Thermotropic polyesters from 2,5-furandicarboxylic acid and vanillic acid: Synthesis, thermal properties, melt behavior, and mechanical performance

Article abstract of DOI:10.1021/ma500433e

In this work, we address the synthesis of novel aromatic-aliphatic biobased polyesters showing thermotropic behavior in the melt. Successful incorporation of different biobased monomers such as 2,5-furandicarboxylic acid (2,5-FDCA), suberic acid (SuA), and vanillic acid (VA) in thermotropic liquid crystalline polymers (TLCPs) is made possible by performing synthesis at low temperatures. The chemical structures, molecular weights, phase transitions, thermal behavior, and mechanical performance of the synthesized polymers are studied using polarization optical microscopy, WAXD, DSC, TGA, DMTA, solid-state NMR spectroscopy, rheology, and tensile tests. It is shown that the incorporation of the rigid, aromatic 2,5-FDCA moiety enhances the formation of blocky copolymers, whereas the VA moiety tends to decrease the block formation. However, when combined, nonblocky TLCPs containing 2,5-FDCA and VA with high aromatic content can be obtained. These materials show a low temperature transition from the crystalline to the nematic phase, and stable nematic phases up to 300 °C and higher. Furthermore, in such polymers, the 2,5-FDCA and VA moieties require more thermal energy to become mobile compared to the phenyl rings in hydroxybenzoic acid, hydroquinone, and 4,4′-biphenol. Mechanical analysis shows that the performance of these polymers is correlated to their crystallinity. Surprisingly, a higher crystallinity results in ductile behavior, whereas a lower crystallinity results in a higher modulus, a higher stress at break, and a lower strain at break.

Full text of DOI:10.1021/ma500433e

Article
pubs.acs.org/Macromolecules
Thermotropic Polyesters from 2,5-Furandicarboxylic Acid and
Vanillic Acid: Synthesis, Thermal Properties, Melt Behavior, and
Mechanical Performance
†
,‡
†
†
§,∥
Carolus H. R. M. Wilsens, Johan M. G. A. Verhoeven, Bart A. J. Noordover, Michael Ryan Hansen,
⊥
⊥,#,7,‡,
Dietmar Auhl, and Sanjay Rastogi
*
†
Laboratory of Polymer Materials, Eindhoven University of Technology, Den Dolech 2, 5600MB, Eindhoven, The Netherlands
Dutch Polymer Institute (DPI), P.O. Box 902, 5600AX Eindhoven, The Netherlands
‡
§
∥
Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany
Interdisciplinary Nanoscience Center (iNANO) and Department of Chemistry, Aarhus University, Gustav Wieds Vej 14, DK-8000
Aarhus C, Denmark
⊥
Department of Biobased Materials, Maastricht University, P.O. Box 616, 6200MD Maastricht, The Netherlands
Department of Materials, Loughborough University, Loughborough, England, U.K.
#
7
Research Institute, Teijin Aramid, Velperweg 76, 6824BM, Arnhem, The Netherlands
*
S Supporting Information
ABSTRACT: In this work, we address the synthesis of novel aromatic−aliphatic biobased polyesters showing thermotropic
behavior in the melt. Successful incorporation of different biobased monomers such as 2,5-furandicarboxylic acid (2,5-FDCA),
suberic acid (SuA), and vanillic acid (VA) in thermotropic liquid crystalline polymers (TLCPs) is made possible by performing
synthesis at low temperatures. The chemical structures, molecular weights, phase transitions, thermal behavior, and mechanical
performance of the synthesized polymers are studied using polarization optical microscopy, WAXD, DSC, TGA, DMTA, solid-
state NMR spectroscopy, rheology, and tensile tests. It is shown that the incorporation of the rigid, aromatic 2,5-FDCA moiety
enhances the formation of blocky copolymers, whereas the VA moiety tends to decrease the block formation. However, when
combined, nonblocky TLCPs containing 2,5-FDCA and VA with high aromatic content can be obtained. These materials show a
low temperature transition from the crystalline to the nematic phase, and stable nematic phases up to 300 °C and higher.
Furthermore, in such polymers, the 2,5-FDCA and VA moieties require more thermal energy to become mobile compared to the
phenyl rings in hydroxybenzoic acid, hydroquinone, and 4,4′-biphenol. Mechanical analysis shows that the performance of these
polymers is correlated to their crystallinity. Surprisingly, a higher crystallinity results in ductile behavior, whereas a lower
crystallinity results in a higher modulus, a higher stress at break, and a lower strain at break.
INTRODUCTION
modifications are required to lower their melting and
■
processing temperatures in order to decrease processing
Thermotropic polyesters are an important class of materials for
high performance applications. Their low melt viscosities, low
thermal expansion coefficients, good thermal stability, and ease
in processing allow for the production of high strength and high
modulus fibers, films, or compression-molded articles. In
general, fully aromatic thermotropic liquid crystalline polyesters
2
costs. Examples of such structural modifications are the
introduction of different (mesogenic) comonomers, monomers
1
Received: February 26, 2014
Revised: April 8, 2014
Published: May 5, 2014
(
TLCPs) have high melting temperatures and structural
©
2014 American Chemical Society
3306
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
Figure 1. Chemical structures and abbreviations of the monomers used in this study.
with bulky substituents, aliphatic or flexible monomers, bended
monomers, or nonlinear monomers.
Flexible spacers used in TLCPs are aliphatic dicarboxylic
preparation of new polymer materials. Understanding the role
of the 2,5-FDCA moiety in the occurrence of thermotropic
behavior in polyesters is of importance for the development of
3
−5
23
acids or aliphatic diols such as ethylene glycol or its
new biobased thermotropic polymers. The influence of the
rigid 2,5-FDCA moiety in aliphatic−aromatic thermotropic
polyesters on the thermal properties, block formation, and
mechanical properties is not known to date. In this paper, we
report on the synthesis and properties of new thermotropic
polyesters containing 2,5-FDCA. Furthermore, copolymeriza-
tion with biobased vanillic acid is performed to reduce the
block formation and to develop biobased TLCPs with stable
LC melts.
6
−9
drivatives.
In particular, the use of sebacic acid and suberic
acid are of interest, since they can be obtained from biobased
feedstock such as vegetable oils and can be used to develop
10,11
12,13
biobased polymers
and TLCPs.
Furthermore, the
presence of these aliphatic dicarboxylic acids drastically
suppresses the melting temperature (crystal to liquid crystal
14
transition) of the TLCPs. For example, Hsiue and colleagues
reported the synthesis of TLCPs containing different amounts
of sebacic acid, hydroquinone, and p-hydroxybenzoic acid (BA)
via melt polymerization. These polymers showed melting
temperatures well below 200 °C. The melting temperatures
increased rapidly when the amount of sebacic acid was
decreased. Though Hsiue reported the presence of homoge-
neous liquid crystalline (LC) melts, the coexistence of both an
isotropic melt and a LC melt in these polymers was reported by
EXPERIMENTAL SECTION
■
Materials and Monomer Preparation. Suberic acid, 1,4-
diacetoxybenzene, and 4-hydroxy-3-methoxybenzoic acid were ob-
tained from Sigma-Aldrich. The monomers 4,4′-diacetoxybiphenyl and
p-acetoxybenzoic acid were obtained from TCI Europe and 2,5-
furandicarboxylic acid (99.5%, GC-MS) was procured from Atomole
1
5
(
China). 4-Hydroxy-3-methoxybenzoic acid was acetylated and
Dingemans and co-workers. The inhomogeneous structure in
this polymer melt can be attributed to nonrandom distribution
of the monomers along the polymer backbone arising due to
recrystallized prior to use. 2,5-Furandicarboxylic acid was purified
through the recrystallization of the dimethyl 2,5-furandicarboxylate
intermediate in methanol. The obtained crystals were hydrolyzed in an
aqueous NaOH solution, neutralized with an aqueous HCl solution,
filtered and dried in vacuo overnight prior to usage. All other
monomers were recrystallized at least once and dried in vacuo
overnight prior to use. The structural formulas of the monomers used
in this study are shown in Figure 1.
