Preparation, characterization and copolymerization of the 12-membered cyclic diamide 1,6-diazacyclododecane-2,5-dione

Article abstract of DOI:10.1002/pola.26975

A 12-membered cyclic diamide monomer for nylon 64 was successfully synthesized in fairly high yield (～45%). The synthesis conditions were varied to see the effect of the diamine and succinyl chloride reactants on yield. Threefold excess of 1,6-hexamethylenediamine (HDA) gave the highest yield, while further increasing the amount of HDA decreased the yield. Using N,N-diisopropylethylamine as acid scavenger resulted in the formation of two different cyclic amides, which were fully analyzed by ¹H and ¹³C solution nuclear magnetic resonance spectrometry and mass spectrometry. Copolymerization of cyclic amides with ε-caprolactam via an anionic route gave a block copolyamide with a two distinct endotherms in the differential scanning calorimetry analysis. However, copolymerization by the hydrolytic route gave only nylon 6 with terminal 64 units.

Full text of DOI:10.1002/pola.26975

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Preparation, Characterization, and Copolymerization of the
12-Membered Cyclic Diamide 1,6-Diazacyclododecane-2,5-dione
Ethem Kaya,1 Eylem Tarkin-Tas,1 Scott Osborn,2 Roger Ayotte,2
Steven Manning,2 Lon J. Mathias1
1Department of Polymer Science University of Southern Mississippi, Hattiesburg, Mississippi 39406-0076
2Ascend Performance Materials, Cantonment, Florida 32533
Correspondence to: L. J. Mathias (E-mail: lon.mathias@usm.edu)
Received 22 July 2013; accepted 24 September 2013; published online 29 October 2013
DOI: 10.1002/pola.26975
ABSTRACT: A 12-membered cyclic diamide monomer for nylon 64
was successfully synthesized in fairly high yield (ꢀ45%). The syn-
thesis conditions were varied to see the effect of the diamine and
succinyl chloride reactants on yield. Threefold excess of 1,6-hex-
amethylenediamine (HDA) gave the highest yield, while further
increasing the amount of HDA decreased the yield. Using N,N-dii-
sopropylethylamine as acid scavenger resulted in the formation of
two different cyclic amides, which were fully analyzed by 1H and
13C solution nuclear magnetic resonance spectrometry and mass
spectrometry. Copolymerization of cyclic amides with e-
caprolactam via an anionic route gave a block copolyamide with a
two distinct endotherms in the differential scanning calorimetry
analysis. However, copolymerization by the hydrolytic route gave
C
V
only nylon 6 with terminal 64 units. 2013 Wiley Periodicals, Inc.
J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 96–103
KEYWORDS: anionic polymerization; block copolymers; e-capro-
lactam; cyclic diamide; cyclic lactams; hydrolytic polymeriza-
tion; macrocyclics; polyamides
because this allows tailoring of properties for specific appli-
cations.12–15
INTRODUCTION Cyclic amides are continuing research topics
of polymer science for two main reasons. They are either
used as monomers for ring-opening polymerization, or they
are generated by normal step-growth polymerization and
ring-opening polymerization processes as minor compo-
nents.1 Cyclic diamides can be obtained in several ways
including by direct cyclization of dicarboxylic acid derivatives
with a diamine,2–4 depolymerization of polyamides,5 and
rearrangement of another cyclic species.6 One of the compli-
cations of homopolymerization of cyclic diamides are that
they usually have high melting points, especially those with
small rings; the actual melting points may be very close to
molecular degradation temperatures.7
Nylon 64 is of interest because of its higher melting point
when compared with commercial nylon 66. Unlike nylon 46,
which has been commercialized by DSM under the trade name
16
R
V
Stanyl , nylon 64 has not yet been commercialized, possibly
due to the side reactions which inhibit obtaining high molecu-
lar weight polymer via polycondensation reaction. Therefore,
an alternate route to high molecular nylon 64 is still needed.
