Development of a triazole-cure resin system for composites: Evaluation of alkyne curatives

Article abstract of DOI:10.1016/j.polymer.2012.04.002

We are developing a resin system that cures via triazole ring formation (cycloaddition reaction of azides with terminal alkynes) instead of the traditional oxirane/amine reaction. The high exothermicity of the azido/alkyne reaction is expected to yield higher extents of reaction under ambient-cure conditions, making the resin system potentially suitable for out-of-autoclave curing processes. The difunctional azide-terminated resin, di(3-azido-2-hydroxypropyl) ether of bisphenol-A, was selected as the baseline diazide. A number of alkyne crosslinkers were synthesized and characterized, including propiolate esters of di- and trifunctional alcohols, propargyl esters of di- and trifunctional carboxylic acids, propargyl ethers of di- and trifunctional alcohols, and N,N,N′,N′-tetrapropargyl derivatives of primary diamines. Commercially available tripropargyl amine was also studied. Those systems employing a propiolate-based alkyne were found to be much more reactive toward the Huisgen 1,3-dipolar cycloaddition than the propargyl species. Curing energetics as a function of alkyne type, investigated through a dynamic differential scanning calorimetry approach, showed a distinct divide between the averaged activation energies of the propiolate and propargyl-type crosslinkers, 69.2-73.6 kJ/mol versus 82.3-86.4 kJ/mol, respectively. Cured network properties were readily manipulated through the incorporation of varying amounts of di- versus tri- and tetra-functional alkynes or through incorporation of soft alkylene and alkyleneoxy versus rigid aromatic polyalkynes. As expected, mechanical properties, e.g., the temperature of the tan δ peak in dynamic mechanical analysis, were found to increase with increasing crosslink density. These results have allowed us to select the most promising systems for scale-up and fabrication of samples of both pure resin and composites for traditional mechanical property testing, which will be reported in a subsequent paper.

Full text of DOI:10.1016/j.polymer.2012.04.002

Polymer 53 (2012) 2548e2558
Contents lists available at SciVerse ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Development of a triazole-cure resin system for composites: Evaluation of alkyne
curatives
*
Irene E. Gorman, Rodney L. Willer, Lisa K. Kemp, Robson F. Storey
School of Polymers and High Performance Materials, The University of Southern Mississippi, 118 College Dr. #5050, Hattiesburg, MS 39406, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We are developing a resin system that cures via triazole ring formation (cycloaddition reaction of azides
with terminal alkynes) instead of the traditional oxirane/amine reaction. The high exothermicity of the
azido/alkyne reaction is expected to yield higher extents of reaction under ambient-cure conditions,
making the resin system potentially suitable for “out-of-autoclave” curing processes. The difunctional
azide-terminated resin, di(3-azido-2-hydroxypropyl) ether of bisphenol-A, was selected as the baseline
diazide. A number of alkyne crosslinkers were synthesized and characterized, including propiolate esters
of di- and trifunctional alcohols, propargyl esters of di- and trifunctional carboxylic acids, propargyl ethers
of di- and trifunctional alcohols, and N,N,N0,N0-tetrapropargyl derivatives of primary diamines.
Commercially available tripropargyl amine was also studied. Those systems employing a propiolate-based
alkyne were found to be much more reactive toward the Huisgen 1,3-dipolar cycloaddition than the
propargyl species. Curing energetics as a function of alkyne type, investigated through a dynamic differ-
ential scanning calorimetry approach, showed a distinct divide between the averaged activation energies
of the propiolate and propargyl-type crosslinkers, 69.2e73.6 kJ/mol versus 82.3e86.4 kJ/mol, respectively.
Cured network properties were readily manipulated through the incorporation of varying amounts of di-
versus tri- and tetra-functional alkynes or through incorporation of soft alkylene and alkyleneoxy versus
Received 8 January 2012
Received in revised form
29 March 2012
Accepted 6 April 2012
Available online 16 April 2012
Keywords:
Triazole crosslink
Alkyne
Azide
rigid aromatic polyalkynes. As expected, mechanical properties, e.g., the temperature of the tan
d peak
in dynamic mechanical analysis, were found to increase with increasing crosslink density. These results
have allowed us to select the most promising systems for scale-up and fabrication of samples of both
pure resin and composites for traditional mechanical property testing, which will be reported in
a subsequent paper.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
reaction can reach higher conversion prior to vitrification, without
external heat [1]. Reaction between a terminal alkyne and an azide
The use of advanced composite materials in high performance
applications has been gaining popularity, due in part to their low
weight and ability to withstand harsh loading conditions, e.g., those
experienced in aerospace and marine applications. A significant
challenge in designing these materials is the development of
systems that can cure out-of-autoclave. In certain situations, the
use of an autoclave is impossible, impractical, or uneconomical.
Examples include the preparation of large composite panels
encountered in marine applications and repair of existing
composite panels incorporated into larger structures.
[2,3], the so-called Huisgen 1,3-dipolar cycloaddition reaction, is an
excellent candidate for this application, and indeed, many attri-
butes of the Huisgen reaction are expected to make it particularly
advantageous toward the synthesis of advanced composite mate-
rials: it is easy to perform; it is tolerant to a wide range of impu-
rities; it is high yielding; no small molecule is evolved, and
most notably, the reaction is highly exothermic, with a
DH
between ꢀ210 kJ/mol and ꢀ270 kJ/mol [4]. The 1,2,3-triazole
heterocycle formed during the cycloaddition has aromatic char-
acter, hydrogen bond accepting ability, and high chemical stability;
it is largely inert to severe hydrolytic, oxidizing, and reducing
conditions, even at high temperature [5].
Curing of composite structures out-of-autoclave is facilitated by
a chemistry that is highly exothermic. Thus, the crosslinking
It should be noted that while the Huisgen 1,3-dipolar cyclo-
addition is extremely exothermic, its activation energy is typically
quite high, and triazole formation is often minimal at room
temperature in the absence of catalyst. However, the reaction is
* Corresponding author. Tel.: þ1 6012664879.
E-mail address: Robson.Storey@usm.edu (R.F. Storey).
0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2012.04.002

I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
2549
exothermically autoaccelerating, whereby the heat generated by
initial reaction events is sufficient to activate further reaction, in
turn generating more heat; this is especially true for large reac-
tion masses with low surface-to-volume ratios. For certain elec-
trophilic alkynes, the activation barrier for reaction is relatively
low, ca. 70 kJ/mol [6], and in such cases the reaction will occur at
low temperatures, including room temperature, even without
catalyst.
5 mm o.d. tubes. Sample concentrations were approximately 25%
(w/v) in either CDCl3, CD3CN, or DMSO-d6 containing 1% TMS as an
internal reference.
Dynamic mechanical analysis (DMA) was conducted using
a Q800 (TA Instruments) instrument. The frequency was set at 1 Hz,
the pre-load static force at 0.005 N, the oscillatory amplitude at
20 mm, and the track setting at 125%.
