Polymer Blends Containing Liquid Crystals: A Review
D. DUTTA, H. FRUITWALA, A. KOHLI,” and R. A. WEISS**
Dept. of Chemical Eng. and Polymer Science Program
University of Connecticut
Storrs, Connecticut 06269-3136
This paper reviews the literature of polymer blends containing low and high
molar mass liquid crystals. Low molar mass liquid crystals have been used as
plasticizers for thermoplastic polymers and in applications such a s electro-optics,
optical recording media, and membranes. High molar mass liquid crystalline
polymers have been primarily used in polymer blends as processing aids and as
an incipient reinforcing phase for “self-reinforced” materials. This review discusses the phase behavior, rheology, and mechanical properties of these blends.
INTRODUCTION
P
olymer blends containing a liquid crystalline
polymers have received considerable attention in
the past decade, and there is now a substantial literature on this subject. Most recently, the topic that
has been most widely investigated is the development
of self-reinforcing blends in which a fibrillar morphology of a dispersed liquid crystalline polymer
phase is sought. Other studies, however, have considered thermodynamic aspects of these blends, with
particular emphasis on the assessment of theoretical
predictions of the phase behavior of mixtures of rodlike and flexible molecules.
This review surveys the published literature concerned with liquid crystalline polymer blends. Also
included are studies of blends of polymers and low
molar mass liquid crystals, for which there are several interesting applications. The subject of molecular composites, though discussed is not covered rigorously in this article.
The article is divided into two major categories: 1)
blends containing low molar mass liquid crystals
(LC’s) and 2) blends containing high molar mass
liquid crystalline polymers (LCP’s).Each topic is further subdivided in order to group studies that had
similar objectives. In many cases, a single paper is
referenced in more than one section.
BLENDS CONTAINING LOW MOLAR MASS
LIQUID CRYSTALS (LC’s)
Monomer liquid crystals (LC)have been known for
over 100 years, and there are now a number of
commercial applications based on these materials,
such as electro-optical displays, thermometry, and
radiation sensors (1). The earliest reference, however, to the addition of an LC to a polymer was by
Gebhard, et al. (2) in 1977. They were interested in
Current Address: IBM Carp.. Endicott. NY 13760-5553
**To whom correspondence should be sent.
the large increase in the orientational birefringence
of N-p-methoxy-benzylidene-p-butylaniline
(MBBA)
above its nematic to isotropic phase transition temperature when it was used to swell trans-1.5-polypentenamer. Since then, a number of other researchers have investigated LC/polymer blends. Most of the
studies have involved the thermodynamics or phase
behavior of the blends, though several have described
applications.
Thermodynamics and Phase Behavior
Flory and coworkers (3-8)provided a theoretical
framework for the thermodynamics of mixtures of
rod-like molecules and polymers of varying flexibility. Considering only the shape anisotropy of the rodlike molecules, they derived expressions for the
phase equilibria in mixtures of low molar mass mesogens with rod-like polymers, semi-rigid polymers,
and flexible coil polymers. The main conclusion of
their theory was that flexible and rigid-rod molecules
have little tendency to mix.
Brochard (9) extended Flory’s theory to networks
made of flexible chains, swollen in a nematogenic
solvent. Above the nematic-to-isotropic transition
temperature, TN+ uniaxial stretching of the gel oriented the solvent and resulted in a n increased orientational birefringence. Below TN-I,
the gel could be
maintained either by rapid cooling or if the gel were
nematic.
Ballauf (10)extended the lattice theory of mixtures
for blends of nematic LC’s and flexible polymers to
include the effect of isotropic interactions between
the components. The theory successfully described
the main features of the phase diagram, Fig. 1, for
mixtures of p-ethoxybenzylidene-bis-4-n-butylaniline (EBBA) and polystyrene (PS). At low polymer
concentrations, a two phase region consisting of a
nematic phase and an isotropic phase was observed
upon cooling the mixture from the single phase, isotropic region. Further cooling led to a triple point,
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, VOI. 30, NO. 17
1005
D. Dutta, H . Fruitwala, A. Kohli, and R. A. Weiss
N*I
3L@
\
below which another nematic phase and a polymerrich isotropic phase were in equilibrium. At higher
polymer concentrations, cooling from a n isotropic
state led first to phase separation of two isotropic
phases, followed by a triple point and finally the
coexistence of a polymer-rich isotropic phase and a
polymer-poor nematic phase.
Noel and coworkers [ 1 1, 12) constructed isobaric
phase diagrams of liquid crystalline copolyesters by
mixing them with reference LC's of known mesophases. The mesophases of the liquid crystalline
polymers (LCP's) were identified by the contact
method, and the phase diagrams were determined
using differential scanning calorimetry (DSC) and
polarized light microscopy. Griffin and Haven (13)
studied the phase behavior of a n LC/LCP blend. They
observed a eutectic point in the solid to nematic
transition at a composition of about 77 weight percent LCP. Cser, et al. (14) measured the heats of
transitions for a variety of LC's and LCP's using DSC.
The enthalpy change at the clearing point was proportional to the polymer content in the mixture, while
the enthalpy change at the solid-LC transition was
proportional to the saturation concentration of the
LC in the LCP.
George, et al. (15) constructed phase diagrams for
blends of linear thermotropic liquid crystalline polyesters with an LC of similar structure. The LC depressed the melting point of the polyester, and at
elevated temperatures transesterification occurred.
This lowered the melt viscosity and improved the
thermal stability relative to that of either of the pure
components.
Several groups have studied mixtures of side chain
LCP's and monomeric LC's. Ringsdorf, et al. (16)
studied the miscibility of mixtures of low and high
molar mass nematic liquid crystals. Miscibility depended on the chemical structure of the component
mesogens. Mixtures with unlike structures phase
separated in the nematic state, while miscibility occurred when the mesogenic side group of the polymer
was similar to the LC. In the latter case, the LC
plasticized the LCP and the crystallization of the LC
was depressed.
1006
Achard, et al. (17) observed a biphasic region for
mixtures of low molar mass nematic solvents and
liquid crystalline polymethacrylates or polyacrylates.
Miscibility in these systems depended on the chemical nature of the side chain and not on the backbone.
That is, similarities between the polymer and monomer mesogen promoted miscibility. In general, the
addition of a mesogenic solvent depressed TN-Iof
the polymer.
Similar results were reported by Sigaud, et al. (18).
who systematically studied the effect of chemical
structure of LC's on their miscibility with a side chain
LCP based on poly(methy1 siloxane) (PMS).Miscibility of the LCP in a nematic solvent depended on
the length of the aliphatic tail of the solvent and
the length of the flexible spacer and tail of the LCP
side group. In some cases, blending also induced the
formation of a smectic A phase.
Kroneberg and co-workers (19) studied the phase
behavior of EBBA with PS and PEO. The width of the
miscibility gap increased with increasing molar mass
of the polymer, and hysteresis was observed in TN.,
between heating and cooling scans. Dubault, et al.
(20)also constructed the phase diagrams of flexible
polymers in low molar mass nematic solvents. They
observed a biphasic region consisting of isotropic and
nematic phases, and its width was increased a s the
polymer molar mass increased. A homogenous nematic phase was not observed when M,,,
of the polymer was greater than 10,000.
Belfiore and coworkers (21, 22) used UNIFAC-FV
(Universal Quasi-Chemical Functional-Group Activity Coefficient including Free Volume Correction)
group contribution thermodynamics to predict the
phase behavior of blends of PEO, with p-hexyloxybenzoic acid (HOBA)and bisphenol-A-polycarbonate
(PC)with p-pentyloxycinnamic acid (POCA).Experimental results based on DSC and W N M R were used
to construct the phase diagrams. Blends of PC/POCA
were partially miscible, whereas PEG/HOBA were
completely miscible.
Dormoy, et al. (23) found that the chain conformation of polystyrene in the isotropic phase of a
nematogen solvent, pentylcyanobiphenyl, was perturbed due to coupling between the local orientational
order of the solvent and the polymer. Lipatov, et al.
(24) used a cobalt gun to irradiate phase-separated
mixtures of cholesteric LC microdomains dispersed
in a polyurethane matrix. Radiation doses varying
from 5 to 50 Mrad improved the modulus and
strength of the mixture. This was attributed to the
formation of a n additional network of microdomains
as the polyurethane crosslink density increased. Optimum strength was achieved with a 20 Mrad dose.
Using ultrasonic impedometric techniques, Martinoty, et al. (25) measured the viscosity coefficients
of dilute solutions of polystyrene in EBBA. Considerable increases in the LC viscosity were achieved by
the addition of small amounts of PS. Moreover, the
increase of the solution viscosity with decreasing
temperature was more rapid than expected from a
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 7990, Vol. 30, NO. 17
Polymer Blends Containing Liquid Crystals: A Review
classical Arrhenius dependence, which was satisfactory for describing the solution viscosity of PS in a n
isotropic solvent.
