Eutectic melting behavior of polyamide 10,T-co-6,T and 12,T-co-6,T copolyterephthalamides

Article abstract of DOI:10.1016/j.polymer.2010.03.045

Semi-aromatic polyterephthalamides 10,T and 12,T were synthesized with 0-60wt-% PA-6,T comonomer by a melt polycondensation process. Molecular weights of the copolymers ranged from 12,000 to 27,000g/mol and all produced tough melt-pressed films. The subst

Full text of DOI:10.1016/j.polymer.2010.03.045

Polymer 51 (2010) 2417e2425
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Eutectic melting behavior of polyamide 10,T-co-6,T and 12,T-co-6,T
copolyterephthalamides
,
b
b
Theodore F. Novitsky a, Christopher A. Lange a, Lon J. Mathias a , Scott Osborn , Roger Ayotte ,
*
Steven Manning b
a The University of Southern Mississippi, 118 College Drive, Hattiesburg, MS 39402, USA
b Ascend Performance Materials, 3000 Old Chemstrand Road, Cantonment, FL 32533, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Semi-aromatic polyterephthalamides 10,T and 12,T were synthesized with 0e60 wt-% PA-6,T como-
nomer by a melt polycondensation process. Molecular weights of the copolymers ranged from 12,000 to
27,000 g/mol and all produced tough melt-pressed films. The substituted aromatic carbon of the 13C NMR
spectra revealed that comonomer sequences are statistical; e.g., 50:50 wt-% PA-12,T-co-6,T copolymer
had 12,T-12,T:12,T-6,T:6,T-6,T sequence ratios of approximately 1:2:1. Copolymers of both PA-10,T and 12,
T exhibited a eutectic melting point at 30 wt-% PA-6,T, with melting points decreasing linearly from 315
and 292 ꢀC to minima of 280 and 272 ꢀC, respectively. Melting enthalpies showed similar minima at ca.
35 wt-% PA-6,T. WAXD and DMA were used to further characterize the eutectic behavior, and
a comprehensive analysis of PA-6,T copolymer melting behavior is given.
Received 29 January 2010
Received in revised form
19 March 2010
Accepted 25 March 2010
Available online 31 March 2010
Keywords:
Copolyamide
Eutectic melting temperature
NMR
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
plotting copolymer melting temperature and enthalpies versus
composition.Ifmeltingtemperaturesdonotdisplayalocalminimum
Semi-aromatic polyamides combine the high melting temper-
atures and heat resistance of wholly aromatic polyamides (e.g.,
Kevlar poly p-phenylene terephthalamide) with the melt process-
ability of aliphatic polyamides such as nylon 6 and 6,6. Much
research has focused on semi-aromatic polyamides synthesized
using aliphatic diamines and terephthalic acid [1,2]. With short
aliphatic diamines (2e7 CH2’s), melting temperatures surpass the
thermal decomposition temperature of polyamides, thus making
melt processing impractical. For example, the melting points of PA-
4,T and PA-6,T are 430 and 370 ꢀC, respectively [3]. Increasing the
length of the aliphatic diamine lowers the melting point into
a processable range. For instance, the melting points of PA-9,T, PA-
10,T, PA-12,T [2] and PA-18,T [4] are approximately 309, 315, 295
and 245 ꢀC, respectively. More important for high temperature
applications, semi-aromatic polyamides have glass transition
temperatures ranging from 100 to 140 ꢀC (as observed by DSC).
Comonomers are commonly used in semi-aromatic polyamides
to alter properties such as melting temperature, processability, and
optical clarity [1]. Copolyamide crystallinity is dependent on degree
of symmetry, the distance between amide bonds and amide bond
orientation (oddeeven effect). These effects are characterized by
as a function of composition, the comonomers are said to be
isomorphous, as is seen for adipic acid and terephthalic acid in nylon
6,6-co-6,T and 4,6-co-4,T copolymers [5e8]. Such co-crystallization
is due to the similar lengths between amide groups of adipic acid
and terephthalic acid units, which allow incorporation into the
same crystal structure while having no change in crystal lattice
dimensions. Co-crystalline copolymers also display no apparent
decrease in crystallinity. Alternatively, when a local minimum, or
eutectic point, is observed, the comonomers are not co-crystalline
and thus do not fit into the same crystalline lattice: the como-
nomer acts as an impurity which decreases the size, perfection,
and melting point of the formed crystals. For example, when
copolymerized with the crystalline PA-6,T, the amorphous PA-6,
I-[poly(hexamethylene isophthalamide)] displays a drastic drop in
melting temperature occurs at 50 wt-%, rendering their copolymers
amorphous [9]. However, when two crystalline polymers are
copolymerized, isodimorphism is known to occur, characterized by
a change in crystal lattice dimensions at the eutectic point [10]. In
this case, the comonomer is incorporated into a crystal structure
different than its own. The degree of melting point depression and
amount of crystallinity at the eutectic point are a function of
structural similarity of the comonomer repeating units.
Research reported on PA-10,T copolymers has been limited to
copolymerizations involving caprolactam and adipic acid [6,11]. It
* Corresponding author. Tel.: þ1 (601) 266 4873; fax: þ1 (601) 266 5504.
E-mail address: lon.mathias@usm.edu (L.J. Mathias).
0032-3861/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.03.045

2418
T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
has been demonstrated that adipic acid and terephthalic acid co-
crystallize when polymerized with decamethylene diamine,
resulting in melting point versus composition trends similar to that
of the nylon 6,6-co-6,T series. No literature currently exists covering
the copolymerization of PA-10,T or 12,T with other linear diamines.
A single reference describes semi-aromatic copolyamides synthe-
sized using terephthalic acid and a combination of linear aliphatic
diamines [12]. It was reported that copolyamides of PA-4,T and 6,T
display a minimum of melting temperature with composition, but
maintain some level of crystallinity throughout all compositions
despite the two monomers being non-isomorphic.
