Full text of DOI:10.1016/S0032-3861(01)00475-X

Polymer 42 ꢀ2001) 9593±9599
www.elsevier.com/locate/polymer
Blends of a thermotropic copolyester and a thermotropic
copolyꢀester±amide): structure and mechanical properties
*
 Â
F.J. Vallejo, J.I. Eguiazabal, J. Nazabal
   Â
Departamento de Ciencia y Tecnologõa de Polõmeros, Instituto de Materiales Polimericos `POLYMAT', Facultad de Ciencias Quõmicas, UPV/EHU,
P.O. Box 1072, 20080 San Sebastian, Spain
Received 19 March 2001; received in revised form 4 June 2001; accepted 19 June 2001
Abstract
Rodrun 5000/Vectra B950 blends were obtained by direct injection moulding at two injections rates. The blends were partially miscible
and unreacted under the processing conditions used. Higher orientations and properties were obtained at the lower injection rate due to the
higher overall orientation obtained. The produced orientation and the negative volume of mixing of the blends led to a range of LCP materials
with intermediate modulus and tensile strength and slightly higher ductility. q 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Rodrun 5000; Vectra B950; Blends
1. Introduction
ꢀVB) containing only HNA [7,10], as well as of other
LCPs with different HNA and HBA contents [11,12] have
shown the development of interchange reactions. These
reactions have also been considered [7,10] as responsible
for the increased orientation of the LCPs in the blends.
The phase structure [6±9,11±14] and the rheological
behaviour [6,8,11,13,15±19] of LCP/LCP blends have
been often studied, but despite their high level, their
mechanical properties have seldom been studied, and
many of the works are patents [20±23]. In the open
literature, only the mechanical properties of blends of two
ethylene terephthalate±HBA copolyesters [9], VA with VB
[7] and with Ultrax ꢀa copolymer based on HBA, terephtha-
loyl and hydroquinone) [6] and only three compositions of
another blend [8] have been studied.
Many other LCP blends are potentially interesting. For
instance, the phase structure of Rodrun LC5000 ꢀR5)/VB
blends has been studied [14], but no systematic analysis of
their mechanical properties has been carried out. This is
despite the fact that R5/VB blends have been reported to
be immiscible [14], and therefore likely to exhibit syner-
gistic mechanical behaviour. Moreover, taking into account
their chemical structures, transesteri®cation reactions can
take place between them, and this could also affect
positively the mechanical properties of their blends.
Thermotropic liquid±crystal polymers ꢀLCPs) have
gained increasing interest in recent years, due to their excel-
lent mechanical properties, good processability and low
thermal expansion coef®cient [1]. Moreover, when mixed
with thermoplastics they are able to act as reinforcements,
due to their ®brillar and highly stiff and resistant structure in
the solid state [2±5].
The number of commercially available LCPs is limited
[1], so that there is a clear need for the development of new
materials with thermotropic behaviour. Blending offers a
valuable method to produce new LCPs due to their simpli-
city and comparatively low cost. The modulus of elasticity
and the tensile strength of blends of two LCPs appear
different from those usually found in conventional thermo-
plastic blends. Thus, they were synergistic in two immisci-
ble blends [6,7], slight synergisms were observed in a
partially miscible blend [8], and a practically linear
behaviour was seen in a miscible blend [9].
Moreover, LCPs are usually copolyesters or copoly-
ꢀester±amides), so that chemical interchange reactions can
take place between the blend components during processing
or thermal treatment in the melt state. In fact, although
exceptions also exist [9], blends of two LCPs, Vectra
A950 ꢀVA) containing 2,6-hydroxynaphthoic acid ꢀHNA)
and p-hydroxybenzoic acid ꢀHBA) with Vectra B950
In this work, R5/VB blends have been prepared across the
full composition range by direct injection moulding. Differ-
ent injection rates have been used because of their effects on
molecular orientation and mechanical properties. The phase
structure of the blends after the blending procedure
* Corresponding author. Tel.: 1349-43-018-218; fax: 1349-43-212-236.
Â
E-mail address: popnaetj@sq.ehu.es ꢀJ. Nazabal).
0032-3861/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0032-3861ꢀ01)00475-X

9594
F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
employed has been analysed by dynamic mechanical±
thermal analysis ꢀDMTA), and possible changes in the
chemical structure have been studied by Fourier Transform
Infrared Spectroscopy ꢀFTIR). The mechanical properties of
the blends have been determined by means of tensile tests
and have been related to both their speci®c volume and
orientation level as measured by X-ray diffraction.