Applications
Huh and coworkers (26) used thermotropic LC's,
EBBA and terephtal-bis-4-n-butyaniline (TBBA),as
plasticizers for polystyrene. EBBA and TBBA were
miscible in PS up to LC concentrations of 20 and 1 1
percent, respectively. Plasticization was demonstrated by decreases in the glass transition temperature, Fig. 2, and melt viscosity with increasing LC
content up to the solubility limits of the LC's in the
polymer. Above the solubility limit, the LC phaseseparated and no further decreases in Tgor viscosity
were observed. TN.,of the LC's decreased as the PS
concentration increased, Fig. 2.
Similarly, Buckley, et al. (27)used LC's to plasticize
polyolefins and polyesters. The melt viscosities of
blends were reduced by as much as 25 to 30 percent
compared to the melt viscosities of the neat polymers.
Several research groups have described electrooptical display technology based on materials containing microdroplets of liquid crystals distributed in
a polymer matrix, i.e., polymer dispersed liquid crystals (PDLC) (28-32). These materials have applications in optical displays, light valve, temperature
sensors, and variable transmission windows. The
principle of operation of such devices is based on the
control of the refractive indices of the two phases,
and hence the light scattering from a PDLC film, by
application of a n electric field. In one design, Fig. 3,
in the absence of an electric field, the PDLC film
appears opaque, due to the random orientation of the
optic axes of LC droplets. This results in a mismatch
in the refractive indices of the LC and the polymer,
which is responsible for the scattering of the incident
light. When a n electric field is applied, the optic axes
of the microdroplets align in a direction parallel to
the electric field, matching the refractive indices, and
the film becomes optically transparent.
A recent paper by West (33)described the following
three ways to prepare PDLC materials: 1) the LC is
dissolved in a polymer precursor and the dispersed
LC-phase forms during polymerization, 2) the LC is
dissolved in a polymer melt and phase separation of
the droplets occurs during cooling of the melt, or 3)
the LC and the polymer are dissolved in a common
solvent followed by evaporation of the solvent, during
which the LC precipitates a s a microphase. The electro-optical properties are influenced by the droplet
morphology. For example, the scattering efficiency
of the PDLC is maximized when the droplet size is of
the order of the wavelength of light. The size of the
droplets, density, and electro-optical properties can
be controlled by varying the cure temperature, rate
of cooling, or rate of solvent removal, depending on
the method used to prepare the PDLC material.
OFF
ON
b
1
0
a
m
m
* EBB&
n
I
m
b
Fig. 2. (a)Phqse diagrams f o r TBBAIPS composite. Reprinted with permission from (26).(b)Phase diagramsf o r
EBBA/PS composite. Reprinted with permissionfrom (26).
.......................
h.i
Fig. 3. Illustration of the principle of optoelectronic operation of the PDLC films. I , = incident light, Is = scattered
light. i T = transmitted light, n, = extraordinary refractive
index, no = ordinary refractive index, n, = polymer refractive index. Reprinted with permission from (33).
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, VOI. 30, NO. 17
1007
D. Dutta, H . Fruitwala, A. Kohli, and R. A. Weiss
Mclntyre and Soane (34)recently described a blend
of EBBA and poly(methy1methacrylate) (PMMA)as a
reversible optical storage medium. This blend exhibits a n upper critical solution temperature, and
therefore, when heated the two components become
miscible. Since the blend is a glass at room temperature, the extent of phase separation that occurs
when the blend is cooled from the one-phase region,
can be controlled by varying the cooling rate-that
is, phase growth is inhibited once the blend vitrifies.
The device that these authors proposed involved a
complicated process using three different incident
light patterns to erase, write, and read the domain
pattern. This system, however, has several potential
advantages over the transition metal and rare earth
element-based systems currently being developed
in that they are less prone to oxidation and thermal expansion problems and can be more easily
fabricated.
LC/polymer blends have also been shown to have
unique transport properties. Kajiyama and coworkers (35-371 described the structure and permeation
properties of membranes made from blends of EBBA
with PVC or PC. The EBBA was molecularly dispersed
in the composite membranes containing up to 30
weight percent LC. For membranes containing 60
percent EBBA, the morphology consisted of a 3-dimensional network of polymer fibrils in a n LC continuous phase. The permeabilities of gases through
the 60 percent EBBA composite membrane were enhanced in the vicinity of the crystalline to nematic
transition of the EBBA, TK-N.
This was attributed to
the onset of molecular motion in the membrane and/
or an increase in pore formation. Permeation was
principally governed by diffusion below TK.N and solubility above T K - N .
Membranes were also prepared from blends of 4cyano-4’-pentylbiphenyl with PC (38).The application of an electric field increased the gas permeability
through the membrane, which was attributed to a
preferential orientation of the LC molecules parallel
to the field. The gas permeability was directly proportional to the magnitude of the applied voltage.
Shinkai, et al. (39),prepared ternary blend composite membranes composed of EBBA, PC, and amphiphilic crown ethers. The ion permeability was
affected by molecular motions of the LC phase. That
is, below TK.Npermeability was low, but it increased
significantly above T K - N . This phenomenon was
found to be thermally reversible and has the potential as a temperature-switchable permselective
membrane.
morphology. Each of these categories is further
subdivided.
Phase Behavior and Thermal Studies
Crystallization P henornena
Various investigators have reported that the rate
and degree of crystallization of flexible-coil polymers
was increased by the addition of an LCP (40-44).
Joseph, et al. (40, 41) studied the thermal behavior
of blends of poly(ethy1ene terephthalate) (PET)and a
liquid crystalline copolyester based on 60 mole percent parahydroxy benzoic acid (PHB) and 40 mole
percent PET (40 PET/6O PHB). The rate of crystallization increased as the LCP content increased in the
blend, which was attributed to nucleation of the PET
crystallization by the LCP. The degree of crystallinity
also increased with the addition of the LCP (Fig. 4).
Similarly, Misra and co-workers (42-43) studied
blends of PET with 2 different LCPs: one, a commercial LCP, based on 6-hydroxy-2-naphthanoic acid
(HNA) and PHB, and the other, a copolymer of PET
and PHB. Like Joseph and coworkers, they also found
that the crystallization rates for the blends were
higher than that of pure PET. Takayanagi, et al. (44)
studied blends of wholly aromatic polyamides with
aliphatic polyamides, and they suggested that the
surface of the rigid polyamide acted as a nucleating
agent for the crystallization of the matrix polymer.
Isothermal crystallization kinetics studies of nematic and smectic liquid crystalline polyesters containing polymethylene flexible spacers in the main
chain and their blends were reported by Yo0 and Kim
(45). They used differential scanning calorimetry
(DSC)and microscopy and found that regardless of
crystallization temperature, the crystallization kinetics followed the Avrami equation up to a high degree
of crystallization. Blends of nematic and smectic
BLENDS CONTAINING LIQUID CRYSTALLINE
POLYMERS (LCP)
The literature dealing with polymer/polymer
blends containing liquid crystalline polymers (LCP’s)
is divided below into three main categories: 1) phase
behavior and thermal studies, 2) rheology and
processing, and 3) mechanical properties and
iooa
0
0
2s
50
75
loo
LC POLYMER CONTENT
Fig. 4. Efect of LCP content on heat of crystallization of
PET. Reprinted with permissionfrom (401.
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, VOI. 30, NO. 17
Polymer B l e n d s Containing Liquid C r y s t a l s : A R e v i e w
effect on either of the T,'s, it did reduce the melting
temperature of the PET in the blend by 42°C.
The phase behavior of solvent cast films of blends
of 40 PET/6O PHB with PET or PC were studied by
Nakai, e t al. (58)and Kyu and Zhuang (59).The films
of PET and the copolyester were prepared by dissolving the polymers in o-chlorophenol and removing the
insoluble part by filtration (59).The as-cast films
were optically clear and homogeneous. They phase
separated, however, into a n anisotropic phase and
an isotropic phase when heated above the melting
temperature of the PET. Similarly, solvent-cast films
prepared from the LCP and PC were optically transparent (59). A cloud point curve characteristic of a
lower critical solution temperature was determined
by annealing the film at elevated temperatures and
identifying the temperature at which phase separation occurred, Fig. 6. Phase separation appeared to
occur via a spinodal decomposition mechanism.
A similar phase separation phenomenon was observed by Chuah, e t a l . (57) in films of rigid rod
molecular composites of poly(p-phenylene benzobisthiazole), (PPBT),and Nylon 66. Microscopic studies
showed that the processed films were optically clear
and homogeneous. However, the molecular composite
phase-separated above the melting temperature of
the polyamide.