1
6,17
clustering of monomers into blocks.
Furthermore, Dinge-
mans and co-workers also showed that the incorporation of one
or two comonomers could enhance the stability of the LC melt.
1
8
De Oca reported the synthesis of bioresorbable TLCPs
containing flexible aliphatic spacers, p-hydroxybenzoic acid,
Preparation of 4-Acetoxy-3-methoxybenzoic Acid (VA). 4-
Hydroxy-3-methoxybenzoic acid (50 g, 304.5 mmol) was placed in a
4,4′sulfonyl bis(2-methylphenol), and vanillic acid synthesized
from solution. These polymers showed low melting temper-
atures between 95 and 165 °C, while no isotropic melt was
observed below 300 °C, indicating that the different monomers
in these polymers are randomly distributed without forming
any blocks.
2
50 mL round-bottom flask on a magnetic stirring plate. Dry acetic
anhydride (50 mL, 530 mmol) was added together with a catalytic
amount of H SO . The mixture was heated to 80 °C and left to react
2
4
for 6 h. After cooling the mixture to 0 °C, 200 mL of water was added,
and the obtained yellow crystals were filtered, washed with water, and
dried in vacuo at 60 °C for 14 h. The resulting powder was
The influence of block formation on the processing and on
the mechanical performance of thermotropic polyesters has
been studied for many decades. For example, the processing
and rheology of different BA and 2-hydroxy-6-naphthoic acid
based copolymers has been studied extensively by Yang and
recrystallized three times from toluene to yield 46.7 g of product
1
(
72.9%). H NMR(MeOD-d , δ, ppm): 7.69 (d, ArH, 1H), 7.64 (dd,
4
ArH, 1H), 7.12 (d, ArH, 1H), 3.86 (s, O−CH , 3H), 2.28 (s, CO−
3
1
3
CH , 3H). C NMR (MeOD-d , δ, ppm): 170.20 (CH −CO),
3
4
3
1
1
69.02 (COOH), 152.62 (ArC−OCHH ), 145.11 (ArC−OAc),
1
9,20
3
Krigbaum.
These authors showed that the flow instability
30.75 (ArC−COOH), 123.88 (ArC), 123.74 (ArC), 114.51 (ArC),
during processing is associated with the presence crystallites
having high melting temperatures, formed via transesterification
in the melt. Furthermore, the mechanical performance of these
spun materials greatly suffered from the presence of such
56.51 (O−CH ), 20.46 (CO−CH ).
3
3
General Polymerization Procedure. All polymerizations were
performed in a 100 mL three-neck round-bottom flask fitted with a
mechanical stirrer. The monomer mixture, between 50 and 100 g in
21
total, was weighed together with 30−60 mg of Zn(AcO) in the round-
2
crystals. Krigbaum and colleagues observed similar behavior
for partially flexible TLCPs based on BA and poly(ethylene
terephthalate) (PET) copolymers.
bottom flask. The monomers were dried in vacuo at 40 °C prior to
usage to eliminate moisture. An oxygen free atmosphere was created
by flushing with Argon and applying reduced pressure for three times
prior to the start of the reaction. The maximum reaction temperature
used in this study is 230 °C for polymerizations using 2,5-FDCA and
260 °C for polymerization without 2,5-FDCA. Acetic acid was distilled
With the development of new methods to produce 2,5-
22
furandicarboxylic acid (2,5-FDCA) on an industrial scale, this
rigid aromatic biobased monomer is readily available for
3
307
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
a
Scheme 1. Acidolysis Reaction Occurring during the Synthesis of the 2,5-FDCA Containing TLCP 4c
a
N.B.: The resulting polymer is a copolymer.
off for 6−8 h, after reaching the maximum reaction temperature.
Reduced pressure was applied to the system for 2−12 h after 90% of
the calculated amount of acetic acid was collected. The final products
were isolated from the reactor flask in the form of a crystalline solid or
polymer melt.
samples were then pressed in the melt and cooled to 50 °C at a
cooling rate of 20 °C/min. Optical micrographs were taken during the
new heating run performed at a heating rate of 20 °C/min up to 350
°C or the degradation temperature.
The stability of the LC melt was followed using a TA Instruments
ARG2 rheometer. Parallel plate geometry of a diameter of 25 mm and
a diameter of 1 mm was used to study the temperature dependence of
the apparent shear viscosity. A fixed strain (1.25%) and frequency (10
Hz) were chosen such that the experiments were performed in the
linear viscoelastic regime. Data points were collected every 1 °C over
the temperature range between 150 and 300 °C and a thermal
expansion of 2.45 μm/K for this geometry was taken into account.
Prior to the collection of data points, the samples were allowed to relax
for 15 min at the starting temperature of the experiments.
Characterization Methods. H and 13C NMR spectroscopy was
performed on a Bruker 400 MHz NMR spectrometer. Samples
measured in NMR analysis were dissolved in HFIP, precipitated in
methanol, and dried in vacuo at 40 °C prior to preparation for NMR
measurements. All samples were measured in a mixture of deuterated
1
chloroform and deuterated trifluoroacetic acid (CDCl /TFA-d) (50/
3
50 v/v) and were referenced against tetramethylsilane (TMS). The
compositions of the polymers were determined by taking the ratio
between the resonance signals corresponding to the different
monomers.
The thermal stability of the polymers was determined using
thermogravimetric analysis (TGA) and was performed on a TA
Instruments TGA Q500 in a nitrogen rich atmosphere. Samples were
heated at 20 °C/min from 20 to 800 °C.
A Bruker AVANCE-III 500 spectrometer employing a double-
resonance H−X probe for rotors with a 4.0 mm outside diameter was
13
1
used to carry out C{ H} insensitive nuclei enhanced by polarization
transfer (INEPT) magic-angle spinning (MAS) and variable-temper-
The glass transition temperature (T ) and the peak transition
g
ature (VT) solid-state 13C{ H} cross-polarization/MAS (CP/MAS)
1
temperature from the crystalline to the LC phase (T ) were
m
determined by differential scanning calorimetry (DSC) using a TA
Instruments Q1000 DSC. The normal heating and cooling-rates of the
samples were 10 °C/min and measurements were performed under a
nitrogen rich atmosphere.
NMR experiments. These experiments utilized a MAS frequency of
1
13
1
0.0 kHz, a 5.0 μs π/2 pulse for H and C, TPPM decoupling during
24
acquisition, and a CP contact time of 3.0 ms. The CP conditions
were preoptimized using L-alanine. The 13C{ H} INEPT MAS spectra
1
Dynamic mechanical thermal analysis (DMTA) was performed on a
TA Instruments DMA Q800. DMTA samples were prepared via
compression molding at temperatures between 220 to 260 °C and
were cut into rectangular shapes of 20 mm × 5 mm × 0.5 mm. All
experiments were performed at a frequency of 1 Hz and at a heating
rate of 2 °C/min. Measurements were continued until the sample was
were recorded using the refocused-INEPT sequence with a J-evolution
1
period of 1/(3JCH) or 1/(6JCH) and assuming a J of 150 Hz, i.e., for
CH
a J-evolution time of 1/(3JCH) the signals from CH and CH groups
3
2
5,26
are positive, while those of CH are negative.