Ring-opening polymerization of the 12-membered cyclic mono-
mer of nylon 64 may offer an alternative way to obtain a high
molecular weight homo- or copolymers of nylon 64.
Herein, we report the synthesis and characterization of the
12-membered cyclic diamide, 1,6-diazacyclododecane-2,5-
dione. The effects of the reactants on yield and products
were studied. Ring-opening reactivity of this cyclic diamide
was also explored. For this purpose, copolymerization of the
12-membered cyclic diamide with CLA was investigated with
different polymerization methods and conditions.
Polyamide 6, or nylon 6, is generally prepared by ring-
opening polymerization of e-caprolactam (CLA). CLA is most
widely polymerized via hydrolytic8 or anionic9,10 polymer-
ization methods. Although it is not commercially used, cati-
onic ring-opening homo- and copolymerization is also of
research interest.11 Polyamide 6 is a semicrystalline ther-
moplastic material with a relatively high melting point. Due
to its good mechanical strength, impact resistance, rigidity,
and low density, it is widely used as an engineering mate-
rial. Although polyamide 6 shows outstanding thermal and
mechanical properties, copolymerization of CLA with vari-
ous monomers are of industrial and academic interest,
EXPERIMENTAL
Materials
All reagents were used as received. Succinyl chloride, N,N-
diisopropylethylamine (DPEA), 4-dimethylaminopyridine
C
V
2013 Wiley Periodicals, Inc.
96
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 1 Synthesis of 1,6-diazacyclododecane-2,5-dione.
(DMAP), succinic anhydride, N-acetylcaprolactam, 2,2,2-tri-
fluoroethanol (TFE), and chloroform d were purchased from
Aldrich Chemical Company. 1,6-Hexamethylenediamine
(HDA) and chloroform were purchased from Fluka. CLA was
purchased from ACROS and recrystallized from acetone; the
crystals were dried at 70 ꢁC under reduced pressure over-
night before use.
in a three-necked round-bottomed flask under nitrogen. After
the reaction was complete, the precipitant was removed by
filtration. Solvent was removed from the filtrate under
reduced pressure at 50 ꢁC and the remaining solid was
washed with acetone several times. The final product was
obtained as white powder. It was recrystallized from acetoni-
trile. The melting point of the 12-membered cyclic diamide
ꢁ
was obtained as 290 C using DSC.
Instrumentation and Techniques
Routine solution 13C nuclear magnetic resonance spectrome-
try (NMR) was performed on Varian MercuryPLUS 200 and
500 MHz spectrometers. Transmission infrared measure-
ments were carried out on a Mattson Galaxy Series FTIR
5000 using KBr pellets and 256 transients. Differential scan-
ning calorimetry (DSC) experiments were performed on a TA
Instruments 2920. The temperature was ramped at a heating
Synthesis of HAD-bisimide
HDA (34 mmol) and succinic anhydride (72 mmol) were dis-
solved in 20 mL of deionized water. The mixture was
refluxed for 2 h and then vacuum was applied to remove
water while the temperature was increased to 150 ꢁC. The
reaction was carried out for another 2 h. After completion,
the crude product was allowed to cool down to room tem-
perature and washed with water several times. After filtra-
tion, the crude product was recrystallized from water twice,
which gave the final product as white crystals in 70% yield.
13C NMR (200 MHz, TFE:CDCl3): 179.1, 38.2, 27.5, 26.6, 25.7
ppm.
ꢁ
rate of 10 C/min under nitrogen.
Wide angle X-ray diffraction (WAXD) measurements were
obtained with a Rigaku Ultima III diffractometer operated at
40 kV and 44 mA. Powder samples were run with a scan
speed of 2ꢁ per minute from 2ꢁ to 45ꢁ.