In several recent reports [6e9], we have described an azide-
modified epoxy resin that cures via triazole ring formation, as
opposed to the traditional oxirane/amine reaction (Fig. 1). This low
temperature-cure composite matrix resin, di(3-azido-2-
hydroxypropyl) ether of bisphenol-A (DAHP-BPA), was first repor-
ted by Fokin, Finn, and coworkers [10] and is based on proven
epoxy (epichlorohydrin/bisphenol-A) backbone chemistry. Kessler
and coworkers have since also published on the subject [11]. In our
work, commercially available epoxy resin (i.e., EPON 825) has been
derivatized to carry azide end groups through a nucleophilic
substitution reaction of the epoxide ring with NaN3. The resulting
liquid, azide-modified, epoxy resin may be cured with a variety of
multifunctional alkynes representing a broad range of reactivities
and chemical structures: propiolate vs. propargyl, aliphatic vs.
aromatic, and di vs. tri- vs. tetra-functionality. Herein we explore
these differences and how they affect the curing profiles and ulti-
mate design of the resin system. Although these initial studies were
conducted in the absence of catalyst, the effect of Cu(I) on the
system will be the subject of subsequent reports. Additionally the
fabrication of carbon filled composites utilizing a propargyl/DAHP-
BPA resin system is currently being explored.
2.3. Synthesis of di(3-azido-2-hydroxypropyl) ether of bisphenol-A
(DAHP-BPA)
To a 500 mL round-bottom flask equipped with a magnetic stir
bar were charged 85.1 g DGEBA epoxy resin (EPON 825) (0.25 mol),
65.7 g ammonium chloride (1.23 mol), 64.3 g sodium azide
(1.0 mol), 167 mL 2-ethoxyethanol (Cellosolve), and 83 mL de-
ionized (DI) water. The reaction was allowed to stir until the
majority of the reagents had dissolved. The reaction was then
brought to reflux by submersing the flask into an oil bath at
a constant temperature of 130 ꢁC. The internal temperature of the
reaction reached 107 ꢁC; the pH remained in the neutral range, ca.
7e8. The reaction was allowed to reflux for 4 h with continuous
stirring. The reaction contents were then cooled and added to
a separatory funnel. To the funnel was added 250 mL DI water, and
the mixture was extracted with ethyl acetate (2 ꢂ 250 mL). The
organic layers were combined and washed with brine (2 ꢂ 250 mL)
and DI water (2 ꢂ 250 mL), dried with MgSO4, filtered, and vacuum
stripped using a wiped film evaporator to yield the final product. 1H
NMR (300 MHz, CDCl3):
d
¼ 1.63 (s, 6H, CH3), 2.55 (s, 2H, OH), 3.51
(m, 4H, CH2eN3), 3.99 (d, 4H, OeCH2), 4.15 (m, 2H, CHeOH), 6.82
(d, 4H, AreC2,6-H), 7.12 (d, 4H, AreC3,5eH) ppm. 13C NMR (75 MHz,
CDCl3):
d
¼ 31.00 (CH3), 41.76 (CeCH3), 53.37 (CH2eN3), 68.97
2. Experimental
(CHeOH), 69.33 (OeCH2), 113.92 (AreC3,5), 127.85 (AreC2,6), 143.86
(AreC1), 156.03 (AreC4) ppm.
2.1. Materials
2.4. Synthesis of polyalkyne curing agents
All reagents were used without further purification. Tetra-n-
butyl ammonium bromide (TBAB, 98%) was purchased from Fluka
Chemicals. The diglycidyl ether of bisphenol-A (DGEBA, EPON 825)
was donated by Hexion Specialty Chemicals. The remaining
chemicals and all of the solvents were purchased from Sigmae
Aldrich.
The multifunctional alkyne curing agents used in this work are
shown in Fig. 2. Tripropargyl amine (TPA) is commercially avail-
able and was obtained from SigmaeAldrich. Syntheses for several
of the propargyl and propiolate esters (DPA, TPTM, HDP, and
TMPTP) were initially developed by Willer, Lin, and Baum at
Fluorchem Inc., Azusa, CA, during development of triazole-cure
propellants [12].
2.2. Instrumentation
Thermograms were recorded using a Q200 (TA Instruments)
differential scanning calorimeter. The furnace atmosphere was
defined by 50 mL/min of nitrogen. Polymer samples ranged from 5
to 12 mg and were analyzed in standard aluminum crucibles.
1H and 13C-NMR spectra were obtained using either a Varian
Mercury 300 or a Varian UNITY-INOVA 500 MHz spectrometer with
2.4.1. Synthesis of oligo(ethylene glycol) dipropiolates
A modification of the procedure of Miller [13] was used to
synthesize three oligo(ethylene glycol) dipropiolates: poly-
(ethylene glycol)-200 dipropiolate (PEGDP), di(ethylene glycol)
dipropiolate (DEGDP), and tetra(ethylene glycol) dipropiolate
(TEGDP). The synthesis of tetra(ethylene glycol) dipropiolate is
representative: tetraethylene glycol (77.6 g, 0.40 mol) was weighed
into a dry 1000 mL round-bottom flask equipped with a magnetic
stir bar, and dried by placing under high vacuum (<0.5 Torr) for 1 h
at 60 ꢁC. Propiolic acid (63 g, 95%, 0.85 mol), benzene (350 mL), and
p-toluenesulfonic acid mono-hydrate (4.0 g) were added to the
flask. A DeaneStark trap and a condenser were attached, and the
reaction was protected by a nitrogen bubbler. The mixture was
heated at reflux for 20 h, and the water of condensation was
separated (bath temp 110 ꢁC). A total of 15.5 mL of water was
collected (theory is 14.4 mL). The mixture was then cooled to below
room temperature and transferred to a 1 L separatory funnel using
two 200 mL diethyl ether rinses to insure complete transfer. The
benzene-diethyl ether solution was washed with 20 wt% sodium
hydroxide solution (1 ꢂ 25 mL). The solution was then washed with
brine (1 ꢂ 25 mL), dried over MgSO4, and filtered, and the solvent
O
O
glycol ether/H2O
NaN3
NH4Cl
+
+
O
EPON 825
OH
O
70/30 (v/v), reflux
-NaCl, -NH3
OH
polyalkyne crosslinker
N3
O
O
N3
di(3-azido-2-hydroxypropyl) ether of bisphenol-A (DAHP-BPA)
H
N
O
N
N
R
O
triazole crosslink
Fig. 1. Synthesis of azide-modified epoxy resin (DAHP-BPA) and its curing reaction
with polyfunctional alkynes to form triazole crosslinks.

2550
I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
≅
≅
Fig. 2. Multi-functional alkyne curing agents used in this work: propiolate, propargyl esters, propargyl ethers, and propargyl amines.
was removed under reduced pressure. 1H NMR (300 MHz, CDCl3):
¼ 3.00 (s, 2H, hCeH), 3.54 (s, 8H, CH2eO), 4.25 (t, J ¼ 6.5 Hz, 4H,
CH2eCH2eOeCO), 4.31 (t, J ¼ 6.5 Hz, 4H, CH2eCH2eOeCO) ppm.
13C NMR (75 MHz, CDCl3):
1712(vs), 1461(m), 1370(m), 1241(vs), 1029(m), 966(m), 955(sh),
913(w), 867(m), 756(s), 710(m) cmꢀ1. 1H NMR (DMSO-d6):
¼ 4.60
(s, 3H, hCeH), 5.23 (s, 6H, CH2eO), 7.41 (s, 3H, AreH) ppm. 13C
NMR (DMSO-d6):
128.89 (AreCH), 136.17 (AreCe), and 152.47 (C]O) ppm. HRMS
(EI); C18H12O6 [M]þ: calcd. 324.0634; found 324.0635 amu.
d
d
d
¼ 65.19, 68.47, 70.53, 70.57 (CH2eCH2),
d
¼ 67.35 (CH2eO), 74.96 (CeCh), 79.95 (hCeH),
74.46 (CeCh), 75.38 (hCeH), 152.59(C]O) ppm.
