Photoreactions and photoinduced molecular orientations of films of a photoreactive polyimide and their alignment of liquid crystals

Article abstract of DOI:10.1021/ma034445u

A soluble photoreactive polyimide (PSPI) with side-group cinnamate (CI) chromophores that forms good quality films through conventional solution spin-casting and drying processes was successfully synthesized with a reasonably high molecular weight: poly(3,3′-bis(cinnamoyloxy)-4,4′-biphenylene hexafluoroisopropylidenediphthalimide) (6F-HAB-CI). This 6F-HAB-CI PSPI is thermally stable up to 340°C, its glass transition temperature (T_g) is 181°C, and it was determined by prism coupling to be positively birefringent. The photochemical reactions of the PSPI in solution and in films, as well as the molecular orientations of the PSPI that are induced by exposure of its films to linearly polarized ultraviolet light (LPUVL), were investigated in detail by nuclear magnetic resonance, ultraviolet-visible, infrared, and Raman spectroscopies and by dissolution testing and optical retardation measurements. It was confirmed that the CI chromophores of the PSPI undergo both photoisomerization and photodimerization. LPUVL exposure was found to induce anisotropic orientations of the polymer main chains and of the CI side groups in the films. Moreover, the LPUVL-irradiated films homogeneously aligned nematic liquid crystal (LC) molecules along a direction at an angle of 107° with respect to the polarization of the LPUVL, which coincides with the orientation direction of the PSPI polymer chains. This result shows that the LC alignment process is principally governed in irradiated PSPI films by the orientations of the polymer main chains and the unreacted CI side groups. Along the director of the LC alignment in the cell, the pretilt angle was measured to be 0-0.15°, depending on the exposure energy and the exposure method. The LC alignment was found to be thermally stable up to 200°C, which is 20°C higher than the T_g of the film. In summary, the LC alignment characteristics of the 6F-HAB-CI PSPI make it a promising candidate material for use as the LC alignment layer in advanced LC display devices with in-plane switching modes that require as low as possible LC pretilt angles.

Full text of DOI:10.1021/ma034445u

Macromolecules 2003, 36, 6527-6536
6527
Photoreactions and Photoinduced Molecular Orientations of Films of a
Photoreactive Polyimide and Their Alignment of Liquid Crystals
Seu n g Woo Lee, Sa n g Il Kim ,† Byeon gd u Lee, Wooyou n g Ch oi, Bok n a m Ch a e,
Seu n g Bin Kim , a n d Moon h or Ree*
Department of Chemistry, Center for Integrated Molecular Systems, BK21 Program,
Division of Molecular and Life Sciences, and Polymer Research Institute,
Pohang University of Science and Technology, San 31, Hyoja-dong, Nam-gu, Pohang 790-784,
The Republic of Korea
Received April 8, 2003; Revised Manuscript Received J une 4, 2003
ABSTRACT: A soluble photoreactive polyimide (PSPI) with side-group cinnamate (CI) chromophores
that forms good quality films through conventional solution spin-casting and drying processes was
successfully synthesized with a reasonably high molecular weight: poly(3,3′-bis(cinnamoyloxy)-4,4′-
biphenylene hexafluoroisopropylidenediphthalimide) (6F-HAB-CI). This 6F-HAB-CI PSPI is thermally
stable up to 340 °C, its glass transition temperature (Tg) is 181 °C, and it was determined by prism
coupling to be positively birefringent. The photochemical reactions of the PSPI in solution and in films,
as well as the molecular orientations of the PSPI that are induced by exposure of its films to linearly
polarized ultraviolet light (LPUVL), were investigated in detail by nuclear magnetic resonance,
ultraviolet-visible, infrared, and Raman spectroscopies and by dissolution testing and optical retardation
measurements. It was confirmed that the CI chromophores of the PSPI undergo both photoisomerization
and photodimerization. LPUVL exposure was found to induce anisotropic orientations of the polymer
main chains and of the CI side groups in the films. Moreover, the LPUVL-irradiated films homogeneously
aligned nematic liquid crystal (LC) molecules along a direction at an angle of 107° with respect to the
polarization of the LPUVL, which coincides with the orientation direction of the PSPI polymer chains.
This result shows that the LC alignment process is principally governed in irradiated PSPI films by the
orientations of the polymer main chains and the unreacted CI side groups. Along the director of the LC
alignment in the cell, the pretilt angle was measured to be 0-0.15°, depending on the exposure energy
and the exposure method. The LC alignment was found to be thermally stable up to 200 °C, which is 20
°C higher than the Tg of the film. In summary, the LC alignment characteristics of the 6F-HAB-CI PSPI
make it a promising candidate material for use as the LC alignment layer in advanced LC display devices
with in-plane switching modes that require as low as possible LC pretilt angles.
In tr od u ction
A representative photoalignment material is poly-
(vinyl cinnamate) (PVCi).6 In fact, this polymer was
introduced in 1959 as a negative photoresist.10 The
cinnamate side groups in the polymer are known to
undergo [2 + 2] photodimerzation by UV light expo-
sure.10 The photodimerization generates cross-links in
the polymer via cyclobutane ring formation, leading to
the insolubilization of the polymer.10 The PVCi and its
derivatives in thin films were reported to have an ability
to align LC molecules in the direction perpendicular to
the polarization axis of the LPUVL when they were
exposed to the LPUVL.5-9 For this phenomenon, con-
troversial debates have been made so far as follows.
Schadt and co-workers6 have proposed a photoalignment
mechanism that the polymer chains are oriented by the
photodimerization of cinnamate side groups in the
polymer film, and the oriented polymer chains induce
LC molecules to align along the axis of the oriented
polymer chains at the surface. Instead, Ichimura and
co-workers7 have suggested a different photoalignment
mechanism. That is, the cinnamate groups in the
polymer film undergo simultaneously photodimerization
and photoisomerization in part. The photoisomerized
cinnamate groups induce the alignment of LC molecules
while the photodimerized polymers enhance the thermal
stability of the LC alignment. Conclusively, the photo-
alignment mechanisms of these polymers has not yet
been understood fully. Moreover, these polymers have
low glass transition temperatures (Tg), so even though
Films of polyimides (PIs) are widely used as liquid
crystal (LC) alignment layers in LC flat-panel display
devices because of their advantageous properties such
as excellent optical transparency, adhesion, heat resis-
tance, dimensional stability, and insulation.1-3 Such PI
film surfaces need to be treated if they are to produce a
uniform alignment of LC molecules with suitable pretilt
angles.1 At present, a rubbing process using velvet
fabrics is the only technique adopted in the LC display
industry for the treatment of PI film surfaces in the
mass production of flat-panel LC display devices. This
process has become the method of choice because it is
simple and enables the control of LC alignment.4
However, the process has some shortcomings, such as
dust generation, electrostatic problems, and poor control
of rubbing strength and uniformity. The search for new
methods that do not suffer from the shortcomings of the
rubbing process has led to the development of several
approaches to polymer alignment layer surface treat-
ment based on irradiation of the polymer with linearly
polarized ultraviolet light (LPUVL).5-9 These techniques
have attracted considerable attention in academia and
industry because they offer the possibility of rubbing-
free production of LC aligning films.
† Present address: Samsung Electronics Company, LCD Divi-
sion, Giheung, Gyeonggi-do, Republic of Korea.
* To whom all correspondence should be addressed: phone 82-
54-279-2120; Fax 82-54-279-3399; e-mail ree@postech.edu.