Synthesis of HAD-bisamide Acid
The molecular weight of cyclic diamide and cyclic tetramide
(structures below) were determined by a ThermoFisher LXQ
Ion-Trap instrument running in positive mode. The samples
were prepared by placing ꢀ10 mg of cyclic diamide or cyclic
tetramide in an Eppendorf tube, then adding 1 mL of 1:1
THF:MeOH with NaCl. The sample was vortexed for 1 min to
ensure a homogenous solution. 100 mL of this sample was
then diluted to 1 mL with 900 mL of the 1:1 THF:MeOH NaCl
solution (the NaCl concentration in the THF:MeOH is approx-
imately 1 mM). The diluted sample was then infused into the
spectrometer at a rate of 10 mL/min until a stable spectrum
was observed. The spectrum was then collected and opti-
mized on the sodium adduct of cyclic diamide or cyclic tetra-
mide using the autotune function in the XCalibur software,
which controls the instrument. The final spectra are the
average of several such collected spectra.
HDA (34 mmol) and succinic anhydride (136 mmol) were
dissolved in 100 mL of deionized water. The mixture was
stirred at room temperature for 6 h. After the reaction was
complete, the precipitant was removed by filtration, and
washed with hot water several times. After washing with
acetone one more time, the final product was allowed to dry
in a vacuum oven overnight. The product was obtained as
white powder in 50% yield. 13C NMR (200 MHz, TFE:CDCl3):
176.8, 174.4, 39.8, 30.3, 29.5, 28.6, 26.1 ppm.
Synthesis of Copolymer Nylon (6-co-64) via the Anionic
Polymerization Method
CLA (2 g) and 10 w % cyclic diamide (0.2 g) were charged
to a test tube. Another test tube was charged with CLA (0.3
g) and dry sodium hydride (0.02 g). Nitrogen was bubbled
through both test tubes for 15 min before heating. Both test
ꢁ
tubes were placed in a previously heated oil bath at 140 C.
Synthesis of 1,6-Diazacyclododecane-2,5-dione
After melting, 0.3 mL of N-acetylcaprolactam was added to
the CLA/cyclic diamide mixture by a syringe and the mixture
was allowed to stir for another 10 min. The caprolactam/
sodium hydride mixture was then added to this mixture, and
the polymerization was carried out at 145 ꢁC until the mix-
ture solidified after 15 min. The product was allowed to cool
down to room temperature under nitrogen, and was then
powdered and washed with refluxing ethanol overnight to
remove unreacted monomers. After washing with ethanol
one more time, the final product was dried in a vacuum
The procedures for synthesis of the cyclic monomers are
given for sample 4. All other samples were synthesized using
the same procedure except the ratio of reactants or the
amounts of solvent were varied. HDA (16.9 mmol), DPEA
(33.9 mmol), and 4-dimethylaminopyridine (DMAP) (1.6
mmol) were dissolved in 50 mL chloroform and charged to a
syringe. A second syringe was charged with succinyl chloride
(16.9 mmol) in 50 mL chloroform. These two syringes were
placed in a syringe pump and their contents added very
slowly (addition rate 0.5 mL/min) to 500 mL of chloroform
ꢁ
oven for 12 h at 100 C (yield ꢀ34%).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103
97

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Synthesis of 1,6-Diazacyclododecane-2,5-dione
Sample #
HDA
Succinyl Chloride
Chloroform (mL)
DPEA
DMAP
Yield (%)
1
2
3
4
5
6
33
53
3
3
3
3
3
3
3
500
–
–
45
26
42
12
40
32
500
–
–
500
23
23
–
–
3
500
0.13
33
3
1000
1000
–
–
23
Attempted Synthesis of Copolymer Nylon (6-co-64) via
Hydrolytic Polymerization Method
reactants were changed with respect to the succinyl chloride.
The data are given in Table 1. The best yield was obtained
for the samples numbered 1 and 3. The catalyst, DMAP,
somehow inhibited the cyclization. Threefold-excess of HDA
gave the best yield. Increasing HDA concentration even fur-
ther decreased the yield of the cyclic diamide, possibly due
to increasing probability that one succinyl chloride could
react with two HDA leading to a linear monomer.