2.4.2. Synthesis of 1,3,5-benzenetrimethanol, 1,3,5-tripropiolate
(TMBTP)
2.4.3. Synthesis of trimethylolpropane tripropiolate (TMPTP)
First, 1,3,5-tri(hydroxymethyl)benzene was synthesized by the
literature procedure developed by Moses [14]. The melting point of
the product was 75e76 ꢁC (lit. 75 ꢁC [14]). Propiolic acid (2.45 g,
0.035 mol) was then weighed into a dry 100 mL round-bottom flask
equipped with a magnetic stirring bar. Toluene (50 mL), 1,3,5-
tri(hydroxymethyl)benzene (1.68 g, 0.010 mol), and p-toluene-
sulfonic acid mono-hydrate (0.2 g) were added to the flask. A
DeaneStark trap and a condenser were attached, and the reaction
was protected by a nitrogen bubbler. The mixture was heated to
reflux for 4.5 h, and 0.7 mL of water was separated (theory is
0.54 mL). Most of the water came off in the first hour. When the
production of water ceased, the mixture was cooled to room
temperature and transferred to a 250 mL separatory funnel. The
toluene solution was washed with 5% sodium hydroxide solution
(2 ꢂ 5 mL) and 10 mL of brine. It was then dried over MgSO4 and
filtered, and the solvent was removed under reduced pressure. The
product slowly crystallized while under vacuum. The yield of crude
product was essentially quantitative. A small sample was recrys-
tallized from methanol to give crystals melting at 62e64 ꢁC. FT-IR
(film, warm): 3404(w), 3253(vs), 2951(m), 2892(m), 2119(vs),
Propiolic acid (70.05 g, 1.0 mol) was weighed into a dry 500 mL
round-bottom flask equipped with a magnetic stirring bar. Toluene
(250 mL), trimethylolpropane (40.25 g, 0.30 mol) and p-toluene-
sulfonic acid mono-hydrate (3.0 g) were added to the flask. A
DeaneStark trap and a condenser were attached, and the reaction
was protected by a nitrogen bubbler. The mixture was heated to
reflux for 4.5 h and 21 mL water was separated (theory is 16.2 mL).
Most of the water came off in the first hour. The mixture was cooled
to room temperature and transferred to a 1 L separatory funnel.
Ethyl acetate (250 mL) was added to keep the product in solution.
The toluene-ethyl acetate solution was washed with 5% sodium
hydroxide solution (2 ꢂ100 mL) and 100 mL of brine solution. It was
then dried over MgSO4 and filtered, and the solvents were removed
under reduced pressure. The product slowlycrystallized with a yield
of 85.72 g (0.296 mol, 98.5%), and a melting point of 71e73 ꢁC. The
product contained traces of toluene, which were removed by
melting the product and casting it into teflon pans as thin sheets and
drying the sheets in a vacuum oven. An analytical sample was
obtained by recrystallization from acetone, and it had a melting
point of 80e82 ꢁC. FT-IR (film, warm): 3283(m), 2973(w), 2121(s),

I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
2551
1717(vs),1466(m),1387(w),1219(vs), 990(m), 752(m) cmꢀ1.1H NMR
(DMSO-d6):
¼ 0.83 (t, J ¼ 7.5 Hz, 3H, eCH3), 1.42 (q, J ¼ 7.5 Hz, 2Η,
CH2eCH3), 4.12 (s, 6Η, CH2eO), 4.58 (s, 3H, hCeH) ppm. 13C NMR
(DMSO-d6):
(CeCh), 80.08 (hCeH), and 152.39 (C]O) ppm. Anal. calcd. for:
C15H14O6: C, 62.07; H, 4.86; Found: C, 62.27; H, 4.84%.
Sodium hydroxide (12 g, 0.30 mol), water (1.0 mL), and TBAB
(0.80 g) were added, and the temperature of the solution was
adjusted to 25 ꢁC. The reaction was protected by a nitrogen bubbler.
Propargyl bromide (45.0 g of 80 wt% solution in toluene, 0.30 mol)
was added drop-wise over the course of 1 h, and then the
temperature was slowly raised to 60 ꢁC over the course of 1 h.
During this time the reaction became homogeneous. The reaction
mixture was stirred at 60 ꢁC overnight. It was then cooled and
diluted with 100 mL of methylene chloride and filtered to remove
salts. The salts were extracted with two 100 mL portions of
methylene chloride, and the extracts were filtered and combined
with the main reaction. The crude product was observed to be
contaminated with some very fine salts that went through the
filter. The solvents were stripped, and the crude product
(yield ¼ 15.01 g) was analyzed by 1H NMR. Except for small
amounts of toluene and propargyl bromide, the product was quite
pure. The remaining toluene and propargyl bromide were removed
under high vacuum, and the product was distilled using a Kugelrohr
apparatus (200 ꢁC @ 0.40 Torr), to yield 10.5 g (42.3 mmol, 84.6%) of
purified product. NMR analysis indicated that the purified product
was 90% tri- and 10% dipropargylated. 1H and 13C-NMR chemical
shifts were in excellent agreement with reported values [18]. FT-IR
(film): 3294(vs), 2880e3000(b, vs), 2117(m), 1721(w), 1475(s),
d
d
¼ 7.51 (eCH3), 22.87 (Ce(CH2)4), 65.62 (eCH2), 74.76
2.4.4. Synthesis of 1,6-hexane dipropiolate (HDP)
1,6-Hexane dipropiolate was prepared by the procedure of Wang
et al. [15]. The melting point was 44e45 ꢁC (lit. 42e45 ꢁC [15]).
2.4.5. Synthesis of dipropargyl adipate (DPA)
Propargyl alcohol (12.0 g, 0.214 mol), pyridine (16.0 g,
0.202 mol), and dry dichloroethane (100 mL) were charged to
a 500 mL round-bottom flask. The resulting solution was main-
tained at 10 ꢁC by means of a salt/ice/water bath. A solution of
adipoyl chloride (18.3 g, 0.10 mol) in 100 mL of dry dichloroethane
was added dropwise over 15 min. The solution was stirred over-
night at room temperature and then transferred to a 1 L separatory
funnel with the aid of a small amount of dichloroethane. The
dichloroethane solution was washed with water (2 ꢂ 100 mL), 5%
HCl (100 mL), and saturated sodium bicarbonate solution (100 mL).
The dichloroethane solution was dried over MgSO4 and filtered, and
the dichloroethane was removed under reduced pressure. The
product was purified by vacuum distillation. The main fraction was
collected at 0.7 Torr, boiling from 130 to 132 ꢁC. The reported
boiling point is 142e145 ꢁC @ 4 Torr [16]. FT-IR (film): 3290(vs),
2947(s), 2874(m), 2129(m), 1740(vs), 1436(s), 1418(sh), 1382(s),
1350(sh), 1311(sh), 1235(s), 1164(s), 1077(m), 1027(m), 996(m),
959(m), 935(m), 680(s), 647(s) cmꢀ1. 1H NMR (300 MHz, CD3CN):
1442(m), 1360(s), 1265(m), 1084(s), 1025(m), 975(m), 912(m) cmꢀ1
.