Transesterification reactions increase the miscibility of a liquid crystalline co-polyesters with isotropic polyesters (62-64). Zhou, e t al. (63)found that
blends of PET and a copolyester of PHB, bisphenol A
and terephthalic acid (TA) exhibited one-phase behavior after transesterification. Paci, e t al. (64) reported that blends of PBT and a liquid crystalline
poly(bipheny1-4.4' xylene sebacate) (PB8) were miscible in the isotropic state and phase separation depended on the rate of cooling. The degree of PBT
crystallinity first increased with the addition of a s
little as 5 weight percent PB8 and then decreased
with increasing LCP. The authors attributed this to
LCPs showed that the crystallization rate of one component was significantly affected by the presence of
the second component.
A theoretical analysis of the effect of adding a nonrigid crystalline solute to a main chain LCP was done
by Dowel1 (46) using the localized mean-field, simple
cubic lattice theory. The results predicted a mixture
of two phases in which the anisotropic and isotropic
phases coexist, and the addition of the flexible solute
causes the disruption of orientational order of the
LCP. This results in a lowering of TN-,.
Miscibility of LCP Blends
The miscibility behavior of blends containing an
LCP have been widely investigated (47-68), using
DSC, scanning electron microscopy (SEM),and X-ray
diffraction (XRD). A great deal of this literature involves the miscibility of the liquid crystalline copolyesters of PET and PHB with other, flexible-chain
polymers (47, 51, 58-62). Kimura and Porter (47)
studied the phase behavior of blends of PET/PHB
with poly(buty1ene terephthalate), PBT, using DSC.
Their studies showed that the PET-rich phase of the
copolyester was miscible with PBT whereas the PHBrich phase was not. PET/PHB itself exhibited two
phases: a PHB-rich phase and a PET-rich phase.
When PC was used instead of PBT, it also was partially miscible with the PET-rich phase of the copolyester (69).According to Luise (50),
miscibility of melt
blends of thermotropic polyesters with isotropic polyesters is governed primarily by polymer asymmetry.
Blends of various PET/PHB copolyesters containing 30, 60, and 80 percent PHB with thermoplastic
polymers such as polystyrene (PS), polycarbonate
(PC), and PET were investigated by Zhuang, e t al.
(51).While PS blends were found to be completely
immiscible, the PC and PET blends were partially
miscible. Immiscibility of 40 PET/GO PHB with PC
was also reported by Nobile, e t al. (68).Hess and coworkers (60-62) however, reported that annealing
the LCP/PC blend in the melt caused transesterification reactions that resulted in a miscible blend a s
evidenced by a single Tg.Fig. 5. Although annealing
:
0
a blend of PET/PHB with PET in the melt
had no
2
2
210
0
200
0
190
0
-
'
0
%
0
0
180
0
170
160
3w J
0
10
m
30
40
so
150
60
0
Annealing time/min
Fig. 5. Effect of annealing time on miscibility of P C I P E T P H B blend. Reprinted with permission from (60).
20
40
80
80
100
PHB-PET (wt%l
Fig. 6. Cloud point p h a s e diagram of P C I P H B - P E T blends
obtained at 2"Clmin. Reprinted with permissionfrom(59).
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, VOI. 30, No. 17
1009
D. D u t t a , H . F r u i t w a l a , A. Kohli, and R . A. Weiss
a cybotactic structure observed in the quenched
blends of PBT and the polyarylate that favored the
cold crystallization of PBT. Transesterification reactions between the two polyesters improved their miscibility. In contrast, blends of a PHB-based thermotropic polyester and Nylon 12 were immiscible
(48, 49).
Miscibility studies on blends of PHB/HNA liquid
crystalline copolyesters differing in comonomer ratios were reported by DeMeuse and Jaffe (52).Using
XRD and DSC results, the authors showed that the
blends were miscible in the solid state.
Weiss, et al. (67) prepared blends of PS and a
thermotropic polyalkanoate of 4,4’-dihydroxy-a,a’dimethylbenzalazine using a two roll mill. DSC studies indicated that the blends were in general immiscible, though there may have been some marginal
miscibility of the LCP in PS of about 1 percent.
Pracella, et al. (65,66)characterized blends of PET
and poly(decamethy1ene4,4’-terephthaloyl dioxy dibenzoate) (HTH10).The phase behavior of the blends
depended on composition and thermal history. The
DSC thermograms of the blends showed a double
melting peak on second heating. The higher melting
peak was attributed to reorganization of the PBT
crystals on heating and its temperature was almost
independent of the composition of the blend. The
temperature of the lower melting peak, however, decreased with increasing HTHlO content, indicating
that HTHlO plasticized the PBT. For blends containing more than 30 weight percent HTH10, coexistence
of the crystal phases of the PBT and HTHlO was
observed. Blends quenched from the isotropic melt
showed one Tg that was intermediate between the
neat polymers, which led the authors to conclude
that PBT and the HTHlO were miscible in the amorphous phase. The presence of HTHlO in the blend
depressed the primary heterogeneous nucleation and
growth rate of the PBT spherulites and caused a
decrease in the crystallization rate of the PBT.
Seurin, et al. (53) used polarized microscopy to
construct a phase diagram of mixtures of a semi-rigid
polymer, hydroxypropyl cellulose (HPC),and a flexible polymer, polyethylene glycol (PEG).The PEG had
no effect on the orientation of the LCP. Low molar
mass PEG was miscible with HPC, which was attributed to intermolecular interactions between the two
polymers. A s the molar mass of the PEG increased,
the intermolecular interactions weakened and eventually phase separation occurred. Increasing the molar mass of the HPC simply shifted the mesophase to
isotropic transition to higher temperatures. Billard,
e t al. (54)also used polarized microscopy to construct
the phase diagrams of mixtures of cholesteric and
nematic thermotropic LCP’s. The mesophases were
identified by the contact method, and the sign of the
twist of the cholesteric phase was determined from
the displacement of the isochromatic lines (55).
The phase behavior of blends containing side chain
liquid crystalline polyesters have also been investigated using DSC, hot-stage microscopy and XRD (56).
1010
Phase Behavior of LCP Blends in a Common
Solvent
In 1978, Flory and Frost (5)proposed a theory that
predicted the phase behavior of two rod-like polymers
dissolved in a common solvent. According to this
theory, a single isotropic phase exists in very dilute
solutions. As the polymer concentration increases,
a n anisotropic phase appears that is in equilibrium
with the isotropic phase. Finally, a single anisotropic
phase is obtained with further increase in concentration. If one of the rod-like polymer is replaced by a
random coil polymer, Flory (7) predicted that above a
critical concentration the mixture separates into two
phases, a n anisotropic phase and a n isotropic phase.
The anisotropic phase is essentially comprised of the
rod-like polymer while the isotropic phase mainly
consists of the random coil polymer. Since Flory’s
work, several experimental papers (70-8 1) have described the phase behavior of rigid-rod polymer/flexible coil (or rigid-rod) polymer/solvent systems. Some
of the ternary systems studied agreed well with Flory’s predictions (72, 74-77) (Fig. 7). but others did
not (70, 71, 73, 78) (Fig. 8).
Aharoni (72) described the phase behavior of ternary mixtures of a rod-like polymer and a flexible
polymer in a common solvent. The flexible polymer
investigated was polystyrene (PS) and the rod-like
polymers were poly(alky1isocyanates). The phase
diagrams were in good agreement with Flory’s theory (5, 7). Similar agreement with theory was also
reported by Bianchi, et al. (74, 75). who studied
ternary systems of a rod-like polymer, a flexible polymer, and a solvent. The systems investigated by them
were poly(p-benzamide)(PBA)/polyterepthalamideof
p-amino benzhydroxide (~-5OO)/N,N-dimethylacetamide (DMAc) + 3 percent LiCl and PBA/polyacrylonitrile (PAN)/DMAc + 3 percent LiCl, Fig. 7. Below
a critical polymer concentration a single isotropic
phase was observed. Above this critical concentration, however, a n anisotropic phase appeared and
OMAc -3%LiCI
0.1
0.3
a5
0.7
09
Fig. 7. Ternary phase diagram comparing the experimental curveibroken lines)with Flory’s theory(full lines).
Reprinted with permissionfrom (75).
