The synthesis and characterisation of reactive poly(p-phenylene terephthalamide)s: A route towards compression stable aramid fibres

Article abstract of DOI:10.1016/j.polymer.2010.03.015

Herein we report on the synthesis of reactive poly(p-phenylene terephthalamide) (PPTA) oligomers and the preparation and characterisation of aramid fibres thereof. Methacrylate and maleimide reactive end-groups were found to be sufficiently stable in H₂SO₄ at 85?°C and they were used to prepare reactive PPTA oligomers. Lyotropic spin-dopes could be prepared with up to 20?wt% of reactive oligomer (M_n?=?3900?g?mol^-1) and this modification did not interfere with the fibre spinning process and had no effect on the fibre tensile properties. The as-spun fibres did indeed show a modest (+0.1?GPa) improvement in compression strength. A high temperature treatment at 380?°C resulted in fibres which all show a significant increase in compressive strength over their as-spun precursors, i.e. from 0.7 to 0.9?GPa. When fibres were treated at 430?°C the compression values of the oligomer-modified fibres dropped somewhat, whereas unmodified PPTA displayed a compressive strength of 1.1?GPa. Other favourable fibre properties such as modulus and tenacity were not compromised.

Full text of DOI:10.1016/j.polymer.2010.03.015

Polymer 51 (2010) 1887e1897
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
The synthesis and characterisation of reactive poly(p-phenylene terephthalamide)s:
A route towards compression stable aramid fibres
,
Alwin Knijnenberg a, Johan Bos b, Theo J. Dingemans a
*
a Delft University of Technology, Faculty of Aerospace Engineering, Kluyverweg 1, 2629 HS Delft, The Netherlands
b Teijin Aramid B.V., Research Institute, P.O. Box 9300, 6800 SB Arnhem, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein we report on the synthesis of reactive poly(p-phenylene terephthalamide) (PPTA) oligomers and
the preparation and characterisation of aramid fibres thereof. Methacrylate and maleimide reactive end-
groups were found to be sufficiently stable in H2SO4 at 85 ꢀC and they were used to prepare reactive PPTA
oligomers. Lyotropic spin-dopes could be prepared with up to 20 wt% of reactive oligomer
(Mn ¼ 3900 g molꢁ1) and this modification did not interfere with the fibre spinning process and had no
effect on the fibre tensile properties. The as-spun fibres did indeed show a modest (þ0.1 GPa)
improvement in compression strength. A high temperature treatment at 380 ꢀC resulted in fibres which
all show a significant increase in compressive strength over their as-spun precursors, i.e. from 0.7 to
0.9 GPa. When fibres were treated at 430 ꢀC the compression values of the oligomer-modified fibres
dropped somewhat, whereas unmodified PPTA displayed a compressive strength of 1.1 GPa. Other
favourable fibre properties such as modulus and tenacity were not compromised.
Received 14 October 2009
Received in revised form
17 February 2010
Accepted 8 March 2010
Available online 15 March 2010
Keywords:
PPTA
Compressive strength
Reactive oligomers
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
0.02 to well above 1.0 GPa, this trend strongly suggests that
stronger lateral interactions indeed increase the fibre compressive
High performance polymeric fibres such as poly(p-phenylene
terephthalamide) (PPTA), poly(pyridobisimidazole), or PIPD (a fibre
which is currently marketed under de name ‘M5’) and ultra high
molecular weight polyethylene fibres (Dyneema) show specific
linear tensile properties based on which they outperform their
inorganic counterparts like carbon and glass fibres [1e4]. A major
concern is, however, their strength under compressive loading
which in general is found to be lower then 1 GPa and this signifi-
cantly limits their use in structural applications [5]. At present,
there is general consensus that the low compressive strength of
polymeric fibres finds its origin in the mediocre lateral interactions
between the polymer chains, i.e. the poor inter-molecular interac-
tions result in a low inter-molecular shear-strength. This idea is
supported by the increasing compressive strengths when
comparing high-modulus fibres of comparable tensile properties
such as UHMWPE, para-aramid, and PIPD fibres. In these fibres, the
lateral interactions between the molecular chains are the results of
van der Waals forces, hydrogen bonds forming one-dimensional
(para-aramid) and hydrogen bonds forming two-dimensional
networks (PIPD). Since the compressive strength increases from
strength. It should be noted that strong lateral interactions are not
only limited to hydrogen bonds, but can also involve other
favourable electrostatic interactions such as dipoleedipole,
pep
and molecular polarizability.
Several attempts to enhance lateral interactions via crosslinking
adjacent polymer chains have been explored in the past. During the
mid-90s a series of articles were published on the synthesis of 1,2-
dihydrocyclobutabenzene-3,6-dicarboxylic acid monomer and its
successful incorporation into PPTA fibres [6e9]. The formation of
crosslinks raised the strain to kink band formation and decreased
the kink band density at a given strain. Although strong evidence
for the control over the compressive behaviour was presented, no
numerical data on the compressive strengths were reported. At the
same time another monomer, disulfonediamine, was explored by
Suter and co-workers [10,11] and incorporated into fibres using this
monomer as a substituent for p-phenylene diamine. The fibres,
however, displayed poor mechanical properties, which is likely to
be the result of the distortion of the crystal structure due to the
bulkiness of the DSDA monomer. Other chemical modifications
include the incorporation of halogenated monomers into PPTA and
PBZT fibres by Sweeny leading to significant deterioration of tensile
properties [12]. Incorporation of methyl-pendant monomers into
PBI and PBZT fibres by Jenkins et al. resulted in higher tensile but
lower compressive strengths upon heat treatment [13], and
* Corresponding author. Tel.: þ31 (0) 15 27 84520; fax: þ31 (0) 15 27 84472.
E-mail address: t.j.dingemans@tudelft.nl (T.J. Dingemans).
0032-3861/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.03.015

1888
A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
diacetylene functionalised PPTA fibres with improved tensile
properties were prepared by Beckham and Rubner [14]. Instead of
introducing monomers capable of forming covalent bonds, Xue and
Hara [15,16], and less successfully by Glomm and co-workers [17],
made use of ionic moieties to enhance the cohesive forces between
adjacent polymer chains. Xue and Hara’s results showed that the
number of kink bands decreased as a result of the presence of
lateral ionic bonds.
crosslinking reactions are allowed. In addition, the conditions for
activating the reactive end-groups should not result in degradation
of the polymer backbone chemistry or disrupt the final fibre
structure and morphology. In order to gain more insight in the
chemical stability, amide model compounds were prepared and
investigated by NMR. An overview of the end-groups that were
selected to prepare these model compounds is shown in Fig. 1.
Maleimides and phenylethynyl end-groups are known for their use
in high performance polymers [24]. Adding substitutes such as
a methyl group, or phenyl ring, possibly shield the reactive double
bond in the acid environment by steric hindrance. In addition, we
also tested a series of acrylates, which were anticipated to be stable
in acidic environments since their synthesis often involves acidic
reaction conditions [25,26], thus suggesting a good chemical
stability. Furthermore, it was believed that the fluorine atoms might
have a shielding effect protecting the reactive double bound. Aniline
was used to prepare oligomers with a non-reactive end-group.