It is the goal of this study to determine the effect of the distance
between amide bonds on the melting behavior of semi-aromatic
copolyterephthalamides by changing the diamine length while
keeping amide orientation fixed by using solely even diamines. PA-
10,T-co-6,TandPA-12,T-co-6,Tcopolymersweresynthesizedbymelt
condensation polymerization of mixed diamine/terephthalic acid
salts. Polymers were characterized using high resolution 13C NMR
spectroscopy for composition analysis, and intrinsic viscosity for
molecular weightestimation. The effect of the PA-6,Tcomonomeron
melting temperature, crystal type, and degree of crystallinity was
studied by differential scanning calorimetry, wide-angle x-ray
diffraction, and dynamic mechanical analysis.
Fig. 1. Synthetic scheme of PA-10,T-co-6,T copolymers, where x ¼ 0e60 wt-%.
loaded into a Parr reactor equipped with a test tube holder, which
was then sealed and purged with nitrogen. The reactor was sur-
rounded by insulation to improve temperature control. Nitrogen
pressure was controlled manually and measured by a gauge on the
reactor head.
The melt condensation polymerization was performed in three
stages involving water boiling, high pressure, and final hold stage
(Fig. 2). Heating steps were programmed into the reactor controller
unit and monitored with a thermocouple placed in one of the test
tubes. Duration and pressure of each stage were determined by
measured temperature of the reaction vessel. The boiling stage
ensures homogenization of the reaction mixture, with little poly-
merization. With approximately 125 kPa nitrogen pressure and
a 190 ꢀC set point, the sample temperature remained at 125 ꢀC until
all water evaporated. At the completion of boiling and a sudden rise
of temperature, the pressure was increased to 1724 kPa to mitigate
the loss of diamine throughout this initial stage of the polymeri-
zation, and the heater set point was changed to 290 ꢀC. When the
reaction temperature reached 280 ꢀC, the pressure was released
and the temperature allowed to rise and hold at 310e315 ꢀC for
15 min while maintaining a nitrogen purge.
Initially, higher polymerization temperatures and hold times
were used, but this process often resulted in polymers that were
neither soluble in sulfuric acid, nor amendable to compression
molding. Dimerization of diamines to form crosslinkable triamine
species is well known to occur at high reaction temperatures. In
order to mitigate this side reaction, a lower maximum temperature
(315 rather than 330 ꢀC) and polymerization time (15 rather than
30 min) were used. This resulted in homo- and copolymers that
were soluble in both sulfuric and dichloroacetic acid, and were
suitable for compression molding into films for testing.
2. Experimental
2.1. Materials
Terephthalic acid, 1,10-diaminodecane, 1,12-diaminododecane,
and 1,6-diaminohexane (HMDA) were purchased from Aldrich.
Terephthalic acid and 1,6-diaminohexane were used as received.
1,10-Diaminodecane, and 1,12-diaminododecane were sublimed at
70 ꢀC and dried at room temperature under vacuum before use. N,
N-dimethylacetamide (DMAc) and triethylamine were distilled
from barium oxide onto molecular sieves before use. Concentrated
sulfuric acid (96%) for viscosity measurements was purchased from
Aldrich and used as received. Solutia provided 6,T salt, siloxane
antifoam, and a commercial antioxidant.
2.2. Laboratory synthesis of polyamide (n,T) salt
Into a 2 L beaker were added 1 L of deionized water, 1,10-dia-
minodecane (49.5 g, 0.287 mol) and terephthalic acid (42.3 g,
0.285 mol), and the slurry that formed was heated to a boil. Addi-
tional water was added and brought to boil to yield a supersatu-
rated, clear salt solution. The hot salt solution was then added to
a 2 L beaker containing 500 mL of reagent alcohol and allowed to
cool to room temperature, followed by cooling in a freezer. The
precipitate was then filtered and washed with reagent alcohol. The
resulting 10,T salt was obtained in near-quantitative yield (95%).
PA-12,T salt was prepared in a similar manner. All terephthalic acid
salts were precipitated from water into reagent alcohol and dried
before use. PA-12,T, 10,T, and 6,T salt melting temperatures were
271.9, 271.6, and 281 ꢀC, respectively, with melting enthalpies
between 430 and 450 J/g.
2.3. Synthesis
2.3.1. Melt condensation of polyamide (n,T) and copolymers
Copolymers were synthesized from their corresponding
diamine/terephthalic acid salts as shown in Fig. 1. Polyamide (n,T)
salt, 3 mol-% excess diamine, and 0.5 wt-% antioxidant were
weighed and mixed with a mortar and pestle. The solid mixture
was then added to a test tube and approximately 50 wt-% deionized
water was added to create a slurry. Twelve such test tubes were
Fig. 2. Melt condensation polymerization temperature and pressure profiles.

T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
2419
2.3.2. Solution polymerization of polyamide 6,T (PA-6,T)
2.5.3. Differential scanning calorimetry (DSC)
HMDA (10.98 g, 0.095 mol) and triethylamine (27 mL, 0.194 mol)
were dissolved in 230 mL of DMAc and added to a Waring blender
equipped with a glass container. Theblender was turned on high and
60 mL of DMAc containing terephthaloyl chloride (19 g, 0.094 mol)
was added as quickly as possible. A white precipitate immediately
formed, followed by a large increase in viscosity. After 5 min,
blending contents were filtered and the precipitate obtained
washed with acetone followed by water, giving PA-6,T (20.7 g, 90%).
Polymer melting temperatures were obtained with
a
TA
Instruments 2920 DSC. Scans were taken at a heating rate of 10 ꢀC/
min. Data from both the first and second-heating scans were
collected. Samples were heated approximately 10e20 ꢀC above
their melting temperature, air cooled to room temperature, and
heated a second time to 350 ꢀC.
2.5.4. Wide-angle x-ray scattering (WAXD)
Diffraction patterns were collected on a Rigaku Ultima II x-ray
diffractometer. Data were collected from 3 to 40ꢀ 2 at a rate of 2ꢀ/
q
2.4. Sample preparation
min. Samples were annealed at 100 ꢀC for 12e24 h prior to analysis
to achieve maximum attainable crystallinity.