10.1021/ma034445u CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/23/2003

6528 Lee et al.
Macromolecules, Vol. 36, No. 17, 2003
the polymer chains in the film are oriented preferen-
tially by exposure to LPUVL, they remain mobile after
treatment and their chain orientation is not stable with
respect to environmental influences such as tempera-
ture change. In addition, these polymer films are known
to have a weak anchoring energy for LCs, resulting in
severe reliability problems for LC devices fabricated
from these polymers.9 Thus, there remain significant
obstacles to the delivery without the use of a rubbing
step of high-performance LC-aligning materials.
In this study, a new photosensitive polyimide (PSPI)
with cinnamate side groups that has a high Tg and is
thus stable both thermally and dimensionally was
synthesized. Its photoreactivity in nanoscaled films was
investigated by ultraviolet-visible (UV-vis), infrared
(IR), Raman, and nuclear magnetic resonance (NMR)
spectroscopies. The LPUVL-induced alignment charac-
teristics of the PSPI film were determined by optical
retardation as well as by polarized UV-vis and IR
spectroscopies. In addition, the alignment behavior of
LC molecules on the surfaces of LPUVL-irradiated PSPI
films was investigated.
Exp er im en ta l Section
Ma ter ia ls. 2,2′-Bis(3,4-dicarboxyphenyl)hexafluoropropane
dianhydride (6F) and 3,3′-hydroxy-4,4′-diaminobiphenyl (HAB)
were supplied by Chriskev Co. All other chemicals were
purchased from Aldrich Company. 6F was purified by recrys-
tallization from acetic anhydride, and HAB was dried at 100
°C under vacuum for 1 day. N-Methyl-2-pyrrolidone (NMP)
and tetrahydrofuran (THF) were distilled over calcium hydride
under reduced pressure. Thionyl chloride was purified by
simple distillation. All other chemicals were used as received.
Syn th esis of th e P h otosen sitive P olym er a n d of a
Mod el Com p ou n d . The photosensitive polyimide 6F-HAB-
CI PSPI was synthesized as follows (Figure 1). First, to
synthesize a polyimide containing hydroxyl side groups, 6F-
HAB PI, equivalent moles of 6F and HAB monomers were
dissolved together with 2 mol equiv of isoquinoline catalyst
in dry NMP. The solution was gently heated with stirring at
70 °C for 2 h and then refluxed with stirring for 5 h. The
reaction solution was then poured into a mixture of methanol
and water (6:4 volume ratio) with vigorous stirring, giving 6F-
HAB PI in the form of a precipitated powder. The precipitated
powder was filtered and dried. Second, the photosensitive
polyimide (PSPI) containing cinnamate side groups, 6F-HAB-
CI PSPI, was synthesized from the 6F-HAB PI and cinnamoyl
chloride. Cinnamic acid of 4 mol equiv was dissolved in excess
thionyl chloride under a nitrogen atmosphere. After heating
this reaction solution at 80 °C for 3 h, the residual thionyl
chloride was removed under reduced pressure. The reaction
product was dissolved in dry THF, followed by the addition of
1 mol equiv of the 6F-HAB PI and pyridine in excess. After
stirring for 3 h at room temperature, the reaction solution was
poured into methanol with vigorous stirring, leading to a
precipitated polymer product. The precipitated polymer powder
was filtered and dried, giving 6F-HAB-CI PSPI.
F igu r e 1. Synthetic scheme for the photoreactive polyimide,
6FDA-HAB-CI PSPI.
The product DP-HAB was further reacted with cinnamoyl
chloride. DP-HAB (8.0 g, 0.017 mol) and cinnamoyl chloride
(6.71 g, 0.04 mol) were dissolved in dry THF and excess
pyridine. After stirring at room temperature for 12 h, the
solution was poured into methanol under vigorous stirring.
The precipitated product was collected by filtration, washed
several times with methanol, and then dried in a vacuum oven,
giving DP-HAB-CI. Yield: 92.8%. 1H NMR (δ, DMSO-d6):
7.99-7.86 (m, 12H, ArH), 7.74-7.65 (m, 8H, ArH,
-Ar-CHdCH-), 7.42-7.37 (m, 6H, ArH), 7.76-6.71 (d, 2H,
-Ar-CHdCH-).
F ilm P r ep a r a tion . Various amounts of 6F-HAB-CI PSPI
were dissolved in cyclohexanone, giving solutions with 1-5
wt % polymer that were filtered with a PTFE membrane of
pore size 0.20 µm before use. The PSPI solutions were then
spin-cast onto glass slides for property measurements, onto
quartz substrates for UV-vis spectral analysis, onto calcium
fluoride (CaF2) windows (25 mm diameter × 2 mm thick) for
FTIR spectral analysis, onto gold-coated silicon substrates for
surface-enhanced Raman scattering (SERS), and onto indium
tin oxide (ITO) glasses for optical retardation and LC cell
assembly, followed by drying on a hot plate at 80 °C for 1 h.
The dried PSPI films were further dried for 2 h in a vacuum
oven at 100 °C. The resulting PSPI films were measured to
have a thickness of around 200 nm using a spectroscopic
ellipsometer (model M2000, J .A. Woollam) and an alpha-
stepper (model Tektak, Veeco). For some of the PSPI films,
UV light irradiations were conducted with and without a linear
dichroic polarizer (Oriel) using a high-pressure Hg lamp
system (1.0 kW, Altech, South Korea) equipped with an optical
filter (Milles Griot), which transmits a band beam of 260-
380 nm wavelength. The irradiations of all samples with
unpolarized UV light or linearly polarized UV light (LPUVL)
In addition to 6F-HAB-CI PSPI, a model photoreactive imide
compound, 3,3′-bis(cinnamate)-4,4′-diphthalimidobiphenyl (DP-
HAB-CI), was prepared according to the reaction scheme in
Figure 2. First, 3,3′-dihydroxy-4,4′-diphthalimidobiphenyl (DP-
HAB) was synthesized as follows. HAB (5.0 g, 0.023 mol),
phthalic anhydride (8.2 g, 0.055 mol), and isoquinoline (0.13
g, 0.001 mol) were dissolved in dried NMP. The reaction
mixture was heated at 170 °C for 2 h and then allowed to cool
to room temperature. Then the solution was poured into 1.0
L of methanol, giving a precipitate powder. The precipitated
product was collected by filtration and washed several times
with hot methanol and then dried in a vacuum oven, giving
1
DP-HAB. Yield: 91.4%. H NMR (δ, DMSO-d6): 10.10 (s, 2H,
-OH), 8.01-7.98 (m, 4H, ArH), 7.94-7.91 (m, 4H, ArH), 7.40-
7.37 (d, 2H, ArH), 7.21-7.18 (d, 2H, ArH).

Macromolecules, Vol. 36, No. 17, 2003
Films of a Photoreactive Polyimide 6529
direction antiparallel to the polarization of the LPUVL by
using 50 µm thick spacers. A nematic LC, 5CB (Aldrich)
containing 1.0 wt % Disperse Blue 1 (Aldrich) as a dichroic
dye was injected into the cell gap, followed by sealing of the
injection hole with an epoxy glue. The LC cells were then heat-
treated for 5 min at 40 °C, which is slightly higher than the
nematic-to-isotropic transition temperature of 5CB, to remove
any flow-induced memory that may have been induced by the
LC injection process. The prepared LC cells were found to be
homogeneous throughout by optical microscopy.
Mea su r em en ts. The synthesized 6F-HAB PI and 6F-HAB-
CI PSPI were dissolved in dimethyl-d6 sulfoxide (DMSO-d6)
and characterized by using a proton NMR spectrometer
(Bruker, model Aspect 300 MHz). Their molecular weights
were measured by using a gel permeation chromatography
(GPC) system (model PL-GPC 210, Polymer Labs, England)
calibrated with polystyrene standards. In these measurements
a flow rate of 1.0 mL/min was employed and THF was used
as the eluent.