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, Vol. 30, NO. 17
Polymer Blends Containing Liquid Crystals: A Review
5
A
I
Fig. 8. P h a s e d i a g r a m s of (a)CHC13 (11, PET127 p e r c e n t
P H B c o p o l y e s t e r (21, PC (31, a n d (b) CHC13 (1). PET135
p e r c e n t P H B c o p o l y e s t e r (2).PC (3).R e p r i n t e d with perm i s s i o nf r o m (70).
the solution became biphasic. The anisotropic phase
consisted only of the high molar mass fraction rodlike polymer, and the isotropic phase contained a low
molar mass fraction of the rigid polymer in addition
to the flexible polymer. Hwang, et al. (76, 77) who
studied systems of a rigid rod poly-p-phenylene benzobisthioazole (PPBT)and a flexible coil polymer in a
solvent also reported similar behavior. Their flexible
coil polymers were poly(2,5(6)-benzimidazole)
(ABPBI)
and poly-paraphenylene-quinoxaline(PPQ).For each
solution there existed a critical concentration below
which the solution was transparent and optically
isotropic. Above this concentration, the solution became opaque or translucent and was optically anisotropic. The anisotropic PPBT domains were dispersed in an isotropic continuous phase, and when
an external shear was applied to the solution, the
anisotropic domains became elongated and aligned
along the shearing direction.
Ternary mixtures investigated by Marsano, et al.
(73)and Russo and Cao (78)did not follow the predictions of Flory and co-workers (5, 7). Marsano, et al.
(73) studied the phase behavior of two semi-rigid
polymers, cellulose acetate (CA),and HPC in DMAc.
The polymers were immiscible over the entire range
of concentration except at very dilute concentrations
where a single isotropic phase was observed. As the
overall concentration of the polymers increased, the
solution first formed two isotropic phases, one rich
in HPC and the other rich in CA. Above a critical
concentration three phases formed, an isotropic HPC
phase, a n anisotropic HPC phase, and a n anisotropic CA phase. Finally, at very high concentrations
the isotropic HPC phase disappeared, leaving only
the two anisotropic phases containing HPC and CA.
The authors attributed the immiscibility in the
anisotropic phase to unfavorable polymer-polymer
interactions and to deviations from a rod-like
conformation.
Russo and Cao (78) determined the ternary phase
diagram for poly-(7-benzyl-a,L-glutamate)/Nylon
6/
m-cresol using optical microscopy. The isotropic
phase excluded the rod-like polymer, and the phase
diagrams were significantly different from those predicted by the Flory theory. The deviations from theory
were attributed to hydrogen-bonding interactions between the three components.
Ambrosino, et al. (79) constructed ternary phase
diagram of HPC, ethyl-cellulose (EC),and acetic acid
(AA) using polarized microscopy. The cellulosic polymers exhibited cholesteric mesophases. Four zones
in the phase diagrams were identified: zone 1 was an
isotropic single phase, zone I1 was a n isotropic-isotropic biphase, zone I11 had an isotropic-anisotropic
and a n anisotropic-anisotropic biphases, and zone IV
was an anisotropic monophase. By changing the
polymer concentration in zone 111 they were able to
control the texture of each phase, and droplets of
either phase could be easily deformed by moving one
of the slides.
Wiff, et al. (82)prepared films of PPBT and Nylon
66 by casting the mixture from a common solvent,
methane sulfonic acid. DSC studies indicated that
PPBT and Nylon 66 formed a two-phase system. TGA
studies indicated that the addition of PPBT enhanced
the thermal stability of Nylon 66.
RHEOLOGY AND PROCESSING
One of the major advantages of blending LCP's with
thermoplastic polymers is that the LCP acts a s a
processing aid. A number of studies (48, 49, 51, 67,
68, 83-98) have shown that extensional flows, such
a s is encountered in the entrance of a capillary, can
preferentially orient the LCP domains in the direction
of flow. The oriented LCP domains smoothly glide
past each other, thus lubricating the polymer melt
and lowering the melt viscosity of the blend. Rheological characterization of LCP/polymer blends have
been done mainly by capillary viscometry, though
some researchers have used cone and plate or parallel
plate rheometers.
An important parameter in utilizing the LCP as a
processing aid for a conventional thermoplastic is
the temperature at which the blend is melt processed.
Cogswell, et al. (83-85) concluded that the temperature range at which conventional polymers are melt
processed should overlap with the temperature range
in which the LCP forms an anisotropic melt. The
viscosity of such melts were found to be much lower
than that of the thermoplastic polymer alone and the
processing temperatures could be reduced. The advantages of this reduction in processing temperatures are reduced energy consumption and less degradation of polymers that are sensitive to higher
temperatures. Moreover, lowering the viscosity facilitates the filling of large or complex molds.
Polymer blends using 40 PET/6O PHB have been
studied by a number of research groups (51, 68, 92,
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, Vol. 30, NO. 17
1011
D. Dutta, H . Fruitwala, A. Kohli, and R. A . Weiss
93, 98, 99). Blizard and Baird (92) studied the rheological property of blends of this LCP with PC and
Nylon 66. The measurements were made with a cone
and plate rheometer at low shear rates and with a
capillary rheometer at higher shear rates. Dynamic
oscillatory and steady shear data showed a significant reduction in the viscosity of the blends by the
addition of the LCP, Fig. 9. Similar results were
reported by Zuang, et al. (51)who studied the blends
of this LCP with PS, PC, and PET. Studies by Acierno,
et al. (93)on blends of 40 PET/6O PHB with PC also
showed a decrease in viscosity with the addition of
the LCP. In contrast to these studies, however, Nobile,
et al. (68)reported that at low shear rates (below0.3/
s) the complex viscosity '7 increased with increasing
LCP content for blends of 40 PET/6O PHB and PC.
At higher shear rates v* decreased with increasing
LCP concentration.
A reduction in the melt viscosity of blends of this
LCP with chlorinated poly(viny1 chloride) was observed by Lee (98).These blends also showed a decrease in the extrudate swell and a n increase in the
spiral mold flow a t low LCP concentrations.
The viscoelastic behavior of blends of 40 PET/6O
PHB with a styrene-butadiene copolymer was reported by Lorenzo, et al. (99).They found that the
barrel temperature had a significant effect on the
melt flow index (MFI) and q* of the blends. For example, when the blends were extruded below the
melting temperature of the LCP, 7" was independent
of LCP concentration. However, when the extrusion
was done above the melting point of the LCP, both
the MFI and 7' exhibited a minima at 10 percent LCP
concentration.
Another liquid crystalline co-polyester that has
been widely used in blends is the copolyester of
PC BLENDS (L/D=75.11,2 6 O o C , lOOO/src)
o
10
20
30
40
so
60
70
eo
so 100
PEQCENT 60 PHB-P E T
Fig. 9. Melt viscosity of LCPIPC blends as a function of
LCP content. Reprinted with permissionfrom (92).
1012
6-hydroxy-2-naphthanoic acid (HNA) and PHB (5:-,
86-88, 97, 98, 100-102). Several grades of this
type of LCP have been commercialized by Hoechstc
Celanese Corp. under the trade name Vectra and by
Imperial Chemical Industries, Ltd. under the trade
name Vitrex. Some of these LCP compositions also
incorporate aromatic ester groups derived from monomers such as terephthalic acid (TA) and hydroquinone (HQ).
Siegmann and co-workers (86)studied the rheological behavior of blends of an HNA/PHB LCP with a n
amorphous polyamide using a capillary rheometer.
All the melts exhibited non-Newtonian flow behavior,
which could be described by a power-law constitutive
equation over a limited shear rate range. The viscosities of the blends were much lower than those of the
thermoplastic polymer. The addition of as little as 5
weight percent LCP resulted in a large reduction in
viscosity from that of the pure polyamide.
Swaminathan and Isayev (88) reported similar
results for blends of a n HNA/PHB copolyester with
poly(ether amide). A reduction in viscosity of
poly(ether sulfone) (PES)by the addition of a n HNA/
PHB LCP was reported by James, et al. (90)and Froix,
et al. (91).James and coworkers (90) characterized
the rheological behavior of the blends by plotting
dynamic viscosity us. shear rate. About a four fold
drop in viscosity was observed with the addition of
only 2 weight percent LCP. The shape of the flow
curves for blends with up to 20 weight percent LCP
resembled that of pure PES, and they were Newtonian until about 16 s-'. The viscosity curve of the 50
percent LCP blend showed shear thinning behavior
and was similar to that of the pure LCP. Chung (4849) found a minimum in viscosity at a concentration
of 10 weight percent LCP and a maximum a t 20
percent LCP for blends of HNA/PHB and Nylon 12.
According to Chung, below 10 percent LCP the LCP
domains were well dispersed in the Nylon 12 matrix
and acted a s a lubricant, thereby reducing viscosity.
At 20 weight percent LCP, however, an interlocking
morphology developed which resulted in a sharp increase in viscosity.