The temperature was
2
controlled using a Bruker temperature control unit in the range of 30
13
1
13
1
to 210 °C. The VT C{ H} CP/MAS and C{ H} INEPT MAS
NMR spectra were recorded under isothermal conditions at selected
temperatures whereas a heating rate of 2 °C/min was employed
between temperatures. Reported temperatures are corrected for
molten or broken. The T values determined from DMTA analysis
g
were obtained by taking the maximum value of the loss modulus
during heating.
207
X-ray diffraction (XRD) measurements were performed on a Rigaku
Geigerflex Bragg−Brentano powder diffractometer using Cu Kα
radiation (λ = 0.154 nm), at 40 kV and 30 mA. The scans were
performed with a 0.01 step in 2θ and a dwell time of 2s. Peak fitting
was performed after background subtraction to determine the
percentage of crystallinity.
Tensile test experiments were performed at a constant deformation
rate of 5 mm/min using a Zwick 100 tensile apparatus and a load cell
of 100 N at room temperature. Samples were prepared via
compression molding and cut into dog-bone shaped samples of 20
mm × 2 mm × 1 mm.
friction-induced heating due to spinning using Pb MAS NMR of
27
Pb(NO ) as a NMR thermometer.
3
2
Molecular weights and polydispersity index (PDI) values were
determined via 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) size exclu-
sion chromatography (SEC). This system is equipped with a Waters
1515 Isocratic HPLC pump, a Waters 2414 refractive index detector
(
40 °C), a Waters 2707 autosampler, and a PSS PFG guard column
followed by 2 PFG-linear-XL (7 mm, 8 × 300 mm) columns in series
at 40 °C. HFIP (Apollo Scientific Limited) with potassium
−
1
trifluoroacetate (3 g L ) was used as an eluent at a flow rate of 0.8
−1
mL min . The molecular weights were calculated against poly(methyl
methacrylate) standards (Polymer Laboratories, M = 580 Da up to
p
6
M = 7.1 × 10 Da). SEC samples were filtered through a 200 μm filter
RESULTS AND DISCUSSION
p
■
prior to injection.
Thermotropic Polyester Synthesis. A series of TLCPs
with varying amounts of 2,5-FDCA and VA were synthesized
according to the general polymerization procedure. Sets of
polymers with three (3a−3b), four (4a−4f), and five (5a−5b)
monomers were synthesized. Both polymers 3a and 3b contain
p-hydroxybenzoic acid (BA), suberic acid (SuA), and hydro-
quinone (HQ) or 4,4′-biphenol (BP). In the synthesis of
Melt morphologies of the synthesized polymers were determined
on a Zeiss Axioplan 2 Imaging optical microscope under crossed
polarizers and CD achorplan objectives (32× zoom). A THMS 600
heating stage connected to a Linkam TMS 94 control unit was
mounted onto the optical microscope. Samples were prepared by
placing a small amount of ground polymer in-between two glass slides
and heating them at a heating rate of 50 °C/min to 250−300 °C. The
3
308
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
Table 1. Compositions and Molecular Weight Distributions Determined for Copolymers Containing Various Amounts of BA,
SuA, 2,5-FDCA, and VA with HQ or BP as Diols
a
composition (mol %)
SuA
SEC
entry
diol
BA
2,5-FDCA
VA
M (g/mol)
PDI
n
3
4
a
a
HQ 33 (33)
HQ 34 (33)
HQ 34 (33)
HQ 32 (33)
HQ 29 (30)
HQ 28 (30)
BP 34 (33)
BP 31 (30)
BP 26 (25)
BP 29 (30)
33 (33)
34 (33)
33 (33)
33 (33)
31 (30)
33 (30)
32 (33)
30 (30)
24 (25)
32 (30)
33 (33)
29 (29)
24 (24)
18 (17)
30 (30)
16 (15)
32 (33)
30 (30)
26 (25)
0 (0)
3 (4)
0 (0)
0 (0)
24 800
10 700
30 500
2.5
2.2
3.5
6.8
2.2
5.1
2.7
2.1
2.4
2.3
4b
9 (8)
0 (0)
b
4
c
17 (16)
0 (0)
0 (0)
27 100
4d
10 (10)
11 (10)
0 (0)
12 800
b
5a
12 (15)
0 (0)
63 400
3b
8200
28 500
14 600
12 000
4e
0 (0)
9 (10)
24 (25)
11 (10)
4
f
0 (0)
5b
17 (15)
10 (15)
a
1
b
Calculated by H NMR analysis in CDCl /TFA-d 50/50. Partially soluble in the HFIP solvent.
3
polymers 4a−4c SuA was replaced with increasing amounts of
reason, to prevent side reactions, the maximum temperature
used in the synthesis of polymers containing 2,5-FDCA was set
to 230 °C.
Interestingly, the PDI of the polymers increases with the
incorporation of 2,5-FDCA, as is evident from the GPC traces
of polymers 3a, 4b, 4c, and 5a shown in Figure 2. The GPC
2
,5-FDCA. For polymers 4d−4f, VA is added as the fourth
monomer and no 2,5-FDCA was used. Polymers 5a and 5b
contain both 2,5-FDCA and VA as additional monomers.
Scheme 1 shows the acidolysis reaction occurring during the
synthesis of polymer 4c.
The composition of the polymers was estimated based on the
1
monomer ratios determined by H NMR spectroscopy. Table 1
shows the obtained compositions after polymerization, whereas
the compositions weighed into the reactor are shown between
brackets. No sublimation of monomer on the reactor wall was
observed during the polymerizations except for the synthesis of
polymers 5a and 5b. These polymers showed slow sublimation
of 2,5-FDCA during the reduced pressure step. Because of this,
the 2,5-FDCA content incorporated in these polymers is lower
than weighed in the reactor. The sublimation of 2,5-FDCA
during the synthesis of polymers 5a and 5b can be attributed to
the extended reaction times required to build up molecular
weight compared to, for example, polymer 4c. Sublimation of
2
,5-FDCA during the synthesis of polymers 4a−4c was
expected, but was not observed due to decreased amounts of
,5-FDCA (4a, 4b) or rapid crystallization during synthesis
4c). Nonetheless, the compositions of the other polymers are
Figure 2. Overlay of the SEC traces of polymer 3a, 4b−4d, and 5a
showing the rise of a high molecular weight tail in 2,5-FDCA
containing copolymers.
2
(
in good agreement with the expected compositions.
It is well established that a decrease of the aliphatic content
in aliphatic−aromatic polymers causes a rapid increase of the
1
4
melting temperature and a decrease of the solubility.
traces in Figure 2 clearly show that the increase in PDI arises
from the presence of a high molecular weight tail. The
replacement of SuA by 2,5-FDCA increases the aromatic
content and apparently promotes the formation of less soluble,
long aromatic sequences. It is expected that these aromatic
sequences show a different hydrodynamic volume during SEC
analysis compared to the more soluble aliphatic sequences and
Furthermore, the formation of long aromatic blocks is
enhanced in these polymers. In this study, we determine the
influence of the nonlinear aromatic 2,5-FDCA monomer on the
formation of aromatic blocks along the main chain. This effect
in TLCPs is studied by gradually increasing the amount of 2,5-
FDCA at the expense of SuA, for example, as is the case for
polymers 3a and 4a−4c in Table 1. The maximum amount of
30
elute at different times. Another explanation for the high
molecular weight tail in the SEC traces is the occurrence of a
branching reaction during polymerization. These branched
chains should exhibit an improved solubility compared to the
linear chains. However, we observe that the solubility of the
polymers that elute a high molecular weight tail in fact
decreases drastically, as is the case for polymers 4c and 5a. For
this reason, we expect that these aromatic blocks are
responsible for the observed high molecular weight tail and
the increased PDI value. This dissolution behavior indicates
that the replacement of flexible spacers with aromatic 2,5-
FDCA enhances the block formation, similar to other aromatic
2,5-FDCA that could be incorporated into the polymer
backbone was 16 mol %, for systems with HQ as the diol
comonomer. The limiting factor for the incorporation of higher
amounts of 2,5-FDCA was the solidification of the reaction
mixture at the chosen low reaction temperatures. The
solidification of the reaction mixture inhibits the desired
molecular weight build-up. Similar solidification problems were
observed for systems containing a combination of 2,5-FDCA
and BP. It is expected that higher amounts of 2,5-FDCA can be
incorporated at higher reaction temperatures. However, an
increase of the reaction temperature is known to result in
28,29
16
partial degradation of the 2,5-FDCA monomer.