1H NMR (CDCl3):
d
¼ 0.88 (t, 3H, J ¼ 4.2 Hz, eCH3), 1.42 (q, 2H,
J ¼ 4.2 Hz, CH2eCH3), 2.42 (t, 3H, J ¼ 2.4 Hz, hCeH), 3.40 (s, 6H,
CH2eO), 4.14 (d, 6H, J ¼ 2.4 Hz, OeCH2eCh) ppm. 13C NMR (CDCl3):
d
¼ 7.42 (eCH3), 22.62 (CH2eCH3), 42.66 (Ce(CH2)4), 58.48
(OeCH2eCh), 70.12 (OeCH2eC), 74.19 (HeCh), 80.09 (CeCh)
ppm.
d
¼ 1.72 (m, 4H, CH2), 2.39 (m, 4H, CH2eCO), 2.50 (t, 2H, hCeH),
4.68 (d, 4H, OeCH2) ppm. 13C NMR (75 MHz, CD3CN):
d
¼ 24.04
2.4.8. Synthesis of oligo(ethylene glycol) dipropargyl ethers
(CH2eCH2-CO), 33.42 (CH2eCO), 51.82 (OeCH2), 74.90 (HeCh),
77.63 (CH2eCh), 172.30 (C]O) ppm.
Two compounds, tetra(ethylene glycol) dipropargyl ether
(TEGDPE) and poly(ethylene glycol)-200 dipropargyl ether
(PEGDPE), were made using a modification of the procedure of
Mekni and Baklouti [17]. The synthesis of tetra(ethylene glycol)
dipropargyl ether is representative: tetraethylene glycol (19.4 g,
0.10 mol) was weighed into a dry 250 mL three-neck round-bottom
flask equipped with a mechanical stirrer, septum, and a condenser.
Sodium hydroxide (16.0 g, 0.40 mol), water (2 mL), and TBAB
(0.40 g) were added to the flask, and the reaction was protected by
a nitrogen bubbler. The mixture was heated to 45 ꢁC, and propargyl
bromide (60 g of 80 wt% solution in toluene, 0.40 mol) was added
drop-wise over 1 h. Following addition, the reaction temperature
was raised to 60 ꢁC and stirred for 6 h. The mixture was cooled and
diluted with 100 mL of methylene chloride. The mixture was filtered
to remove salts; the salts were extracted with methylene chloride
(2 ꢂ 50 mL), and the combined solutions were stripped of solvents.
FT-IR spectroscopy showed that no hydroxyl remained, indicating
reaction completion. The yield of crude product was 32.28 g. The
remaining toluene and propargyl bromide were removed under
high vacuum, and the resulting product was Kugelrohr distilled at
200 ꢁC @ 0.50 Torr in two batches. The reported boiling point was
110 ꢁC @ 0.1 Torr [17]. The yield of purified product was 26.46 g
(98 mmol, 98%). FT-IR (film): 3251(m), 2870(s), 2113(w), 1458(w),
1350(m), 1270(w), 1250(w), 1102(vs), 1033(m), 919(w), 842(w)
2.4.6. Synthesis of tripropargyl trimesate (TPTM)
Propargyl alcohol (18.0 g, 0.32 mol) and pyridine (24 g, 0.30 mol)
were dissolved in 150 mL of anhydrous dichloroethane in a 500 mL
round-bottom flask equipped with a magnetic stirring bar. This
reaction was maintained at 10 ꢁC by means of a salt/ice/water bath.
A solution of trimesoyl chloride (26.6 g, 0.10 mol) in 100 mL of
anhydrous dichloroethane was added drop-wise over 20 min. The
solution was stirred overnight at room temperature. The next day
the solution was transferred to a 1 L separatory funnel with the aid
of a small amount of dichloroethane. The dichloroethane solution
was washed with water (2 ꢂ 100 mL), 5% HCl (100 mL), and 5%
sodium hydroxide solution (100 mL). The dichloroethane solution
was then dried over MgSO4 and filtered, and the dichloroethane
was removed under reduced pressure. The product was purified by
recrystallization from acetone. The yield was 29.16 g (0.09 mol,
90%). The melting point was 82 ꢁC. FT-IR (film): 3296(vs), 3288(s),
3271(m), 3090(w), 2125(w), 1731(vs), 1609(w), 1445(s), 1433(w),
1375(s), 1364(m), 1329(s), 1277(sh), 1229(vs), 1151(m), 1103(m),
1022(s), 1001(s), 959(m), 930(m), 732(vs), 700(s), 684(s), 649(m)
cm-1. cmꢀ1. 1H NMR (DMSO-d6)
d
¼ 3.65 (t, J ¼ 2.4 Hz, 3H, hCeH),
5.03 (d, J ¼ 2.4 Hz, 6H, OeCH2), 8.62 (s, 3H, AreCH) ppm. 13C NMR
(DMSO-d6)
d
¼ 51.84 (OeCH2), 74.90 (HeCh), 77.64 (CH2eCh),
cmꢀ1
OeCH2eCh), 3.60 (s, 8H), 3.40 (s, 8H), 2.74 (t, J ¼ 2.4 Hz, 2H, HeCh)
ppm. 13C NMR (75 MHz, CD3CN):
.
1H NMR (300 MHz, CD3CN):
d
¼ 4.16 (d, J ¼ 2.4 Hz, 4H,
124.05 (AreCH), 133.44 (AreCe), 172.33 (C]O) ppm. HRMS (EI);
C18H12O6 [M]þ: calcd. 324.0634; found 324.0641.
d
¼ 57.82, 69.05, 70.02, 70.02,
70.25 (OeCH2eCh), 74.94 (HeCh), 80.08 (CeCh) ppm.
2.4.7. Synthesis of trimethylolpropane tripropargyl ether (TMPTPE)
This compound was synthesized using a modification of the
procedure of Mekni and Baklouti [17]. Trimethylolpropane (6.71 g,
0.050 mol) was weighed into a dry 100 mL round-bottom flask
equipped with a mechanical stirrer, condenser, and septum.
2.4.9. Synthesis of N,N,N0,N0-tetrapropargyl-2,20-(ethylenedioxy)
bis(ethylamine) (TPEDEA)
To a 3-neck 500 mL round-bottom flask, equipped with
a
mechanical stirrer, were charged 2,20-(ethylenedioxy)

2552
I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
bis(ethylamine) (14.82 g, 0.10 mol), water (40 mL), sodium
hydroxide (17.6 g, 0.44 mol), dichloroethane (80 mL) and TBAB
(0.322 g). A reflux condenser equipped with a nitrogen bubbler
was attached, and a room temperature bath was placed around
the flask. Propargyl bromide (65.46 g of 80 wt% solution in
toluene, 0.44 mol) was added drop-wise over the course of 1 h,
and the mixture was stirred overnight. The mixture was trans-
ferred to a 1 L separatory funnel with the aid of 400 mL of
methylene chloride. The pH of the aqueous layer was checked and
found to be slightly basic. The aqueous layer was separated, and
the organic layer was dried over magnesium sulfate, filtered, and
then solvent-stripped to yield the crude product (32.5 g). The
crude product was chromatographed on 250 g of silica using an
80/20 solution of methylene chloride/ethyl acetate as the eluent;
150 mL fractions were collected. A very dark band eluted as
fraction 3, with fractions 5e8 containing the desired compound.