Several research groups have studied blends of
HNA/PHB type copolyesters with PC. Isayev and
Modic (87) observed a crossover point in the flow
curves of the PC and the LCP melts, and this crossover point had a special significance in that maximum fibrillation of a blend occurred at this point
during flow. Malik, et al. (101)studied similar blends
and found that the melt viscosities were similar to
that of the unblended PC, but more shear thinning
and less viscous. They also investigated time-dependent behavior of the blends by solid-state relaxation
measurements and found that the relaxation modulus increased with the addition of the LCP.
Kohli, et al. (102) studied blends of PC and a lower
melting point LCP based on HNA/PHB/TA/HQ and
found that the addition of about 5 percent LCP resulted in a 68 percent drop in viscosity. Moreover,
the viscosity decreased monotonically with increas-
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, Vol. 30, NO. 17
Polymer Blends Containing Liquid Crystals: A Review
3 g LCP content over the entire range of blend composition. They observed that the general shape of the
viscosity us. shear rate curve was dominated by the
continuous phase of the blend, Fig. 10. At low LCP
concentrations, the PC constituted the continuous
phase. Here, the flow curves were similar to that of
PC, except that shear thinning occurred a t lower
shear rates as the LCP concentration increased. A
phase inversion occurred at about 40 to 50 percent
LCP content and above this composition the flow
curves were more like that of the LCP. No Newtonian
region was observed and the viscosities were strongly
shear thinning.
Most investigators have reported a decrease in viscosity of blends with the addition of a n LCP. Weiss,
et al. (67,94-97),however, reported that the addition
of a thermotropic LCP based on 4,4’-dihydroxy,a,a’dimethylbenzalazine to P S raised both the steady
shear and dynamic viscosities a t low shear rates
(1 s-’) or frequencies. This increase in viscosity was
attributed to tumbling and rotation of phase-separated LCP domains. At higher shear rates, however,
the viscosity decreased with increased LCP content.
For example, a composition of 10 weight percent LCP
resulted in about a 40 percent reduction in viscosity.
The lowering of viscosity at high shear rates was
explained by a n extensional deformation of the LCP
domains during flow in the entrance region of the
capillary viscosometer. An important point in this
research was that the lower shear rate data were
obtained with a cone and plate rheometer and, therefore, the blends experienced only simple shear flow.
The increase in viscosity under these conditions was
consistent with the theoretical predictions for the
simple shear flow of two-phase fluids. The higher
viscosity data required the use of a capillary viscometer, and in this case, the simple shear flow in the
capillary is preceded by a n extensional flow in the
entrance region. The extensional flow can align and
deform the LCP domains in the flow direction, and it
I
2
I
1
I
1
I
0
I
2
3
Fig. 10. Viscosity us. s he ar rate d a t a f o r LCPIPC blends
a t 2 7 0 ° C . (0) 0, (015. RI 10,(01 20,(el 40.(A1 60,I*) 80.
(Vj 100% LCPj. Reprinted w i t h permissionf r o m (1021.
is this change in the melt morphology that is largely
responsible for the decrease in the blend viscosity.
Most studies of LCP-containing polymer blends
have used a thermotropic LCP together with a flexible, thermoplastic polymer. In contrast, DeMeuse and
Jaffe (52) studied blends in which both the components were LCP copolyesters based on PHB and HNA.
The rheological behavior of the blends depended on
the composition of the individual copolymers. For
example, when the compositions of the copolymers
were similar, the viscosity of the blend followed a
rule of mixtures relation similar to that for a miscible
blend,
In t = wlln t1 + wzln tz
(1)
where 9 is the viscosity of the blend, w, and w1are
the mass fractions of the two polymers, and t1 and
92 are the viscosities of the individual components in
the blend. When the difference in the composition of
the copolymers was significant, the viscosity of the
blend followed a n inverse rule of mixtures expression
formulated for immiscible blends,
110 = w1/q1 + wzltz
(2)
The rheology of blends of rod-like and flexible polymers was modeled by Kim, et al. (103).They studied
the blends below a critical concentration that was
defined as the concentration above which phase separation occurred. The rheological behavior of these
blends could be represented by a single integral constitutive equation of the BKZ type.
MECHANICAL PROPERTIES AND MORPHOLOGY
One of the primary objectives of blending LCPs
with thermoplastic polymers has been to use the LCP
as a reinforcement for flexible thermoplastic polymers. Mechanical properties of blends have been
reported in a number of studies (48. 49, 67, 86-89,
92.93-97, 102, 104-109). Most of these researchers
have attempted to explain the changes in mechanical
properties in terms of the morphology of the LCP
domains in the blends. The most widely used technique to study the morphology has been scanning
electron microscopy, (SEM),though XRD and SAXS
have also been employed (68, 100, 110, 1 11) to find
the degree of order of the LCP phase in the blend.
In nearly all the studies of LCP/polymer blends,
the two component polymers were immiscible. The
size, shape, and distribution of the LCP phase
depended on many factors such a s composition,
processing conditions, viscosity ratio of the component polymers, and the rheological characteristics of
the matrix polymer (i.e., viscoelastic, pseudoplastic,
or Newtonian). For example, Acierno, et al. (93)found
that for a blend of 60 PHB/40 PET and PC containing
10 percent LCP, extrudates at 260°C contained
spherical LCP domains dispersed in the PC matrix.
This was true even for drawn blends. But blends
drawn at 2 10°C had fibrillar LCP domains oriented
in the direction of flow. The difference in the mor-
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 7990, Vol. 30, No. 17
1013
D . D u t t a , H . F r u i t w a l a , A. Kohli, and R. A. W e i s s
phologies was due to different viscosity ratios a t the
two temperatures. Similar results were also obtained
by Nobile, e t al. (68) with the same system. They
reported that although the tensile modulus increased
with draw ratio for fibers drawn at 220°C there was
very little effect of draw ratio for fibers drawn at
260°C. In general, drawing or stretching a blend
extrudate results in the formation of LCP microfibrils that are highly oriented in the direction
of stretch (67, 84, 97, 102, 109, 112), Fig. 1 1 .
Ramanathan, e t al. (113),however, studied blends
of poly(pheny1ene sulfide) and a variety of LC copolyesters and found that the shape of the LCP domains (i.e. droplets us. fibrils) in a n extrudate depended on the composition of the LCP and not on
draw ratio.
Increased orientation of the dispersed LCP microfibrils formed by drawing can result in greatly improved mechanical properties of the blend (68, 101,
102, 104, 109, 112, 114, 115). Malik, e t al. (101)
studied blends of HNA/PHB and PC and found that
the modulus of the pure LCP extruded from a capillary viscometer increased up to shear rate of 150 s-’,
but then decreased above this deformation rate.
Wellman, e t al. (104) investigated the reinforcement of the flexible coil-like, amorphous polymer,
ABPBI, with the rod-like polymer, PPBT. Stretching
a blend film resulted in a n increase in the modulus.
Blends containing up to 30 percent PBT had saucershaped domains, which the authors concluded were
undesirable. In a n attempt to remove the LCP aggregates, a blend containing 60 percent PPBT was made.
The modulus in this case did not change much upon
stretching. The high LCP content made it difficult to
control the morphology and failed to improve the
orientation of the domains. Similar studies (114, 1 15)
were carried out by blends containing a variety of
rod-like aromatic heterocyclic polymers and a flexible
coil-like amorphous polymer. Thin films were obtained by vacuum casting and by precipitation from
dilute solutions. The modulus increased on stretching when the rod-like polymer was present as a minor
component in the blend.
Weiss, e t al. (109)observed a n order of magnitude
increase in modulus a s the draw ratio was increased
from 1 to 1000 for fibers extruded from blends of a n
HNA/PHB/HQ/TA LCP with PC, Fig. 12. The enhancement of the modulus due to melt drawing was
Draw Ratio
- 50 (2KXI
Draw Ratio
- 500 (2Kx]
Fig. 11. Microphotographs showing the effect of draw on
the morphology of 10 percent LCPISO percent PC blend.
Reprinted with permission from (1 02).
1014
4
D
2
o l . ,
0
200
.
,
400
. ,
SO0
. , . I
800
1000
DRAW RATIO
Fig. 12. Effect of draw ratio (DRJon the tensile moduli of
20 percent LCPI80 percent PC blend.
a result of improved orientation and fibrillation of
the dispersed LCP domains. They also found that the
LCP chains within the domains were highly oriented
in the direction of flow (109).
Sun, et al. (112) extruded sheets of PPS and 40
PET/6O PHB and 20 percent PET/80 percent PHB
copolyester and PEI/Thermoplastic polyimide and
HNA/PHB LCPs and found that the modulus increased with drawing. SEM micrographs revealed
fibrillar morphologies of the LCP domains. A high
degree of molecular orientation within the LCP was
also confirmed by XRD.