Solid polymers obtained from the test tubes were immersed in
liquid nitrogen for approximately 10 min and then ground using
a Waring blender with a stainless steel mixing jar. The samples were
then dried at 80 ꢀC under vacuum for 24 h, and stored in a desiccator
until further characterization. Dried samples were compression
molded into 0.8 or 0.3 mm films at a temperature approximately
10e20 ꢀC above the melting temperature of the polymer as deter-
mined by DSC. The pellets were placed between two woven Teflon
sheets and held at this temperature between two Carner press hot
plates for 3 minwith no pressure and then for 2 min under a pressure
of 2700 kPa. Hot plates were then taken from the press and immersed
in a room temperature water bath. For a higher rate of cooling, films
were removed from the hot plates and placed underneath a room
temperature steel plate. Films were dried at 80 ꢀC under vacuum for
12 h. Annealing was performed at 200 ꢀC for 12 h under vacuum.
2.5.5. Dynamic mechanical analysis (DMA)
Tensile mode DMA data were obtained using a Thermal Analysis
Q800 series instrument. Samples of 0.8 mm thickness were
prepared by melt pressing, followed by annealing and drying at
200 ꢀC prior to testing. Data were collected from 30 to 250 ꢀC at
a heating rate of 2 ꢀC/min with a 20 um amplitude, 0.05 N preload
force, and 125% force track.
3. Results
3.1. Intrinsic viscosity and molecular weight
In studies published separately, it was found that PA-12,T
samples synthesized with
1 and 3 mol-% excess 1,12-dia-
minododecane had the lowest total end group concentration, the
most balanced acid/amine end groups, and the highest molecular
weights, using these melt polymerization conditions [14]. In order to
obtain suitable molecular weights for copolymer analysis, 3 mol-%
excess deca- or dodecamethylene diamine was added to each
reaction. Single point intrinsic viscosities (IV) were determined to
estimate the relative number average molecular weights between
copolymer samples. IV values for several of the copolymers studied
are given in Table 1.
Number average molecular weights for PA-12,T-co-6,T and PA-
10,T-co-6,T copolymers were estimated from MarkeHouwink
constants developed for PA-12,T homopolymer. Assuming that the
solution properties of PA-10,T, 12,T and 6,T are similar, an estimate
of Mn for the copolymers was determined using the equation;
2.5. Characterization
2.5.1. Nuclear magnetic resonance spectroscopy (NMR)
Solutions for NMR analysis having a 10 wt-% polymer concen-
tration were prepared using
a 3:1 volume ratio of hexa-
fluoroisopropanol (HFIP) to CDCl3. Samples were first dissolved in
HFIP, followed by addition of CDCl3. Solution 13C NMR spectra were
collected on a Varian UNITYINOVA NMR spectrometer operating at
a frequency of 125.7 MHz. Routine acquisitions at 25 ꢀC involved
a 1.3 s acquisition time, a 45ꢀ pulse width of 2.9
ms, and a 1 s recycle
delay. The number of accumulated transients ranged from 15,000 to
30,000, requiring 12e24 h collection times. Raw data were zero-
filled up to 256k points and filtered with 1 Hz of line broadening
prior to application of Fourier transformation. Baselines were cor-
rected using a 10th order polynomial.
½
h
ꢂ ¼ 0:000558ðdL=gÞðMnÞ^ð0:81Þ
Note that these constants were developed using single point IV
2.5.2. Viscometry
values, and this relationship is only applicable for 0.5 g/dL
concentrations dissolved in 96% concentrated H2SO4 at 25 ꢀC [14]. K
and a values were developed from PA-12,T with NMR calculated Mn,
and therefore, Mn instead of Mv values are reported.
Solutions containing 0.5 g/dL of polymer in concentrated sulfuric
acid were prepared by adding 0.25 g of polymer and 25 mL of
concentrated sulfuric acid into a 50 mL flask. After 12 h of mixing via
magnetic stirring, the solutions were diluted with an additional
25 mL of sulfuric acid, and stirring continued for another 12 h.
Solutions were visually inspected for gel particles, and passed
through a funnel packed with glass fiber if gel was present.
Measurements were obtained using a Cannon viscometer in a 25 ꢀC
controlled waterbath. The viscometer was washed with sulfuric acid
and a portion of the next sample to be tested before measurements
were recorded. Flow times were an average of three values that
agreed within ꢁ0.2 s. Flow times of concentrated sulfuric acid and
each of the polymer solutions were used to calculate the specific and
relative viscosities. Single point intrinsic viscosities were then
determined using the Solomon and Ciuta relationship:
Table 1
Intrinsic viscosities, estimated Mn values, and 1st heating DSC melt temperatures of
homo- and copolymers.
a
PA-n,T-co-6,T (wt-%)
IV
Mn (g/mol)
Tm (ꢀC)
PA-10,T
100:0
85:15
70:30
50:50
PA-12,T
100:0
85:15
70:30
50:50
1.68
1.41
1.26
1.68
19,700
15,900
13,800
19,700
327
316
283
324
1.14
1.21
1.24
2.17
12,200
13,100
13,500
27,000
295
287
274
328
h ꢀ
ꢁ
i.
1=2
½
h
ꢂ ¼
2
hsp ꢃ lnðhrel
Þ
C
where, hsp is the specific viscosity, hrel is the relative viscosity, and C
a
K ¼ 0.000558 dL/g and
a
¼ 0.81 in 96% H2SO4 at 25 ꢀC.
is the concentration [13].

2420
T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
There are two main effects that create Mn differences in the
small peak shift changes; therefore peaks were aligned based on the
-amide carbon. The -amide carbon (4) shows two distinct peaks
copolymers: differential diamine volatility and the physical state of
the polymer during the reaction. Excess diamine is required in
order to maintain balanced stoichiometry. The amount of excess
needed is a function of the volatility of the diamine during the
polymerization and the polymerization conditions used. Since,
hexa-, deca-, and dodecamethylene diamine have different melting
points and vapor pressures, using a constant 3 mol-% excess
diamine for all copolymer compositions will not result in the same
stoichiometric balance under identical polymerization conditions.