Based on results of the aforementioned analyses, chemically
stable end-groups were selected and used to end-cap aramid
oligomers, which were incorporated in aramid fibres. These fibres
were extensively tested to determine the effect of our approach e i.e.
incorporating reactive oligomers into fibres and subjecting them to
a thermal post-treatment e on the linear tensile properties and in
particular on the critical compressive stresses. Since literature
results showed that a high concentration of crosslinkable func-
tionalities (>5 mol%) resulted in a modest increase in compressive
strength but a drop in other favourable fibre properties such as
tenacity and elongation at break, we decided to keep the concen-
tration of the reactive oligomers low.
Other attempts to enhance lateral interactions via crosslinking
rely on the addition of a crosslinking agent or infiltration of
crosslinkable polymers rather than on changing the polymer
backbone chemistry. The most successful attempt at present is the
development of a new polymer, PIPD, which is capable of forming
a 2 dimensional hydrogen bonded network. As a result of the high
lateral interactions between polymer chains, compressive strengths
as high as 1.7 GPa were reported [18]. This polymer is rather
expensive and the compressive properties appear to be sensitive to
water, which reduces compressive performance over time [19e21].
We report here on a novel approach towards the synthesis of
compression stable aramids by incorporating reactive end-groups
on the poly(p-phenylene terephthalamide) (PPTA) chain ends. This
new concept is based on PPTA-compatible reactive oligomers,
which are added to the standard PPTA spin-dope prior to the fibre
spinning process and which are activated in a thermal post-pro-
cessing step. This approach was inspired by our previous work on
reactive all-aromatic thermotropic oligomers. The processing and
toughness of all-aromatic liquid crystal polymers could be
improved significantly when crosslinkable end-groups were intro-
duced at the polymer chain ends [22,23]. In case of PPTA, the olig-
omers with reactive end-groups were designed to match the
molecular packing preferences of PPTA-based fibres and not to
interfere with the lyotropic spinning process. In particular, we
describe here a systematic approach for careful selection of suitable
end-groups and a route to incorporate them into PPTA fibres. The
spinning process carried out in sulphuric acid (>98.5%) at 85 ꢀC
requires any end-group to be incorporated into the PPTA fibre to be
chemically stable in 85 ꢀC sulphuric acid for at least the duration of
the spinning process. i.e. no chemical degradation or premature
2. Experimental
2.1. Materials
All materials were obtained from SigmaeAldrich and used as-
received unless stated otherwise. Acetic anhydride 97% was
obtained from J.T. Baker, p-phenylene diamine pellets from DuPont,
terephthaloylchloride and the methylmaleimido benzamide end-
Fig. 1. Selection of reactive end-groups used for the preparation of amide model compounds.

A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
1889
Scheme 1. Synthetic route (A) towards maleimidobenzoyl chloride (1) and (B) the coupling of (1) with p-phenylene diamine to prepare the amide model compound (I).
groups from Teijin Aramid B.V. Tigloyl chloride (98þ%) was
purchased from TCI Europe and 4-(trifluorovinyloxy)benzoyl
chloride from Fluorochem Ltd.
consists of a primary 60 mm Göbel focussing mirror, a parabolic Ni/
C multilayer device, providing Cu-K
radiation, and 0.12ꢀ Soller
slits. The diffractometer is equipped with scintillation detector and
auto-changer.
a
2.2. Measurements
2.3. Synthesis of reactive end-groups
1H NMR and 13C NMR spectra were recorded on a Varian Unity
INOVA 300 spectrometer operated at 300 and 75 MHz respectively.
Chloroform-d (CDCl3) and dimethyl sulfoxide-d6 (DMSO-d6) were
used as solvents and sulphuric acid-d2 (98 wt% in D2O, with
99.5 atom% D) was used as the solvent for determining the chemical
stability of the model compounds under acid conditions at 85 ꢀC.
For such measurements, reference spectra were recorded at room
temperature in DMSO-d6 and D2SO4. Next, the sample was heated
to 85 ꢀC within the spectrometer and 3 subsequent spectra were
recorded at time intervals of 10 min.
In order to acquire information on the oligomer molecular
weight distribution, size exclusion chromatography (SEC) experi-
ments were performed. The oligomers were dissolved in 100%
sulphuric acid (0.1 mg mLꢁ1) and eluted over a Zorbax GPC column
(250 ꢂ 6.2 mm). For detection a UV detector (340 nm) was used.
The molecular weight values were calculated using Cirrus version
1.1 GPC software (Polymer Laboratories). As references, a high
molecular weight PPTA (yarn type 1010) and a PPTA-based trimer
were used.
Thermal properties of end-groups and model compounds were
investigated using a PerkinElmer Pyris Diamond TG/DTA and
a PerkinElmer Sapphire DSC. Unless stated otherwise, samples of
typically 5 mg were investigated while heating at 10 ꢀC minꢁ1 and
using nitrogen as a purge gas. Curing experiments were done using
a Linkham THMS hot-stage pre-heated to the desired temperatures
and purged with nitrogen.
The linear fibre properties were determined on an Instron 5543
tensile tester. For each fibre 10 filaments with a gauge length of
100 mm were tested at a speed of 10 mm minꢁ1. Experiments were
performed under standard laboratory conditions of 21 ꢀC and 65%
relative humidity. The compressive properties of single filaments
were determined by the elastica single filament compression test
[27,28].
2.3.1. Maleimidobenzoyl chloride (1)
The synthesis of the maleimide reactive end-group was per-
formed using standard synthetic procedures as summarised in
Scheme 1. The maleimidobenzoyl chloride was prepared from
maleic anhydride and p-aminobenzoic acid according to a literature
procedure [29].
The title compound was purified via recrystallisation (CH2Cl2:
hexane, 50:50) or sublimation. 1H NMR (CDCl3)
d(ppm): 6.9 (s, 2H),
7.6 (d, J ¼ 8.7 Hz, 2H), 8.2 (d, J ¼ 8.7 Hz, 2H); 13C NMR (CDCl3)
d
(ppm): 125.1, 131.7, 132.2, 134.6, 137.7, 167.5, 168.6; Tm ¼ 157 ꢀC. The
melting point was found to be in agreement with literature [30].
2.3.2. Phenylmaleimidobenzoyl chloride (3)
The phenyl-substituted maleimide was synthesised from phe-
nylmaleic anhydride and p-aminobenzoic acid employing
the previously described [25]. 1H NMR (DMSO-d6)
d
(ppm): 7.5e7.6
d(ppm): 126.2, 127.1, 129.4,
(m, 6H), 8.1 (m, 4H). 13C NMR (DMSO-d6)
129.6, 130.4, 130.6, 131.8, 136.3, 143.6, 167.4, 169.5, 169.8; Tm ¼ 261 ꢀC.
The acid functionality was converted into an acid chloride by
refluxing the product in SOCl2 and removing the excess by distilla-
tion. The acid chloride was used immediately in the next step.