These fractions were combined and solvent-stripped to give 17 g
(57 mmol, 57%) of a slightly yellow-brown product. Thin layer
chromatography (TLC) showed that fractions 9 and 10 might also
contain the desired compound. They were combined and solvent-
stripped, yielding an additional 3.1 g (10 mmol, 10%). NMR analysis
showed both products were the desired tetra-functional
compound. FT-IR (film) ¼ 3300 (vs), 2924(m), 2821(m), 2103(w),
1672(w), 1642(w), 1440(m), 1352(m), 1328(m), 1124(m), 1101(m),
(HeCh), 78.66 (CeCh) ppm. HRMS (CI): C14H17N2 [MH]þ: calc.
213.1392; found 213.1383 amu.
2.5. Neat linear polymerization
Neat, linear polymerizations were conducted by combining
difunctional alkynes DEGDP and DPA with DAHP-BPA such that the
molar ratio of azide:alkyne was 1:1. A high density polyethylene
scintillation vial (20 mL) served as the polymerization reactor. A
mixture of two alkyne monomers was used to create three different
curing conditions: A) 50/50 DEGDP/DPA, cured at 100 ꢁC for 2 h, B)
50/50 DEGDP/DPA, cured at RT for 12 h, then at 100 ꢁC for 2 h, and
C) 65/35 DEGDP/DPA, cured at 100 ꢁC for 2 h. All systems were
mixed for 5 min, and then 0.5 g of the mixture was charged to a vial.
The polymerization conditions were chosen to limit molecular
weight and maintain solubility of the product in common solvents.
2.6. Differential scanning calorimetry sample preparation
Differential scanning calorimetry (DSC) samples were prepared
by combining azide-modified epoxy resin DAHP-BPA and poly-
alkyne crosslinker(s) such that the ratio of azide to alkyne func-
tionality was 1:1 and the total reaction mass was approximately
0.3 g. When a combination of a solid and a liquid alkyne
component was used, prior to combining with DAHP-BPA, the
solid alkyne component was dissolved in the liquid alkyne
component by warming the mixture. Alkyne and azide compo-
nents were then combined and mixed for 5 min, after which
5e12 mg of the mixture was charged to a standard, capped
aluminum DSC pan. Once the pans were prepared either a DSC
run was immediately conducted or the sample was placed in the
freezer until DSC was conducted. Those samples placed in the
freezer were run within 12 h.
1055(w), 977(w), 902(m), 832(w), 806(w), 636(s) cmꢀ1
.
1H NMR
(CDCl3):
d
¼ 2.45 (t, J ¼ 2.4 Hz, 4H, HeCh), 2.68 (t, J ¼ 7 Hz,
NCH2eCH2O), 3.45 (d, J ¼ 2.4 Hz, 8H, NeCH2eCh), 3.55 (s, 4H,
OCH2CH2O), 3.60 (t, J ¼ 7 Hz, 4H, NCH2eCH2O) ppm. 13C NMR
(CD3CN):
d
¼ 42.46 (NeCH2eCh), 52.40 (NeCH2eCH2), 69.51
(OeCH2CH2), 70.32 (OeCH2CH2), 73.73 (HeCh), 79.41 (CeCh)
ppm. HRMS (NCI): C18H25N2O2 [MH]þ: calc. 301.1916; found
301.1904 amu.
2.4.10. Synthesis of N,N,N0,N0-tetrapropargylethylenediamine
(TPEDA)
The ASTM E698 method [20] was applied to DAHP-BPA resin/
alkyne crosslinker formulations derived from all eleven alkyne
crosslinker types shown in Fig. 2. Each sample was equilibrated in
the DSC for 2 min at 0 ꢁC and then heated to 300 ꢁC at a pre-
This compound was prepared using a modification of the
procedure of Yamada and Aoki [19]. To a 3-neck 500 mL round-
bottom flask, equipped with a mechanical stirrer, were charged
ethylene diamine (6.0 g, 0.10 mol), water (40 mL), sodium
hydroxide (17.6 g, 0.44 mol), dichloroethane (80 mL), and TBAB
(0.322 g). A reflux condenser equipped with a nitrogen bubbler was
attached, and a room temperature bath was placed around the
flask. Propargyl bromide (65.46 g of 80 wt% solution in toluene,
0.44 mol) was then added drop-wise over 1 h, and the reaction was
stirred for 22 h at room temperature. The mixture was then
transferred to a 500 mL separatory funnel with the aid of 100 mL of
dichloroethane and 100 mL of water. The pH of the aqueous layer
was checked to ensure it was basic. A very stable emulsion formed;
so the mixture was filtered through a fiberglass filter to remove any
polymeric impurities before returning it to the separatory funnel.
The organic layer was separated, dried over magnesium sulfate,
filtered, and then solvent-stripped to give the crude product
(28.2 g). TLC showed at least three products. The crude product was
chromatographed on 200 g of silica using methylene chloride as the
eluent, collecting 150 mL fractions. A very dark band eluted at
fractions 3 and 4 and contained a solid material, perhaps a quater-
nary salt. Fractions 6e11 contained the desired crude compound
(8.5 g). Fractions 6e11 were combined, solvent-stripped, and
Kugelrohr distilled (175 ꢁC @ 0.5 Torr) to give 5.9 g (28 mmol, 28%)
of slightly yellow product. FT-IR (film) ¼ 3292(vs), 2900(m),
2870(s), 2821(m), 2375(w), 2344(w), 2103(w), 1734(w), 1442(m),
1352(m), 1328(m), 1247(m), 1117(m), 902(m), 839(w), 802(w),
determined heating rate, b. Each formulation was analyzed at five
heating rates (
b
¼ 2, 5,10,15, and 20 ꢁC/min), and each formulation/
heating rate combination was analyzed in duplicate. A follow-up
dynamic scan was conducted at 10 ꢁC/min to ensure that there
was no residual exotherm and that the curing reaction had gone to
completion. The maximum exotherm temperature, Tmax, was
determined for each run from curves similar to those in Fig. 3.
ꢀ1
Linear plots of ln (
b
) versus (T ) were then constructed and used
max
to calculate Ea from the slope.
2.7. Rheological sample preparation
Viscosity of developing networks was measured using an ARES
(TA Instruments) Rheometer. All tests were conducted at room
temperature and at a frequency of 1 Hz. Uncured resin systems
were mixed according to the above DSC specifications and carefully
poured onto parallel 25 mm aluminum plates.
3. Results and discussion
3.1. Synthesis of di(3-azido-2-hydroxypropyl) ether of bisphenol-A
(DAHP-BPA)
Sharpless and coworkers first reported the synthesis of DAHP-
BPA in 2007 [10]. The structure of their product was identical to
that shown in Fig. 1 (attack of azide ion on the less substituted
carbon of the epoxide ring). They reported the use of NaN3/LiClO4 in
acetonitrile at 90 ꢁC overnight, followed by the addition of water,
635(s) cmꢀ1. 1H NMR (CDCl3):
2.68 (s, 4H, CH2eCH2), 3.45 (d, J ¼ 2.4 Hz, 8H, NeCH2eCh) ppm.