The glass transition temperatures (Tg) of the films of the
two polymers were measured in the range 25-400 °C using a
differential scanning calorimeter (DSC) (model DSC 220CU,
Seiko, J apan). During the measurements, dry nitrogen gas was
used for purging; a flow rate of 80 cm3/min and a ramping rate
of 10.0 °C/min were employed. In each run, a sample of about
5 mg was used. The value of Tg was taken as the onset
temperature of the glass transition in the thermogram. The
degradation temperatures (Td) of the polymer films were
measured in the range 50-800 °C using a thermogravimeter
(model TGA7, Perkin-Elmer). During these measurements, dry
nitrogen gas was used for purging; a flow rate of 100 cm3/min
and a ramping rate of 5.0 °C/min were employed. Refractive
index measurements were performed on PI and PSPI films
approximately 5.0 µm in thickness using a prism coupler11
equipped with a He-Ne laser source (632.8 nm wavelength).
The refractive index in the film plane (nxy) was measured in
the transverse electric mode, and the refractive index in the
out-of-plane (nz) was obtained in the transverse magnetic
mode. All these measurements were performed using a cubic
zirconia prism of n ) 2.1677 at a wavelength of 632.8 nm.
UV-vis spectra were recorded using an HP 8452 Hewlett-
Packard spectrometer with and without an Oriel linear dich-
roic polarizer. FTIR spectroscopic measurements were carried
out on a Bomem DA8 FTIR spectrometer with and without a
polarizer (single diamond polarizer, Harrick Scientific). Samples
were installed perpendicular to the incident beam direction.
While rotating the polarizer, UV-vis spectra were recorded
at 0.98 nm resolution as a function of the angle of rotation.
While rotating the polarizer, IR spectra were recorded as a
function of the angle of rotation at 4 cm-1 resolution with a
liquid nitrogen cooled mercury cadmium telluride (MCT)
detector under vacuum, and 256 interferograms were ac-
cumulated. Solid-state 13C NMR experiments were carried out
with a Bruker DSX 400 instrument (Bruker Analytische
Messtechnik GmbH, Germany) at 9.4 T and room temperature.
The pulse repetition delay, proton flip pulse, and contact time
for the cross-polarization magic angle spinning (CP-MAS)
experiments were 3 s, 5 µs, and 2 ms, respectively. A typical
spinning rate employed was 12 kHz. Carbon chemical shift
was referenced to external neat tetramethylsilane (TMS).
Optical phase retardation was measured using a phase
retardation analyzer built in our laboratory. In these measure-
ments, the laser beam was incident normal to the film surface,
and the transmitted light intensity [) (in-plane birefringence)
× (phase)] was monitored as a function of the angle of rotation
of the film sample with respect to the surface normal.
For the SERS measurements, silver colloids were prepared
by a method reported previously.12 Silver nitrate was added
into deionized water with stirring, and then the mixture was
heated to boiling. Then a solution of sodium citrate (1.0 wt %)
was rapidly added to the reaction mixture under vigorous
stirring, giving silver colloids. The resulting silver colloids were
directly deposited onto the PSPI films coated onto gold-coated
silicon substrates; here some PSPI films were irradiated with
unpolarized UV light, and the others were used without any
F igu r e 2. Synthetic scheme for the photoreactive model
compound, 3,3′-bis(cinnamate)-4,4′-diphthalimidobiphenyl (DP-
HAB-CI).
were conducted under vacuum; here, samples were installed
perpendicular to the incident beam direction. In particular,
for film samples to fabricate LC cells, LPUVL irradiations were
conducted in two different ways: (i) some films were installed
perpendicular to the incident beam direction, and (ii) the other
films were positioned with a tilt angle of 45° where the tilt
angle is the angle between the film plane and the propagation
plane of the light. The intensity of the unpolarized UV light
was 50 mW/cm2, while the intensity of the linearly polarized
UV light (LPUVL) was 10 mW/cm2. The exposure dose was
measured using an International Light photometer (model IL-
1350, International Light) with a sensor (model SED-240,
International Light). Some PSPI films irradiated with LPUVL
at 1.5 J /cm2 were thermally annealed in an accumulative step
manner in the range 100-200 °C: 100 °C/10 min, 120 °C/10
min, 140 °C/10 min, 160 °C/10 min, 180 °C/10 min, and 200
°C/10 min; these films were used for LC cell assemblies. For
thermal analysis, dissolution testing, prism coupling, and
solid-state NMR spectra, 2-10 µm thick films of the PSPI were
additionally prepared on precleaned glass slides by casting
concentrated solutions and subsequent drying in a vacuum
oven at 100 °C for 2 days.
LC Cells. Some of the LPUVL-exposed PSPI films on glass
substrates were cut into 2.5 × 2.5 cm pieces. Paired pieces
from the same glass substrate were assembled together in the

6530 Lee et al.
Macromolecules, Vol. 36, No. 17, 2003
F igu r e 3. 1H NMR spectra of a polyimide and its photoreac-
tive derivative: (a) 6FDA-HAB; (b) 6FDA-HAB-CI.
F igu r e 4. 1H NMR spectra of DP-HAB-CI in dimethyl-d6
sulfoxide (DMSO-d6) (0.005 g/mL concentration): (a) unex-
posed; (b) exposed to UV light (260-380 nm) at 10.0 J /cm2.
UV light irradiation. The SERS measurements were conducted
using a Renishaw model single grating spectrometer equipped
with a charge coupled device (CCD) detector and an Olympus
microscope. An Omnichrome argon ion laser (514.5 nm wave-
length) was used with the laser power fixed at 25 mW.
proton peak (-CHdCH-Ph) and all aromatic proton
peaks (7.1-8.3 ppm), the conversion yield of the CI
incorporation reaction is estimated to be 99%. These
results indicate that the hydroxyl groups of the 6F-HAB
PI have almost completely reacted with cinnamoyl
chloride, giving 6F-HAB-CI PSPI with two CI side
groups per repeat unit of the polymer backbone. Taking
this reaction yield and the Mw of 6F-HAB PI into
account, the synthesized 6F-HAB-CI PSPI is estimated
to have 76 000 Mw. The resulting PSPI formed good
quality thin films through conventional solution spin-
casting and drying processes.
The LC alignment in an LC cell containing Disperse Blue 1
dichroic dye was examined using an optical setup2 that was
equipped with a He-Ne laser (632.8 nm wavelength), a
polarizer, a photodiode detector, and a goniometer. In these
measurements, the laser beam was incident normal to the
surface of the LC cell mounted on the goniometer, and these
components were placed between the polarizer and the detec-
tor. Light absorption of the dichroic dye molecules, which tend
to align parallel to the LCs in the cell, was then monitored as
a function of the angle of rotation of the cell. The pretilt angle
R of the LC molecules was measured using a crystal rotation
apparatus.2
The 6F-HAB PSPI film was found to have a Tg of 181
°C and a Td of 340 °C, whereas the 6F-HAB PI film had
a Td of 440 °C but no glass transition in the range 50-
500 °C. Overall, the values of Tg and Td of the 6F-HAB
PI were lowered by the incorporation of CI side groups.