2.3.3. 4-Phenylethynylaniline (9)
4-Phenylethynylaniline was prepared via a palladium-catalysed
coupling of ethynylbenzene and 4-iodoaniline [31,32]. The product
was recrystallised from n-heptane/ethylacetate (9:1) yielding
brown flakes. TLC(4/1 hexane/ethyl acetate) tr ¼ 0.10 (one spot); 1H
NMR (CDCl3)
d
(ppm): 3.8 (b, eNH2) 6.6 (d, J ¼ 8.6, 2H), 7.23e7.35
(m, 5H), 7.47e7.51 (m, 2H); 13C NMR (CDCl3)
d
(ppm): 87.35, 90.15,
112.61,114.75,123.92,127.65,128.27,131.35,132.95,146.66. MS (m/z):
193 (Mþ).
X-ray diffraction was used to determine the crystal unit cell
parameters and the apparent lateral crystal size. The XRD
2.4. Synthesis of model compounds
measurements were carried out on a Bruker D8 diffractometer in
q/
In order to further investigate the end-groups with respect to
2
q
geometry, equipped with parallel beam optics. The optics
their stability in strong acidic environments, well-defined amide-

1890
A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
Scheme 2. Synthetic route towards poly(p-phenylene terephthalamide) oligomers with reactive maleimide (1) or methacryloyl (5) end-groups.
based model compounds were synthesised via a condensation
coupling of the amine, or acid (chloride) functionalised end-
groups, with either terephthaloylchloride (TDC) or p-phenylene
diamine (PPD). As a representative example, the synthesis of
a maleimide (I) end-capped model compound is described below.
2.4.2. Methylmaleimide amide (II) or N,N0-1,4-phenylenebis[4-(2,5-
dihydro-3-methyl-2,5-dioxo-1H-pyrrol-1-yl)-benzamide
Texo ¼ 376/417 ꢀC; 1H NMR (D2SO4)
d(ppm): 2.08 (6H, s,), 6.70
(4H, s,), 7.41 (4H, d,), 7.69 (4H, d,), 8.13 (4H, s,)
All other model compounds were prepared using
procedure.
a similar
2.4.3. Phenylmaleimide amide (III) or N,N0-1,4-phenylenebis[4-
(2,5-dihydro-3-phenyl-2,5-dioxo-1H-pyrrol-1-yl)-benzamide
Tm ¼ 363 ꢀC, Texo ¼ 376/379 ꢀC. 1H NMR (D2SO4)
d(ppm): 7.4e8.3
2.4.1. Maleimide (MIBC) amide (I) or N,N0-1,4-phenylenebis[4-(2,5-
dihydro-2,5-dioxo-1H-pyrrol-1-yl)-benzamide [33]
(m, 24H)
Maleimidobenzoyl chloride 1.55 g (6.6 mmol) was charged into
a 100 mL round bottom flask and immediately dissolved in 40 mL
anhydrous CH2Cl2. After cooling to 0 ꢀC, 0.328 g (3.0 mmol) p-
phenylenediamine (PPD) dissolved in 30 mL CH2Cl2 was slowly
added (step B in Scheme 1) to the maleimide benzoyl chloride
solution, resulting in an instant change of the reaction mixture to
an opaque yellow/brown colour. Anhydrous pyridine (2 mL,
24.8 mmol) was added as an acid scavenger, and the mixture was
allowed to warm up to room temperature overnight. Finally the
solvent was evaporated and the product was precipitated in
demineralised water and collected by filtration. The final product
was refluxed in excess tetrahydrofuran (THF) in order to remove
any remaining starting material. The beige coloured solids were
then collected by filtration and dried under vacuum (50 ꢀC).
Texo ¼ 289/315 ꢀC (onset to/peak of reaction exotherm); 1H NMR
2.4.4. Acrylate amide (IV) or N,N0-p-phenylenebisacrylamide
Tm ¼ 276 ꢀC (242 ꢀC [34]), Texo ¼ 282/289 ꢀC. 1H NMR (DMSO-d6)
d
(ppm): 5.8 (m, 2H), 6.3 (m, 2H), 6.5 (m, 2H), 7.6 (s, 4H); 13C NMR
(DMSO-d6) d(ppm): 119.6, 126.4, 131.5, 134.4, 162.8
2.4.5. Methylmethacrylate (MMA) amide (V) or N,N0-p-
phenylenebis methacrylamide
Tm ¼ 256 ꢀC (248 ꢀC [34]) 1H NMR (DMSO-d6)
d(ppm): 2.0 (s,
6H,), 5.5 (s, 2H, vinyl group), 5.8 (s, 2H), 7.6 (s, 4H), 9.8 (s, 2H);
13C NMR (DMSO-d6)
d(ppm): 18.4, 119.6, 120.2, 134.3, 140.0,
166.4.
2.4.6. 3,3-Dimethacrylate amide (VI) or N,N0-p-phenylenebis-3,3-
dimethacrylamide
Tm ¼ 264 ꢀC; 1H NMR (DMSO-d6)
d(ppm): 1.9 (s, 6H), 2.1 (s, 6H),
(D2SO4) d(ppm): 6.27 (4H, s), 6.71 (4H, d), 6.99 (4H, d), 7.44 (4H, s);
5.8 (s, 2H), 7.5 (s, 4H), 9.8 (s, eNH); 13C NMR (DMSO-d6)
d(ppm):
13C NMR (D2SO4)
135.9, 171.8, 174.7
d(ppm): 126.3, 129.1, 131.1, 132.5, 133.5, 133.7,
19.3, 26.9, 119.1, 119.2, 134.7, 150.7, 164.2.
Table 1
Starting materials used for the synthesis of poly(p-phenylene terephthalamide) oligomers end-capped with reactive groups.
Oligomer
Selected end-group
PPD
TDC
PPTA-ANIL-5K
PPTA-MMA-5K
PPTA-MIBC-5K
PPTA-ANIL-2.5K
PPTA-ANIL-10K
Aniline
Methacryloyl chloride
Maleimidobenzoyl chloride
Aniline
Aniline
Methacryloyl chloride
Methacryloyl chloride
Methacryloyl chloride
Methacryloyl chloride
Maleimidobenzoyl chloride
Maleimidobenzoyl chloride
0.94 g (0.010 mol)
1.05 g (0.010 mol)
2.07 g(0.0087 mol)
1.97 g (0.021 mol)
0.46 g (0.005 mol)
4.80 g (0.045 mol)
2.27 g (0.022 mol)
0.79 g (0.008 mol)
2.35 g (0.022 mol)
11.41 g (0.048 mol)
3.12 g (0.013 mol)
21.60 g (0.20 mol)
21.60 g (0.20 mol)
19.46 g (0.18 mol)
21.17 g (0.20 mol)
21.17 g (0.20 mol)
13.00 g (0.12 mol)
10.46 (0.10 mol)
9.31 g (0.09 mol)
12.07 g (0.11 mol)
10.47 g (0.10 mol)
7.14 g (0.07 mol)
40.56 g (0.20 mol)
40.56 g (0.20 mol)
36.54 g (0.18 mol)
39.73 g (0.20 mol)
39.74 g (0.20 mol)
12.20 g (0.06 mol)
14.74 (0.07 mol)
15.89 g (0.08 mol)
20.40 g (0.10 mol)
14.74 g (0.07 mol)
12.07 g (0.06 mol)
PPTA-MMA-0.5K-Ia
PPTA-MMA-1K-Ia
PPTA-MMA-2.5K-Ia
PPTA-MMA-2.5K-IIb
PPTA-MIBC-1.2K-Ia
PPTA-MIBC-2.5K-IIb
a
Calculations based on the exact molecular weight of an oligomer backbone.