For this
monomers such as hydroxybenzoic acid.
3
309
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
Table 2. Thermal Data of TLCPs Containing Variable Amounts of Vanillic Acid and 2,5-FDCA
composition
VA
DSC
POM
DMTA
TGA
a
b
entry
diol
2,5-FDCA
Tg (°C)
Tm (°C)
Tlc (°C)
Tg (°C)
Tons (°C)
3
4
4
4
4
5
3
4
4
5
a
a
b
c
d
a
b
e
f
HQ
HQ
HQ
HQ
HQ
HQ
BP
0
0
0
3
42
43
56
e
193
184
165
164
210
205
240
300
150
280
280
220
210
170
46
47
57
82
47
79
c
384
384
390
409
370
412
410
402
406
406
0
9
0
17
0
d
10
11
0
45
66
44
63
63
73
126
12
0
186
189
191
BP
9
0
67
63
82
d
BP
24
11
0
148
b
BP
10
145
a
b
Peak melting temperature obtained from the second DSC heating run. Temperature where all crystals are molten and a liquid crystalline melt is
obtained, determined using polarization optical microscopy. Thus, the T represents the end melting temperature determined from DSC as depicted
lc
by arrows in Figure 3. Melting refers to the crystal to liquid crystal transition. In all the polymers investigated in this publication, melting proceeds
c
from crystal to liquid crystal where the liquid crystal to isotropic transition normally occurs close to the thermal degradation. Not possible to
d
e
compression mold samples due to a low melt-viscosity. Cold-crystallization is observed prior to melting. Not observed.
Interestingly, the samples containing VA show PDI values
between 2.1 and 2.4 (e.g., SEC trace 4d in Figure 2), which is
close to the theoretical PDI value of 2.0 expected for
polycondensation polymers. The copolymerization of VA is
known to lead to lower PDI values, as is reported by Li et al. for
polarization optical microscopy (POM), and dynamic mechan-
ical thermal analysis (DMTA) were employed to identify the
effect of the chemical composition and block formation on the
thermal properties of the polymers (Table 2).
An onset temperature in the weight loss (Tons) of 380 °C was
found in TGA analysis for polymer 3a. An increase of Tons to
405−410 °C was observed for the polymers containing 2,5-
FDCA, which is thought to be the result of the increased
aromatic content. Similarly, the replacement of diol from HQ
to BP for the same monomer composition increases the Tons by
several degrees.
30
copolymerizations of VA with PET and BA based TLCPs.
The methoxy group of VA improves the solubility of the final
polymer and is thought to be responsible for the decrease of the
PDI, as the solubility also affects the hydrodynamic volume in
3
1
solution. This implies that VA is able to increase the solubility
of the aromatic blocks, while 2,5-FDCA only enhances the
formation of aromatic blocks. In particular, the PDI of 2.3 for
sample 5b is interesting since this sample contains 10 mol % of
Figure 3 shows the second DSC heating curves of polymers
a, 4b, 4c, 4d, 4e, 4f, 5a and 5b. The polymer 3a, containing no
3
2
,5-FDCA and was expected to show a high molecular weight
tail. This example clearly shows that the copolymerization of
VA prevents the formation of poorly soluble aromatic blocks.
However, other factors such as reactivity, solubility, and length
of the mesogenic unit also affect the formation of aromatic
sequences. For example, the addition of VA to polymer 4c
decreased the PDI from 6.8 to 5.1 for the polymer 5a. The high
PDI of the polymer 5a indicates that poorly soluble aromatic
sequences are still present in this system. If we now compare
the PDI values of the polymers 5a and 5b, it appears that the
choice of the diol also plays an important role in the formation
of insoluble aromatic sequences (or blocks).
Beside the formation of insoluble aromatic blocks, we
noticed that the composition of the polymers also influences
the time required to build up molecular weight. For example,
when using 2,5-FDCA, over 10 h of reduced pressure are
required to achieve a molecular weight of 10 000 g/mol for
polymer 5b. In contrast, for the polymer 4e containing no 2,5-
FDCA a molecular weight well over 20 000 g/mol was obtained
after only 2 h of processing under reduced pressure. More
information regarding the effect of the composition on the
molecular weight build-up is supplied in the Supporting
Information.
Figure 3. Second heating curves of polymers 3a, 4b, 4c, 4d, 4e, 4f, 5a,
and 5b obtained during heating at 10 °C/min. The peak melting
temperature Tm of the polymers is indicated with an ∗, whereas the
arrows denote the Tlc as observed in polarization optical microscopy
(POM). The dotted lines in the DSC traces are added to illustrate the
temperature regions where melting of the polymers occur.
2,5-FDCA, shows a peak melting temperature (T ) at 193 °C.
m
Interestingly, the incorporation of 2,5-FDCA results in an
Thermal Properties of the TLCPs. The previous section
showed that both 2,5-FDCA and VA influence the formation of
aromatic sequences along the polymer chains. This variation of
the chemical microstructure as a function of chemical
composition is expected to influence both the melting behavior
and the processability of these polymers. Thermogravimetric
analysis (TGA), differential scanning calorimetry (DSC),
decrease of the T , but also to a broadening of the melting
transition. For example, the incorporation of 17 mol % of 2,5-
FDCA (4c) yields a polymer that shows a mobile liquid
m
crystalline phase (T ) above 300 °C, as observed in POM. The
lc
DSC trace of polymer 4c shows the presence of a weak
endotherm with a T at 165 °C (denoted with an ∗ in Figure
m
3). This indicates that, although the peak melting temperature
3
310
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
Figure 4. DMTA traces of (left) E′ and (right) E″ of polymers 3a, 4c, 4d, 4e, 4f, 5a, and 5b showing the effect of the copolymerization of VA and
,5-FDCA on the T and melting behavior. The arrow in the E′ versus temperature plot indicates the decrease of T and mechanical stability at
2
g
m
higher temperatures upon copolymerization of VA. The arrow in the E″ versus temperature plot shows the increase of T upon copolymerization of
g
2,5-FDCA.
of this polymer decreases compared to polymer 3a, the melting
range of this polymer increases resulting in an increase of Tlc
from 210 °C (3a) to 300 °C (4c).
have seen earlier, this is clearly not the case for blocky polymer
5a, which also contains 10 mol % VA.
Polymers 4e and 4f show Tm values below 200 °C, but
exhibit a Tlc above 210−220 °C. This indicates that these
polymers are able to form NPL crystallites, but without the high
melting temperatures observed for polymers 4c and 5a. It is
likely that the length of the crystallizing aromatic blocks in
these polymers is smaller than those in polymers 3b, 4c, and 5a,
resulting in lowered melting temperatures. Therefore, the
resulting melt behavior of these polymers is neither character-
istic for a random nor a blocky polymer.