For example, PA-12,T, 85:15 and 70:30 PA-12,T-co-6,T had intrinsic
viscosities of 1.14, 1.21, and 1.24, respectively. Increasing the
amount of PA-6,T comonomer has a greater impact on stoichio-
metric imbalance, because a higher fraction of the more volatile
HMDA (compared to deca- and dodecamethylene diamine) is
present. If equal molecular weights were desired, the amount of
excess diamine would need to be increased corresponding to the
amount of PA-6,T comonomer present in order to compensate for
these differences in volatility of the two diamine monomers.
Another cause of molecular weight variation is the physical state
of the polymer during the reaction, as determined by the rela-
tionship between the crystalline melting temperature of the poly-
mer and the maximum reaction temperature. Table 1 lists the DSC
melting temperatures of several of the copolymers. The 1st heating
DSC data are related to the crystalline state of the polymer formed
during the reaction. Comparison of the IV and DSC data shows that
when the melting temperature of the polymer is higher than the
maximum polymerization temperature, the intrinsic viscosity is
notably higher. The melting temperatures of PA-10,T (327 ꢀC),
50:50 10,T:6,T (324 ꢀC) and 50:50 12,T:6,T (328 ꢀC) are substantially
higher than the final reaction temperature of 315 ꢀC. These poly-
mers also have the highest intrinsic viscosities (1.68, 1.68, and 2.17)
and molecular weights (19,700, 19,700, and 27,000). Comparison of
a molten (PA-12,T e Tm ¼ 295 ꢀC) and crystallized polymer (50:50
PA-12,T-co-6,T e Tm ¼ 328 ꢀC) shows a near doubling of intrinsic
viscosity from 1.14 to 2.14 dL/g. This behavior is presumed to be due
to solid state polymerization that occurs after the crystallization
takes place within the reactor, a well-known phenomenon in nylon
and polyester synthesis. Crystallization of polymer from the reac-
tion increases the concentration of reactive groups in the melt and
will drive the polymerization to higher conversions relative to
a fully molten sample under the same polymerization conditions.
Despite the Mn differences discussed above, the copolymers
could all be compression molded into tough, flexible films. This
indicates that all copolymers were obtained with molecular
weights above the critical value needed for good cumulative
interactions and chain entanglement; thus, molecular weight
variation effects in the obtained results are considered negligible. It
is noteworthy to mention that the compression molding of 50:50
PA-12,T-co-6,T (the highest Mn) was the most difficult, requiring
a much higher temperature and pressure compared to the other
copolymers. Semi-aromatic polyamides are known to have mark-
edly higher melt viscosities than all-aliphatic polyamides, and an IV
of 2.16 (Mn w 27,000) may be approaching the upper limit with
regard to processability of these materials.
a
a
representing the fraction of PA-12,T and PA-6,T in the copolymers.
Their peak heights were found to correspond to the monomer feed
ratios. Note that samples were prepared in wt-%, e.g., 50:50 wt-% is
approximately 45:55 mol-% PA-12,T-co-6,T. This is significant since
NMR spectral intensities reflect molar compositions.
In addition to the substituted aromatic carbon (2) peaks (Fig. 4,
left) for PA-12,T and 6,T at 137.74 and 137.62 ppm respectively, two
new peaks appear for all copolymers. The new peaks are due to the
formation of PA-12,TePA-6,T alternating sequences. This is further
confirmed by comparing the 50:50 wt-% copolymer and
a 50:50 wt-% mixed solution of the two homopolymers shown in
Fig. 5. The physical mixture of the two homopolymers shows only
two peaks for the substituted aromatic carbon, while the 50:50
copolymer has four peaks (Fig. 5, a and b). The new peaks are due to
PA-12,TePA-6,T alternating sequences and have identical peak
heights. These data are consistent with completely random copol-
ymers [15]. For example, for the 50:50 wt-% PA-12,T-co-6,T copol-
ymers, the ratio of 12,T-12,T:12,T-6,T:6,T-6,T units is approximately
1:2:1, as expected for statistically random copolymers.
The relative peak intensities for varying amounts of PA-12,T, PA-
6,T, and PA-12,TePA-6,T sequences formed by varying the PA-6,T
content of the copolymer from 15 to 50 wt-% are shown in Fig. 4
(left). At 15 wt-% PA-6,T (Fig. 4b), a small broad peak between the
peaks for pure PA-6,T and the 12,T-6,T unit is seen. Comparison
with the corresponding peak of the PA-12,TePA-6,T sequence
indicates little if any homo-sequences of PA-6,T exist at a 15 wt-%
loading. At 30 wt-% PA-6,T, the peak due to PA-6,T homo-sequences
increases, but the number of these sequences is still small
compared to the 12,T and 12,T-6,T sequences. At 50 wt-% PA-6,T, the
peak for PA-6,T homo-sequences is increased to approximately the
same number as seen for 12,T homo-units and half that of 12,T-6,T
sequences, what would be expected for a random melt condensa-
tion polymer. Since crystallinity is partially dependent on regularity
of the polymer structure, the behavior of the substituted aromatic
carbon should provide sequence information needed to help
understand the eutectic melting behavior of the copolymer series
as discussed below.
3.3. Differential scanning calorimetry
A series of second-heating DSC thermograms of PA-10,T-co-6,T
(top trace) and 12,T-co-6,T (bottom trace) copolymers are shown in
Fig. 6. Melting temperatures and enthalpies are plotted versus wt-%
PA-6,T in Fig. 7. PA-10,T and PA-12,T homopolymers have melting
temperatures of 315 and 292 ꢀC, respectively. Copolymers showed
3.2. 13C NMR spectroscopy
High resolution NMR spectroscopy was used to determine
copolymer composition and comonomer distribution. The 13C NMR
spectrum for PA-12,T is shown in Fig. 3. NMR spectra of the PA-10,T
(12,T), 6,T copolymers showed multiple peaks for carbon atoms 1, 2,
3, and 4. Fig. 4 shows expanded regions of the a-amide carbon (4)
and substituted aromatic carbon (2) of the related PA-12,T, PA-6,T
copolymers. Slight variations in sample viscosity and solvent caused
Fig. 3. 13C NMR spectrum of PA-12,T.