Calculations based on Carothers general equation modified for end-capping both polymer chain ends.
b

A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
1891
Table 2
Chemical and thermal properties of the reactive model compounds.
Tm (ꢀC)b
Texo (ꢀC)c
a
Molecular structure
Stable in H2SO4
I
Y
e
289/315
376/417
II
N
N
e
III
363
376/379
IV
N
Y
N
Y
Y
276
256
264
230
e
282/289
V
e
e
e
e
VI
VII
VIII
IX
X
N
N
N
391
e
394/404
274/301
e
XI
345
e Not observed.
a
Chemically stable in 85 ꢀC sulphuric acid (>98.5%) for 30 min.
Onset to melt.
Onset/Peak value for exotherm.
b
c
2.4.7. Tigloyl amide (VII) or N,N0-p-phenylenebis-2,3-dimethacrylamide
2.4.8. Trichloroacrylate amide (VIII) or N,N0-p-phenylenebis-2,3,3-
Tm ¼ 230 ꢀC; 1H NMR (DMSO-d6)
d
(ppm): 1.8 (m, 12H), 6.4 (m,
trichloroacryl-amide
2H), 7.6 (s, 4H), 9.5 (s, eNH); 13C NMR (DMSO-d6)
d
(ppm): 13.2,
1H NMR (DMSO-d6) (ppm): 7.62 (s, 4H), 11.00 (s, NH); 13C NMR
d
14.5, 120.9, 130.7, 133.3, 135.5, 168.2.
(DMSO) d(ppm): 120.22, 121.86, 124.35, 134.26, 158.21.

1892
A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
2.4.9. Phenylethynyl amide (IX) or N,N0-diphenylethynyl-1,4-
which resulted in a spin-dope with a maximum PPTA content of
19.8%. In a twin-screw extruder the spin-dope was heated to 85 ꢀC
and pushed though a spinneret, stretched within the air-gap and
finally coagulated in water with a temperature of around w5 ꢀC.
After drying, fibres were annealed at temperatures at either 380 or
430 ꢀC in order to induce crosslinking via the reactive end-groups.
All fibres were spun at a line speed of 200 m minꢁ1. The fibres were
labelled according to the end-group used. For example, ANIL20-200
refers to a fibre based on 20 wt% aniline terminated oligomer spun
at a line speed of 200 m minꢁ1, whereas ANIL20-200HT refers to the
same fibre but heat-treated at 380 ꢀC.
benzenedi-carboxamide
Tm ¼ 392 ꢀC, Texo ¼ 394 ꢀC. 1H NMR (D2SO4)
d(ppm): 7.3e8.4 (m,
22H)
2.4.10. 4-(Trifluorovinyloxy) amide (X) or N,N0-1,4-phenylenebis[4-
[(trifluoroethenyl) oxy]-(9 cl) benzamide
13H NMR (DMSO-d6)
d
(ppm): 7.5 (d, 4H), 7.7 (s, 4H), 8.1 (d, 4H),
10.3 (s, eNH). 13C NMR (DMSO-d6)
(ppm): 106.7, 116.2, 121.4, 130.0,
132.6, 135.6, 157.0, 164.8. 19F NMR (DMSO-d6)
d
d
(ppm): ꢁ117.2 (1F,
dd, cis CF]CF2), ꢁ125.3 (1F, dd, trans CF]CF2), ꢁ134.4 (1F, dd, CF]
CF2).
3. Results and discussion
2.4.11. Aniline (ANIL) amide (XI) or N,N0-diphenyl-1,4-
benzenedicarboxamide
3.1. Synthesis and characterisation of end-groups, model
compounds and oligomers
Tm ¼ 345 ꢀC. 1H NMR (DMSO-d6)
d(ppm): 7.13 (t, 2H), 7.38 (t, 4H),
7.81 (d, 4H), 8.11(s, 4H), 10.39 (s, 2H). 13C NMR (DMSO-d6)
d(ppm):
121.2, 124.6, 128.4, 129.4, 138.2, 139.7, 165.5.
The model compounds were synthesised following a straight-
forward condensation coupling of the acid chloride or amine
functionalised end-group with either p-phenylenediamine or ter-
ephthaloylchloride. The end-groups that were used are readily
available or easy to synthesise via standard literature procedures.
Final products were purified by recrystallisation or by washing
them with hot THF to remove any remaining starting materials or
side products. NMR using either D2SO4 or DMSO-d6 as a solvent
confirmed the structure of all model compounds.
Thermal analysis revealed that maleimide (I) and methyl (II)
maleimide end-capped model compounds, did not show a well-
defined melt-transition and that the phenyl maleimide melted at
363 ꢀC. The starting temperature for the endothermic reaction of the
maleimide functionality (289 ꢀC) in the unsubstituted maleimide (I),
is influenced by the presence of a substituent such as a methyl (II) or
phenyl (III) group and increases towards 376 ꢀC for the latter ones e
a shift of approximately 90 ꢀC. An overview of all model compounds
and their thermal properties is provided in Table 2.
In contrast, melting temperatures for acrylates were rather low
and in a range of 230e276 ꢀC. Immediately upon melting, the
unsubstituted acrylate (IV) shows an exothermal peak in the DSC
curve whereas no exothermal reactions could be detected for the
substituted acrylates (VeVIII). Both the phenylethynyl (IX) and the
trifluorovinyloxy (X) compound show distinct curing reactions in
the liquid and solid phase respectively. The aniline (XI) end-capped
model compound, synthesised as inert reference only melted upon
heating.
2.5. Synthesis of oligomers
All oligomer variations were prepared according to the synthetic
procedure as shown in Scheme 2. As a representative example, we
describe the preparation of a 5000 g molꢁ1 maleimidobenzoyl
chloride end-capped oligomer. In preparation of the synthesis, all
glassware were pre-dried in an air-circulation oven at 120 ꢀC for at
least 60 min. NMP was dried over and distilled from P2O5 and
stored under a continuous nitrogen flow. CaCl2 was dried in
a vacuum oven at 200 ꢀC for at least 12 h prior to the experiments.
The final solvent mixture generally contained less then 200 ppm
H2O.
PPTA-MIBC-5K. A 2 L reactor vessel equipped with a mechanical
stirrer, nitrogen inlet and vacuum connection was charged with
w270 mL NMP/CaCl2 (8.5 wt%) and 19.46 g (0.18 mol) p-phenylene
diamine (PPD). After closing the reactor, a vacuum was applied on
the reaction mixture followed by a nitrogen purge. This procedure
was repeated at least 3 times in order to remove all traces of
moisture and air. This mixture was stirred (150 rpm) for 30 min
while heating at 60 ꢀC and sonicating for 10 min in order to dissolve
all PPD monomer. This mixture was cooled to w0 ꢀC and 36.54 g
(0.18 mol) terephthaloylchloride (TDC) and 2.07 g (0.0087 mol)
maleimidobenzoyl chloride (MIBC) were added. The stirring rate
was increased to 320 rpm and the reaction mixture was stirred for
at least 20 min while monitoring torque and temperature. Directly
following the addition of the TDC/MIBC components the temper-
ature increased to around 37 ꢀC and the reaction mixture formed
a yellow thick slurry, which transformed into a claylike substance
towards the end of the reaction.