It is anticipated that long aromatic sequences present in
polymer 4c can aggregate more easily in the melt resulting in
crystallization upon cooling. However, the crystals formed in
such systems are nonperiodic layer (NPL) crystallites rather
32
than regular chain-folded crystals. These NPL crystallites are
formed when polymer chains with similar monomer sequences
come in close contact, enabling chain stacking and local
crystallization. The melting temperature of these crystals thus
depends on both the composition of the polymer chains and on
the chain length that participates in the crystallization. It is
anticipated that the chains with long aromatic sequences in
polymer 4c are able to form NPL crystallites with high melting
temperatures, whereas the flexible chains form crystallites with
lower melting temperatures. For this reason, we assume that the
It should be noted that it is difficult to determine whether the
polymers synthesized in this study are truly of a blocky nature
or of a random nature. The determination of the chemical
microstructure is normally performed via 13C NMR or H
NMR analysis. A shift in aromatic protons or carbonyl
resonances is expected for monomers with different neighbors
and the chemical microstructure can be estimated with the ratio
1
endotherm observed below 200 °C, including the T at 164 °C,
m
30
is the melting transition of crystallites containing the more
flexible polymer chains, whereas the part of the transition
between 200 and 300 °C corresponds with the melting of
crystals containing more rigid polymer chains. Such a broad
melting temperature range is characteristic for polymers that
are of a blocky nature. A similar melting behavior was observed
between these shifted signals. Unfortunately, overlapping of
signals observed in the NMR spectra obtained in CDCl /TFA-
3
d mixtures did not allow us to determine the sequence
distribution of the polymer synthesized in this study. Another
route to determine the presence of species with different
17
chemical microstructure is via solvent extraction. In this case,
the polymer chains with different solubility can be extracted
and analyzed separately. Unfortunately, extraction of the
different phases in solvents such as boiling chloroform or
toluene was not successful. The poor solubility of the polymers
in common solvents did not allow us to find a suitable solvent
or solvent mixture to extract different polymer fractions.
Furthermore, it is possible that polymers with a random
chemical microstructure still contain blocks with long aromatic
sequences. For example, as mentioned before, the exchange of
flexible SuA for 2,5-FDCA in polymers 3a and 4c promotes the
formation of large aromatic sequences, simply due to the
increased statistics for aromatic monomers to have an aromatic
neighbor.
for polymers 3b and 5a, which showed a T in DSC at 189 and
m
1
86 °C respectively, but both showed a T melt in POM at 280
lc
°
C.
Contrary to the effects observed when incorporating 2,5-
FDCA, the copolymerization of VA results in a decrease of the
melting temperature. For example, polymer 4d shows a weak
melting endotherm between 90 and 140 °C and a thermotropic
melt above this temperature. Similar melt behavior is observed
for polymer 5b, which shows an LC melt above 170 °C. Such
low T and T are the characteristics for random polymers. As
m
lc
stated earlier, NPL crystals formed in these systems are a result
of agglomeration or stacking of chains with similar sequences.
The formation of these NPL crystals decreases with the
increasing randomness in the monomer sequence distribution,
yielding perturbed crystals with lower melting temperatures.
Furthermore, it is anticipated that the steric hindrance of the
methoxy groups of VA also decreases the ability of the polymer
to crystallize. It should be noted that such perturbation of the
crystallinity can only occur if the methoxy groups are
distributed randomly along the polymer backbone. As we
Nonetheless, the data obtained so far indicates that polymers
3b, 4c, and 5a exhibit characteristic behavior expected for
blocky polymers, whereas polymers 4d and 5b exhibit
characteristic behavior of nonblocky polymers. Thus, in the
following sections of the paper the polymers 3b, 4c, and 5a will
be considered to be blocky, whereas the polymers 4d and 5b
will be considered to be random (or nonblocky).
3
311
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
It is expected that the nonblocky polymers have much lower
crystallinities and show more pronounced glass transitions (Tg)
in DSC and DMTA analysis. Indeed, this is the case when we
For the polymers 4d and 5b, no phase transition from the LC
phase to the isotropic phase is observed prior to degradation of
the polymers. Instead, a threaded texture is observed over the
whole temperature range. Also the partially blocky polymers 4e
and 4f show no isotropization transition in POM. Apparently, a
decrease in the “blockiness” in these TLCPs improves the
homogeneity of the LC melt. Figure 5 shows an example of the
phase separation observed for polymer 3a upon heating and the
stable LC melt during heating for polymer 5b.
compare the T value of polymer 4c with those of polymers 4d
g
or 5b in Figure 3. Figure 4 shows the temperature dependent
behavior of the storage (E′) and loss (E″) moduli for the
TLCPs obtained by DMTA analysis. The decrease in the
crystallinity and the melting temperature can clearly be
detected for the nonblocky polymers 4d and 5b, indicated by
the decrease of the plateau modulus value above T and the
g
rapid deterioration of the E′ modulus at higher temperatures.
As expected, the blocky polymers 4c and 5a have higher
crystalline plateau modulus values above T and broader melt
g
transitions. As already anticipated from the DSC analysis,
polymers 4e and 4f show a partial blocky behavior; these
polymers do show an onset of melting at low temperatures, but
are mechanically more stable at higher temperatures.
Surprisingly, the copolymerization of vanillic acid does not
seem to influence the T to a significant extent, for example, for
g
polymers 3a (T = 42 °C, no VA, DSC) and 4d (T = 45 °C,
g
g
1
0% VA, DSC). In contrast, literature shows that the
copolymerization of VA in PET and BA containing TLCPs
33
did result in an increase of the T with increasing VA content.
g
Interestingly, the copolymerization of 2,5-FDCA does increase
the T of the polymers to values around 80 °C as is detectable
g
by the shift of the peak value of E″ for polymers 4c, 5a, and 5b.
The shift in T is attributed to the increased aromatic content in
g
these polymers.
Figure 5. Optical micrographs obtained, between crossed polarizers,
during constant heating at 10 °C/min. The micrographs depict (A)
the LC phase of polymer 3a observed at 220 °C, (B) the coexistence
of the LC phase and the isotropic phase observed for polymer 3a at
300 °C, (C) the LC phase of polymer 5b at 200 °C, and (D) the LC
phase of polymer 5b at 300 °C.
Melt Behavior of the TLCPs. It is well established that the
homogeneity and the transition temperature of an LC phase are
dependent on both the total amount of aromatic monomers in
the polymer backbone and the chemical microstructure of the
polymer chains. For example, Jackson reported the isotropiza-
tion temperature for compounds with three and four
oxybenzoate units connected with one acetic acid and one
To obtain more information on the stability of the LC melt,
rheological studies using plate−plate geometry as a function of
temperature while heating up to 300 °C were performed on
polymers 3a, 4e, 4d, and 5b. The complex viscosity curves of
the melts thus obtained are depicted in Figure 6. The plate−
plate configuration was chosen over a cone−plate configuration
because it is easier to compensate for the thermal expansion
7
ethanol end-group. The author showed that the isotropization
temperature of the compound with three oxybenzoate units was
282 °C, while the compound with four oxybenzoate units
showed no isotropization prior to degradation above 400 °C.
For processing purposes, the stability of the LC melt is of great
importance, since spinning from an entangled isotropic state
leads to a loss of mechanical performance. Furthermore, in
order to produce a highly oriented product, the presence of
crystals is not desired. For these reasons, the occurrence of a
homogeneous LC melt at the processing temperature is
preferred.