CLA (2 g), 10 w % cyclic diamide (0.2 g), 10 mol % amino-
caproic acid, and 1 mL deionized water were charged to a
test tube. Nitrogen was bubbled through the mixture for 15
min before heating. The temperature was then maintained at
250 ꢁC for 4 h in a sand bath. The product was allowed to
cool down to room temperature under nitrogen, and was
then powdered and washed with refluxing ethanol overnight
to remove unreacted CLA and cyclic diamide. After washing
with ethanol one more time, tꢁhe final product was dried in a
vacuum oven for 12 h at 100 C.
The 12-membered ring cyclic diamide obtained as a white
powder was not soluble in chloroform, and a TFE:CDCl3 mix-
ture was used for NMR analysis. The 13C solution NMR spec-
trum of 1,6-diazacyclododecane-2,5-dione in TFE:CDCl3 is
shown in Figure 1. The carbonyl carbon peak of the cyclic
amide appears at 173.3 ppm. This peak and the peak
assigned as “a” are strong indicators of the absence of linear
compounds in the final product.
RESULTS AND DISCUSSION
The overall synthetic procedure for the 12-membered cyclic
diamide is given in Scheme 1. The synthesis of cyclic dia-
mides of diamines and diacid chlorides is usually carried out
with high-dilution methods or slow addition of reactants
into a solvent to decrease the possibility of linear oligomer
or polymer formation.3
The 1H solution NMR spectrum of 1,6-diazacyclododecane-
2,5-dione is shown in Figure 2. All peaks are assigned based
on the molecular structure expected for the cyclic diamide.
The amide NH can be seen around 6.6 ppm and there is a
small shoulder on this peak at 6.9 ppm. The protons a to the
amide NH can be observed at 3.3 ppm, and this peak also
has a small shoulder around 3.2 ppm. The shoulder may be
related to higher order macrocycle formation, which cannot
be observed easily with 13C solution NMR. No free acid or
amine end-groups were observed, confirming macrocycle
formation.
It was of interest here to explore the effects of reactants,
ratio of reactants, and amount of solvent used for synthesis
of the 12-membered cyclic diamide on yield. For this pur-
pose, the amount of succinyl chloride was kept almost con-
stant (ꢀX 5 17 mmol) and the amounts of the other
FIGURE 1 13C solution NMR spectrum of 1,6-diazacyclodode-
FIGURE
dione. The sample was dissolved in TFE:CDCl3.
2
1H solution NMR of 1,6-diazacyclododecane-2,5-
cane-2,5-dione. The sample was dissolved in TFE:CDCl3.
98
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
higher order cyclic formation. However, the same reaction
with DPEA gives a 50/50 mixture of cyclic diamide and
cyclic tetramide. Cyclic diamide is the same 12-membered
cyclic compound obtained without DPEA. Cyclic tetramide,
on the other hand, can be obtainedꢁnearly pure as a white
powder with a melting point of 250 C after recrystallization
of the 50/50 mixture from acetonitrile. Using mass spec-
trometry, the molecular weight of cyclic diamide and cyclic
tetramide were determined to be 198.18 and 396.41 g/mol,
respectively, corresponding to mono- and biscyclic amides.