A number of studies have considered the effect of
LCP concentration on the mechanical properties and
morphology of blends (43, 51, 66, 68, 86, 87, 100,
101, 108, 113, 116, 117). KO and Wilkes (100) reported that the modulus of blends of HNA/PHB and
PET increased only at high LCP content (SO percent
LCP). When this same LCP was blended with a n
amorphous polyamide, however, the mechanical
properties increased monotonically with increasing
LCP content (86). Fig. 13. For blends of 40 PET/6O
PHB and PET, Bhattacharya, e t a l . (43) found that
the initial modulus and tenacity of melt spun fibers
increased with increasing LCP content, while elongation to break decreased. A s the LCP content was
increased, the mode of failure changed from brittle
to ductile.
Zhuang, e t al. (51) found that adding small
amounts of 40 PET/6O PHB to PET, PC. or PS increased the modulus and tensile strength of compression molded films, extrudates, and melt spun filaments. Similarly, Froix and Park (116) showed that
the mechanical properties of PC were improved by
incorporating melt processible, wholly aromatic liquid crystalline polyesters. Amano and Nakagawa
( 108) investigated the drawing behavior of blends of
PET and a 40 PET/6O PHB copolyester using conventional and microwave heating. The optimum drawing
tension, drawing temperature, and draw ratio necessary for attaining an optimum modulus decreased
with increasing LCP concentration.
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, VOI. 30, NO. 17
Polymer Blends Containing Liquid Crystals:A Review
8
5
I0
0
15
20
25
0
LCP content (%)
F i g . 13. Effect of LCP content on elastic modulus and
ultimate strength of injection molded LCPIPA blends. Reprinted with permissionfrom (861.
Isayev and Modic (87) reported that injection
molded and extruded blends of a n HBA/HNA and PC
containing greater than 25 weight percent LCP had
spherical LCP domains dispersed in the PC matrix.
Blends with 10 percent LCP content, however, had a
fibrillar LCP morphology. A similar observation was
also reported by Pracella, et al. (66). who examined
the morphology of blends of PBT and a liquid crystalline polyester, decamethylene 4,4’-terephthaloydioxy dibenzoate, quenched from melt. The fracture
surfaces of blends containing up to 50 weight percent
LCP revealed the presence of rod-like structures of
the LCP component oriented perpendicular to the
fracture plane. For blends containing more than 50
weight percent LCP, the morphology was homogeneous and LCP domains were not observed.
Skin-core morphologies are common in blends of
LCP’s and thermoplastic polymers (48, 49, 67, 86,
88, 90, 92, 97, 107, 114, 118), and they play a n
important role in determining the properties of extruded and injection molded samples. Normally,
LCP’s in the skin have a higher degree of orientation
than in the core when the blends are extruded or
injection molded (86, 90, 106). Baird and coworkers
(92, 107) observed a skin-core morphology in blends
of Nylon 66 with HBA/HNA and 40 PET/6O PHB and
20 percent PET/80 percent PHB copolyesters. In both
systems more LCP fibers were present in the skin
than in the core. Swaminathan and Isayev (88)also
observed a shell-core structure in the fracture surfaces of injection molded blends of a n HNA/PHB
liquid crystalline copolyester and poly(etherimide).
Weiss, et al. (67, 97). however, found a higher
concentration of LCP microfibrils near the core in
blends of PS and LCP based on 4,4’-dihydroxy-a,a’dimethylbenzalazine. They attributed this behavior
to a two way migration effect in which drops initially
at the center of the tube migrated towards the wall
and drops initially at the walls migrated towards the
tube axis. With increasing shear rates, the inward
migration dominated and the equilibrium position
shifted towards the tube axis. A similar phenomenon
was observed by Chung (48, 49) in his studies of
blends of copolyesters and Nylon 12. For melt drawn
blends of this LCP with PES, however, a n LCP phase
oriented in the draw direction was found in the skin
region (90).
The fact that LCPs can act as a reinforcing agent
in a blend has led some investigators to model the
mechanical behavior of the blends according to composite theories (87, 102).Isayev and Modic (87)based
their calculations on the assumption that blends of
PHB/HNA and PC could be treated as unidirectional
fiber-reinforced composites. Their experimental data
approached the calculated properties. Kohli, et al.
(102) in their studies on blends of PHB/HNA/HQ/TA
and PC showed that the moduli of highly drawn melts
could be modeled effectively by the simple rule of
mixtures:
where E, is the composite modulus, El and E2are the
moduli of the reinforcing LCP and the matrix, respectively, and V1 and Vz are their volume fractions,
Fig. 14. A major reason for the good agreement in
this study between the data and the model was the
high alignment of the LCP microfibrills LCP in the
draw direction and their high aspect ratios. They also
POLYMER ENGlNEERlNG A N D SCIENCE, MID-SEPTEMBER 1990, V O ~30,
. NO. 17
1015
D. D u t t a , H . F r u i t w a l a , A. K o h l i , and R. A. Weiss
1) it reduced phase separation between the thermotropic oligomer and the polymer and 2) it helped
improve the dispersion of the oligomer in the polymer. The resulting blends had increased tensile modulus, tensile strength, and abrasion resistance than
when the particulate was not used.
A
CONCLUSIONS
w
‘
0
02
0.4
0.6
0.8
1.0
VLCP
Fig. 14. Moduli us. uolumef raction LCP data of compression molded (0)and melt drawn samples (0)of PCILCP
blends compared with theoretical predictionsf r o m composite analogy (solid lines). Reprinted with permission
from (102).
modeled the modulus of compression molded samples
where the LCP phase was spherical according to the
inverse rule of mixtures:
(4)
Takayanaki, et al. (38, 44, 1 1 9 , 1 2 0 ) investigated
blends of rigid, wholly aromatic polyamides such as
poly(p-phenylene terephthalamide) (PPTA), poly-pbenzamide (PBA) with aliphatic nylons, PVC, NBR,
and ABS. The Young’s modulus and yield stress increased when the rigid polyaramides were dispersed
in the aliphatic polyamides (38).The modulus and
yield stress increased with increasing molar mass of
the polyaramide, but the elongation to break decreased. This was due to the formation of more perfect and stronger microfibrils with increasing molar
mass which restrained the ductile deformation of the
blend. In a n attempt to improve the ultimate elongation block copolymers of a wholly aromatic polyamide
and a n aliphatic polyamide were blended with the
aliphatic polyamide. The elongation to break increased while the high modulus and strength were
retained. PPTA and ABS blends had superior mechanical properties than the ABS alone (1 1 9 ) . Outstanding reinforcement at concentrations less than
5 percent of the rigid component was obtained. This
was due to the formation of a microfibrillar network.
The properties of the network were anisotropic, and
a quasi-three-dimensional lattice model was proposed by Takayanagi (120) to explain the anisotropy.
One of the drawbacks of using LCP’s a s reinforcements is their poor adhesion to the matrix. To overcome this problem Akkapeddi, et al. (12 1) described
a process of blending thermotropic oligomers and
isotropic polymers in the presence of a particulate
material such a s talc, silica, carbon black, and
quartz. The particulate material served two purposes:
1016
LCP/polymer blends may represent a feasible
method for exploiting some of the desirable features
of LCPs, but at a reduced cost. The addition of small
amounts of a n LCP to a thermoplastic polymer can
result in significant improvements in processability.
At the same time, a suitable processing history, i.e.,
one that includes extensional flow, can yield a reinforcing, microfibrillar morphology of the dispersed
LCP-phase. Moduli similar to those of short glass
fiber-reinforced plastics have been achieved. One
deficiency of these blends is the poor interphase
adhesion, which limits the strength of the material.
It is also difficult to prepare samples that contain
other than a unidirectional microfibril orientation.
This limits the applications to those requiring high
mechanical anisotropy. Future research is expected
to be focused at improving these deficiencies.
NOMENCLATURE
ABPBI = Poly(2,5(6))-benzimidazole.
= Acrylonitrile-butadiene-styrene.
CA
= Cellulose acetate.
DMAC = N,N-dimethyl acetamide.
EBBA = P-eihoxy benzylidene-bis-4-n-butylaniline.
HNA
= 6-hydroxy-2-naphthanoic acid.
HPC
= Hydroxy propyl cellulose.
HOBA = P-hexyloxybenzoic acid.
HQ
= Hydroquinone.
MFI
= Melt flow index.
NBR
= Acrylonitrile-butadiene rubber.
PC
= Polycarbonate.
PEI
= Polyetherimide.
PET
= Poly(ethy1eneterephthalate).
PEG
= Poly(ethy1eneglycol).
PMMA = Poly(methy1methacrylate).
PAN
= Polyacrylonitrile.
PPTA = Poly(p-phenylene terephthalamide).
PPBT = Poly-p-phenylenobenzobisthiazole.