13C NMR (CDCl3):
¼ 42.47 (CH2eCH2), 49.89 (NeCH2eCh), 73.19
d
¼ 2.21 (t, J ¼ 2.4 Hz, 4H, HeCh),
d

I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
2553
3.0
2.5
2.0
1.5
1.0
0.5
0.0
first class of alkynes species to be synthesized, followed by the
propargyl esters, and then the propargyl ethers and amines. As
a general summary, the propiolate esters were readily synthesized
by condensing the appropriate polyol with a slight excess of pro-
piolic acid in refluxing toluene and in the presence of a catalytic
amount of p-toluenesulfonic acid. The propargyl esters were
synthesized by the reaction of the appropriate acid chloride with
a slight excess of propargyl alcohol in toluene or dichloroethane
using one equivalent of pyridine as an acid acceptor. The propargyl
ethers were synthesized by reaction of the corresponding polyol
with a large (2X) excess of propargyl bromide under forcing phase-
transfer conditions. And finally, the propargyl amines were
synthesized by reaction of the corresponding amine with only
a slight excess of propargyl bromide to avoid formation of quater-
nary salts.
Crosslinkers that are solid at room temperature proved trou-
blesome to mix with the azide-modified epoxy resin, yielding
formulated resins with higher that optimal viscosity for fabrication
of composite materials. Furthermore, the susceptibility to hydro-
lytic degradation of the ester linkage found in the propiolates and
propargyl esters was of some concern. All of the propargyl ether
and propargyl amine compounds have proven to be liquids, cir-
cumventing both issues. The propargyl amines have the added
advantage of high functionality and low equivalent weight, thus
offering a low-viscosity curative with the potential for higher
crosslink density in the final thermoset, and making them excellent
candidates for scale-up and fabrication of both pure resin and
composite samples. Among the amines, TPA has the additional
advantage of commercial availability.
2o C/min
5o C/min)
10o C/min
15o C/min
20o C/min
0
100
200
300
Temperture (°C)
Fig. 3. Dynamic DSC curves for the DAHP-BPA/TEGDPE system at heating rates of 2, 5,
10, 15, and 20 ꢁC (exothermal direction up). The exotherm temperature maxima, Tmax
and H were determined.
,
D
and extraction of the product into ethyl acetate. The extraction was
washed with water and the product isolated by the removal of the
solvent. We used a modification of this procedure in which the
acetonitrile was replaced with a mixture of 2-ethoxyethanol (Cel-
losolve) and water (3:1, v/v), with NH4Cl charged to the reaction at
the beginning to serve as a proton source and to help catalyze the
epoxy ring opening [21]. The pH was monitored over the course of
the reaction to ensure it remained in a neutral range, ca. 6e9;
typically, no adjustment was necessary. When outside this pH
range the reaction has been reported to become sluggish [22]. Use
of 2-ethoxyethanol enabled a higher reaction temperature (107 ꢁC),
and this modification, coupled with the addition of ammonium
chloride, lead to an increase in reaction rate. These conditions
produced complete reaction in about 4 h, yielding a single product
with 1H and 13C-NMR chemical shifts identical to those reported by
Sharpless et al. [10]. This procedure was also successfully scaled to
produce 2 kg of DAHP-BPA in a single reaction.
More recently Kessler and coworkers reported the synthesis of
DAHP-BPA by reaction of EPON 828 with NaN3/NH4Cl for 12 h in
refluxing methanol [11]. However, these authors reported that the
azide group was substituted at C2 of the glycidyl group (attack of
azide ion on the more substituted carbon of the epoxide ring) and
that two diasteromers were obtained (major and minor). Whether
Kessler et al. obtained a different regioisomer is questionable since
the chemical shifts they reported for their major isomer are iden-
tical to those observed by us and reported by Sharpless et al. Both
DAHP-BPA and the EPON reactant should consist of two diaste-
reomers, i.e., a pair of enantiomers and a meso compound, but the
ratio should be 50:50. We observed only one set of peaks in the 13C-
NMR spectrum, suggesting that the chemical shift differences
between diastereomers are vanishingly small.
3.3. Linear polymerizations
The reaction between azide-modified epoxy resin and the
alkyne curatives was initially investigated using difunctional
alkynes, which yielded linear polymer products. In this initial study,
a mildly elevated curing temperature of 100 ꢁC was employed.
Although room temperature-cure remained the ultimate goal of the
project, here we wanted to ensure that the monomers would in fact
polymerize, and the use of elevated cure temperatures was the
easiest pathway to that information. Mixtures of two alkyne
monomers, DEGDP and DPA, were used to create three different
resin systems, all at a theoretical 1:1 azide:alkyne stoichiometric
ratio. Molecular weight was limited by controlling the time held at
100 ꢁC, i.e., by limiting conversion, in order to maintain solubility in
common GPC solvents. Fig. 4 shows gel permeation chromatograms
of the starting epoxy resin, the azide-modified resin, and the chain-
extended polymers. The chromatograms show that the addition of
azide to the epoxy resin results in a significant increase in hydro-
dynamic volume, while the peak shape remains the same, indi-
cating the lack of degradation, fractionation, or chain coupling/
extension during reaction with the azide. Upon polymerization the
peaks shift to shorter elution times, and broaden, clearly indicating
an increase in molecular weight. Table 1 lists the quantitative
results obtained from GPC-MALLS. Average degree of polymeriza-
tion, Xn, was calculated as number average molecular weight, Mn,
obtained from MALLS, divided by the average molecular weight of
the structural units in the polymer (assuming that the propiolate
and propargyl ester monomers entered the polymer in proportion
to their relative abundances in the feed). The greatest Xn was
obtained for the system containing the higher proportion (65 mol%)
of the more reactive propiolate monomer. The two systems
formulated at equimolar ratio of propiolate/propargyl produced the
same Xn, but the sample immediately reacted at 100 ꢁC produced
a narrower distribution of degrees of polymerization. Triazole
formation was confirmed by 1H NMR analysis; Fig. 5 shows
3.2. Synthesis of alkyne crosslinkers
Two basic building blocks were employed to create polyfunc-
tional alkyne crosslinkers: propioloyl and propargyl. A number of
the alkyne species have been previously described in literature, and
were synthesized by applying modifications to these methods
[13,15,17,19]. Two alkyne compounds are novel and previously
undescribed: TPEDEA and TMBTP. The propiolate esters were the

2554
I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
and 8). A total of 7 systems were investigated, containing relative
ratios of DPA/HDP of 100/0, 80/20, 60/40, 50/50, 40/60, 20/80, and
0/100, respectively. All systems were formulated so that the molar
ratio of total azide to total alkyne was 1:1. The DSC peak exotherm
shifted from ca. 138 ꢁC for the system containing 100% DPA to ca.
89 ꢁC for the system containing 100% HDP; all systems containing
a mixture of the two alkyne species displayed a bimodal thermo-
gram. In addition to the two exotherms observed at ca. 89 ꢁC and
138 ꢁC, a third exotherm was observed ca. 250 ꢁC for those systems
incorporating the HDP alkyne. The size of the exotherm at 250 ꢁC
increased as the HDA content increased, and was not observed in
the system containing 100% DPA, indicating that the HDA cures by
a more complex mechanism. Overall, we observed that a very high
rate of heat release can result in uncontrollable reactions, as seen in
some propiolate systems. The bimodal curing profiles displayed by
the DPA/HDP mixtures offer a potential solution to this problem by
releasing the heat in two fractions.