3.2. Chemical stability of end-groups in strong acid environment
Good chemical stability of the end-groups in 85 ꢀC sulphuric
acid is important to survive the fibre spinning process and to be
effective in the subsequent heat treatment process. This means that
the end-groups should not undergo any premature reactions that
prevent proper crosslinking in the latter heat treatment step or
even interfere with the fibre spinning process, resulting in deteri-
oration of tensile properties. The spin-dope has to remain stable for
at least 30 min at 85 ꢀC before the fibres are spun, coagulated in
water and washed. The most important technique that we
employed for investigating the chemical stability was the evolution
of 1H NMR spectra, recorded every 10 min with D2SO4 as the
solvent at 85 ꢀC. Where possible or required, additional NMR
methods were employed, i.e. 13C NMR and 19F NMR.
The final product was coagulated in demineralised water and
washed 4 times before being collected by filtration. The yellow
fibrous product was then dried for 24 h in an 80 ꢀC vacuum
(w25 mbar) oven and another 24 h in a 50 ꢀC vacuum oven
(<1 mbar). After drying, 37 g (83%) oligomer was collected for
further processing and characterisation. The quantities of starting
materials used for the syntheses are summarised in Table 1. The
yields were generally found to be above 80%.
2.6. Fibre spinning
In preparation of the dry-jet wet spinning process for the
manufacturing of the fibres, spin-dopes containing 20 wt%
5000 g molꢁ1 oligomer and 80 wt% high molecular weight
(Mn ¼ 10,000 g molꢁ1) PPTA polymer were prepared. Both oligomer
and polymer were ground, filtered and at a temperature of ꢁ20 ꢀC
mixed with sulphuric acid (99.8%) flakes and stirred overnight,
The 1H NMR reference spectrum of the methacrylate (V) end-
capped model compound, as shown in Fig. 2(a), was obtained at
24 ꢀC immediately after dissolving this compound in D2SO4. Upon
inspection of the proton spectra all-aromatic and -allylic peaks can
be accounted for, i.e. a single peak for the methyl protons, two

A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
1893
Fig. 2. 1H NMR spectra of the methacrylate (V) end-capped model compound in D2SO4 recorded at t ¼ 0 (a), t ¼ 10 (b), t ¼ 20 (c) and t ¼ 30 (d) minutes, at temperatures of 24 ꢀ
C
(a) and 85 ꢀC (b, c and d) respectively.
single peaks that can be assigned to the vinyl protons and a single
peak for the aromatic protons. Compared to the DMSO-d6 spectrum
(Experimental section) a small downfield chemical shift for the
peaks can be detected, which can be explained by the change of
solvent, i.e. DMSO-d6 to D2SO4. It is well-known that the chemical
shifts are solvent dependent and in this case, the downfield shift is
the result of an increased de-shielding of the protons due to an
increase in solvent polarity [35]. The results, however, clearly
indicate that upon dissolving of the model compound, no or at least
no fast reactions take place between the sulphuric acid and the
reactive methacrylate terminated model compound.
electrophilic addition type reaction [36]. The reaction is slow,
however, and only takes place at elevated temperatures and based
on the integration the change in concentration is well below 5% and
therefore the methacrylate (V) end-cap seems to be a suitable
candidate as a potential crosslink group in aramid fibres.
A second example is the maleimide (I) end-capped model
compound, which is shown in Fig. 3. The proton NMR experiments
at 85 ꢀC in D2SO4 show that the model compound is affected at two
different sites. In the first place, there are noticeable changes of the
two doublets corresponding to the phenyl ring adjacent to the
maleimide end-group (Fig. 3b). It is possible that this phenyl ring is
susceptible towards sulfonation. This reaction, however, will not
affect the ability of the maleimide group to form crosslinks or to
undergo chain extension and is therefore not considered a problem.
The second reaction, as indicated by the appearance of small peaks
around the 7.0 ppm maleimide proton peak, indicates a reaction at
the aforementioned maleimide reactive side (the maleimide double
bond).
Extended testing of the chemical stability of the maleimide
bond showed that the model compound remains stable in sul-
phuric acid even after a 24-h exposure time at room temperature,
see Fig. 3c. This result combined with the fact that the reaction at
the maleimide double bond is relatively slow even at 85 ꢀC indi-
cates that the maleimide group is another suitable candidate for
crosslinking aramid fibres.
Immediately after recording a reference spectrum at 24 ꢀC/t ¼ 0,
the temperature of the sample was increased to 85 ꢀC. After
increasing the temperature, new spectra were recorded after 10, 20
and 30 min, respectively. The spectra for the methacrylate (V) end-
capped model compound as a function of time at 85 ꢀC are shown
in Fig. 2 (bed).
To acquire information on the effects of temperature and time
on sulphuric acid exposures, spectra (a) to (d) (Fig. 2) are compared.
The 7.6 ppm peak, corresponding to the central phenyl ring, was
selected as a reference peak, since this aromatic moiety is relatively
stable under these acidic conditions at elevated temperatures. The
first noticeable change is the peak shift from 7.6 to 7.8 ppm of the
reference peak after heating the sample from room temperature to
85 ꢀC. A similar shift is also present for all other peaks and is due to
the increase in temperature. A second, and more important
observation is the slow change of the appearance of small peaks at
7.04 and 2.29 ppm. This is indicative of the fact that acid catalyzed
chemistry might be taking place at the reactive double bond at
elevated temperatures. It is possible that the double bond
undergoes an addition reaction with the sulphuric acid. It is known
that a double bond (SP2eSP2) structure can react with HSOꢁ3 via an
Following the approach as described above, all reactive model
compounds were investigated on their chemical stability and the
results are summarised in Table 2. Based on their ease of synthesis
and commercial availability, maleimidobenzoyl chloride (1) and
methacryloyl (5) chloride, were incorporated in PPTA and spun into
fibres. In addition aniline (11) was used as a non-reactive end-
group.

1894
A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
Table 3
Solubility of sulphuric acid stable model compounds (I and V) and an aniline (XI)
end-capped reference compound, as measured in DMSO-d6 after a 30 s heat treat-
ment (under Ar) at various temperatures.
Type of end-cap
300 ꢀC
350 ꢀC
400 ꢀC
Aniline (XI)
Maleimidobenzoyl chloride (I)
Methyl methacryloyl chloride (V)
þ
ꢁ
þ
þ
ꢁ
þ
þ
ꢁ
þ
(ꢁ not soluble, þ soluble).
compounds. Before the heat treatment, fresh samples of approxi-
mately 10 mg were prepared in aluminium cups. Samples were
heated for 30 s at three different temperatures, i.e. of 300, 350 and
400 ꢀC using an Argon (Ar) purged hot-stage. When removed from
the hot-stage all samples were quenched to room temperature and
transferred to a sample vial where w2 mg DMSO was added. The
samples were stirred at room temperature for at least 24 h prior to
recording their solubility, as summarised in Table 3.