POCA = P-pentyloxy cinnamic acid.
PS
= Polystyrene.
PVC
= Poly(vinylch1oride).
TA
= Terephthalic acid.
TGA = Thermogravimetric analysis.
TK-N = Crystal to nematic transition temperature.
TN.,
= Nematic to isotropic transition temperature.
o*
= Complex viscosity.
ABS
REFERENCES
1 . G. W. Gray, ed., Thermotropic Liquid Crystals, John
Wiley and Sons, Chincester (1977).
2. G. Gebhard. G. Rehage, and J. Schwartz. Non-Crystalline Solids, 81,177 (1977).
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, Vol. 30, NO. 17
Polymer Blends Containing Liquid Crystals: A Review
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
P. J . Flory, Macromolecules, 11, 11 19 (1977).
P. J. Flory, Macromolecules, 11, 11 12 (1977).
P. J . Flory. Macromolecules, 11, 1126 (1977).
P. J. Flory, Macromolecules, 11. 1134 (1977).
P. J . Flory. Macromolecules, 11, 1138 (1977).
P. J. Flory, Macromolecules, 11. 1141 (1977).
F. Brochard, J.P h y s . , 40, 1049 (1979).
M. Ballauff, Molec. C r y s t . Liq. C r y s t . , 4, 15 (1986).
C. Noel and J. Billard, Molec. C r y s t . Liq. C r y s t . L e t t . .
41, 269 (1978).
B. Fayolle, C. Noel, and J. Billard, J . P h y s . , C3, 485
(1979).
A. C. Griffin and S. J. Havens, J. Polyrn. Sci., Polyrn.
Lett., 18, 259 (1980).
F. Cser, K. Nyitrai. G. Hardy, J. Menezel, and J. Vorga,
J . Polyrn. Sci., Polyrn. Syrnp., 69, 91 (1981).
E. R. George. R. S. Porter, and A. C. Griffin, Molec.
C r y s t . Liq. C r y s t . , 110, 27 (1984).
R. Ringsdorf, H. W. Schmidt, and A. Schneller, M a k rornol. Chern.. R a p i d Cornrnun., 3 , 745 (1982).
M. F. Achard, G. Sigaud, P. Keller, and F. Hardouin.
Int. Conf. Liq. Cryst. Polym., Bordeaux, France, Paper
11P3 (July 1987).
G. Sigaud, M. F. Achard, F. Hardouin. and H. Gasparous. Molec. C r y s t . Liq. C r y s t . , 155, 443 (1988).
B. Kronberg, I. Bassignana. and D. Patterson, J. P h y s .
Chern., 8 2 (15) 1714 (1978).
A. Dubault, C. Casagrande, and M. Veyssie, Molec.
C r y s t . Liq. C r y s t . , 72, 189 (1982).
L. A. Belfiore. Am. Chern. SOC., D i v . Polyrn. Chern.,
Polyrn. Prepr., 28, 158 (1987).
A. A. Patwardhan and L. A. Belfiore, Polyrn. Eng. Sci.,
28, 916 (1988).
Y. Dormoy. J. L. Gallani, and P. Martinoty, Int. Conf.
Liq. Cryst. Polym., Bordeaux, France, Paper 13P5
(July 1987).
Yu. Lipatov, V. V. Tsukruk, V. V. Shilov, and G. Ya.
Boyarski, Polyrn. Sci. U.S.S.R., 28, 60 (1986).
P. Martinoty, A. Dubault, C. Casagrande, and M. Veyssie, J. P h y s . L e t t . . 44, 935 (1983).
W. Huh, R. A. Weiss, and L. Nicolais, Polyrn. Eng.
Sci.. 25, 779 (1983).
A. Buckley, A. B. Conciatori, and G. W. Calundann,
U S . Patent, 4.434.262 (1984).
J. W. Doane, N. A. Vaz, B. G. Wu. and S. Zumer, A p p l .
P h y s . L e t t . , 48, 269 (1986).
N. A. Vaz, G. W. Smith, and G. P. Montgomery, Jr.,
Eleventh International Liquid Crystal Conference,
Berkeley, Calif. (June 1986).
G. P. Montgomery, J r . and N. A. Vaz, A p p l . O p t i c s ,
26, 738 (1987).
P. S. Drzaic, J. A p p l . P h y s . , 60, 2142 (1986).
Search, Tech. Information, GM Laboratories, 22
(1987).
J . L. West, Molec. C r y s t . Liq. C r y s t . . 157,427 (1988).
W. D. Mclntyre and D. S. Soane, A m . Chern. SOC..D i v .
Polyrn. Chern., Polyrn. Prepr.. 29, 197 (1988).
T. Kajiyama, Y. Nagata, S. Washizu, and M. Takayanagi, J. Mernbr. Sci., 11,39 (1982).
T. Kajiyama, S. Washizu, and M. Takayanagi. J. A p p l .
Polyrn. S c i . . 29, 3955 (1984).
S. Washizu, I. Terada, T. Kajiyama, and M. Takayanagi, Polyrn. J.,16, 307 (1984).
T. Kajiyama, S. Washizu, A. Kumano, I. Terada, M.
Takayanagi, and S. Shinkai, J. A p p l . Polyrn. Sci.,
A p p l . Polyrn. Syrnp., 41, 327 (1985).
S. Shinkai, S. Nakamura, K. Ohara, S. Tachiki. 0.
Manabe, and T. Kajiyama, Macromolecules, 20, 21
(1987).
E. G. Joseph, G. L. Wilkes, and D. G. Baird, in P o l y m e r
Liquid C r y s t a l s , A. Blumstein. ed., Plenum Press, NY
(1984).
E. G. Joseph, G. L. Wilkes. and D. G. Baird. A m . Chern.
SOC., Div. Polyrn. Chern.. Polyrn. Prepr.. 24, 304
(1983).
42. S. K. Sharma, A. Tendolkar, and A. Misra, Molec.
C r y s t . Liq. C r y s t . , 157. 597 (1988).
43. S. K. Bhattacharya, A. Tandolkar, and A. Misra, Molec. C r y s t . Liq. C r y s t . , 153, 501 (1987).
44. M. Takayanagi, T. Ogata, M. Morikawa. and T. Kai, J .
Macrornol. Sci.. P h y s . , B17 (4). 591 (1980).
45. Y . D. Yo0 and S. C. Kim, Polyrn. J., 20, 12, 1117
(1988).
46. F. Dowell. Molec. C r y s t . Liq. C r y s t . , 155,457 (1988).
47. M. Kimura and R. S. Porter, J . Polyrn. S c i . , Polyrn.
P h y s . E d . , 22, 1697 (1984).
48. T. Chung, P l a s t . E n g . . 43, 39 (1987).
49. T. Chung, SPEANTEC T e c h . P a p e r s , 33,1404 (1987).
50. R. R. Luise, Polyrn. Liq. C r y s t . . Faraday Diss., 79.
290 (1985).
51. P. Zhuang, T. Kyu, and J . L. White, SPE ANTEC T e c h .
P a p e r s , 34. 1237 (1988).
52. M. T. DeMeuse and M. Jaffe, Molec. C r y s t . Liq. C r y s t . ,
157, 535 (1988).
53. M. J . Seurin. J. M. Gill, A. T. Bosch, and P. Sixou,
P o l y m e r , 25, 1073 (1984).
54. J. Billard, A. Blumstein, and S. Vilasagar, Molec.
C r y s t . Liq. C r y s t . , 72, 163 (1982).
55. J. Malthete, N. Isaert, and B. Soulestein, Molec. C r y s t .
Liq. C r y s t . , 27, 321 (1976).
56. B. Bresci, M. Paci, D. Lupinacci, and P. L. Magagnini,
Int. Conf. Liq. Cryst. Polym., Bordeaux, France, paper
19P3 (July 1987).
57. H. H. Chuah, T. Kyu, and T. Halminiak, P o l y m e r , 28,
2 130 (1987).
58. A. Nakai, T. Shiwaku, H. Hasegawa, and T. Hashimoto, Macromolecules, 19. 3008 (1986).
59. T. Kyu and P. Zhuang, Polyrn. Cornrnun., 29, 4, 99
(1988).
60. K. Friedlich, M. Hess, and R. Kosfield, Makrornol.
Chern., Macrornol. Syrnp., 16, 251 (1988).
61. R. Kosfield, K. Friedlich, and M. Hess. 18th Europhysics Conference on Macromolecular Physics, 1l c ,
81 (1987).
62. M. Hess. K. Friedlich, and R. Kosfield, Int. Conf. Liq.
Cryst. Polym., Bordeaux, France, Paper 7P8 (July
1987).