Triethylamine, which is commonly used in synthesis to
obtain cyclic diamides18 instead of DPEA, gives negligible
yield of cyclic product with the monomers used here. This
may be due to higher reactivity of succinyl chloride when
compared with other aliphatic diacid chlorides evaluated
previously.3
Figure 4 shows the 13C solution NMR spectra of the macro-
cycle mixture and of cyclic tetramide in TFE:CDCl3 solvent
mixture. All peaks are assigned based on the expected struc-
tures of the cyclic monomers from molecular weight deter-
mination by mass spectrometry. It can be seen that there are
two set of peaks with the same number of peaks of roughly
equal intensities. However, the peak intensities of the car-
bonyl carbons are not the same, perhaps due to different
relaxation times for these carbons. The peaks associated
with cyclic diamide appear exactly at the same chemical
shifts as for the cyclic monomer obtained without DPEA and
are assigned with the same letters. The carbonyl carbon
peak of cyclic tetramide appears at 173.5 ppm. When com-
pared to the cyclic diamide, peaks related to cyclic tetramide
are shifted significantly. The peak a to the amide groups of
cyclic diamide assigned as “b” (observed at 36.7 ppm) shifts
to 38.4 ppm for the cyclic tetramide. An upfield shift is
observed for the methylene group a to the carbonyl which
moves from 34.5 to 31.4 ppm. The significant difference in
FIGURE 3 FTIR spectrum 1,6-diazacyclododecane-2,5-dione.
The infrared spectra for the 12-membered cyclic diamide
displayed the characteristic peaks of amide groups and
methylene groups, as shown in Figure 3. Peaks were found
at 3313 cm21 (NAH stretching), 3095 cm21 (overtone of
NAH in-plane bending), 2834–3000 cm21 (CAH stretching),
1644 cm21 (C@O stretch), and 1538 cm21 (CAN stretch and
COANAH bend).17
For the reaction of HDA with succinyl chloride, only one kind
of cyclic product was obtained (samples 1, 2, 5 in Table 1).
The second product, which gives a peak around 6.9 ppm in
the 1H NMR spectrum, was speculated as being due to
FIGURE 4 13C solution NMR of the 50/50 mixture (top) and cyclic tetramide (bottom). Each sample was dissolved in TFE:CDCl3.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103
99

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
for TFE broaden and shifted in the bottom spectra probably
due to a trace amount of water.
FTIR spectra of the 50/50 mixture and purified cyclic tetra-
mide are shown in Figure 6. As expected, the two spectra
are almost identical with only minor differences when com-
pared with cyclic diamide. Most changes are observed for
the stretching vibration of NH groups, as seen by comparing
the FTIR spectrum of cyclic diamide (Fig. 3) with the spec-
trum of cyclic tetramide. For diamide, this band is narrower
and shows only one peak at 3313 cm21 with a small
shoulder, while two peaks at 3301 and 3334 cm21 are
observed for the tetramide. This band is very sensitive to
hydrogen bonding and generally enhanced vibration is a
result of conformational disorder, which leads to a broader
peak, while an ordered structure will show a narrower
band.19,20 Based on the shape and frequency of this band, it
appears that cyclic diamide is more ordered in the solid
state when compared with cyclic tetramide. Alternatively, the
difference may also be related to the types of inter- and
intramolecular hydrogen bonding of the different cyclic
structures.2
FIGURE 5 1H solution NMR of a 50/50 mixture (top) and puri-
fied tetramide (bottom). The sample was dissolved in
TFE:CDCl3.
chemical shifts for these two cyclic amides is consistent with
different ring strains and/or conformational mobility for
these two monomers.
Crystallinities of these cyclic diamides were investigated
using WAXD and found to be completely different for each
cyclic monomer when compared to the mixture. WAXD pat-
terns recorded at room temperature of the 50/50 mixture,
and purified cyclic diamide and cyclic tetramide, are shown
in Figure 7. Cyclic tetramide was purified by recrystallization
from the 50/50 mixture. The WAXD pattern for each of the
diamides is different from each other and both are very dif-
ferent when compared to the 50/50 mixture. This may indi-
cate that the mixture crystallizes as a 1:1 molecular complex
rather than into separate crystallites for each of the two
cyclic structures. Further work is needed to clarify and
understand this behavior.