The average refractive index [nav ) (2nxy + nz)/3] and
out-of-plane birefringence [∆xy-z ) nxy - nz] are 1.599
and 0.038 for the 6F-HAB PI and 1.641 and 0.032 for
the 6F-HAB-CI PSPI, respectively. The 6F-HAB-CI
PSPI film exhibits a higher refractive index than that
of the 6F-HAB PI. This might be due to a decrease in
the fluorine fraction of the polymer. The polymers ex-
hibit a positive out-of-plane birefringence, regardless of
the incorporation of the CI side groups. In thin films,
long-chain polymer molecules generally tend to lie in
the film plane, resulting in an out-of-plane birefringence
in the films.11 Therefore, we interpret the measured
Resu lts a n d Discu ssion
Syn th esis a n d P r op er ties. As shown in Figure 1,
6F-HAB-CI PSPI was synthesized in two major steps:
synthesis of soluble 6F-HAB PI followed by incorpora-
tion of cinnamate (CI) side groups into the polymer.
Soluble 6F-HAB polyimide was synthesized directly
from the polycondensation of the respective monomers
using isoquinoline as a catalyst. The obtained polymer
1
1
was characterized by H NMR spectroscopy. In the H
NMR spectrum (Figure 3a), the proton peak of the
hydroxyl side groups appears at 10.0 ppm, while the
proton peaks of the aromatic rings on the polymer
backbone appear in the range 7.1-8.3 ppm. Amino
protons originating from possible residues of partially
imidized 6F-HAB poly(amic acid) were not detected,
suggesting that the PI polymer we obtained was of high
molecular weight. The PI polymer was measured
∆
xy-z
values to be evidence that the polymer chains in
both the PI and the PSPI films are preferentially
oriented in the plane of the film. The fact that the ∆xy-z
values are positive indicates that both 6F-HAB PI and
6F-HAB-CI PSPI are positively birefringent polymers
whose polarizations are larger along their polymer chain
axes than along the directions normal to the polymer
chain axes.
to have a weight-averaged molecular weight (Mw) of
53 400 and a polydispersity (PDI) of 1.87 by GPC
analysis. In conclusion, the soluble 6F-HAB PI was
synthesized with a reasonably high molecular weight.
Cinnamate (CI) side groups were incorporated in the
soluble 6F-HAB PI, giving 6F-HAB-CI PSPI. As shown
in Figure 3b, the 1H NMR spectrum of 6F-HAB-CI PSPI
does not contain any peaks originating from the hy-
droxyl groups of 6F-HAB PI. From the integration of
the vinyl proton peak (6.6-6.8 ppm, doublet,
-CHdCH-Ph) with respect to those of the vinylene
P h otor ea ctivity. Figure 4a shows the 1H NMR
spectrum of the model imide compound DP-HAB-CI in
DMSO-d6 (0.005 g/mL concentration) before UV light
irradiation. The observed proton peaks were assigned
with the aid of results previously reported for cinnamic

Macromolecules, Vol. 36, No. 17, 2003
Films of a Photoreactive Polyimide 6531
F igu r e 5. 1H NMR spectra of 6F-HAB-CI PSPI in dimethyl-
d6 sulfoxide (DMSO-d6) (0.005 g/mL concentration): (a) un-
exposed; (b) exposed to UV light (260-380 nm) at 10.0 J /cm2.
F igu r e 6. 13C CP/MAS spectra of 6F-HAB-CI PSPI films: (a)
unexposed; (b) exposed to UV light (260-380 nm) at 10.0 J /cm2.
The assignments and the structure are shown. Asterisks
denote spinning sidebands.
acid.13 In this spectrum, the peak (doublet) at 6.73 ppm
is assigned to the -CHdCH-Ph of the trans-isomeric
CI group while that (doublet) at 7.68 ppm is assigned
to the -CHdCH-Ph of the trans-isomeric CI group.
This spectrum indicates that the CI groups of the DP-
HAB-CI are only present in the trans-isomeric form in
solution. However, after the solution had been irradi-
ated with UV light at 10.0 J /cm2 exposure energy, the
at 12.03 mol %, as estimated from the integrations of
its proton peaks with respect to those of the trans-
isomer. On the other hand, no signal due to the
photodimerization of the CI groups was detected in the
NMR spectra. These results suggest that in solution the
CI groups of the DP-HAB-CI molecules favorably un-
dergo photoisomerization.
In addition, some insoluble particles were found in
the solution of the PSPI after it had been irradiated with
UV light, as was observed for the DP-HAB-CI solution
after it had been irradiated with UV light. This suggests
that the PSPI molecules also undergo some photodimer-
ization in solution.
Dissolution studies were also performed for the PSPI
films before and after UV light irradiation. Unexposed
films of the PSPI are soluble in cyclohexanone. The
PSPI films become insoluble in cyclohexanone after
irradiation with UV light at an exposure energy of 1.0-
1.5 J /cm2, suggesting that the PSPI films become cross-
linked after irradiation because of the photodimeriza-
tions of the CI side groups.
The composition of the PSPI films before and after
UV light irradiation were further investigated by solid
13C NMR spectroscopy. The 13C NMR spectra are shown
in Figure 6. As shown in Figure 6a, an unexposed PSPI
film exhibits several carbon peaks in the range 100-
170 ppm. These carbon peaks were assigned with the
aid of results previously reported for poly(vinyl cin-
namate) films.14 The peaks at 118, 147, and 165 ppm
correspond to the -CHdCH-Ph, -CHdCH-Ph, and
CdO groups of the PSPI polymer, respectively. The
peaks for all the aromatic carbons of the PSPI appear
in the range 120-145 ppm. After the PSPI film had
been irradiated with UV light at 10 J /cm2, a new, broad
peak appeared at 47 ppm, as shown in Figure 6b; this
new peak is due to cyclobutane ring carbons generated
by the photodimerization of CI groups in the PSPI film,
as reported previously for poly(vinyl cinnamate) films
1
DP-HAB-CI exhibited additional peaks in its H NMR
spectrum at 6.06 ppm (doublet) and 7.10 ppm (doublet)
(Figure 4b), which correspond to the -CHdCH-Ph and
the -CHdCH-Ph of the cis-isomeric CI group, respec-
tively. In this case, the cis-isomer is formed at 20.6 mol
%, as estimated from the integrations of its proton peaks
with respect to those of the trans-isomer. On the other
hand, no signal due to the photodimerization of the CI
groups was detected in the NMR spectra. These results
suggest that in solution the CI groups of the DP-HAB-
CI molecules favorably undergo photoisomerization.
In fact, the DP-HAB-CI solution used in this study
was optically clear before UV light irradiation. However,
after the solution had been irradiated with UV light it
was found to contain some insoluble particles; these
insoluble particles might be products cross-linked by
photodimerizations of the CI groups. This suggests that
the DP-HAB-CI molecules also undergo some photo-
dimerization in solution.
1
Figure 5 shows the H NMR spectra of 6F-HAB-CI
PSPI in DMSO-d6 (0.005 g/mL concentration) measured
before and after UV light irradiation at 10.0 J /cm2
exposure energy. As shown in Figure 5a, the spectrum
for the PSPI also exhibits peaks typical of vinyl protons
(a doublet peak at 6.73 ppm and another doublet peak
at 7.88 ppm), as observed for the model compound DP-
HAB-CI, which indicates that prior to UV light irradia-
tion the CI groups of the PSPI are only present in the
trans-isomeric form in solution. However, after the
solution had been irradiated with UV light at 10.0 J /cm2
1
exposure energy, the H NMR spectrum for the PSPI
contained additional peaks at 6.08 ppm (doublet) and
7.13 ppm (doublet) (Figure 5b), which correspond to the
-CHdCH-Ph and the -CHdCH-Ph of the cis-iso-
meric CI group, respectively. The cis-isomer is formed

6532 Lee et al.
Macromolecules, Vol. 36, No. 17, 2003
F igu r e 7. Raman spectra of 6F-HAB-CI PSPI films: (a)
unexposed; (b) exposed to UV light (260-380 nm) at 10.0 J /cm2.