Polymer Laboratories DMTA at a frequency of 1 Hzin
the ¯exural mode and at a heating rate of 48C/min. The
samples were obtained in the orientation direction from
the central section of the injection moulded tensile speci-
mens. The chemical structure of the blends was studied by
FTIR, using a Nicolet 5DXC spectrophotometer. Speci®c
volume values were measured in a Mirage SD-120-L elec-
tronic densitometer with a resolution of 0.0005 cm3/g using
butyl alcohol as the immersion liquid.
The tensile tests were carried out using an Instron 4301 at
a cross-head speed of 10 mm/min on 1.9 mm thick ASTM
D-638 type IV injection moulded specimens. The specimens
were gated at the outside end of the shoulder of the dumb-
bell-shaped moulding cavities. The mechanical properties
ꢀYoung's modulus, E; break stress, sb; and ductility
measured as the break strain, 1b) were determined from
the load-elongation curves. A minimum of eight specimens
were tested for each reported value.
2. Experimental
The polymers used in this work were Rodrun LC5000w
ꢀR5) from Unitika Ltd. and Vectra B950w ꢀVB), commer-
cialised by Celanese ꢀTicona). R5 is a copolyester based on
p-hydroxybenzoic acid ꢀHBA) and ethylene terephthalate
ꢀET). VB is a copolyꢀester±amide) composed of hydroxy-
naphthoic acid ꢀHNA), aminophenol ꢀAP) and terephthalic
acid ꢀTA). The chemical structures of R5 and VB are:
Both LCPs were dried before processing for 8 h at 1358C
in order to avoid moisture-induced degradation reactions
during processing. The dried pellets were directly melt
mixed and injection moulded over the entire composition
range in a Battenfeld BA230E reciprocating screw injection
moulding machine. A melt temperature of 3008C, a mould
temperature of 208C, and an injection pressure of 2450 bar
were used. The low mould temperature was selected in order
to obtain a high cooling rate and, thus, to improve the LCP
orientation in the moulded specimens. The rapid cooling
meant that a high injection pressure was required. In order
to analyse the effect of the injection rate on the structure and
properties of the blends, the moulding process was carried
out at two injection rates, 1 and 7 cm3/s.
The WAXS diffraction patterns of some samples of the
neat components and of the blends were obtained using a
Statton ¯at camera ꢀW.H. Warhus, Co.) and CuKa ꢀl ˆ
ꢀ
1:542 A† radiation. The samples ꢀlength ˆ 20 mm† for
X-ray diffraction were cut from the central section of the
tensile specimens.
3. Results and discussion
3.1. Phase behaviour
The thermal transitions of the neat components and of the
blends were not clearly observed by differential scanning
calorimetry ꢀDSC). Consequently, the thermal behaviour
The DMTA were carried out from 30 to 2008C in a

F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
9595
Fig. 1b shows the tan d ±temperature plots for neat R5
and VB and for the VB-rich blends. As can be seen, and as
was less clearly seen in Fig. 1a, the height of the tan d peak
of the VB is greater in the blends than in the neat0 compo-
nents. This was accompanied by the fact that the E drop in
the glass transition region was also greater in the blends than
in VB. This has been attributed in miscible blends [14] to
changes in the liquid crystalline alignment of the blends
compared to that of the neat components. The same
phenomenon is also observed in the partially miscible
blends of this paper. As can also be seen in Fig. 1b and
Table 1, the Tg of the VB-rich phase of the VB-rich blends
also decreased upon the addition of increasing amounts of
R5. As stated above, this indicates the presence of some R5
in the VB-rich phase. However, no peak close to that of pure
R5 appeared in the VB-rich blends. This is helped by the
fact that the Tg of R5 is wide and weak, and therefore
dif®cult to observe in the blends. The absence of the Tg of
an LCP component of a blend is not unusual, even in
sensitive DMTA measurements. For example, in extruded-
injection moulded VA/Ultrax [6] and in directly injection
moulded VA/VB [7] blends, the Tg of VA could not be
observed. However, after a more careful observation of
Fig. 1b, we realise that the minimum in tan d of pure VB
between 80 and 1208C does not appear in the blends. This
effect is more clearly seen as the R5 content increases, and
gives rise to tan d peaks which are broader, as the composi-
tion tends to be equal amounts of both components. This
suggests the presence of a single mixed phase, or micro-
phase separation where the phase sizes are too small to be
resolved by DMTA into separate peaks. Thus, although the
VB-rich blends of Fig. 1b are close to miscibility, the
presence of two partially mixed phases in the R5-rich blends
of Fig. 1a indicates the existence of partial miscibility
between R5 and VB.