Epoxy Resin
(EPON 825)
DAHP-BPA
DEGDP/DPA (50/50) 100oC, 2 h
DEGDP/DPA (50/50) RT 12 h; 100oC, 2 h
DEGDP/DPA (65/35) 100oC, 2 h
Chain-extended Polymers
10
15
20
Elution Volume (mL)
The same DPA/HDP blends were reacted with DAHP-BPA
isothermally at room temperature and viscosity was monitored as
a function of time (Fig. 8). For the system containing 100% DPA,
viscosity was found to remain relatively constant over the course of
the experiment, with no appreciable increase. This indicated that
resin compositions consisting of DAHP-BPA and propargyl-type
alkynes are shelf-stable for long periods (days) in the absence of
catalyst. In comparison, the system utilizing 100% HDP was found to
vitrify after ca. 600 min. The rheological profiles of those systems
containing a mixture of DPA and HDP fell between these two
extremes, predictably in accordance with the propiolate/propargyl
ratio. Triazole formation at room temperature was confirmed by 1H
NMR analysis for the systems composed of 100% HDP and the 20/80
DPA/HDP system. Samples were mixed according to the standard
procedure and were allowed to react at room temperature for 24 h.
Fig. 9 shows 1H NMR spectra for the resulting polytriazoles, where
the chemical shifts associated with the propiolate triazole proton,
for the 1,4- and 1,5-regioisomer, were found at 8.69 and 8.27 ppm,
respectively. For the 20/80 DPA/HDP system, the spectrum was
dominated by the propiolate triazole protons, and no peaks asso-
ciated with the propargyl triazole were detected, which are pre-
dicted to occur at 8.04 and 7.68 ppm. This was expected since the
propiolate is much more reactive than the propargyl species.
We concluded our studies of relative reactivity by determining
the averaged activation energy, Ea, of our polyalkyne crosslinkers in
the absence of catalyst, using the ASTM E698 method [20] (Table 2).
This method is based on the variable program rate method of
Ozawa, which requires three or more experiments at different
heating rates, typically between 1 and 20 ꢁC/min. The Ozawa
approach assumes Arrhenius behavior, and that the extent of
reaction at the peak exotherm is constant and independent of
heating rate. In combination with Doyle’s approximation [23], this
approach yields the following equation:
Fig. 4. GPC chromatograms (differential refractometer) of epoxy resin (EPON 825),
DAHP-BPA, and three linear polytriazoles obtained by the reaction of DAHP-BPA with
a mixture of DEGDP and DPA.
a representative spectrum. The proton resonances associated with
the triazole rings are clearly visible at 8.69 and 8.27 ppm for the
propiolate-based triazoles, and at 8.04 and 7.68 ppm for the
propargyl-based triazoles. The triazole protons of the 1,4-
regioisomers appear downfield relative to those of the 1,5-
regioisomers. For both the propiolate- and propargyl-based tri-
azoles, the 1,4-regioisomer was the major product.
In Fig. 6, dynamic DSC scans, conducted at a scan rate of 10 ꢁC/
min, compare the curing rate of azide-modified epoxy resin with
a representative alkyne of each type: TMBTP (propiolate ester
type), TPTM (propargyl ester type), TMPTPE (propargyl ether type),
and TPA (propargyl amine type), as well as a commercial aerospace
resin system composed of EPON 825 and an amine crosslinker, 4,40-
diaminodiphenyl sulfone (4,40-DDS). The significantly higher
reactivity of the propiolate type, as compared to the propargyl and
the EPON 825/4,40-DDS system, is evident from the earlier (lower
temperature) onset of the exothermic curing reaction. The initial
propiolate ester exotherm occurred at a relatively low temperature
of ca. 18 ꢁC. In comparison, the three propargyl-based crosslinkers
showed much lower reactivity and did not begin to exotherm until
75 ꢁC, while the EPON 825/4,40-DDS system did not begin to exo-
therm until 150 ꢁC. The low temperature reaction onset of a pro-
piolate could be used to kick start the reaction of a propargyl
species when the two monomers are used in tandem. For all azide/
alkyne polymerizations, the total heat of reaction was approxi-
mately the same, ca. 700 J/g.
A systematic study was conducted to compare formulations
containing varying amounts of propiolate and propargyl species,
specifically HDP and DPA. These two alkynes were chosen because
they have equivalent molecular weights. The samples were studied
by dynamic DSC and through rheological measurements (Figs. 7
1:052 Ea
lnð
b
Þ ¼ A0 ꢀ
RTmax
Table 1
GPC results comparing linear polymerization to the modified and unmodified resins.
a
Mna (g/mol)
Mw (g/mol)
PDI
Theor. Mn (g/mol)
Xn
EPON 825
DAHP-BPA
340
450
350
460
e
1.03
1.02
e
340.41
426.47
210.18
222.24
e
e
e
e
e
Diethylene glycol dipropiolate (DEGDP)
Dipropargyl adipate (DPA)
DEGDP/DPA (50/50) 100 ꢁC, 2 h
DEGDP/DPA (50/50), RT 12 h, 100 ꢁC 2 h
DEGDP/DPA (65/35) 100 ꢁC 2 h
e
e
e
e
2830
2900
3720
4390
6100
6730
1.55
2.11
1.81
8.8
9.0
11.6
e
e
a
GPC-MALLS.

I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
2555
N
N
N
N
N
N
N
N
N
N
N
N
a
a'
b
b'
C
H
H
H
C
H
H
H
H
O
O
H
a
b
a'
b'
8.8
8.6
8.4
8.2
ppm
8.0
7.8
7.6
c,d
e
10
9
8
7
6
5
4
3
2
1
0
-1
ppm
1
Fig. 5. H NMR spectrum (d6-DMSO, 23 ꢁC) of a representative linear polytriazole from bulk polymerization of DAHP-BPA with 50/50 DEGDP/DPA conducted at 100 ꢁC for 2 h. The
triazole protons associated with the 1,4- and the 1,5-regioisomer, for the propiolate and propargyl linkages, have been denoted (a) and (a0) and (b) and (b0), respectively.
where
b
is the scan rate (K/min), A0 is a constant, Ea is activation
rates of 2, 5, 10, 15, and 20 ꢁC/min are shown in Fig. 3, and it can be
energy (J/mol), Tmax is the temperature at maximum exotherm (K),
and R is the universal gas constant (8.314 J/mol K). Activation
seen that peak temperature, Tmax, shifts to higher temperatures as
the heating rate, b, is increased. For all samples the linear fit coef-
energy Ea can be calculated from the slope of linear plots of ln(
b
)
ficient of determination (R2) was greater than 0.99. Furthermore,
from the follow-up scan, all systems had gone to completion, as
evidenced by the absence of any residual exotherm.
versus 1/Tmax. Typical dynamic DSC curves at a range of heating
3.0
2.5
2.0
1.5
1.0
0.5
0.0
147.6
4.5
131.2
83.0
0 DPA/100 HDP
EPON 825/4,4-DDS
TMPTPE
TMBTP
TPTM
TPA
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
20 DPA/80 HDP
40 DPA/60 HDP
50 DPA/50 HDP
60 DPA/40 HDP
80 DPA/20 HDP
100 DPA/0 HDP
142.0
227.5
0
100
200
300
400
Temperature (oC)
0
50
100
150
200
250
300
Temperature (°C)
Fig. 6. Dynamic DSC scans obtained at 10 ꢁC/min (exothermal direction up),
comparing the curing rates of DAHP-BPA with a representative from each class of
alkyne species: TMBTP (propiolate ester), TPTM (propargyl ester), TMPTPE (propargyl
ether), and TPA (propargyl amine).