F igu r e 9. FTIR spectra of 6F-HAB-CI PSPI films exposed to
UV light (260-380 nm) with varying exposure energy.
decrease rapidly as the UV exposure energy increases.
Moreover, as the exposure energy increases, the inten-
sity of the band at 1737 cm-1 (due to the conjugated
CdO stretching vibration) decreases. These drops in
intensity of the conjugated CdC and CdO stretching
bands could be caused by two factors: the loss of
π-conjugation due to photodimerization of the CI chro-
mophores and/or the photoisomerization of the CI chro-
mophores from trans-isomers to cis-isomers. According
to the report of Chakrabartk et al.,16 the shift in position
of the conjugated CdO stretching band due to trans-
cis photoisomerization is small. In fact, such a small
position shift cannot be resolved in the present study
because of the extensive overlapping at higher exposure
energy of the band at 1737 cm-1 with the stretching
vibrational band of the nonconjugated CdO formed as
a result of the photodimerization of the CI groups. This
fact supports the suggestion that the intensity drop of
the conjugated CdO stretching band originates princi-
pally from the photodimerization of the CI chro-
mophores and possibly in part from the trans-cis
photoisomerization of the CI chromophores. However,
the UV absorption results as described above imply that
the drops in intensity of the conjugated CdC and CdO
stretching bands result from the loss of CI chromophores
by photodimerization.
The spectroscopic results and the dissolution tests all
point to the conclusion that the photoreactive CI side
groups of the 6F-HAB-CI PSPI undergo both trans-cis
photoisomerization and [2 + 2] photodimerization in
solution as well as in films; the cis-isomerization product
is easily detected in solution, while the dimerization
product is mostly identified in films.
The concurrence of photoisomerization and photo-
dimerization of CI chromophores were also observed for
PVCi and its derivatives in films.7,17 For these polymers,
the fractions of the photoisomerization and photodimer-
ization products were however found to depend on the
regioisomerization and substituents of CI chromophore7
as well as the connection modes of CI chromophore to
the polymer backbone and substituents of the polymer
backbone.7,17 Thus, the fractions of the photoreaction
products in the 6F-HAB-CI PSPI may be different from
those observed for PVCi and its derivatives.
F igu r e 8. UV spectra of 6F-HAB-CI PSPI films exposed to
UV light (260-380 nm) with varying exposure energy.
irradiated with UV light.14 In contrast to the solution
1H NMR spectral analysis, the cis-isomers of the CI
groups formed by photoisomerization could not be
distinguished from the trans-isomers in the solid 13C
NMR spectra.
Figure 7a shows the Raman spectrum of an unex-
posed PSPI film. The observed Raman peaks were
assigned according to the results previously reported for
cinnamic acid and its derivatives.15 The peaks at 956
and 990 cm-1 correspond to the vinyl C-H out-of-plane
and in-plane bendings, respectively, while the other
peaks at 621, 665, and 683 cm-1 are due to the out-of-
plane vibration and the deformation of the aromatic
ring. After the PSPI film had been irradiated with UV
light at an exposure energy of 10.0 J /cm2, additional
peaks were observed in the Raman spectrum (Figure
7b): cyclobutane ring breathing and stretching modes
appeared at 971 and 1053 cm-1, respectively, and
cyclobutane ring deformations appeared at 672 and 694
cm-1. These results again confirm that the CI groups
of the PSPI chains in the film favorably undergo
photodimerization.
Figure 8 shows the UV absorption spectra of PSPI
films after irradiation with UV light at various exposure
energies. Before UV light irradiation, the PSPI films
exhibit an absorption maximum at 278 nm () λmax),
which is due to the CI chromophores in the side groups.
The absorption peak at 278 nm decreases abruptly in
intensity in the early stages of photoreaction, and then
its intensity decreases more slowly with increasing
energy exposure.
P h otor ea ction -In d u ced Molecu la r Or ien ta tion .
For the PSPI films irradiated with linearly polarized
UV light (LPUVL) at various exposure energies, UV-
vis spectra were measured while rotating the UV-vis
polarizer. Figure 10a shows a polar diagram of the
intensities of the absorption peak (λmax ) 278 nm) of a
PSPI film irradiated with LPUVL at 0.5 J /cm2, which
Figure 9 shows the FTIR absorption spectra of PSPI
films irradiated with UV light at varying exposure
energies. As shown in the figure, the intensities of the
bands at 1633 and 981 cm-1 (due to the vinylene CdC
stretching vibration and the trans-vinylene C-H out-
of-plane bending in the CI chromophores, respectively)

Macromolecules, Vol. 36, No. 17, 2003
Films of a Photoreactive Polyimide 6533
F igu r e 10. (a) Polar diagram of the intensities of absorption
peak (λmax ) 278 nm) of a 6F-HAB-CI film irradiated with
linearly polarized UV light (LPUVL) (260-380 nm) at 0.5
J /cm2, measured with a linearly polarized probing beam as a
function of the angle of rotation of the film. (b) UV dichroic
ratios determined for 6F-HAB-CI PSPI films exposed to
LPUVL (260-380 nm) with varying exposure energy.
F igu r e 11. Polar diagrams of the absorption intensities of
some specific IR peaks of a 6F-HAB-CI film irradiated with
linearly polarized UV light (LPUVL) (260-380 nm) at 0.5
J /cm2, measured with a linearly polarized probing beam as a
function of the angle of rotation of the film: (a) vinylene CdC
stretching band at 1633 cm-1; (b) imide N-C stretching band
were measured as a function of the angle of rotation of
the film. As seen in the figure, the absorption of the
PSPI film is less intense when the polarization of the
incident probing beam is parallel to the polarization
direction of the LPUVL used in the exposure. Similar
anisotropic absorptions were observed for the other
PSPI films irradiated at different exposure energies.
Taking the results above into account, for the LPUVL-
irradiated PSPI films, the dichroic ratio [) (A - A|)/
(A + A|)] was calculated from the absorbance A at the
maximum absorption wavelength (λmax), as measured
with UV-vis light linearly polarized perpendicular to
the polarization direction of the LPUVL, and A|, as
measured with UV-vis light linearly polarized parallel
to the polarization direction of the LPUVL. The calcu-
lated dichroic ratios are displayed in Figure 10b. As
shown in the figure, these values are all positive for
exposure energies < 3.5 J /cm2. The dichroic ratio
increases rapidly with increasing exposure energy up
to 0.5 J /cm2 and then very slowly increases with further
increases in the exposure energy.
These results indicate that the CI chromophores
located parallel to the polarization direction of the
incident LPUVL are consumed more rapidly than those
positioned perpendicular to the polarization direction
of the LPUVL. Thus, the directionally selective photo-
reaction of CI chromophores in the PSPI films due to
LPUVL exposure leaves CI chromophores unreacted
along the direction perpendicular to the polarization of
the LPUVL. This preferential orientation of unreacted
CI chromophores is easily obtained with a LPUVL
exposure energy of only 0.5 J /cm2.
at 1377 cm-1
.
mophores and of the polymer chain backbones. Figure
11 shows the measured peak intensities of two selected
IR bands for a PSPI film irradiated with LPUVL at 0.5
J /cm2, plotted as polar diagrams with respect to the
angle of rotation of the film.
As shown in the shaded areas of Figure 11a, the
vinylene CdC stretching band at 1633 cm-1 is less
intense when the polarization of the incident IR beam
is at angles in the ranges 340-30° and 160-210° with
respect to the polarization direction of the LPUVL used
in the exposure. This result indicates that the CI
chromophores with vinylene units positioned parallel
to and in the vicinity of the polarization direction of the
LPUVL (see the shaded areas in Figure 11a) are
consumed more rapidly by photoreaction than those
positioned at other orientations in the film. Moreover,
this result suggests that the LPUVL exposure induces
a directionally selective photoreaction of the vinylene
units in the CI side groups, although the directional
photoreaction selectivity of the LPUVL is not high.