63. Z. Zhou, X. Wu. and M. Yu, 18th Europhysics Conference on Macromolecular Physics, 1l c , 8 2 (1987).
64. M. Paci, C. Barone, and P. L. Magagnini, J. Polyrn.
Sci., Polyrn. P h y s . , 25, 1595 (1987).
65. M. Pracella. D. Dainelli, G . Galli, and E. Chiellini,
Makrornol. Chern., 187. 2387 (1986).
66. M. Pracella, E. Chiellini. G . Galli, and D. Dainelli,
Molec. C r y s t . Liq. C r y s t . , 153. 525 (1987).
67. R. A. Weiss, W. Huh, and L. Nicolais. Polyrn. Eng.
Sci., 27.684 (1987).
68. M. R. Nobile. E. Amendola, L. Nicolais. D. Acierno,
and C. Carfagna, Polyrn. Eng. Sci., 29, 244 (1989).
69. S. Huang, A. C. Griffin. and R. S. Porter, Polyrn. Eng.
S c i . . 29, 55 (1989).
70. F. Schubert, K. Friedlich, M. Hess, and R. Kosfield,
Molec. C r y s t . Liq. C r y s t . , 155, 477 (1988).
71. R. Kosfield, K. Friedlich, and M. Hess, 18th Europhysics Conference on Macromolecular Physics, 1l c ,
30 (1987).
72. S. M. Aharoni, P o l y m e r , 21, 21 (1980).
73. E. Marsano, E. Bianchi. and A. Ciferri, Macrornolecules, 17, 1126 (1984).
74. E. Bianchi. A. Ciferri, and A. Tealdi, Macromolecules.
15, 1268 (1982).
75. E. Bianchi, A. Ciferri, G. Conio, E. Marsano, and A.
Teladi, Macromolecules, 17, 1526 (1984).
76. W. F. Hwang. D. R. Wiff, C. L. Benner, and T. E.
Helminiak. J. Macrornol. Sci.. P h y s . . B22.231 (1983).
77. W. F. Hwang, D. R. Wiff, and C. Verschoor, P o l y m .
Eng. Sci.. 23, 789 (1983).
78. P. S. RussoandT. Cao. Molec. C r y s t . Liq. C r y s t . , 157,
501 (1988).
79. S. Ambrosino, T. Khallala, M. J. Seurin, A. T. Bosch,
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, Vol. 30, NO. 77
1017
D. D u t t a , H . F r u i t w a l a , A. Kohli, a n d R. A. Weiss
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
1018
F. Fried, P. Maissa, a n d P. Sixou. J . Polyrn. Sci.,
Polyrn. Lett., 25, 351 (1987).
S. Ambrosino, A. T. Bosch, a n d P. Sixou, 18th Europhysics Conference on Macromolecular Physics, 1lc,
83 (1987).
S. Ambrosino a n d P. Sixou, Int. Conf. Liq. Cryst.
Polym., Bordeaux, France. paper 6P3 (July 1987).
D. R. Wiff, W. F. Hwang, H. H. Chuah, a n d E. J.
Soloski, Polyrn. Eng. Sci., 27, (1987).
F. N. Cogswell. B. P. Griffin, a n d J. B. Rose, U S .
Patent, 4,386,174 (1983).
F. N. Cogswell, B. P. Griffin, a n d J . B. Rose, U S .
Patent, 4,433,083 (1984).
F. N. Cogswell, B. P. Griffin, a n d J. B. Rose, U.S.
Patent, 4,438,236 (1984).
A. Siegmann, A. Dagan, a n d S. Kenig. Polymer, 26,
1325 (1985).
A. I. Isayev a n d M. J . Modic, Potyrn. Cornpos.. 8, 158
(1987).
S. Swaminathan a n d A. 1. Isayev, Proc. ACS Div. of
Polym. Mater., 57, 330 (1987).
G. Kiss, Polyrn. Eng. Sci., 27, 410 (1987).
S. G. J a m e s , A. M. Donald, and W. A. Macdonald,
Molec. Cryst. Liq. Cryst., 153, 491 (1987).
M. F. Froix. M. Park, a n d N. Trouw, U S . Patent
4,460,736 (1984).
K. G . Blizard a n d D. G. Baird, Polyrn. Eng. Sci., 27,
653 (1987).
D. Acierno, E. Amendola, C. Carfagna, L. Nicolais,
a n d R. Nobile, Molec. Cryst. Liq. Cryst., 153, 533
(1987).
R. A. Weiss, W. Huh, a n d L. Nicolais, Int. Conf. Liq.
Cryst. Polym.. Bordeaux, France, paper 8P8 (1987).
R. A. Weiss, W. Huh, L. Nicolais, a n d A. Kohli, Conference on Emerging Technologies in Materials,
AIChe, paper CO1.50 (August 1987).
R. A. Weiss, W. Huh, L. Nicolais, a n d P. Yanisko, S P E
RETEC Tech. P a p e r s , p 267, 1987.
R. A. Weiss, W. Huh, a n d L. NicQlais, i n High Modulus
Polymers, A. E. Zachariades a n d R. S. Porter, eds., p.
145, Marcel Dekker Inc., New York (1988).
B. Lee, SPEANTEC Tech. P a p e r s , 34, 1088 (1988).
L. Lorenzo, S . K. Ahuja, a n d H. Chang, Am. Chern.
Soc., Div. Polyrn. Chern., Polyrn. Prepr., 29, 488
(1988).
C. U. KO and G. L. Wilkes, J. Polyrn. Sci., 37, 3063
(1989).
101. T. M. Malik, P. J . Carreau. a n d N. Chapleau, Polyrn.
Eng. Sci., 29, 600 (1989).
102. A. Kohli, N. Chung, a n d R. A. Weiss, Polyrn. Eng. Sci..
29, 573 (1989).
103. C. Kim, V. Sullivan, a n d G. Berry, SPE ANTEC Tech.
Papers, 34, 990 (1988).
104. M. Wellman, G. Hussman, A. K. Kulshreshtha, T.
Helminiak. D. Wiff, C. Benner, a n d W. F. Hwang, ACS
Organic Coat. Plast. Chern., 43, 783 (1980).
105. 0.A. Nehme, C. A. Gabriel, R. J. Farris. E. L. Thomas,
a n d M. F. Malone, J . Appl. Polyrn. Sci., 35, 1955
(1988).
106. E. G. Joseph, G. L. Wilkes, a n d D . G. Baird, Am. Chern.
Soc., Div. Polyrn. Chern., Polyrn. Prepr., 25, 94
( 1984).
107. R. Ramanathan. K. G. Blizard, a n d D. G. Baird, S P E
ANTEC Tech. P a p e r s , 33, 1399 (1987).
108. M. Amano a n d K. Nakagawa, Polymer, 28,263 (1987).
109. R. A. Weiss, A. Kohli, N. S. Chung, and D. Dutta,
unpublished work.
110. Y. S. Lipatov, V. V. Tsukruk, 0. A. Lokhonya, V. V.
Shilov, Y. B. Amerik, I. I. Konstantinov, a n d V. S.
Grebneva, Polymer, 28, 1370 (1987).
111. S. U. Hongand J. Blackwell, Polymer, 30,225 (1989).
112. T. S u n , D. S. Done, a n d D. G. Baird, Abstracts of t h e
Polymer Processing Society Summer Meeting, Amherst, Mass., 9 B (1989).
113. R. Ramanathan, K. Blizard. a n d D. Baird, SPEANTEC
Tech. P a p e r s , 34, 1123 (1988).
114. G. Husman, T. Helminiak, W. Adams, D. Wiff, a n d C.
Benner, ACS Symposium Ser., EMS. 132,203 (1980).
115. G. Husman, T. Helminiak. M. Wellman, W. Adams, D.
Wiff, a n d C. Benner, Tech. Rep. AFWAL-TR-80-4034
(1980).
116. M. F. Froix a n d M. Park, U.S. Patent 4,460,735(1984).
117. D. Berry, A. Siegmann, a n d S. Kenig, J . Mater. Sci.
Lett.. 7, 1071-1073 (1988).
118. P. G. Hedmark, J. Manuel Rego Lopez, M. Westdahl,
P. E. Werner, J. F. J a n s s o n , a n d U. F. Gedde, Polyrn.
Eng. Sci., 28, 1248 (1989).
119. M. Takayanagi a n d K. Goto, J . Appl. Polyrn. Sci., 29,
2547 (1984).
120. M. Takayanagi, P u r e Appl. Chern., 55,819 (1983).
121. M. K. Akkapeddi. T. DeBona, H. L. Li. a n d D. C.
Prevorsek, U S . Patent, 4.61 1,025 (1986).
POLYMER ENGINEERING AND SCIENCE, MID-SEPTEMBER 1990, V d . 30, NO. 77