Fig. 1. Tan d vs. temperature plots for: ꢀa) R5 ꢀX), VB ꢀB) and the 80/20
ꢀW) and 60/40 ꢀA) blends and ꢀb) R5 ꢀX), VB ꢀB) and the 40/60 ꢀW), 30/70
ꢀA), 20/80 ꢀS) and 10/90 ꢀK) blends.
and the phase structure were analysed by DMTA, which is a
more sensitive technique than DSC.
Fig. 1a shows the tan d ±temperature plots for neat R5
and VB and for the R5-rich blends, obtained at an injection
rate of 7 cm3/s. The residence time of the blends in the melt
state was similar for both injection rates, so a possible effect
on the phase structure can be neglected. As can be seen, the
Tg of R5 was close to 858C, while that of VB appeared near
1508C. The slight peak in the plot of VB near the Tg of R5
indicates the presence of a secondary transition probably
due to the motion of the HNA unit. Its location seems to
be invariant with the blend composition in spite of the fact
that in previous DMTA studies [24] this transition was
dependent on copolymer composition. The plots of the
blends showed two tan d peaks that overlapped in the
60/40 composition. The Tg values are collected in Table 1
together with those of the VB-rich blends that will be
discussed later. Although an exact measurement of the Tgs
from the maxima of the tan d ±temperature curves is dif®-
cult, the Tgs of the blends are intermediate between those of
the pure components and approach each other slightly as the
content of the two components of the blend equilibrates. This
indicates the presence in the R5-rich blends of Fig. 1a of two
mixed phases, each of them rich in one of the components.
The presence of wide single Tg peaks must be due to
either partial miscibility or development of transesteri®ca-
tion reactions during processing [7]. For this reason, FTIR
analyses of the blends were carried out to check the possible
development of interchange reactions between R5 and VB.
The results are shown in Figs. 2a,b, where the experimental
FTIR spectra of the 80/20 and 20/80 blends are compared
with those calculated by combining the weighted spectra of
Table 1
Glass transition temperatures ꢀTg) of the pure components and of the blends
at an injection rate of 7 cm3/s
Tg ꢀ8C)
R5
84
89
99
±
±
80/20
60/40
40/60
30/70
20/80
10/90
VB
138
137
134
135
136
143
147
±
±
±
±

9596
F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
Fig. 3. Young's modulus of the blends injected at 1 cm3/s ꢀX) and at 7 cm3/s
ꢀW) vs. composition.
behaviour is common in LCPs, and is attributed [25±28]
to the increase in the thickness of the oriented skin [29] as
the injection rate decreases, although at decreasing injection
rates the mechanical properties of the skin also appear to
improve. The higher mechanical properties at low injection
rates have also been attributed to cooling of the melt during
¯ow that leads to a viscosity increase [27±30] and, there-
fore, both increased shear and molecular orientation.
To ®nd out the reasons for the modulus behaviour, the
orientation of the blends and of the pure components were
measured. Additionally, the speci®c volume of the blends
may change when some miscibility is present [31] and also
in¯uences the modulus values, and therefore, was also
studied. The thickness of the skin could not be clearly deter-
mined, due to the presence in half of each specimen of four
layers of different thicknesses.
Fig. 2. Experimental and calculated FTIR spectra for the: ꢀa) 80/20 and
ꢀb) 20/80 blends.
The plot of the speci®c volume of the blends against
composition is shown in Fig. 4. As can be seen, the differ-
ence in speci®c volume of the blends at the two injection
rates is not signi®cant. Hence a single curve is plotted.
However, decrease in speci®c volume due to blending is
clear at both injection rates. The decreases are quantitatively
large enough ꢀmean value 1 £ 1023 cm3/g) [32] to lead to
an increase in the modulus of elasticity. Therefore, the
the pure components at each composition. The spectra of the
80/20 blend of Fig. 2a are representative of the R5-rich
blends, and those of the 20/80 blend of Fig. 2b are
representative of the VB-rich blends. As can be seen, the
experimental and calculated spectra are practically identical
in both composition regions. Therefore interchange
reactions are not the reason for the Tg changes, indicating
that the R5/VB blends are partially miscible.
3.2. Mechanical properties
The Young's modulus±composition relationships of the
blends are shown in Fig. 3. As can be observed, in the blends
injected at 1 cm3/s, the modulus followed approximately the
linear rule of mixtures; whereas in the case of the R5-rich
blends injected at 7 cm3/s, a very slight synergism ꢀmaxi-
mum approximately 4%) was seen. These modulus results
agree with the tendency towards slight synergisms or
linearity found in partially miscible and miscible LCP1/
LCP2 systems [8,9].