The results in Table 3 show that the “inert” aniline-based
reference model compound (XI) remains soluble over the whole
range of heat treatment temperatures. The aniline end-group,
therefore, is a suitable end-cap to be used in synthesis as a non-
reactive reference compound. In addition it becomes clear that the
aramid-like structure does not undergo any thermally induced
polymerisation or crosslinking by itself. Unlike the aniline end-
capped compound (XI), model compound (I) is the only compound
to become insoluble after the heat treatment. This indicates that at
relatively high concentrations the maleimide end-group effectively
forms an insoluble network after heat treatment in the solid-state.
These results are supported by the literature, from which it is
known that maleimideemaleimide crosslinks are formed at
temperatures in excess of 250 ꢀC [24,37]. The third model
compound (V) is found to remain soluble after the heat treatments.
A possible explanation for this result is decomposition of the
sample during the heat treatments, which was confirmed by TG/
DTA experiments, where we observed rapid decomposition after
melting at 256 ꢀC. As will be discussed in more detail later, meth-
acrylate (V) end-capped oligomers display improved thermal
stabilities over their low molecular weight analogs, and appear
more suitable for our solid-state crosslinking experiments.
3.4. Oligomer synthesis and characterisation
3.4.1. Molecular weight analysis
Large-scale synthesis (w40 g scale) of the oligomers could be
performed without noticeable problems and with yields above 80%.
The number average molecular weight (Mn) for all (non)-reactive
oligomers used for fibre spinning experiments are summarised in
Table 4. For all oligomers the measured number average molecular
weight (Mn) seems to be lower than the targeted Mn and is probably
due to the presence of moisture during the reaction or premature
phase separation. Water will hydrolyse the acid chloride-based
monomers, which reduces their reactivity and finally results in
a lower molecular weight of the oligomers.
Fig. 3. 1H NMR spectra of the maleimide (I) end-capped model compound recorded in
D2SO4 at (a) room temperature, (b) after 15 min at 85 ꢀC and (c) after 24 h at room
temperature in D2SO4.
3.3. Curing behaviour of model compounds
3.4.2. Thermal analysis
In order to investigate the solid-state thermal curing of the
reactive end-groups, as will be the case once the reactive groups are
incorporated in PPTA fibres, qualitative solubility experiments were
conducted before and after heat treatment. For the three
compounds, i.e. maleimidobenzoyl chloride (1) and methacrylate
(5) end-capped model compounds and the aniline (11) end-capped
reference, dimethyl sulfoxide (DMSO) was found to be a suitable
solvent to prepare solutions of 0.5 wt% of the selected model
Since our fibres will be exposed to a thermal post-treatment
step, in order to activate the reactive end-groups, it is important to
investigate whether our oligomers display the required thermal
stability. Thermogravimetrꢁic1 experiments (TGA), performed at
a heating rate of 10 ꢀC min under a nitrogen atmosphere, show
that severe degradation of the oligomers starts at temperatures
well above 500 ꢀC which is comparable to their high molecular
weight PPTA counterpart [38]. The small amount of weight lost

A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
1895
Table 4
Table 5
Typical values found for the number average molecular weights as measured by SEC
(H2SO4 was used as the eluent).
Solubility of oligomers in NMP/CaCl2 (5 wt%) before and after a 60 s/350 or 380 ꢀC
heat treatment under nitrogen atmosphere. (þsoluble, ꢃ partially soluble and ꢁ
insoluble).
Oligomer
Mn (g molꢁ1
)
Mw (g molꢁ1
)
D
Oligomer
No treatment
350 ꢀC
380 ꢀC
PPTA-ANIL-2.5K
PPTA-ANIL-5K
PPTA-ANIL-10K
PPTA-MMA-5K
PPTA-MIBC-5K
Twaron 1010 ref.
1400
2100
2900
3900
3900
2700
5133
7500
9833
10 867
31 400
2.0
2.4
2.6
2.5
2.6
3.0
PPTA-ANIL-2.5K
PPTA-MMA-2.5K-II
PPTA-MIBC-2.5K-II
þ
þ
þ
ꢁ
ꢁ
ꢃ
ꢁ
ꢁ
ꢃ
10 300
For all oligomers the measured number average molecular weight (Mn) seems to be
lower than the targeted Mn and is probably due to the presence of moisture during
the reaction or premature phase separation. Water will hydrolyse the acid chloride-
based monomers, which reduces their reactivity and finally results in a lower
molecular weight of the oligomers.
solutions. Upon heat treatment,
a decreased solubility was
observed for all oligomer modifications, where both PPTA-ANIL-
2.5K and PPTA-MMA-2.5K-II became fully insoluble and the mal-
eimide end-capped PPTA-MIBC-2.5K-II became partially insoluble.
These results are in contrast with the solubility experiments of the
model compounds, as discussed in Section 3.3. Especially aniline,
which was concluded to be an “inert” end-cap, appears to be
reactive when used in an oligomer. A major difference is the higher
thermal stability (Tdec > 500 ꢀC) of the oligomer when compared to
the model compound (Tdec ¼ 333 ꢀC). The high post-treatment
temperatures are likely to be sufficient to cause chain extension/
crosslinking via un-reacted acid- and amine-functionalities, leading
to an insoluble product.
(approximately 1%) upon heating at 100 ꢀC is related to the loss of
water. This was confirmed by DSC, where the endothermic peaks
around 100 ꢀC disappeared upon a second heating. Due to the low
concentration of reactive end-groups, we could not detect
exothermic events associated with the chain extension or crosslink
chemistry.
3.4.3. Nuclear magnetic resonance spectroscopy
Heat treatment of the methacrylate end-capped PPTA-MMA-
2.5K-II also resulted in an insoluble product, likely to be the result of
crosslinking. The thermal stability is significantly higher for the
oligomer and therefore the vinyl end-group can be thermally
activated before degradation starts. PPTA-MIBC-2.5K-II becomes
partially insoluble after the various heat treatments. Thermal
degradation at high temperatures competing with chain extension/
crosslinking could be the reason that part of the materials dissolves
even after heat treatment [39]. In general, however, it is shown here
that oligomers with an Mn of w2000 g molꢁ1 are able to undergo
polymerization, i.e. via chain extension or crosslinking.
The presence of the end-groups in the oligomers was investi-
gated by 1H NMR analysis. Because of their limited solubility,
3900 g molꢁ1 oligomers were not suitable to prepare proper
samples. Instead, low molecular weight equivalents were used as
a proof of principal. The spectra, recorded in D2SO4 at room
temperature, are shown in Fig. 4. For PPTA-MMA-0.5K three peaks
could be observed and assigned to the methacrylate end-group.
Those are the peaks at 6.3 and 6.1 ppm, which can be assigned to
the vinylic protons, and the peak at 2.1 ppm, which belongs to the
protons of the methyl group. Evidence for the presence of the
maleimide moiety in the spectrum of the PPTA-MIBC-1.2K oligomer
is found at 7.0 ppm, and can be assigned to the maleimide double
bond.