The solution 1H NMR spectra of the mixture and of cyclic
tetramide are shown in Figure 5. Two sets of peaks with the
same number and roughly equal intensities are also
observed in the 1H NMR spectra. All peaks are assigned
based on the molecular structure of cyclic amides. The peaks
associated with the protons a to the amide NH, which are
hardly observed in Figure 2, are of equal intensities. Splitting
of the peaks associated the same protons of cyclic diamide is
seen at 3.3 ppm. The amide NH, which can be observed
around 6.6 ppm for cyclic diamide, shifts downfield to 6.9
ppm for cyclic tetramide. Similar downfield shifts were
observed for various ring size cyclic oligo(undecanamide)s.1
However, the shifts are not as pronounced for cyclic oli-
go(undecanamide)s of different sizes. This may be due to the
fact that the ring strain is more crucial for the cyclic amides
in this study. The exchangeable hydrogen peak at 4.39 ppm
Homopolymerizations of these cyclic diamides in the molten
state were difficult due to the fact that the melting points of
these monomers are so high, especially forꢁthe 12-membered
cyclic diamide, which melts at about 290 C, in the range of
nylon 64. For this reason, we first explored the copolymer-
ization of CLA with these cyclic monomers to investigate
FIGURE 6 Partial FTIR spectra of the 50/50 mixture (solid line) and cyclic tetramide (dashed line).
100
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
amides used for polymerization mostly contained a single
type of cyclic diamide.
The 13C solution NMR spectrum of a typical copolymer syn-
thesized by anionic polymerization is shown in Figure 8. The
spectrum of homopolymer nylon 6 is included for compari-
son. The carbonyl carbon peak of the cyclic amide observed
at 173.3–173.5 ppm shifts downfield to 174 ppm in the
polymer backbone. The peaks assigned as a and c of cyclic
diamide are observed at 34.05 and 25.4 ppm, respectively,
while these peaks appear at 31.5 and 28.8 ppm, respectively,
in the copolymer. These chemical shift values are very close
to those of these same carbons of the cyclic tetramide. This
is a good indication that cyclic tetramide possesses lower
ring strain when compared with diamide, as would be
expected for larger cycle size. Based on the intensity of the
carbonyl carbon peak, the incorporation of nylon 64 seg-
ments is calculated as being 12 mol %. The actual intensity
of the carbonyl carbon of nylon 64 units is divided by two
based on the assumption that there are two carbonyl car-
bons for nylon 64 units while there is only one for nylon 6
units. The copolymer number-average molecular weight was
calculated to be 8927 g/mol using NMR data with the appli-
cation of previously published equations,21,22 and peak
assignments.23
FIGURE 7 WAXD patterns of 50/50 mixture (bottom), cyclic tet-
ramide (middle), and cyclic diamide (top).
ring-opening capabilities and the effect of nylon 64 repeat
units on possible improvement of thermal and mechanical
properties of nylon 6. Polymerizations of monomer mixtures
could be carried out at temperatures lower than those
required for pure monomers, offering a pathway to non-
degraded and more easily processed products.
Copolymerization of 1,6-diazacyclododecane-2,5-dione with
CLA was attempted with both anionic and hydrolytic poly-
merization methods to investigate the effect of the reaction
mechanism on final properties. The details of the copolymer-
ization are described in the experimental section. As anionic
polymerization is faster and carried out at lower tempera-
ture, it was expected to give fewer side products. Cyclic
As expected, copolymerization using the hydrolytic polymer-
ization method is more complicated and results in several
side products. To identify the peaks of such copolymers, two
novel model compounds were synthesized and the chemical
shifts of copolymers analyzed based on the chemical shift of
these compounds in the same solvent system (3:1
FIGURE 8 13C solution NMR of neat nylon 6 (top) and nylon (6-co-64) (bottom) by anionic polymerization method; 3:1 TFE:CDCl3
NMR solvent.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103
101

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 9 Chemical structures of (a) HAD-bisimide and (b) HAD-bisamide acid.