The effect of the injection rate on the modulus is also
shown in Fig. 3. As can be seen, the blends and the neat
components injected at the low injection rate show modulus
values higher than those of the high injection rate. This
Fig. 4. Speci®c volume of the blends vs. composition. Symbols as in Fig. 3.

F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
9597
Fig. 5. X-ray diffraction patterns of: ꢀa) R5, ꢀb) 80/20 blend, ꢀc) 20/80 blend and ꢀd) VB injected at 1 cm3/s.
decrease in speci®c volume upon blending will have a
positive effect on the modulus of elasticity of the blends.
The possible differences in the overall orientation of the
blends were analysed by X-ray diffraction. Some represen-
tative diffraction patterns for samples injected at 1 cm3/s are
shown in Fig. 5. As is seen from the arcing levels of the
equatorial diffraction of the samples, the orientation is lower
in R5 ꢀFig. 5a) than in VB ꢀFig. 5d). The diffraction pattern
of the 80/20 R5/VB blend ꢀFig. 5b) indicates an orientation
level similar or slightly lower than that of R5, while that of
the 20/80 R5/VB blend of Fig. 5c approaches that of neat
VB. Thus, the orientation of the blends could have a slight
negative effect on the modulus values. The positive effect on
the modulus of the negative volume of mixing should
counteract the slight negative effect of the orientation, lead-
ing to the rather linear behaviour of the plot of modulus of
elasticity at 1 cm3/s of Fig. 3.
high injection rates, there is a lower overall orientation in
VB and in the blends, that agrees with their lower modulus.
The ductility of the blends is shown in Fig. 7. As can be
seen, due to the brittle nature of the LCPs, the ductility is
very low both for the neat components and for the blends.
Moreover, it shows an overall positive deviation with
The in¯uence of the orientation on the modulus values is
also seen when the effect of the injection rate on the orienta-
tion is studied. Fig. 6 shows the diffraction pattern of the
neat VB injected at 7 cm3/s. The arcing is larger than that of
the low injection rate of Fig. 5d. The same effect was less
clearly seen in most of the blends. This indicates that, at
Fig. 6. X-ray diffraction pattern of VB injected at 7 cm3/s.

9598
F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
the rule of mixtures ꢀmaximum approximately 10%), which
can be attributed to the practically additive modulus ꢀFig. 3)
and the generally slight negative deviation of the ductility
ꢀFig. 7). The positive behaviour of both, the modulus and
the ductility of the blends injected at 7 cm3/s led to a
positive deviation that reached a maximum of 25% in the
R5-rich composition region. Thus, contrary to thermoplastic
blends and to previous [6] clear positive synergisms in
immiscible blends, and in agreement with other partially
miscible blends [8], the partial miscibility of the blends
leads to intermediate behaviours in the tensile strength
and in the modulus of elasticity of these LCP blends.
Fig. 7. Ductility of the blends vs composition. Symbols as in Fig. 3.
4. Conclusions
respect to the rule of mixtures between the values of R5 and
VB at the high injection speed, and a mixed behaviour at the
low injection speed. The high ductility value of the 80/20
blend at the low injection speed is a repetitive value ꢀdouble
amount of specimens tested), but it is not very signi®cant,
taking into account, the values of the rest of the blends and
the lack of change of the measured morphological
characteristics. The partial miscibility between the blend
components and the improved interfacial adhesion probably
led to the positive behaviour of Fig. 7.
As can also be seen in Fig. 7, the low injection rate leads
to slightly higher ductilities. Ductility data of LCPs as a
function of the injection rate are seldom reported in the
literature and the relation between the injection rate and
the ductility of LCPs is still not clear. This is because
both a tendency for the ductility to increase [25] at lower
injection rates, as in this work, and the opposite trend [27]
have been seen.
The tensile strength plots of the blends are shown in
Fig. 8. Its behaviour is rather parallel to that of the modulus.
This is despite the fact that the ductility must affect the
tensile strength in these brittle materials. The samples
moulded at the low injection rate show clearly higher tensile
strength, in agreement with their higher orientation and
modulus behaviour and with the results obtained by other
authors [25,27,33]. The tensile strength of the samples
injected at 1 cm3/s shows a slight negative deviation from
R5/VB blends are partially miscible. At the processing
conditions of this work, the blends are unreacted as seen
from FTIR spectra. Lower injection rates lead to clearly
higher modulus of elasticity and tensile strength due to the
overall greater orientation produced. The tensile strength
values were close to those predicted by the rule of mixtures
and were a consequence of the trends in the modulus and
ductility. The modulus values of the blends at an injection
rate of 1 cm3/s were close to those predicted by the rule of
mixtures. They were a consequence of the slightly smaller
orientation of the components in the blends, partially
counteracted by the positive effect of the negative excess
volume of mixing.