3.5. Fibre spinning and characterisation
3.4.4. Curing behaviour
Incorporation of (reactive) oligomers was shown to be fully
compatible with the ‘classic’ PPTA fibre spinning process [40]. Spin-
dopes could be prepared following the traditional procedure and
the fibre spinning process itself was also carried out without
noticeable problems. Typical stressestrain curves for the as-spun
and heat-treated experimental fibres are shown in Fig. 5. Fig. 5a
indicates that the incorporation of 20 wt% reactive PPTA oligomers
is of minor influence on the stressestrain behaviour. Careful
observation reveals an initial linear behaviour up to the point of
yielding where the slope of the curves changes. Upon further
straining of the fibre, an increase in slope is observed. This tensile
The ability of the oligomer to form crosslinks was investigated
following a similar approach as was used for investigating the
model compounds. An NMP/CaCl2 (5 wt%) mixture was preferred
over sulphuric acid to prevent possible degradation of the cross-
links that were formed during the heat treatment. Oligomers with
Mn ¼ 3900 g molꢁ1 were found to have a too low solubility (much
lower than 0.5 wt%), therefore oligomers with an Mn of
2000e2500 g molꢁ1 were used for these experiments instead. The
results are summarised in Table 5. Before any heat treatment, all
oligomers were readily soluble and formed fully transparent
Fig. 4. 1H NMR spectra in D2SO4 of (a) PPTA-MMA-0.5K at room temperature and (b) PPTA-MIBC-1.2K at 85 ꢀC. The spectra confirm the presence of the methacrylate and maleimide
end-groups in the oligomers.

1896
A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
2500
2000
1500
1000
500
2500
a
b
2000
PPTA200-HT1b
PPTA200
ANIL20-200-HT1b
MMA20-200-HT1b
MIBC20-200-HT1b
ANIL20-200
MMA20-200
MIBC20-200
1500
1000
500
0
0
0,0 % 0,5 % 1,0 % 1,5 % 2,0 % 2,5 % 3,0 % 3,5 %
Strain
0,0 % 0,5 % 1,0 % 1,5 % 2,0 % 2,5 % 3,0 % 3,5 %
Strain
Fig. 5. Stressestrain curves as measured for as-spun (a) and heat-treated at 380 ꢀC (HT) (b) filaments based on standard PPTA polymer (PPTA-200) or on a mixture of 80 wt%
standard PPTA polymer and 20 wt% of a 5000 g molꢁ1 (reactive) end-capped oligomer spun at a line speed of 200 m minꢁ1
.
behaviour was explained by Northolt and co-workers and attrib-
uted to the onset of a sequential and plastic orientation mechanism
of the chains brought about by the resolved shear stress [3,41,42]. In
other words yielding occurs when the stress increases and the
corresponding shear stresses within a domain reaches a critical
shear level. One could argue that the addition of low molecular
weight oligomers could have been of influence on this critical shear
level resulting in a more distinct yielding behaviour. This is,
however, not the case for our fibres as shown by the curves in
Fig. 5a. In line with the underlying theory, yielding becomes even
less pronounced (see Fig. 5b) as the initial orientation of the
domains in the fibre is higher due to the heat treatment at 380 ꢀC.
Mechanical data of as-spun fibres, as summarised in Table 6,
shows that the incorporation of oligomers did not affect the char-
acteristics of the stressestrain curves but resulted in a modest
decrease (max 9%) of the tenacity while leaving the moduli largely
unaffected. The fibre based on the 20 wt% aniline end-capped
oligomer performs the worst and shows a drop in tenacity of 17%.
This is likely to be the result of the average molecular weight being
lower than the anticipated 5000 g molꢁ1 resulting in less effective
stress-transfer.
moduli increase upon heat treatment. No correlations were found
between fibre strength or fibre stiffness on one hand, and oligomer
chemistry, or oligomer concentration on the other hand. Data, as
obtained via the ‘elastica’ single filament compression test indi-
cated that the as-spun fibres did indeed show a modest (þ0.1 GPa)
improvement in compression strength. When exposed to a high
temperature (380 ꢀC) heat treatment, both the reference PPTA
fibres and the oligomer-modified fibres showed a significant
increase in compression strength, i.e. from 0.7 to 0.9 GPa.
Increasing the post-treatment temperature to 430 ꢀC, however,
reduces the compressive properties again, with the sole exception
of the PPTA reference fibre, where we measured a compression
strength of 1.11 GPa. This value comes close to the 1.29 GPa
reported for the M5 fibre [43], which is considered the polymeric
fibre with the best compressive strength to date. We believe at this
point that the concentration of end-groups is too low to have
positive contribution on the compressive strength. This was
confirmed by ‘elastica’ single filament compression test, as well as
by Raman experiments [44]. WAXS experiments (Table 6) showed
that the incorporation of oligomers has no measurable effect on
the fibre crystal structure. The values for the apparent crystallite
size, i.e. L 110 and L 200, are similar for the fibres based on 20 wt%
3900 g molꢁ1 oligomers when compared to the PPTA reference
fibres. We are currently exploring whether higher concentrations
of reactive oligomers, i.e. 40 and 50%, results in an improvement in
compression strength.
The heat treatment had an overall effect on all fibres, i.e.
decreasing the tenacity with an average of 8% compared to the as-
spun precursors. The strength retention of the PPTA reference fibre
annealed at 430 ꢀC shows that high temperature post-treatments
are possible without a penalty in fibre strength. As expected, all
Table 6
Physical and mechanical properties of fibres based on high molecular weight PPTA and 3900 g molꢁ1 end-capped PPTA oligomers. Fibres subjected to a thermal post-treatment
at 380 ꢀC are encoded HT. All fibres were spun at a line speed of 200 m minꢁ1
.
Fibre
LD
(dtex)
Tenacity
Modulus
(GPa)
Elongation
at break (%)
Compressive
L 110
L 200
(mN texꢁ1
)
strength (GPa)b
(nm)c
(nm)d
PPTA200
1.75
1.78
1.73
1.74
2195
1879
2019
2080
91.2
87.0
93.5
93.7
3.22
2.95
3.01
3.12
0.65 ꢃ 0.04
0.74 ꢃ 0.11
0.71 ꢃ 0.07
0.77 ꢃ 0.07
55
55
55
54
52
50
51
50
ANIL20-200
MMA20-200
MIBC20-200
PPTA200-HT
1.69
1.65
1.65
1.64
2019
1787
1860
1911
115.4
114.7
113.9
121.1
2.47
2.19
2.30
2.24
0.90 ꢃ 0.06
0.98 ꢃ 0.11
0.90 ꢃ 0.06
0.98 ꢃ 0.08
90
87
91
89
67
65
67
66
ANIL20-200-HT
MMA20-200-HT
MIBC20-200-HT
PPTA-250-HT2a
1.27
2079
137
2.1
1.11 ꢃ 0.17
96
72
a
PPTA reference fibre spun at 250 m minꢁ1 and heat-treated at 430 ꢀC.
Average of 5 experiments.
L 110: apparent lateral crystal size in the direction perpendicular to the plane through the diagonal of the basal plane and the c-axis.
L 200: apparent lateral crystal size in the a-direction.
b
c
d

A. Knijnenberg et al. / Polymer 51 (2010) 1887e1897
[7] Jones MCG, Martin DC. J Mater Sci 1997;32(9):2291e303.