T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
2421
Fig. 4. Expanded 13C NMR spectrum of substituted aromatic (left) and
a-amide (right) carbon atoms of (a) PA-12,T, (b) 85:15 wt-% PA-12,T-co-6,T, (c) 70:30 wt-% PA-12,T-co-6,T,
(d) 50:50 wt-% PA-12,T-co-6,T, and (e) PA-6,T.
a linear decrease in melting temperature with increasing PA-6,T
comonomer content, with the lowest melting temperatures of 280
and 272 ꢀC observed for PA-10,T-co-6,T and 12,T-co-6,T copolymers
at ca. 30 wt-% PA-6,T, respectively. Similarly, melting enthalpies of
both sets of copolymers displayed a linear decrease with increasing
PA-6,T comonomer content. Melting enthalpies of PA-10,T-co-6,T
and PA-12,T-co-6,T copolymers showed a minimum at 35 and
30 wt-% PA-6,T, respectively, decreasing from 70 to 24 J/g for PA-10,
T-co-6,T and 60 to 26 J/g for PA-12,T-co-6,T copolymers. In addition,
the melting endotherms of copolymers containing 30e40 wt-% PA-
6,T were very broad compared to those of the respective
homopolymers.
The data indicate that both PA-10,T and PA-12,T are not co-
crystalline with PA-6,T. The 6 or 8 carbon difference in the diamine
portion of the repeat unit of the former increases the length
between amide units, and therefore inhibits entrance of the PA-6,T
repeat unit into PA-10,T and 12,T crystal structures (and vice versa).
The decreasing melting temperatures up to 30 wt-% PA-6,T are
consistent with the formation of fewer and smaller or less perfect
crystals, while the decrease in melting enthalpies is due to the
overall decrease in the total crystallinity and crystal perfection as
well. This implies that PA-6,T sequences up to 30 wt-% act as
impurities, and inhibit the formation of PA-10,T or PA-12,T crystals.
13C NMR spectroscopy has shown that with 15 wt-% PA-6,T, most
PA-6,T units are statistically incorporated as PA-6,TePA-12,T alter-
nating units with no PA-6,T homo-sequences. This maximizes the
number of irregularities along a given polymer chain, and therefore
has a negative effect on crystal content as well as size and perfec-
tion of crystallites that do form. The broadening of the melting
endotherms is also associated with a larger distribution of crystal
size and perfection. Further support for this interpretation comes
from the fact that compression molded films in this range of
compositions have increased optical clarity with increasing PA-6,T
content, consistent with the reduced crystallinity, and with 30 wt-%
PA-6,T transparent materials are obtained.
The eutectic points for both PA-10,T-co-6,T and PA-12,T-co-6,T
copolymer series were found at 30 wt-% PA-6,T. The depression of
melting temperature and enthalpy may extend to lower values and
result in a totally amorphous composition if 1e2 wt-% increments
were used in the vicinity of the eutectic point. Increasing PA-6,T
comonomer content above 30 wt-%, increased the melting
temperature in a manner that was largely independent of which
comonomer was present, that of PA-10,T or 12,T. NMR results
indicated that, at 30 wt-% and higher PA-6,T content, longer PA-6,T
sequences start being formed. The crystalline melting temperature
of the PA-6,T homopolymer is 372 ꢀC, and the homo-6,T crystallites
may have reduced size and perfection that determine the melting
point values. That is, on increasing above 30 wt-% 6-T content, the
PA-10,T and 12,T homo-sequences and alternating sequences act as
defects to the formation of PA-6,T crystals. This effect is reduced as
the amount of PA-6,T comonomer was increased further. Melting
temperatures of PA-10,T, 6,T and PA-12,T-co-6,T copolymers
increased from 272 and 280 at 30 wt-% PA-6,T to 325 ꢀC and 320 ꢀC
at 60 wt-% PA-6,T, respectively. The fact that an almost identical
plot of the melting temperatures is seen for both copolymer series
further confirms that formation of PA-6,T crystallites is the
controlling factor in the solid polymers. In addition, melting
Fig. 5. Peak assignments of substituted aromatic 13C NMR chemical shifts of the 50:50
PA-12,T-co-6,T copolymer (bottom spectra) and a 50:50 mixture (top spectra) of the
two homopolymers.

2422
T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
Fig. 7. Second-heating DSC melting temperatures (upper plot) and enthalpies (lower
plot) of PA-10,T (,) and PA-12,T (C) versus wt-% PA-6,T. Labeled regions of crystal
type are identified in Section 3.4.
Fig. 6. Second-heating DSC thermographs of PA-10,T (top) and PA-12,T (bottom)
containing 0 (a), 10 (b), 20 (c), 30 (d), 40 (e), 50 (f), and 60 (g) wt-% PA-6,T.
five times larger than that of the un-annealed sample, confirming
that phase separation and annealing are occurring during melt
pressing of the 40:60 PA-10,T-co-6,Tcopolymers. Since there was no
mixing during the reaction and the maximum temperature was
316 ꢀC, this phase separationcould be aresultof the crystallization of
PA-6,T rich phases during the polymerization reaction. If crystalli-
zation during the reaction occurs, the randomness of the resulting
polymer would be directly affected, thus physically changing the
microscopic composition and morphology of the multiphase system
generated. Despite having such separated crystal domains, tie
molecules and material flow during compression molding were
apparently sufficient to allow formation of tough films.
enthalpies showed a corresponding increase with increasing PA-6,T
content greater than 35 wt-%.
Copolymers exceeding 60 wt-% PA-6,T displayed phase separa-
tion, producing crystalline domains having melting points
exceeding 350 ꢀC. Since PA-6,T homopolymer has a melting point of
370 ꢀC, the separated phase consists mostly of 6,T homo-sequences.