Thus, this selective photoreaction induces a preferen-
tially molecular orientation of the vinylene CdC bonds
of unreacted CI chromophores in directions defined by
the angle ranges 50-140° and 230-325° with respect
to the polarization direction (0° T 180°) of the LPUVL.
As shown in Figure 11b, the imide N-C stretching
band at 1377 cm-1 is more intense when the polarization
of the incident beam is perpendicular to the polarization
of the LPUVL. This result indicates that the imide N-C
bonds are oriented by the LPUVL exposure to an
alignment perpendicular to the polarization of the
LPUVL. The imide N-C bonds are part of the PSPI
main chain, so this result provides information on the
orientation of the PSPI chains induced by LPUVL
exposure.
The molecular orientations in LPUVL-irradiated PSPI
films were further investigated by using transmission
FTIR spectroscopy with a linearly polarized IR light
source and measuring the absorbance while rotating the
IR polarizer. This procedure enables us to deduce the
extents of the orientations of the unreacted CI chro-

6534 Lee et al.
Macromolecules, Vol. 36, No. 17, 2003
F igu r e 13. Polar diagram of absorbances measured from a
parallel LC cell assembled with a 6FDA-HAB-CI film irradi-
ated with linearly polarized UV light (LPUVL) (260-380 nm)
at 0.5 J /cm2 as a function of the angle of rotation of the LC
cell.
(parallel) or 90° (perpendicular) with respect to the
polymer main chain. Therefore, the long axis of CI group
and the vinylene bond axis are not parallel to that of
the polymer chain. Because of this geometrical arrange-
ment of the CI side groups and the polymer main chain,
the selective photoreactions of the CI side groups by
LPUVL might explain the orientation direction of the
polymer chains shown in Figure 12a.
In Figure 12a, the optical retardation was determined
from the maximum transmitted light intensity values
along the direction 287° T 107°. The above determina-
tion of optical retardation for the PSPI films irradiated
with LPUVL was extended to various exposure energies.
The resulting retardation values are plotted in Figure
12b as a function of exposure energy. As shown in
Figure 12b, the retardation rapidly increases with
exposure energy up to around 0.5 J /cm2 and then slowly
increases with further increases in the exposure energy.
This result suggests that a preferential orientation of
the PSPI polymer chains in films is induced by LPUVL
exposure only at 0.5 J /cm2, which is consistent with our
conclusion from the dichroic UV-vis spectroscopy mea-
surements described above.
LC Align m en t. All the antiparallel LC cells fabri-
cated with PSPI films irradiated with LPUVL at various
exposure energies were found by optical microscopy to
be homogeneous throughout. Figure 13 displays a
representative polar diagram, which was constructed for
an LC cell fabricated with a PSPI film irradiated with
LPUVL at 0.5 J /cm2. As shown in the figure, the main
director of the LC molecules lies along the direction 287°
T 107°, which is at an angle of 107° to the polarization
direction (0° T 180°) of the LPUVL used in the UV
exposure. This result indicates that the LC molecules
in contact with the film surface are induced to homo-
geneously align in a direction at an angle of 107° to the
polarization of the LPUVL. The director of the LC
alignment was same for the other PSPI films irradiated
at various exposure energies in the range 0.1-5.0 J /cm2.
We now consider the interactions between the LC
(5CB) molecules and the PSPI polymer chains that
might affect the alignment of the LC molecules.
F igu r e 12. (a) Polar diagram of the transmitted light
intensity [) (in-plane birefringence) × (phase)] as a function
of the angle of rotation of the film observed in the optical phase
retardation measurements of a 6F-HAB-CI PSPI film irradi-
ated with linearly polarized UV light (LPUVL) (260-380 nm)
at 0.5 J /cm2. (b) Variation of the optical retardation [) (in-
plane birefringence) × (film thickness)] of a 6F-HAB-CI film
irradiated with LPUVL (260-380 nm) with varying exposure
energy.
Figure 12a displays a polar diagram of the transmit-
ted light intensity [) (in-plane birefringence) × (phase)]
as a function of the angle of rotation of the film, obtained
from optical phase retardation measurements of a PSPI
film irradiated with LPUVL at an exposure energy of
0.5 J /cm2. This polar diagram shows that the irradiated
film exhibits a maximum in transmitted light intensity
along the direction 287° T 107°, which is at an angle of
107° with respect to the polarization direction (0° T
180°) of the LPUVL used in the UV exposure, but
exhibits a minimum transmitted light intensity value
along the direction 197° T 17°, which is at an angle of
17° with respect to the polarization direction of the
LPUVL. On the other hand, the PSPI films that were
not irradiated exhibited isotropic polar diagrams of the
transmitted light intensity in the optical retardation
measurements (data not shown).
Given that the PSPI chain is positively birefringent,
as described earlier in the “Synthesis and Properties”
section, the anisotropic form of the polar diagram
indicates that the PSPI polymer chains are oriented
preferentially along the direction at an angle of 107° to
the polarization direction of the LPUVL used in the
exposure. This preferential orientation direction is
somewhat different to that of the imide N-C bonds as
well as to that of the vinylene CdC bonds, both
described above. These differences in orientation direc-
tion are attributed to the geometrical structure of the
PSPI polymer. As shown in Figure 1, the polymer chain
has a kink at every 6F moiety, and hence its long chain
axis is not parallel to the N-C bond axis. The CI side
groups are attached at some angle rather than 0°
First, the 5CB molecule has a length of approximately
1.8 nm and a diameter of approximately 0.25 nm, which
are smaller than the dimensions of the chemical repeat
unit of the main chain backbone but slightly larger than
those of the CI side group. Therefore, both the polymer
main chains and the CI side groups in the PSPI film
may influence the LC alignment on the film surface.

Macromolecules, Vol. 36, No. 17, 2003
Films of a Photoreactive Polyimide 6535
Second, it has been previously suggested that for
conventional rubbing-type polyimide alignment layer
materials the major intermolecular interaction between
polyimide and LC molecules is the π-π interaction
between the phenyl rings of the polymer and those of
the LC molecule.18 Taking this into account, the main
chain backbone, which has four phenyl rings per chemi-
cal repeat unit, should have stronger intermolecular
interactions with the biphenyl ring of the LC molecule
than do the two side groups, which have only two phenyl
rings per chemical repeat unit (see Figure 1). Moreover,
the main chain backbone has a biphenyl unit, as does
the mesogen unit of the LC molecule. Thus, the biphenyl
units in the polymer main chain may be significantly
involved in interactions with the LC molecules.
tions with the oriented polymer main chains, is a
significant departure from the LC alignment observed
for PVCi and its derivatives reported so far,5-9 for which
LC alignment is always induced mainly by anisotropic
interactions with the unreacted CI chromophores.
For the LC cells, the pretilt angle of the LCs was
determined along the director of the LC alignment in
the cell by using the crystal-rotation technique. The LC
pretilt angle was measured to be 0° for the LC cells
fabricated with the PSPI films irradiated at the position
perpendicular to the incident LPUVL direction. Instead,
the LC cells, which were prepared with the PSPI films
irradiated at the position with a tilt angle of 45° with
respect to the propagation plane of LPUVL, show a LC
pretilt angle ranged from 0.05° to 0.15°, depending on
the exposure energy; the films irradiated at higher
exposure energy gave larger LC pretilt angle.