1897
4. Conclusions
[8] Walker KA, Markoski LJ, Deeter GA, Spilman GE, Martin DC, Moore JS. Polymer
1994;35(23):5012e7.
We have synthesised and characterised a series of amide model
compounds end-capped with various types of reactive end-groups.
It was shown by NMR spectroscopy using D2SO4 as a solvent that
maleimide benzoyl chloride, methacryloyl chloride, tigloyl chloride
and trichloro acryloyl chloride could be coupled to PPD and that the
resulting end-groups are chemically stable in 85 ꢀC sulphuric acid
(>98.5%). Maleimide benzoyl chloride and methacryloyl chloride
were successfully used for end-capping aramid oligomers with
molecular weights up to 3900 g molꢁ1. The addition of up to 20 wt%
oligomers did not interfere with the spinning process and fibre
tensile strengths were maintained between 1879 and
2080 mN texꢁ1 for as-spun fibres based on 20 wt% oligomers, which
is somewhat lower then the 2195 mN texꢁ1 found for the PPTA
reference fibre. Upon heat treatment, a decrease in tensile strength
was measured with values of 1787e1911 mN texꢁ1 for oligomer-
based fibres. Fibre moduli remained unaffected by the incorpora-
tion of oligomers but we found that the compressive strengths
improved significantly upon heat treatment. The reference fibres
based on high molecular weight PPTA polymer, however, showed
similar improvements suggesting that the incorporation of up
20 wt% reactive oligomers does not result in an additional
measurable increase in compressive strength. We found that the
PPTA-based fibres without reactive oligomers and post treated at
430 ꢀC exhibit compressive strengths of 1.11 GPa. We believe that
this significant improvement will expand the application range of
aramid fibres and closes the performance gap with other, more
expensive, high-modulus fibres such as M5 and PBI.
[9] Walker KA, Markoski LJ, Moore JS. Synthesis 1992;12:1265e8.
[10] Glomm B, Rickert C, Neuenschwander P, Suter UW. Macromol Chem Phys
1994;195(2):525e37.
[11] Rickert C, Neuenschwander P, Suter UW. Macromol Chem Phys 1994;195
(2):511e24.
[12] Sweeny W. J Polym Sci Part A Polym Chem 1992;30(6):1111e22.
[13] Jenkins SE, Jacob KI, Polk MB, Kumar S, Dang TD, Arnold FE. Macromolecules
2000;33:8731e8.
[14] Beckham HW, Rubner MF. Polymer 1991;32(10):1821e5.
[15] Xue Y, Hara M. Macromolecules 1997;30:3803e12.
[16] Xue Y, Hara M, Yoon H. Macromolecules 2001;34:844e51.
[17] Glomm BH, Oertli AG, Rickert C, Neuenschwander P, Suter UW. Macromol
Chem Phys 2000;201(12):1374e85.
[18] Lammers M, Klop EA, Northolt MG, Sikkema DJ. Polymer 1998;39
(24):5999e6005.
[19] Leal AA, Deltzel JM, McKnight SH, Gillespie JW. Polymer 2009;50(13):
2900e5.
[20] Zhang T, Jin J, Yang S, Li G, Jiang J. Appl Mater Interfaces 2009;1(10):
2123e5.
[21] Zhang T, Jin J, Yang S, Li G, Jiang J. Carbohydr Polym 2009;78(2):364e6.
[22] Iqbal M, Norder B, Mendes E, Dingemans TJ. J Polym Sci Part A Polym Chem
2009;47:1368e80.
[23] Knijnenberg A, Weiser E, StClair T, Mendes E, Dingemans TJ. Macromolecules
2006;39:6936e43.
[24] Hergenrother PM. High Perform Polym 2003;15:3e45.
[25] Bauer W. Methacrylate acid and derivatives. Ullmann’s encyclopedia of
industrial chemistry. Wiley-VCH: Weinheim; 2005.
[26] Ohara T, Sato T, Shimizu N, Prescher G, Schwind H, Weiberg O, et al. Acrylic
acid and derivatives. Ullmann’s encyclopedia of industrial chemistry. Wiley-
VCH: Weinheim; 2003.
[27] Greenwood JH, Rose PG. J Mater Sci 1974;9(11):1809e14.
[28] Sinclair D. J Appl Phys 1950;21:380e6.
[29] Cava MP, Deana AA, Muth K, Mitchell MJ. Org Synth 1961;41:93e4.
[30] Augustin M, Müller W. J Prakt Chem 1985;327(5):857e64.
[31] Endo Y, Songkram C, Yamasaki R, Tanatani A, Kagechika H, Takaishi K, et al.
J Org Chem 2002;657(1e2):48e58.
[32] Takahashi S, Kuroyama Y, Sonogashira K, Hagihara N. Synthesis
1980;8:627e30.
Acknowledgements
[33] Liu FJ, Munukutla S, Levon K, Tesoro G. J Polym Sci Part A Polym Chem 1992;30
(1):157e62.
This research was funded by and carried out in close cooperation
with Teijin Aramid BV (Arnhem, The Netherlands).
[34] Al-Fulaij OA, Elassar AZA, El-Dissouky A.
J Appl Polym Sci 2006;101
(4):2412e22.
[35] Reichardt C. Solvents and solvent effects in organic chemistry. Weinheim:
Wiley-VCH Verlag GmbH & Co. KGaA; 2003.
[36] Fessenden RJ, Fessenden JS. Organic chemistry. 4th ed. Pacific Grove: Brooks/
Cole Publishing Company; 1990.
[37] Curliss DB, Cowans BA, Caruthers JM. Macromolecules 1998;31:6776e82.
[38] Li X-G, Huang M-R. J Appl Polym Sci 1999;71(4):565e71.
[39] Mison P, Sillion B. Adv Polym Sci 1999;140:137e79.
[40] Northolt MG, Sikkema DJ. Adv Polym Sci 1991;98:115e77.
[41] Northolt MG, Baltussen JJM, Schaffers-Korff B. Polymer 1995;36(18):3485e92.
[42] Northolt MG, Baltussen JJM. J Appl Polym Sci 2002;83(3):508e38.
[43] Knijnenberg A. Compressive Failure Behaviour of Novel Aramid Fibres, PhD
Thesis, Delft University of Technology; 2009.
References
[1] Chae HG, Kumar S. J Appl Polym Sci 2006;100:791e802.
[2] Hearle J. High-performance fibres. Cambridge: Woodhead Publishing Ltd;
2001.
[3] Northolt MG, den Decker P, Picken SJ, Baltussen JJM, Schlatmann R. Adv Polym
Sci 2005;178:1e108.
[4] Hu Xiao-Dong, Jenkins SE, Min BG, Polk MB, Kumar S. Macromol Mater Eng
2003;288(11):823e43.
[5] Kozey VV, Jiang H, Mehta VR, Kumar S. J Mater Res 1995;10(4):1044e61.
[6] Jiang T, Rigney J, Jones M-CG, Markoski LJ, Spilman GE, Mielewski DF, et al.
Macromolecules 1995;28(9):3301e12.
[44] Knijnenberg A, Young RJ, van der Zwaag S. In preparation.