21
The stability and the homogeneity of the LC melt of the
synthesized polymers in this study are determined via polarized
optical microscopy (POM). Upon heating at 20 °C/min, above
300 °C the formation and growth of isotropic droplets is
observed in the LC melt for polymers 3a and 4a. Also blocky
polymers 3b and 4b show a similar phase separation with the
onset at 340−350 °C. However, none of these materials show a
fully isotropic phase prior to degradation. The stability of the
LC melt of blocky polymers 4c and 5a is difficult to investigate
due to the high melting temperatures of these polymers, and
the rapid degradation of the polymer above 350 °C.
Nevertheless, no phase separation of the melt is observed
upon heating to and beyond the degradation temperature.
Apparently, the increased aromatic content, due to the
copolymerization of 2,5-FDCA, stabilizes the LC melt, even
though these polymers are of a blocky nature.
Figure 6. Complex viscosity versus temperature plot showing an
increase of the viscosity above 230 °C for polymer 3a, indicating that a
phase transition is occurring. A continuous decrease of the complex
viscosity is observed for polymers 4d, 4e, and 5b, indicating that a
higher mobility of the LC phase is achieved, but no phase transition is
occurring. The letters A−D correspond to the temperatures where
micrographs A−D in Figure 5 were taken for samples 3a and 5b.
3
312
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
occurring due to the wide temperature range of these
experiments. Similar rheological experiments were performed
by Gao and co-workers on polymers based on sebacic acid, HQ,
Scheme 2. Labels of the Carbon Atoms in the Structural
Formula of the Monomer Residues Present in Polymer 5b
3
4
and BA. These authors showed that the transition from the
crystalline to the nematic phase resulted in a decrease of the
viscosity, whereas the transition from the nematic to the
isotropic phase at higher temperatures increased the viscosity
again. For some polymers in this study, the viscosity measured
in the nematic phase falls below the transducer limit and might
no longer be accurate (visible for polymer 5b in Figure 6).
However, since we are only interested in the dependence of the
viscosity as a function of phase transitions, we can use this data
as a qualitative indicator.
From Figures 5 and 6, it can be seen that polymer 3a exhibits
a stable nematic phase above its melting temperature. However,
the increase in viscosity with temperature indicates occurrence
of a phase transition from the nematic to the isotropic phase.
The POM micrograph obtained for polymer 3a at 300 °C
(
micrograph b, Figure 5) confirms the coexistence of the
isotropic and the LC phase. Similar to our observation in POM,
polymers 4e and 5b do not show the formation of the isotropic
phase, and the viscosities of these polymers only decrease over
temperature. Interestingly, polymer 4d does show a slight
increase in complex viscosity above 250 °C indicating that a
slow phase transition, not observed in POM, might be
occurring. However, other phenomena such as slow degrada-
tion, the formation of gaseous compounds, and cross-linking
might also occur at these elevated temperatures and can also
affect the viscosity during these measurements. Nonetheless,
these data clearly indicate that nonblocky polymers exhibit
stable LC melts over a wide range of temperatures.
Furthermore, these data also show that biobased thermotropic
polyesters with high aromatic contents such as 5b can be
synthesized without exhibiting blocky behavior and with stable
nematic melts up to 300 °C.
polymers 3a, 4c, and 4d are provided in the Supporting
Information (Figures S2−S4).1
Figure 7b shows that no 3C signals are observed in the
13
1
C{ H} INEPT MAS NMR experiments below 130 °C,
indicating that the polymer chains in sample 5b are rigid below
1
13
this temperature; i.e., the polarization transfer from H to
C
via the J-coupling is inhibited due to large 13C− H dipolar
couplings, leading to short, effective T values (Figure 7b).
1
3
6
2
Thus, the material can be described as crystalline and/or rigid
13
amorphous. Upon heating to 130 °C, the C resonances at
1
32.0 (B ), 127.9 (A ) and 122.6 ppm (A and B ) start to rise.
β
β
α
β
These signals appear due to the increased flexibility of the
aromatic rings for BA and BP via rotational motion, which
1
13
37
From the rheological data, it is apparent that the polymers
effectively averages the H− C dipolar couplings. This allows
3
a, 4d, and 4e (without 2,5-FDCA) show the stable complex
for an effective transfer via the J-coupling. The resonances at
viscosity in the LC melt, whereas the polymer 5b (with 2,5-
FDCA) show continuous decrease in the complex viscosity
with the increasing temperature. The difference in the
rheological response of the polymer 5b could be attributed to
the presence of 2,5-FDCA in the polymer. To investigate the
cause of this difference what follows is the solid state NMR
studies performed on these TLCPs.
5
7.4 (V ), 34.9 (S ), 29.5 (S ), and 25.6 (S ) ppm,
α α β γ
corresponding to the aliphatic carbons become apparent
upon further heating to 150 °C. Simultaneously, a weak signal
at 20.8 ppm arises that corresponds to the aliphatic carbon of
the acetoxy end-group. The 13C{ H} CP/MAS spectra of
polymer 5b (Figure 7a) show a similar behavior as the C{ H}
INEPT MAS NMR at low temperature; the peaks sharpen with
increasing temperature and increase in intensity above the glass
transition temperature.
1
1
3
1
Influence of 2,5-FDCA and VA on the Molecular
35
Mobility of the TLCPs. Burgess and co-workers recently
showed that the furan ring exhibits much slower ring rotation/
flipping in PEF compared to the phenyl ring in PET. In this
study, Burgess correlated the slow ring flipping of the furan ring
to the improved barrier properties. It is expected that similar
slowed flipping kinetics of the 2,5-FDCA moiety occurs in our
TLCPs and that the 2,5-FDCA ring therefore requires more
energy to become mobile. Furthermore, we expected that the
methoxy group limits the rotational motion of VA, causing VA
to require a higher temperature to rotate freely too. To
investigate this behavior, temperature-dependent solid-state
As anticipated, the carbon signal of 2,5-FDCA moiety at
1
20.6 ppm (F ) is not visible at 150 °C in Figure 7b. At this
α
temperature, the system is almost fully molten and all
monomers can freely move. However, the F resonance only
α
becomes visible at 170 °C, which is well above the melting
temperature for this polymer (145 °C, DSC analysis, Table 2).
Similarly, the rotational motion of the vanillic acid ring starts at
170 °C, indicated by the development of the signals at 123.9
(V ) and 121.8 (V ) ppm. This data shows that both the 2,5-
FDCA and the vanillic acid moiety require more thermal
energy to become mobile in the LC melt phase. The delayed
mobility in these two moieties might be the cause for the
continuous decrease in the complex viscosity of the polymer
5b, see Figure 6.