TFE:CDCl3). The structures of these compounds are shown
in Figure 9.
while the higher endotherm is almost the same as the melt-
ing point of nylon 64, which was reported as 278 ꢁC.1 The
two melting endotherms were still observed even after sev-
eral washing and reprecipitation processes.
The 13C solution NMR spectrum of the copolymer with 10
wt % feed of cyclic diamide monomer synthesized by hydro-
lytic polymerization is shown in Figure 10. All peaks are
assigned based on the chemical shifts of model compounds
and nylon (6-co-64) obtained by anionic polymerization.
Four distinct carbonyl peaks are observed including the car-
bonyl carbon of nylon 6. The carbonyl peak of acid end-
groups appears at 178 ppm, while the carbonyl peak of
imide-end groups appear at 179.8 ppm. The peak associated
with nylon 64 carbonyl is observed at 173.9 and split into
two; one probably in the backbone, and the other the car-
bonyl carbon close to the carboxylic acid end-group. The
splitting can be observed easily for the methyl carbon a to
the carbonyl assigned as a. There are three peaks observed
for this carbon in which one is associated with the carbon a
to carboxylic acid end-groups and the others are associated
with carbons a to backbone amides and amides close to car-
boxylic acid end-groups. Considering the intensities of the
peaks, the cyclic diamide appears to have reacted only once
with nylon 6 units to form either imide or carboxylic acid
terminal units.
In addition to 13C solution NMR spectrum of the copolymer
synthesized by the anionic method, which shows one peak
and no splitting for nylon 64 repeat units, two distinct peaks
in the DSC similar to each homopolymer strongly suggests
that the copolymer is a block copolyamide, and that the
reactivity of 1,6-diazacyclododecane-2,5-dione and CLA are
sufficiently different to minimize alternating monomer incor-
poration. Similar results were reported by Roda et al.,25,26
for the copolymerization of CLA and x-laurolactam by differ-
ent initiator systems. The copolymer prepared by sodium
salt of caprolactam in combination with N-benzoyl-e-capro-
lactam initiator system showed a random character with a
single melting endotherm and one crystalline form, while the
copolymer prepared by e-caprolactam magnesium bromide
initiator showed two melting endotherms and two crystalline
forms (a and c). Unlike nylon (6-co-64) reported here show-
ing two distinct blocks, the copolymer of CLA and x-
ꢁ
laurolactam shows a melting endotherm around 140 C cor-
responding to the melting of a random copolymer of CLA
and x-laurolactam, whereas the second endotherm around
214 C corresponds to the melting of nylon 6.
ꢁ
The DSC heating thermograms of Nylon (6-co-64) prepared
by hydrolytic polymerization and anionic polymerization is
shown in Figure 11. The latter shows two distinct peaks at
217 ꢁC and 280 ꢁC indicating two different crystal popula-
tions. The lower endotherm is similar to that of nylon 6,24
Two distinct peaks at 195 ꢁC and 218 ꢁC are also observed
in the DSC heating thermogram of nylon (6-co-64) obtained
by hydrolytic polymerization. No endotherm is observed at a
FIGURE 10 13C solution NMR of nylon (6-co-64) by hydrolytic
polymerization method. The sample was dissolved in 3:1
TFE:CDCl3.
FIGURE 11 DSC heating thermograms of nylon (6-co-64) by
hydrolytic polymerization (dashed line) and anionic polymeriza-
tion (solid line).
102
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
7 Y. Iwakur, K. Uno, K. Haga, M. Akiyama, J. Polym. Sci. Part
A-1: Polym. Chem. 1969, 7, 657–666.
temperature higher than the melting point of nylon 6. The
higher endotherm is presumed to be the melting point of
nylon 6 segments, while the lower endotherm appears to be
related to the melting of terminal nylon 64 units at the sur-
face of nylon 6 crystallites, or of very short nylon 6 chains
with acid and/or imide terminal units.