The thermal stabilities of the anisotropically oriented
polymer chains and their LC alignment ability were also
examined. For these studies, a series of LC cells were
prepared with PSPI films irradiated with LPUVL at 1.5
J /cm2 and followed by thermal annealing from 100 to
200 °C via the following steps: 100 °C/10 min, 120 °C/
10 min, 140 °C /10 min, 160 °C/10 min, 180 °C/10 min,
and 200 °C/10 min. The directors of the alignments of
LCs on these annealed PSPI films were found to be
same as that for the unannealed PSPI film, regardless
of the annealing history. In particular, for the films
annealed above the Tg (181 °C), the LC alignment
directors were the same as that of the unannealed film.
Only very small changes in the LC pretilt angle R were
observed. Collectively, these results indicate that the
oriented polymer chains and CI side groups that induce
LC alignment are thermally stable up to 200 °C, which
is approximately 20 °C higher than the Tg of the film.
Third, we consider the unreacted CI side groups left
by LPUVL exposure and their contribution to the
intermolecular interactions of the polymer chains with
the LC molecules. The vinylene CdC bonds of the
unreacted CI chromophores in the film were found to
reorient into the orientations defined by the angles
50-140° and 230-325° to the polarization direction
(0° T 180°) of the LPUVL, as described earlier (see
Figure 11a). As shown in Figure 1, the axis of the
vinylene CdC bond is at some angle, rather than 0° or
90°, to the long axis of the CI chromophores. Thus, the
direction of the oriented CI chromophores might lie
along a direction at some angle, rather than 0° or 90°,
to that of the oriented vinylene CdC bonds. In addition,
the dichroic UV spectroscopy results (Figure 10) suggest
that the unreacted CI chromophores lies preferentially
along a direction perpendicular to the polarization of
the LPUVL. Collectively, the IR and UV results suggest
that the unreacted CI chromophores are oriented in a
direction that is not parallel to the polarization of the
LPUVL. Such preferentially oriented CI chromophores
might contribute positively to the observed LC align-
ment.
In summary, the LC alignment characteristics of the
thermally stable 6F-HAB-CI PSPI make it a promising
candidate material for use as an LC alignment layer in
advanced LC display devices, in particular devices with
an in-plane switching mode that require as low as
possible LC pretilt angles.
Finally, the preferential orientation of the polymer
main chains and their contribution to the intermolecular
interactions of the polymer chains with the LC mol-
ecules are considered. In the PSPI films irradiated with
LPUVL, the polymer chains are preferentially oriented
along the direction at an angle of 107° with respect to
the polarization of the LPUVL, as determined in the
optical retardation measurements above. The preferen-
tial orientation of the polymer chains leads to anisotro-
pic interactions with adjoining LCs, which in turn leads
to LC alignment along the orientation direction (i.e., at
an angle of 107° with respect to the polarization of the
LPUVL). As shown in Figures 12a and 13, the main
director of the preferentially oriented polymer chains
exactly coincides with the main director of the LC
alignment. This is a good sign that the PSPI polymer
chains oriented anisotropically by LPUVL exposure play
a major role in alignment of the LC molecules in contact
with the irradiated film.
Con clu sion
A soluble polyimide with photoreactive CI chro-
mophores in side groups, 6F-HAB-CI PSPI, was suc-
cessfully synthesized with a reasonably high molecular
weight (76 000 Mw). The PSPI solutions give good
quality films through conventional spin-casting and
drying processes. The PSPI polymer exhibits a Tg of 181
°C and a Td of 340 °C, which are higher than those of
the conventional rubbing-type polyimides currently used
in the LC display industry and much higher than those
of photoalignable poly(vinyl cinnamate) and its deriva-
tives. The PSPI was determined to be positively bire-
fringent by prism coupling.
The photochemical reactions of the PSPI in solution
and in films were investigated in detail by NMR, UV-
vis, IR, and Raman spectroscopies as well as by dis-
solution testing; the CI chromophores of the PSPI were
confirmed to undergo both trans-cis photoisomerization
and [2 + 2] photodimerization.
Taking the orientations of molecular groups and their
possible interactions with the LC molecules into ac-
count, we conclude that the anisotropically oriented
polymer main chains and the unreacted CI side groups
work together to induce a homogeneously uniaxial align-
ment of LC molecules, although the oriented polymer
main chains make the dominant contribution to the LC
alignment.
The molecular orientations in PSPI films that are
induced by LPUVL exposure were determined by lin-
early polarized UV spectroscopy, linear polarized IR
spectroscopy, and optical retardation measurements.
LPUVL exposure was found to induce anisotropic
This LC alignment at the irradiated 6F-HAB-CI PSPI
films, which is induced mainly by anisotropic interac-

6536 Lee et al.
Macromolecules, Vol. 36, No. 17, 2003
Binger, D. R.; Hanna, S. Liq. Cryst. 1999, 26, 1205. (h) Ge,
J . J .; Li, C. Y.; Xue, G.; Mann, I. K.; Zhang, D.; Wang, S.-Y.;
Harris, F. W.; Cheng, S. Z. D.; Hong, S.-C.; Zhuang, X.; Shen,
Y. R. J . Am. Chem. Soc. 2001, 123, 5768. (i) Ge, J . J .; Xue,
G.; Li, F.; McCreight, K. W.; Wang, S.-Y.; Harris, F. W.;
Cheng, S. Z. D.; Zhuang, X.; Hong, S.-C.; Shen, Y. R.
Macromol. Rapid Commun. 1998, 19, 619.
orientations of the polymer main chains and of the CI
side groups in the film.
The LPUVL-irradiated films homogeneously aligned
LC molecules along a direction at an angle of 107° with
respect to the polarization of the LPUVL. This LC
alignment result, along with our conclusions in regard
to the orientation of polymer chains in irradiated films,
shows that the oriented polymer chains in the irradiated
films interact anisotropically with LC molecules and
align the LC molecules along the orientation direction
of the polymer chains. This LC alignment process is
principally governed in the irradiated PSPI films by the
orientations of the polymer main chains and of the
unreacted CI side groups, whose directionally anisotro-
pic interactions contribute to the alignment of the LC
molecules. Along the director of the LC alignment in
the cell, the pretilt angle was 0-0.15°, depending on
the exposure energy and the exposure method. This LC
alignment was found to be thermally stable up to 200
°C, approximately 20 °C higher than the Tg of the film.
In summary, the present study has revealed the
homogeneous, uniaxial LC-aligning ability of 6F-HAB-
CI PSPI. The properties of this PSPI make it a promis-
ing candidate material for use as an LC alignment layer
in advanced LC display devices, in particular in devices
with an in-plane switching mode that require as low as
possible LC pretilt angles.
(4) (a) J anning, J . L. Appl. Phys. Lett. 1972, 21, 173. (b) Cognard,
J . Alignment of Liquid Crystals and Their Mixtures; Gorden
Breach: London, 1982. (c) Mauguin, C. Bull. Soc. Fr. Mineral.
1911, 34, 71. (b) Geary, J . M.; Goodby, J . W.; Kmetz, A. R.;
Patel, J . S. J . Appl. Phys. 1987, 62, 4100.
(5) (a) O’Neill, M.; Kelly, S. M. J . Phys. D: Appl. Phys. 2000, 33,
R67. (b) Ichimura, K. Chem. Rev. 2000, 100, 1847.
(6) (a) Schadt, M.; Schmitt, K.; Kozinkov, V.; Chigrinov, V. J pn.
J . Appl. Phys. 1992, 31, 2115. (b) Schadt, M.; Seiberle, M.;
Schuster, A.; Kelly, S. M. J pn. J . Appl. Phys. 1995, 34, L764.