The degree of phase separation was found to be affected by the
thermal history of the polymer. For example, compression molding
of the 40:60 10,T-co-6,T copolymer produced a film with different
regions of clarity, similar to a stained glass pattern. Fig. 8 shows first
heating DSC thermographs of both relatively (a) opaque white and
(b) clear yellow domains in the compression molded film; the two
domains were physically cut from the film and analyzed separately.
The white sections (a) of the blotchy film have a larger PA-6,T rich
phase melting enthalpy compared to the clearer area (b) of the film.
It appears that compression molding at 340 ꢀC induces phase
separation and/or annealing of this higher melting phase, resulting
in the blotchiness of the film. To confirm the annealing of the higher
melting phase, DSC annealing of the bulk reaction material at 340 ꢀC
was performed for 30 min and compared to a control sample (c)
heated to 340 ꢀC in the DSC and immediately air cooled. Heating
scans after these thermal treatments are shown in Fig. 8. The
annealed sample displayed a higher melting peak approximately
3.4. WAXD
X-ray diffraction was used to further confirm crystal composi-
tion of PA-10,T-co-6,T and PA-12,T-co-6,T copolymers. Copolymers
of PA-10,T with 0e60 wt-% PA-6,T are shown in Fig. 9. For brevity,
WAXD trends of PA-12,T-co-6,T copolymers are omitted, but show
similar diffraction patterns with increasing PA-6,T content.
Diffraction patterns of PA-10,T and 12,T compression molded films
display sharp diffraction peaks at approximately 19.5, 20.5 and 20ꢀ
2q<GK>. Note that the PA-6,T diffraction pattern was obtained
from a powder, whereas all others are from compression molded
films. Three regions of diffraction patterns were observed as labeled
on the phase diagram in Fig. 7:

T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
2423
it cannot be said if PA-10,T and 6,T crystals are both formed in
this intermediate range.
ꢄ 50e60 wt-% PA-6,T e The reappearance of sharp diffraction
peaks is observed between 50 and 60 wt-% PA-6,T, giving peaks
similar to those of the PA-6,T homopolymer synthesized by
solution polymerization. This further confirms that, beyond the
eutectic point, PA-6,T crystals are being formed.
X-ray diffraction patterns and DSC thermographs show good
agreement, confirming the change of crystal type through the
eutectic point. However, there is discrepancy between the two sets
of data. PA-10,T copolymers with 15 and 45 wt-% PA-6,T have
similar melting enthalpies (w43 J/g), while showing distinct
differences in their diffraction patterns. It appears from the WAXD
data, that crystallinity is higher in the former due to sharp
diffractions peaks, while the latter displays only an amorphous
halo. Since PA-6,T crystals have a higher amide density and thus
a higher melting enthalpy, it can be inferred that the degree of
crystallinity for the 45 wt-% PA-6,T copolymer is less, despite
having the same melting enthalpy as the 15 wt-% copolymer.
Fig. 8. DSC thermographs of white (a) and clear (b) regions of compression molded
film, and un-annealed (c) and annealed (d) non-compression molded samples of the
40:60 wt-% PA-10,T-co-6,T copolymer (vertical line represents compression molding
and DSC annealing temperature).
3.5. Dynamic mechanical analysis
Change in the degree and type of crystallinity provided by the
variation of comonomer sequence consequently impacts the
composition of the amorphous phase and thermomechanical
ꢄ 0e20 wt-% PA-6,T e Sharp diffraction peaks of PA-10,T broaden
and their intensity decreases, falling into the amorphous halo
when the PA-6,T content is increased to 20 wt-%. All diffraction
peaks are consistent with PA-10,T homopolymer diffraction
patterns. This confirms that the addition of low levels of PA-6,T
disrupt PA-10,T crystal formation, and new types of crystals are
not formed.
ꢄ 25e45 wt-% PA-6,T e No sharp diffraction peaks are observed
between 25 and 45 wt-% PA-6,T, although DSC results show
melting enthalpies between 24 and 30 J/g. The absence of sharp
diffraction peaks is due to the low crystallinity of the samples
and only broad amorphous scattering is observed. Increasing
from 25 to 45 wt-%, the width of this broad diffraction
increases from that of PA-10,T to PA-6,T. At 40e45 wt-% PA-6,T
crystals are forming, but are still less intense than the amor-
phous phase halo. Due to the low crystallinity of these samples
properties. Fig. 10 displays the tan
d of compression molded and
annealed PA-10,T-co-6,T copolymer films obtained from dynamic
mechanical analysis. All the DMA data, including the storage
modulus at 75 and 200 ꢀC and the maximum tan
d
, are summarized
in Table 2. The glass transition temperatures (Tg), as determined by
the tan maximum for PA-10,T and PA-12,T were found to be 148
d
and 137 ꢀC, respectively. PA-12,T has two additional flexible
methylene groups per repeat unit when compared to PA-10,T, thus
slightly decreasing the amide density and Tg. Both PA-10,T and 12,T
with 15 wt-% PA-6,T display a small second transition at approxi-
mately 190 and 182 ꢀC, respectively. This is consistent with the
formation of a separate second phase. Although there exists no data
in the literature describing the miscibility of these polymers, this
work suggests that at 15 wt-% PA-6,T content, phase separation
occurs resulting in this higher Tg transition. WAXD has shown that
6,T segments do not participate in homo-sequence crystallization at
15 wt-% PA-6,T. With this exclusion from the crystalline domain,
the local content of PA-6,T in the amorphous domain increases
upon crystallization of the 10,T or 12,T sequences. This results in
a concentration and distribution of sequences that favor phase
separation. At 30 and 50 wt-% PA-6,T, this shoulder disappears but
an overall increase in Tg is observed. This slight increase in Tg is due
to the increase of more rigid 6,T segments in the amorphous
domain. Due to lack of crystallinity at these PA-6,T levels, the tan
peak intensity is larger compared to the homopolymers.