γ
δ
1
3
1
13
1
C{ H} INEPT MAS and C{ H} CP/MAS NMR spectros-
copy was performed on polymer 5b. Scheme 2 shows the
structural formula of monomer residues present in polymer 5b
and the labels of the corresponding carbon atoms, while Figure
1
3
1
7
summarizes the data obtained from the C{ H} CP/MAS
The delayed mobility of the 2,5-FDCA moiety compared to
regular phenyl based monomers is attributed to the energetic
penalty of the dipole moment induced by the free electron pairs
13
1
NMR and the C{ H} INEPT MAS NMR measurements. The
C{ H} INEPT NMR and C{ H} CP/MAS NMR results of
1
3
1
13
1
3
313
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
Figure 7. Temperature-dependent solid-state 13C{ H} CP/MAS NMR (A) and C{ H} INEPT MAS NMR (B) results obtained for sample 5b
1
13
1
showing the delayed mobility of the 2,5-furandicarboxylate and vanillic acid moieties occurring above 130 °C. Signals are assigned according to
1
Scheme 2 while the asterisks denote spinning sidebands. All experiments were recorded at 11.75 T (500.13 MHz for H) using a MAS frequency of
1
1
0.0 kHz and H decoupling during acquisition.
of the oxygen heteroatom in the furan ring; thus it hinders
rotational motion. Another possibility for the higher thermal
energy requirement for the furan ring to become mobile is the
melting of NPL crystals with different compositions, as is
possible for blocky polymer 4c. However, since 5b is already
molten before the furan moiety becomes mobile, this is not the
case (see Supporting Information). This data clearly indicates
that both VA and 2,5-FDCA become mobile at higher
temperatures compared to the phenyl-based monomers such
as BA and HQ (4c) or BP (5b). Similar changes in mobility as
a function of temperature were observed for VA in polymer 4d
Table 3. Correlation between the Mechanical Performance
and Crystallinity of the Compression Molded Samples
a
WAXD
tensile tests
b
crystallinity
(%)
E-modulus
(GPa)
stress at break
(MPa)
strain at break
(%)
entry
4
3
5
4
e
a
a
c
51.9 (0.4)
49.5 (0.6)
36.9 (0.1)
36.6 (1.6)
30.3 (0.3)
c
0.64 (0.07)
0.50 (0.04)
1.12 (0.11)
1.22 (0.14)
1.64 (0.10)
1.25 (0.12)
1.23 (0.10)
21.3 (2.9)
17.4 (3.3)
32.0 (6.4)
50.8 (10.3)
51.4 (4.0)
42.0 (5.6)
29.6 (5.2)
10.0 (2.42)
15.1 (4.66)
3.20 (0.70)
7.38 (1.79)
5.16 (0.75)
4.73 (0.95)
2.83 (0.61)
4
d
4
f
(Figure S4), and for 2,5-FDCA in polymer 4c (Figure S3). As
5
b
c
expected, polymer 3a did not show such behavior and only
showed a gradual increase and sharpening of signals with
increasing temperature (Figure S2). These molecular motions
as a function of temperature can be correlated to the
rheological response of these materials, see Figure 6. This
information is interesting for future development and design of
new biobased thermotropic polyesters, containing 2,5-FDCA,
vanillic acid, or other comparable biobased monomers. The
higher threshold in terms of the thermal energy required to
become mobile can potentially contribute to the development
a
b
Standard deviations are displayed between brackets. Obtained from
tensile tests at 20 °C, averaged from 10 tensile tests performed at 5
mm/min at RT of compression molded samples of 1 mm ×12 mm ×2
mm. Polymers showed no diffraction peak after compression molding,
c
only an amorphous halo.
of materials with an increased T value, a better dimensional
g
stability at higher temperatures, and a better high temperature
performance compared to polymers containing regular phenyl-
based monomers.
Mechanical Performance of the TLCPs. To determine
the mechanical performance of these materials, tensile tests
were performed on compression molded samples. All samples
were compression molded at 10 °C above the temperature at
which they were fully molten into the LC phase. Wide angle X-
ray diffraction (WAXD) was performed in the reflection mode
on the compression molded samples to determine the
crystallinity (Table 3). Tensile tests were performed at room
temperature (20 °C) at a constant stretching rate of 5 mm/
min. Figure 8 and Table 3 show the characteristic tensile
behavior observed for these polymers.
Figure 8. Characteristic tensile behavior of tensile tests the TLCPs
synthesized in this study, performed at 5 mm/min at room
temperature of compression molded samples of 1 mm × 12 mm ×
2
mm.
In general, no yielding and low strain at break are anticipated
in thermotropic polyester materials, since the LC domains do
not allow for elongation due to a lack of entanglements. The
polymers 4d, 4f, and 5b, show this expected behavior for
TLCPs. However, polymers 3a, 4c, and 4e exhibit larger strain
at break values. To recall, the polymers 4d, 4f, and 5b are of
nonblocky nature i.e. they are thought to exhibit a
homogeneous distribution of monomers along the main
chain. The nonblocky nature only allows the formation of
small and defected nonperiodic layer (NPL) crystals. This
3
314
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
becomes evident by the broad WAXD peak at 2θ = 20° and low
crystallinity (4d) or absence of crystallinity (5b), as is shown in
Figure 9 and Table 3. It is acknowledged by the authors that the
of vanillic acid appears to decrease the block formation in these
TLCPs. Polymers with a range of melting temperatures and
stable LC melts are obtained. When combined, the application
of 2,5-FDCA and VA can yield polymers that have a high
aromatic content, low melting temperatures, and stable LC
melts up to 300 °C. These polymers show overall low
crystallinity and a good mechanical performance for compres-
sion molded samples. Furthermore, the 2,5-FDCA ring and the
vanillic acid moiety require higher thermal energy to rotate the
aromatic ring, which is anticipated to enhance the dimensional
stability and high temperature performance of these polymers.
Overall, it is concluded that, although 2,5-FDCA is unstable at
temperatures above 250 °C and requires a low temperature
synthetic route for successful incorporation in TLCPs, 2,5-
FDCA is a viable monomer for use in aliphatic−aromatic
thermotropic polyesters.
ASSOCIATED CONTENT
Supporting Information
Additional information regarding the molecular weight build-up
during reduced pressure, of the polymers synthesized in this
■
Figure 9. 1D-WAXD diffractograms showing the difference in
crystallinity (between brackets) and the width of the diffraction peak
around 20° 2θ for the polymers 5b, 4f, 4c, 5a, and 3a. Polymers 5b and
*
S
4
f do not show a distinct diffraction peak, which indicates that they are
13
1
amorphous like, having short-range structural correlation that cannot
be easily detected by WAXD. The amorphous content also increases in
the compression molded isotropic samples.
study, additional temperature-dependent C{ H} CP/MAS
13
1
and C{ H} INEPT MAS NMR experiments for polymers 3a,
c, and 4d. This material is available free of charge via the
4
Internet at http://pubs.acs.org.
rapid cooling of the nonblocky polymers limits crystallization in
the isotropic samples even further, explaining the amorphous
like diffraction patterns.
AUTHOR INFORMATION
Corresponding Author
(S.R.) E-mail: s.rastogi@lboro.ac.uk; sanjay.rastogi@
teijinaramid.com.
■
*
The polymers 3a, 4c, and 4e are of a blocky nature; i.e., they
exhibit a heterogeneous distribution of monomers along the
main chain. This segregation promotes crystallization leading to
the appearance of a sharper WAXD peak at 2θ = 20° (polymer
Notes
The authors declare no competing financial interest.
3a, Figure 9) and higher crystallinity (Table 3). The presence
of noncrystalline domains between the crystalline segments
likely act as a soft component that allow a larger strain at break.
This is pronounced in polymer 3a, having the least amount of
aromatic component. With the addition of 2,5-FDCA in
polymer 4c, i.e., with increasing amount of the aromatic
component, the polymer changes its ductile behavior to brittle
and decreases in crystallinity from 50 to 37% while maintaining
its blocky nature. This implies that the mechanical performance
of these TLCPs is governed by both the aromatic content and
the randomness of the distribution of monomers along the
chain.
From these results, with the exception of the polymer 4e, it
can also be concluded that the presence of vanillic acid
promotes the homogeneous distribution of monomers and the
presence of aromatic 2,5-FDCA enhances the modulus of the
polymer.
ACKNOWLEDGMENTS
■
This work is part of the research program of the Dutch Polymer
Institute (DPI), #739 BioLCP. M.R.H. acknowledges financial
support from the Villum Foundation under the Young
Investigator Programme (VKR023122).