8 R. Laura, R. Saverio, M. Orietta, B. Alessandro, B. Federica,
Polymer 2005, 46, 6810–6819.
9 A. Bernaskova, D. Chromcova, J. Brozek, J. Roda, Polymer
2004, 45, 2141–2148.
10 E. Tarkin-Tas, L. J. Mathias, Macromolecules 2010, 43, 968–
974.
CONCLUSIONS
11 J. M. Warakomski, Chem. Mater. 1992, 4, 1000–1004.
The 12-membered cyclic diamide has been successfully syn-
thesized in a reasonable yield. The yield and the ring size of
cyclic diamide were found to depend on the reactants used
and their feed ratios. The synthesis of cyclic diamide with
DPEA gave two different cyclic structures, which were easily
separated by recrystallization. The crystal packing as well
as the melting points of each diamide is very different.
These cyclic diamides were shown to copolymerize with
e-caprolactam, resulting in block copolymers of nylon 6
and 64.
12 T. Whelan, In Polymer Technology Dictionary, 1st ed.; Chap-
man & Hall: London, 1994; pp 275–276.
13 N. Toncheva, R. Jerome, R. Mateva, Eur. Polym. J. 2011, 47,
238–247.
14 X. Fang, R. Hutcheon, D. A. Scola, J. Polym. Sci. Part A:
Polym. Chem. 2000, 38, 1379–1390.
15 T. F. Novitsky, L. J. Mathias, J. Polym. Sci. Part A: Polym.
Chem. 2011, 49, 2271–2280.
16 I. Vulic, R. W. W. Hooft, J. Kroon, J. Crystallograph. Spec-
trosc. Res. 1993, 23, 917–920.
17 X. Cui, D. Yan, J. Polym. Sci. Part B: Polym. Phys. 2005, 43,
2048–2060.
ACKNOWLEDGMENT
18 J. H. Glans, M. K. Akkapeddi, Macromolecules 1992, 25,
5526–5527.
The authors thank Douglas Masterson for obtaining mass spec-
trometry data on the cyclic dimers.
19 D. J. Skrovanek, P. C. Painter, M. M. Coleman, Macromole-
cules 1986, 19, 699–705.
20 Q. Zhang, Z. Mo, H. Zhang, S. Liu, S. Z. D. Cheng, Polymer
2001, 42, 5543–5547.
REFERENCES AND NOTES
21 R. D. Davis, S. J. Steadman, W. L. Jarrett, L. J. Mathias,
Macromolecules 2000, 33, 7088–7092.
1 P. Peng, P. Hodge, Polymer 1998, 39, 981–990.
2 P. Salanski, J. W. Krajewski, P. Gluzinski, S. Kasprzyk, T.
Stankiewicz, J. Jurczak, J. Crystallograph. Spectrosc. Res. 1991,
21, 195–200.
22 C. Bennett, E. Kaya, A. M. Sikes, L. J. Mathias, J. Polym.
Sci. Part A: Polym. Chem. 2009, 47, 4409–4419.
23 R. D. Davis, W. L. Jarrett, L. J. Mathias, Polymer 2001, 42,
2621–2626.
3 O. Vogl, A. C. Knight, Macromolecules 1968, 1, 311–315.
4 O. Vogl, A. C. Knight, Macromolecules 1968, 1, 315–318.
24 C. Bennett, J. Zeng, S. Kumar, L. J. Mathias, J. Appl. Polym.
Sci. 2006, 99, 2062–2067.
5 S. V. Levchik, E. D. Weil, M. Lewin, Polym. Int. 1999, 48,
532–557.
25 J. Budin, J. Brozek, J. Roda, Polymer 2006, 47, 140–147.
6 Y. Iwakur, K. Uno, K. Haga, K. Nakamura, J. Polym. Sci.
Polym. Chem. Ed. 1973, 11, 367–376.
26 J. Budin, J. Brozek, J. Roda, Macromol. Symp. 2006, 240,
78–82.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 96–103
103