d
Crystallites can act as reinforcing agents to tie the amorphous
chains together in a physically crosslinked morphology; therefore,
modulus should be somewhat proportional to the degree of
crystallinity. Below Tg (75 ꢀC), the amorphous chains are in a rigid
state and this effect is not apparent. However, above Tg (200 ꢀC)
amorphous chains are mobile and the degree of crystallinity plays
a dominant effect on the modulus of the material. Consistent with
this interpretation, PA-10,T and 12,T homopolymers have the
highest values of melting enthalpy and modulus at 200 ꢀC. In
relation to melting enthalpy, from 0 to 30 wt-% PA-6,T, the storage
modulus above Tg decreases. This is consistent with the gradually
decreasing degree of crystallinity. However, the melting enthalpy
of the 50 wt-% PA-6,T copolymers significantly recovers from the
eutectic point, while the storage modulus above Tg does not. For
Fig. 9. X-ray diffraction patterns of PA-10,T-co-6,T copolymers.

2424
T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
All other PA-6,T copolymer units are not isomorphic, display-
ing a minimum melting temperature as a function of composi-
tion. Fig. 11 shows that the eutectic melting behaviors of these
copolyamides are distinct in terms of the minima location and
shape of the plot. For example, PA-12,T-co-6,T, PA-10,T-co-6,T
copolymers display a smaller melting point depression with
a minimum occurring at higher amounts of PA-6,T, compared to
the nylon 6-co-6,T copolymers. At the eutectic point, the total
decrease in melting temperature from nylon 6 homopolymer is
88 ꢀC with 16 mol-% PA-6,T, while for PA-12,T-co-6,T and PA-10,
T-co-6,T copolymers these values are only 20 ꢀC at 34 mol-% and
35 ꢀC at 37 mol-%, respectively. This implies that PA-6,T disrupts
nylon 6 crystallinity to a greater extent than those of PA-10,T and
PA-12,T. This behavior can be attributed to greater similarities in
PA-10,T, 12,T, and 6,T repeat unit structures. These tereph-
thalamide copolymers maintain
a regular hydrogen bonding
orientation and placement, and all have an even number of
carbon atoms between amide bonds. Alternatively, PA-6 includes
a reverse amide bond due to the head-to-tail linkage of the AeB
structure with an odd number of carbon atoms between amide
bonds. Therefore, addition of PA-6,T has a larger effect on the
eutectic melting behavior of nylon 6 than 10,T or 12,T. The degree
of the eutectic depression has been discussed for poly(ethylene
terephthalate-co-p-benzoates) copolyesters [16]. Since there was
a noticeably smaller drop in melting temperature at the eutectic
point compared to poly(ethylene terephthalate-co-adipate) and
poly(ethylene terephthalate-co-sebacate) it was concluded that
the copolymers were partially isomorphous(or isodimorphic).
The effect of chain orientation can be understood by
comparison with PA-6,I, 6,T copolymers. PA-6,I is amorphous due
to its kinked chain. Its copolymers display no crystallinity up to
43 mol-% PA-6,T [17]. Both the amide bond length and chain
orientation are affected by inclusion of the isophthalamide
structure. The data show the PA-6,T crystals can have exceedingly
low melting temperatures (<250 ꢀC) in such copolymers [18],
although the eutectic behavior of the fully terephthalamide
polymers does not show such reduced melting points for the
onset of PA-6,T crystallinity. For example, PA-10,T-co-6,T, and 12,
T-co-6,T copolymers have respective lowest melting points of 281
and 270 ꢀC despite having different lengths between amide
bonds. Therefore, the symmetry of the aromatic substitution
(meta versus para) and its effect on chain orientation and packing
prevail over the distance between amide bonds, and this affects
the eutectic melting temperature trends. Correspondingly, PA-4,T,
Fig. 10. DMA tan
d of PA-10,T (a) with 15 (b), 30 (c), and 50 (d) wt-% PA-6,T.
example, the 15:85 and 50:50 PA-12,T-co-6,T copolymers have
similar melting enthalpies, but have a 20% difference in modulus
value. Therefore, the rise in melting enthalpy from 30 to 50 wt-%
PA-6,T is not solely due to the increase of the degree of crystal-
linity, but also corresponds to the change of crystal type from PA-
12,T to PA-6,T. That is, the equilibrium heat of fusion of PA-6,T is
higher than that of PA-12,T (10,T) because of the increased
concentration of amide groups per gram of crystals. Therefore,
comparing melting enthalpies to the left and the right side of the
eutectic point does not properly describe relative degrees of
crystallinity but is also related to other contributions to inter- and
intramolecular interactions in these materials. This is in agree-
ment with WAXD data as noted earlier, where the PA-10,T crystals
at 15 wt-% PA-6,T were found to show sharper peaks than those
for PA-6,T homopolymer crystals and the 50 wt-% PA-6,T sample.
4. Discussion
Copolyamide crystallization behavior is a result of the symmetry
of amide distance and orientation of the comonomer repeat unit
structure. For example, nylon 6,6-co-6,T copolymers exhibit co-
crystalline (isomorphic) behavior because they are both AeA BeB,
eveneeven monomers, and the distance between amide groups
formed from adipic and terephthalic acids has only a 0.3 angstrom
difference in length [5]. Therefore, they display an almost linear
trend of melting temperature with increasing content of the PA-6,T
comonomer. These data are shown in Fig. 11 with PA-12,T-co-6,T,
PA-10,T-co-6,T and several other PA-6,T copolymers as a function of
PA-6,T comonomer content.
Table 2
DMA storage modulus and maximum tan
d values plus DSC enthalpy values for
PA-10,T-co-6,T and PA-12,T-co-6,T copolymers.