ABBREVIATIONS
,5-FDCA, 2,5-furandicarboxylic acid; BA, p-hydroxybenzoic
■
2
acid or 4-acetoxybenzoic acid; BP, 4,4′-biphenol or 4,4′-
diacetoxybenzoic acid; CP/MAS, cross-polarization/magic-
angle spinning; DMTA, dynamic mechanical thermal analysis;
DSC, differential scanning calorimetry; HQ, hydroquinone or
diacetoxybenzene; INEPT, insensitive nuclei enhanced by
polarization transfer; LC, liquid crystalline; PDI, polydispersity
index; POM, polarization optical microscopy; SuA, suberic
acid; TGA, thermogravimetric analysis; T , glass transition
g
CONCLUSIONS
temperature; TLCP, thermotropic liquid crystalline polyesters;
Tm, peak melting temperature; VA, 4-hydroxy-3-methoxyben-
zoic acid or vanillic acid; WAXD, wide-angle X-ray diffraction
■
In this study, we developed novel biobased, thermotropic
polyesters containing 2,5-FDCA and vanillic acid. It is shown
that 2,5-FDCA was successfully incorporated into the polymer
backbone via the developed low-temperature synthesis route.
Once 2,5-FDCA is incorporated in the chain it remains
thermally stable. The incorporation of 2,5-FDCA as an
additional aromatic component yields polymers that seem to
be of a blocky nature. Furthermore, the T and T increase
upon the replacement of the flexible suberic acid by the rigid
,5-FDCA. The increase of the aromatic content also leads to
an increase of the mechanical performance of compression
molded articles. Opposite to 2,5-FDCA, the copolymerization
REFERENCES
■
(
(
(
1) Han, H.; Bhowmik, P. K. Prog. Polym. Sci. 1997, 22, 1431−1502.
2) Economy, J. Mol. Cryst. Liq. Cryst. 1989, 169, 1−22.
3) Chung, R. S.; Calundann, G. W.; East, A. J. Encycl. Eng. Mater.
1
989, 2, 65.
g
lc
(4) Collyers, A. A. Mater. Sci. Technol. 1989, 5, 309.
(5) Gray, G. W.; Spiess, H.; Vill, V. Handbook of liquid crystals;
2
Demus, D., Goodby, J., Eds.; Wiley-VCH: New York, 1998; Vol. 3.
(6) Liu, L.; Lu, X. Plast., Rubber Compos. 2002, 21, 283−288.
(7) Jackson, W. Br. Polym. J. 1980, 12 (4), 154−162.
3
315
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316

Macromolecules
Article
(
8) Bello, P.; Bello, A.; Riande, E.; Heaton, N. J. Macromolecules
2
(
(
001, 34, 181−186.
9) Kuhfuss, H.; Jackson, W. J., Jr. US Patent 3,804,805, 1974.
10) Jasinska-Walc, L.; Villani, M.; Dudenko, D.; van Asselen, O.;
Klop, E.; Rastogi, S.; Hansen, M. R.; Koning, C. E. Macromolecules
012, 45, 2796.
11) Jasinska-Walc, L.; Dudenko, D.; Rozanski, A.; Thiyagarajan, S.;
2
(
Sowinski, P.; van Es, D.; Shu, J.; Hansen, M. R.; Koning, C. E.
Macromolecules 2012, 45, 5653.
(
(
12) Zhang, M. CN101318892(A), 2008.
13) Ayorinde, F. O.; Osman, G.; Shepard, R. L.; Powers, F. T.
JAOCS, J. Am. Oil Chem. Soc. 1988, 65, 1774−1777.
14) Hsiue, L. T.; Ma, C. C. A.; Tsai, H. J. Appl. Polym. Sci. 1995, 56,
71−476.
15) De Ruijter, C.; Bos, J.; Boerstoel, H.; Dingemans, T. J. Polym.
Sci., Part A: Polym. Chem. 2008, 46, 6565−6574.
16) Nicely, V. A.; Dougherty, J. T.; Renfro, L. W. Macromolecules
987, 20, 573−578.
17) Quach, L.; Hornogen, E.; Volksen, W.; Economy, J. J. Polym.
Sci., Part A: Polym. Chem. 1989, 27, 775−784.
18) De Oca, H. M.; Wilson, J. E.; Penrose, A.; Langton, D. M.;
(
4
(
(
1
(
(
Dagger, A. C.; Anderson, M.; Farrar, D. F.; Lovell, C. S.; Ries, M. E.;
Ward, I. M.; Wilson, A. D.; Cowling, S. J.; Saez, I. M.; Goodby, J. W.
Biomaterials 2010, 31, 7599−7605.
(
1
(
1
19) Yang, D. K.; Krigbaum, W. R. J. Polym. Sci., Part B: Polym. Phys.
989, 27, 1937−1851.
20) Yang, D. K.; Krigbaum, W. R. J. Polym. Sci., Part B: Polym. Phys.
989, 27, 819−835.
21) Muramatsu, H.; Krigbaum, W. R. J. Polym. Sci., Part B: Polym.
(
Phys. 1987, 25, 2303−2314.
(
(
22) Shaikh, A.; et al.. US 2012/0302772 A1, 2012.
23) Wilsens, C. H. R. M., Noordover, B. A. J., Rastogi, S.
DOI:10.1016/j.polymer.2014.03.033.
24) Bennett, A. E.; Rienstra, C. M.; Auger, C. M.; Lakshmi, K. V.;
Griffin, R. G. J. Chem. Phys. 1995, 103, 6951.
(
(25) Morris, G.; Freeman, R. J. Am. Chem. Soc. 1979, 101, 760−762.
(26) Burum, D. P.; Ernst, R. R. J. Magn. Reson. 1980, 39, 163−168.
(27) Bielecki, A.; Burum, D. P. J. Magn. Reson., Ser. A 1995, 116,
2
(
15−220.
28) Gruter, G-J. M.; Sipos, L.; Dam, M. A. Comb. Chem. High
Throuhput Screening 2012, 15, 180−188.
(
(
29) Drewitt, J. G. N.; Lincoln, J. Br. Patent, 621971, 1946.
30) Li, X. G.; Huang, M.-R.; Guan, G.-H.; Sun, T. Angew. Makromol.
Chem. 1995, 227, 69−85.
(
(
31) Nagata, M. J. Appl. Polym. Sci. 2000, 78, 2474−2481.
32) Windle, A. H.; Viney, C.; Golombok, R.; Donald, A. M.;
Mitchell, G. R. Faraday Discuss. Chem. Soc. 1985, 79, 55−72.
33) Li, X. G.; Huang, M.-R.; Guan, G.-H.; Sun, T. J. Appl. Polym. Sci.
996, 59, 1−8.
34) Gao, P.; Lu, X. H.; Chai, C. K. Polym. Eng. Sci. 1996, 36 (22),
771−2780.
35) Burgess, S. K.; Leisen, J. E.; Kraftschik, B. E.; Mubarak, C. R.;
(
1
(
2
(
Kriegel, R. M.; Koros, W. J. Macromolecules 2014, DOI: 10.1021/
ma5000199.
(
36) Schmidt-Rohr, K.; Spiess, H. W. Multidimensional Solid-State
NMR and Polymers; Academic Press: New York, 1994.
37) Hansen, M. R.; Graf, R.; Spiess, H. W. Acc. Chem. Res. 2013, 46,
996−2007.
(
1
3
316
dx.doi.org/10.1021/ma500433e | Macromolecules 2014, 47, 3306−3316