.-H. Yoon and M. Ree, Polymer, 2012, 53, 1703;
T. J. Lee, Y.-G. Ko, H.-J. Yen, K. Kim, D. M. Kim, W. Kwon,
S. G. Hahm, G.-S. Liou and M. Ree, Polym. Chem., 2012, 3,
1276; Y.-G. Ko, W. Kwon, D. M. Kim, Y.-S. Gal and M. Ree,
Polym. Chem., 2012, 3, 2028; Y.-G. Ko, W. Kwon, H.-J. Yen,
J. Mater. Chem. C
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
C.-W. Chang, D. M. Kim, K. Kim, S. G. Hahm, T. J. Lee,G.-S. Liou 30 K. L. Jessen, J. Vac. Sci. Technol., B: Microelectron. Nanometer
and M. Ree, Macromolecules, 2012, 45, 3749; K. Kim, H.-J. Yen,
Y.-G. Ko, C.-W. Chang, D. M. Kim, W. Kwon, S. G. Hahm,
G.-S. Liou and M. Ree, Polymer, 2012, 53, 4135.
Struct.–Process., Meas., Phenom., 2003, 21, 1528; J. Frenkel,
Phys. Rev., 1938, 54, 647; C. Laurent, E. Kay and N. Souag,
J. Appl. Phys., 1988, 64, 336.
19 W. W. Chung and C. L. Hong, Macromolecules, 2006, 39, 31 G. Liu, Q.-D. Ling, E. Y. H. Teo, C.-X. Zhu, D. S.-H. Chan,
7232.
K.-G. Neoh and E.-T. Kang, ACS Nano, 2009, 3, 1929; C. Ye,
Q. Ping, M. Li, J. Luo, Z. Tang, J. Pei, J. Chen, Z. Shuai,
L. Jiang and Y. Song, J. Am. Chem. Soc., 2012, 134, 20053.
32 L. Deng, P. T. Furuta, S. Garon, J. Li, D. Kavulak,
20 (a) G. Wang, C. Yuan, H. Wu and Y. Wei, J. Appl. Phys., 1995,
78, 2679; (b) S. Jerzy, B. Dariusz, W. Marek and B. Adam,
Chem. Mater., 2002, 14, 89.
´
21 A. Kra, A. C. Grimsdale and A. B. Holmes, Angew. Chem.,
Int. Ed., 1998, 37, 402; D. Y. Kim, H. N. Cho and C. Y. Kim,
M. E. Thompson and J. M. J. Frechet, Chem. Mater., 2006,
18, 386.
Prog. Polym. Sci., 2000, 25, 1089; L. Akcelrud, Prog. Polym. 33 J. Bolze, M. Ree, H. S. Youn, S.-H. Chu and K. Char,
Sci., 2003, 28, 875.
22 P. Wang, C. Chai, Q. Yang, F. Wang, Z. Shen, H. Guo,
Langmuir, 2001, 17, 6683; Y. Hwang, K. Heo, C. H. Chang,
M. K. Joo and M. Ree, Thin Solid Films, 2006, 510, 159.
X. Chen, X. Fan, D. Zou and Q. Zhou, J. Polym. Sci., Part A: 34 F. Salah, B. Harzallah and A. van der Lee, J. Appl. Crystallogr.,
Polym. Chem., 2008, 46, 5452. 2007, 40, 813.
23 D. D. Evanoff Jr, J. B. Carroll, R. D. Roeder, Z. J. Hunt, 35 J. Yoon, K.-W. Kim, J. Kim, K. Heo, K. S. Jin, S. Jin, T. J. Shin,
J. R. Lawrence and S. H. Foulger, J. Polym. Sci., Part A:
Polym. Chem., 2008, 46, 7882.
24 K. Tsuchiya, K. Sakaguchi, A. Kawakami, H. Taka, H. Kita,
T. Shimomura and K. Ogino, Synth. Met., 2010, 160, 1679.
25 Y.-K. Fang, C.-L. Liu and W.-C. Chen, J. Mater. Chem., 2011,
21, 4778.
26 W. Kwon, B. Ahn, D. M. Kim, Y.-G. Ko, S. G. Hahm, Y. Kim,
H. Kim and M. Ree, J. Phys. Chem. C, 2011, 115, 19355.
27 L. G. Parratt, Phys. Rev., 1954, 95, 359.
28 P. Mark and W. Helfrich, J. Appl. Phys., 1962, 33, 205.
29 A. J. Campell, D. D. C. Bradley and D. G. Lidzey, J. Appl. Phys.,
1997, 82, 6326.
B. Lee, Y. Rho, B. Ahn and M. Ree, Macromol. Res., 2008, 16,
575; B. Lee, Y.-H. Park, Y. Hwang, W. Oh, J. Yoon and M. Ree,
Nat. Mater., 2005, 4, 147; B. Lee, W. Oh, Y. Hwang,
Y.-H. Park, J. Yoon, K. S. Jin, K. Heo, J. Kim, K.-W. Kim
and M. Ree, Adv. Mater., 2005, 17, 696.
36 J. Yoon, K. S. Jin, H. C. Kim, G. Kim, K. Heo, S. Jin, J. Kim,
K.-W. Kim and M. Ree, J. Appl. Crystallogr., 2007, 40, 476;
J. Yoon, S. W. Lee, S. Choi, K. Heo, K. S. Jin, S. Jin, G. Kim,
J. Kim, K.-W. Kim, H. Kim and M. Ree, J. Phys. Chem. B,
2008, 112, 5338; G. Kim, S. Park, J. Jung, K. Heo, J. Yoon,
H. Kim, I. J. Kim, J. R. Kim, J. I. Lee and M. Ree, Adv.
Funct. Mater., 2009, 19, 1631.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C