E0 at
75 ꢀC (MPa)
E0 at
200 ꢀC (MPa)
tan
dmax
DHmelting
(J/g)
PA-10,T
1800
1890
1850
1880
1820
1690
1729
1810
470
330
260
260
450
360
240
280
148
146
149
149
137
140
142
144
70
43
30
52
60
38
26
42
15 wt-% PA-6,T
30 wt-% PA-6,T
50 wt-% PA-6,T
PA-12,T
15 wt-% PA-6,T
30 wt-% PA-6,T
50 wt-% PA-6,T
Fig. 11. PA-10,T(12,T)e6,T copolymer melting data overlaid with several additional
PA-6,T copolymers versus mol-% PA-6,T.

T.F. Novitsky et al. / Polymer 51 (2010) 2417e2425
2425
6,T terephthalamide copolymers display a similar shape (Fig. 11)
as PA-10,T-co-6,T and 12,T-co-6,T, although the composition at
which the eutectic point occurs is reversed because PA-6,T is the
lower melting component here. Distinct from the isophthalamide
copolymers, terephthalamide sequences maintain a symmetrical
chain orientation in both homo and alternating segments. Their
unique melting point trends are associated with the ability of
alternating units to crystallize or participate in homopolymer
crystal formation in the eutectic well. Thus, they do not display
a large reduction in the melting point of PA-6,T crystals as is seen
for PA-6,I, 6,T copolymers.
A final observation relates the isomorphic PA-6,6, 6,T copoly-
mers with isodimorphic PA-10,T-co-6,T and 12,T-co-6,T copolymers
since their melting behaviors at compositions greater than 50 mol-
% PA-6,T are similar. In general, the melting temperatures of
isomorphic copolymers are a function of crystal composition, while
isodimorphic crystalline melting is a function of crystal amount and
size. Consequently, the density of PA-6,6, 6,T isomorphic copolya-
mides were found to rise with increasing PA-6,T content [5],
whereas the crystallinity of PA-10,T(12,T), 6,T copolymers was
shown to fall and then rise with increasing PA-6,T content. Despite
these differences in behavior, PA-6,6, 6,T and PA-10,T(12,T)e6,T
copolymers exceeding 50 mol-% PA-6,T have different crystalline
properties, yet have the same melting temperature.
independent of the comonomer composition. Past the eutectic
point, pure 6,T sequences become sufficiently abundant to crys-
tallize alone. NMR analysis confirmed increases in PA-6,T homo-
sequences above 50 wt-% PA-6,T. X-ray diffraction in this compo-
sition range also displayed sharp diffraction peaks that correspond
to PA-6,T crystals.
It has been demonstrated that melting temperatures and the
degree of crystallinity can be tuned by changing the amount of PA-
6,T comonomer in PA-10,T-co-6,T and PA-12,T-co-6,T copolymers.
These properties affect the processability, degree of crystallinity
and optical clarity of the resulting copolymers, while maintaining
higher glass transition temperatures than pure aliphatic
polyamides.
Acknowledgments
The authors would like to thank Dr. W.L. Jarrett for 13C NMR help
with interpretation, Dr. Kenneth Mauritz for the use of his DMA
equipment, and Ascend Performance Materials for providing
funding and starting materials, and the use of their Parr equipment.
References
[1] Kohan MI. Nylon plastics handbook. New York: Carl Hanser Verlag; 1995.
p. 372e4 and 594e9.
5. Conclusions
[2] Morgan PW, Kwolek SL. Macromolecules 1975;8(2):104e11.
[3] Shashoua VE, Eareckson WM. Journal of Polymer Science 1959;40:343e58.
[4] Kaitian X, Jarrett WL, Mathias LJ. Polymer Preprints 2005;46(1):789e90.
[5] Edgar OB, Hill R. Journal of Polymer Science 1952;8:1e22.
[6] Yu AJ, Evans RD. Journal of Polymer Science 1960;42:249e57.
[7] Harveym ED, Hybart FJ. Polymer 1971;12:711e6.
[8] Gaymans RJ, Aalto S, Maurer FHJ. Journal of Polymer Science, Part A: Polymer
Chemistry 1989;27:423e30.
[9] Togashsi O, Umetsu H, Iwamoto M. European patent 0448221; 1991.
[10] Wunderlich B. Macromolecular physics. New York: Academic Press; 1973.
p. 147e61.
[11] Liu H, Yang G, He A, Wu M. Journal of Applied Polymer Science
2004;94:819e26.
[12] Rulkens R, Crombach RCB. United States patent 6,747,120; 2004.
[13] Solomon OF, Ciuta IZ. Journal of Applied Polymer Science 1962;6:683.
[14] Novitsky TF, Lange CA, Mathias LJ, Osborn S, Manning S, Ayotte R. Journal of
Applied Polymer Science 2010;116(6):3388e95.
[15] Aerdts AM, Eersels KLL, Groeninckx G. Macromolecules 1996;29:1041e5.
[16] Meesiri W, Menczel J, Gaur U, Wunderlich B. Journal of Polymer Science, Part
B: Polymer Physics 1982;20:719e28.
[17] Blaschke F, Ludwig W. United States patent 3,382,216; 1968.
[18] Lum FG, Carlston EF. Industrial Engineering and Chemistry 1952;44
(7):1595e600.
Melt condensation polymerization of PA-10,T and 12,T with PA-
6,T comonomer has been demonstrated to produce statistical
copolymers. The 13C chemical shift of the substituted aromatic
carbon is sensitive to the comonomer connectivity at different
compositions, providing key chemical information in describing
the physical changes of the above mentioned copolymers. It was
found that at 15 and 30 wt-% PA-6,T, few pure PA-6,T sequences
exist, and the majority of the PA-6,Tcomonomer is found in PA-10,T
(or 12,T) ePA-6,T alternating sequences. Since the comonomers are
not co-crystalline, PA-10,T (or 12,T) ePA-6,T alternating units act as
impurities that inhibit the formation of PA-10,T or (12,T) crystals.
This is supported by a decrease in melting temperature and
enthalpy by DSC, the loss of sharp x-ray diffraction peaks up to
30 wt-% PA-6,T, and the increased clarity of compression molded
films. However, PA-6,T content greater than 30 wt-% increases the
melting temperature of the corresponding copolymer in a manner