Acknowledgements
The ®nancial support of the University of the Basque
Country ꢀProject number G46/98) is gratefully acknowl-
edged. F.J. Vallejo also acknowledges the University of
the Basque Country ꢀUPV/EHU) for the award of a grant.
The authors also acknowledge Dr InÄaki Iribarren for the
X-ray measurements.
References
[1] Brown CS, Alder PT. In: Folkes MJ, Hope PS, editors. Polymer
blends and alloys. Glasgow: Chapman & Hall, 1993. Chapter 8.
 Â
[2] Vallejo FJ, Eguiazabal JI, Nazabal J. Polymer 2000;41:6311±21.
Â
Â
[3] Bastida S, Eguiazabal JI, Nazabal J. Polymer 2001;42:1157±65.
[4] La Mantia FP. Thermotropic liquid crystal polymer blends. Lancaster:
Technomic Publishing Co., 1993.
[5] Acierno D, La Mantia FP. Processing and properties of liquid crystal-
line polymers and LCP based blends. Ontario: Chemtec Publishing,
1993.
[6] Akhtar S, Isayev AI. Polym Engng Sci 1993;33:32±42.
 Â
[7] Vallejo FJ, Eguiazabal JI, Nazabal J. Polym Engng Sci 2001 ꢀin
press).
[8] Kenig S, DeMeuse MT, Jaffe M. Polym Adv Technol 1991;2:25±30.
Â
Â
[9] Vallejo FJ, Eguiazabal JI, Nazabal J. Polym J 2001;33:128±32.
[10] Dreher S, Seifert S, Zachmann HG, Moszner N, Mercoli P, Zanghel-
lini G. J Appl Polym Sci 1998;67:531±45.
Fig. 8. Tensile strength of the blends vs composition. Symbols as in Fig. 3.
[11] McLeod MA, Baird DG. Polymer 1999;40:3743±52.

F.J. Vallejo et al. / Polymer 42 ;2001) 9593±9599
9599
[12] McCullagh CM, Blackwell J, Jamieson AM. Macromolecules
1994;27:2996±3001.
[23] Isayev AI, Subramanian PR. US Patent 5,070,157, 1991.
[24] Troughton MJ, Davies GR, Ward IM. Polymer 1989;30:58±62.
[25] Boldizar A. Plast Rubber Process Appl 1988;10:73±8.
[13] Lin YG, Winter HH. Polym Engng Sci 1992;32:773±8.
[14] Hsieh T, Tiu C, Simon GP. Polymer 2000;41:4737±42.
[15] Hsieh TT, Tiu C, Simon GP, Andrews SR, Williams G, Hsieh KH,
Chen CH. In: Shonaike GO, Simon GP, editors. Polymer blends and
alloys. New York: Marcel Dekker, 1999. Chapter 12.
[26] Hsiung CM, Tian J, Cakmak M. Int Polym Process 1993;8:164±77.
[27] ZuÈlle B, Demarmels A, Plummer CJG, Kausch HH. Polymer
1993;34:3628±37.
[28] Kingma JA, Geijselaers H. In: Lemstra PJ, Kleintjens LA, editors.
Integration of fundamental polymer science and technology, vol. 4.
London: Elsevier, 1990. p. 361±6.
[16] Ding R, Isayev AI. J Thermoplast Compos Mater 1995;8:208±24.
[17] DeMeuse MT, Jaffe M. In: Weiss RA, Ober CK, editors. Liquid
crystalline polymers, ACS Symp. Ser. No. 435, 1990. Chapter 30.
[18] McLeod MA, Baird DG. J Appl Polym Sci 1999;73:2209±18.
[19] Done D, Baird DG. J Appl Polym Sci 1990;30:989±95.
[20] Kiss G. US Patent 4,567,227, 1986.
[29] Kenig S. Polym Compos 1986;7:50±5.
[30] Suokas E. Polymer 1989;30:1105±12.
[31] Hsieh TT, Tiu C, Hsieh KH, Simon GP. Korea Polym J 1998;6:44±
52.
[21] Matsumoto M, Kaneda T. US Patent 4,837,268, 1989.
[22] Froix MF. US Patent 4,267,289, 1981.
[32] Barlow JW, Paul DR. Polym Engng Sci 1981;21:985±96.
[33] Kenig S, Trattner B, Anderman H. Polym Compos 1988;9:20±6.