(c) Schadt, M.; Seiberle, H.; Schuster, A. Nature (London)
1996, 381, 212.
(7) Ichimura, K.; Akita, Y.; Akiyama, H.; Kudo, K.; Hayashi, Y.
Macromolecules 1997, 30, 903.
(8) (a) Iimura, Y.; Kobayashi, S.; Hashimoto, T.; Sugiyama, T.;
Katoh, K. HEICE Trans. Electron. 1996, E39, 1040. (b) Han,
K. Y.; Chae, B. H.; Yu, S. H.; Song, J . K.; Park, J . G.; Kim,
D. Y. AM-LCD’96/ IDW’96 1996, 403.
(9) (a) Vilfan, M.; Olenik, I. D.; Mertelj, A.; Copic, M. Phys. Rev.
E 2001, 63, 061709. (b) Klopcar, N.; Olenik, I. D.; Copic, M.;
Kim, M. W.; Rastegar, A.; Rasing, Th. Mol. Cryst. Liq. Cryst.
2001, 368, 395. (c) Li, X. T.; Pei, D. H.; Kobayash, S.; Iimura,
Y. J pn. J . Appl. Phys. 1997, 36, L432. (d) Lee, J .-W.; Kim,
H.-T.; Sung, S.-J .; Park, J .-K. Synth. Met. 2001, 117, 267. (e)
Kim, E. J .; Park, O O.; Feng, L. H.; Kawanami, Y.; Furue,
H.; Kobayashi, S. AM-LCD’97/IDW’97 1997, 105. (f) Yamagu-
chi, R.; Goto, Y.; Sato, S. J pn. J . Appl. Phys. 2002, 41, L889.
(10) Minsk, L. M.; Smith, J . G.; van Deusen, W. P.; Wright, J . F.
J . Appl. Polym. Sci. 1958, 2, 302.
(11) (a) Kim, S. I.; Shin, T. J .; Pyo, S. M.; Moon, J . M.; Ree, M.
Polymer 1999, 40, 1603. (b) Pyo, S. M.; Kim, S. I.; Shin, T.
J .; Park, Y. H.; Ree, M. J . Polym. Sci., Part A: Polym. Chem.
1999, 37, 937. (c) Ree, M.; Shin, T. J .; Park, Y.-H.; Kim, S. I.;
Woo, S. H.; Cho, C. K.; Park, C. E. J . Polym. Sci., Polym.
Phys. 1998, 36, 1261. (d) Ree, M.; Kim, K.; Woo, S. H.; Chang,
H. J . Appl. Phys. 1997, 81, 698. (e) Ree, M.; Chu, C. W.;
Goldberg, M. J . J . Appl. Phys. 1994, 75, 1410. (f) Ree, M.;
Shin, T. J .; Lee, S. W. Korea Polym. J . 2001, 9, 1. (g) Goh,
W. H.; Kim, K.; Ree, M. Korea Polym. J . 1998, 6, 241.
(12) Lee, P. C.; Meisel, D. J . Phys. Chem. 1982, 86, 3391.
(13) Hanai, K.; Kuwae, A.; Takai, T.; Senda, H.; Kunimoto, K.-K.
Spectrochim. Acta, Part A 2001, 57, 513.
Ack n ow led gm en t. This study was supported by the
Korea Research Foundation (project contracted with
Postech Polymer Research Institute in 2002) and by the
Ministry of Education (Brain Korea 21 Project). The
authors thank Prof. T. Chang for the molecular weight
measurements of polymers.
Refer en ces a n d Notes
(1) (a) Collings, P. J ., Ed. Liquid Crystals; IOP Publishing, Ltd.:
Bristol, 1990. (b) Collings, P. J ., Patel, J . S., Eds. Handbook
of Liquid Crystal Research; Oxford University Press: Oxford,
1997.
(2) (a) Lee, K.-W.; Paek, S.-H.; Lien, A.; During, C.; Fukuro, H.
Macromolecules 1996, 29, 8894. (b) van Aerle, N. A. J .; Tol,
J . W. Macromolecules 1994, 27, 6520. (c) Chae, B.; Kim, S.
B.; Lee, S. W.; Kim, S. I.; Choi, W.; Lee, B.; Ree, M.; Lee, K.
H.; J ung, J . C. Macromolecules 2002, 35, 10119. (d) Kim, S.
I.; Ree, M.; Shin, T. J .; J ung, J . C. J . Polym. Sci., Part A:
Polym. Chem. 1999, 37, 2909. (e) Kim, S. I.; Pyo, S. M.; Ree,
M.; Park, M.; Kim, Y. Mol. Cryst. Liq. Cryst. 1998, 316, 209.
(f) Uchida, T.; Hirano, M.; Sakai, H. Liq. Cryst. 1989, 5, 1127.
(3) (a) Toney, M. F.; Russell, T. P.; Logan, J . A.; Kikuchi, H.;
Sands, J . M.; Kumar, S. K. Nature (London) 1995, 374, 709.
(b) Samant, M. G.; Stohr, J .; Brown, H. R.; Russell, T. P.;
Sands, J . M.; Kumar, S. K. Macromolecules 1996, 29, 8334.
(c) Cossy-Favre, A.; Diaz, J .; Liu, Y.; Brown, H. R.; Samant,
M. G.; Stohr, J .; Hanna, A. J .; Anders, S.; Russell, T. P.
Macromolecules 1998, 31, 4957. (d) Mori, N.; Morimoto, M.;
Nakamura, K. Macromolecules 1999, 32, 1488. (e) Weiss, K.;
Woll, C.; Bohm, E.; Fiebranz, B.; Forstmann, G.; Peng, B.;
Sheumann, V.; J ohannsmann, D. Macromolecules 1998, 31,
1930. (f) van der Vegt, N. F. A.; Muller-Palthe, F.; Gelebus,
A.; D. J ohannsman, D. J . Chem. Phys. 2001, 115, 9935. (g)
(14) Skloss, T. W.; Haw, J . F. Macromolecules 1994, 27, 6998.
(15) (a) Allen, S. D. M.; Almond, M. J .; Bruneel, J .-L.; Gilbert, A.;
Hollins, P.; Mascetti, J . Spectrochim. Acta, Part A 2000, 56,
2423. (b) Ghosh, M.; Mandal, T. K.; Chakrabarti, S.; Misra,
T. N. J . Raman Spectrosc. 1998, 29, 263. (c) Ghosh, M.;
Mandal, T. K.; Chakrabarti, S.; Misra, T. N. J . Raman
Spectrosc. 1998, 29, 807.
(16) Chakrabartk, S.; Maity, A. K.; Misra, T. N. J . Polym. Sci.,
Part A: Polym. Chem. 1992, 30, 1625.
(17) (a) Egerton, P. L.; Pitts, E.; Reiser, A. Macromolecules 1981,
14, 95. (b) Coqueret, X. Macromol. Chem. Phys. 1999, 200,
1567. (c) Chae, B.; Lee, S. W.; Ree, M.; J ung, Y. M.; Kim, S.
B. Langmuir 2003, 19, 687.
(18) (a) Sakamoto, K.; Arafune, R.; Ito, N.; Ushioda, S.; Suzuki,
Y.; Morokawa, S. J . Appl. Phys. 1996, 80, 431. (b) Stohr, J .;
Samant, M. G. J . Electron. Spectrosc. Relat. Phenom. 1999,
98, 189. (c) Stohr, J .; Samant, M. G.; Cossy-Favre, A.; Diaz,
J . Macromoleules 1998, 31, 1942. (d) Wei, X.; Hong, S.;
Zhuang, X.; Goto, T.; Shen, Y. R. Phys. Rev. E 2000, 62, 5160.
MA034445U