Fig. 7. Dynamic DSC scans obtained at 10 ꢁC/min (exothermal direction up), comparing
the curing rates of DAHP-BPA with various blends of DPA and HDP, ranging from 100%
DPA to 100% HDP.

2556
I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
determined DH were within the predicted DH range for the Huisgen
1E8
1E7
DPA/HDP
1,3-dipolar cycloaddition, ꢀ210 and ꢀ270 kJ/mol. These numbers
indicate that the Huisgen reaction is extremely exothermic among
organic reactions generally.
0/100 (mol/mol)
20 / 80
1000000
100000
10000
1000
100
The highly exothermic nature of the Huisgen cycloaddition
represents an opportunity in terms of out-of-autoclave curing but
also a significant challenge. Too rapid heat release during cure can
lead to burning and decomposition of the network. The reported
heat capacity of epoxy resins is in the range 1.3e2.1 J/g-K [24].
Assuming that the heat capacity of our resin is similar (for example,
1.6 J/g-K), then the DAHP-BPA/TPA system at 1:1 stoichiometry
would have a heat capacity of about 411 J/mol-K, and therefore,
a reaction exotherm of about 130 kJ/mol would be sufficient to
bring the system from room temperature to the decomposition
temperature of 340 ꢁC [25] under adiabatic conditions. This clearly
shows that the total heat of reaction from the Huisgen cycloaddi-
tion is more than sufficient to cause severe decomposition of the
reacting mass, at a typical equivalent weight used herein. There-
fore, the rate of reaction must be carefully balanced with the rate of
heat loss, which is dependent on section thickness of the reacting
mass, heat conductivity of the surroundings, etc. Fortunately, rate
of reaction can be controlled by alkyne choice and catalyst, and in
fact, the bimodal curing profiles displayed by the propargyl/pro-
piolate mixtures offer the flexibility of releasing the heat in
multiple stages.
40 / 60
50 / 50
60 / 40
80 / 20
100 / 0
10
1
0
200
400
600
800
1000
1200
1400
1600
Time (min)
Fig. 8. Viscosity vs. time for mixtures of DAHP-BPA with various DPA/HDP blends,
ranging from 100% DPA content to 100% HDP. Data were collected at 1 Hz and room
temperature.
As expected, a propiolate compound, TMBTP, was determined to
have the lowest averaged Ea (69.2 kJ/mol) of all crosslinkers studied.
Furthermore all propiolate-based crosslinkers displayed approxi-
mately the same averaged Ea, with values ranging from 69.2 to
73.6 kJ/mol. In contrast, TPTM crosslinker, a propargyl type, dis-
played the highest averaged Ea, 86.4 kJ/mol. All of the propargyl-
type structures displayed relatively high values ranging from 82.3
to 86.4 kJ/mol. Although a distinct divide was evident between the
propiolate and propargyl crosslinkers, no significant difference in
Ea was observed among the propargyl ether, propargyl amine, and
propargyl ester subdivisions.
The normalized enthalpy of reaction,
for all systems by integration of the dynamic DSC peaks for each
scan rate. The
H values ranged from ꢀ200 to ꢀ264 kJ/mol.
Determination of H suffers from higher error, due in part to
baseline selection and baseline irregularities, as compared to Tmax
which is not significantly affected by baseline shifts. In general the
3.3.1. Crosslink density
In addition to reactivity, crosslink density and resulting physical
properties can be manipulated though selective blending of
different alkyne types; in particular, glass transition temperature of
the resulting network can be strongly affected by incorporation of
linear polyethylene glycol based alkyne species. Fig. 10 displays
plots of tan
d versus temperature, obtained by DMA, of DAHP-BPA
cured with various mixtures of PEGDP and TMPTP. A total of 5
systems were investigated, containing relative ratios of PEGDP/
TMPTP of 100/0, 80/20, 70/30, 60/40, and 50/50. The systems were
stirred for 5 min, molded, and allowed to react at 70 ꢁC for 2 h. All
systems were formulated so that the molar ratio of total azide to
total alkyne was 1:1. For each curve, the peak maximum represents
the glass transition temperature of the cured resin at that particular
DH, was also determined
D
D
N
N
CH
CH
2N
N
2N
C C
N
C C
a
a'
H
C
C
H
O
O
A
B
a
a
a'
a'
10
9
8
7
6
5
4
3
2
1
0
-1
ppm
1
Fig. 9. H NMR spectra (d6-DMSO, 23 ꢁC) of polytriazoles from bulk polymerization of DAHP-BPA with (A) 0/100 DPA/HDP, and (B) 20/80 DPA/HDP. Polymerizations were conducted
at room temperature for 24 h with a theoretical 1:1 azide:alkyne stoichiometric ratio. The triazole protons associated with the 1,4- and the 1,5-regioisomer, for the propiolate
linkage, have been denoted (a) and (a’), respectively.

I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
2557
Table 2
based alkynes were initially synthesized. Their excessive reac-
tivity led to the synthesis of slower acting propiolate esters. The
latter compounds displayed more favorable reactivity profiles, but
they still contained undesirable ester linkages and many were solid,
leading to poor processability. The final two groups of alkynes,
propargyl ethers and propargyl amines, circumvented all issues;
they were found to possess relatively low reactivity, and they
contain no labile linkages and are liquids. The propargyl amines
showed the greatest promise for scale up and fabrication of both
filled and unfilled composite samples because, in addition to the
aforementioned advantages, they have high functionality and low
equivalent weight. TPA was particularly attractive in this work,
since it could be obtained commercially. Overall, the propiolate
species were found to have Ea in the range of 70e74 kJ/mol; while
the propargyl species were found to have Ea in the range of
82e87 kJ/mol. Though the propiolate species proved too reactive
on their own, they could be combined with the less reactive
propargyl species to achieve reactivity and properties tailored to
specific applications.
Activation energies and other kinetic parameters by the ASTM E698 method for the
reaction of DAHP-BPA with the indicated alkyne.
Alkyne
Ea (kJ/mol)
Avg
D
H (J/g)
Avg DH
(kJ/mol)
Propiolate
Propargyl
TMBTP
HDP
TEGDP
TMPTP
DPA
TMPTPE
TEGDPE
TPEDA
TPA
TPEDEA
TPTM
69.2
70.0
72.3
73.6
82.3
82.5
83.3
84.3
84.7
85.9
86.4
ꢀ660 ꢃ 60
ꢀ600 ꢃ 50
ꢀ651 ꢃ 8
ꢀ850 ꢃ 10
ꢀ780 ꢃ 20
ꢀ810 ꢃ 26
ꢀ704 ꢃ 6
ꢀ930 ꢃ 35
ꢀ960 ꢃ 43
ꢀ825 ꢃ 8
ꢀ740 ꢃ 10
ꢀ210 ꢃ 19
ꢀ200 ꢃ 17
ꢀ236 ꢃ 03
ꢀ264 ꢃ 03
ꢀ242 ꢃ 06
ꢀ248 ꢃ 08
ꢀ249 ꢃ 02
ꢀ246 ꢃ 09
ꢀ250 ꢃ 11
ꢀ238 ꢃ 02
ꢀ237 ꢃ 03
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
100/0
80/20
Similarly, we demonstrated that by mixing varying amounts of
difunctional ethylene glycol-based alkynes with tri- or tetra-
functional species, crosslinking density and resulting physical and
thermal properties of the systems can be manipulated. For
example, by incorporating 20% PEGDP into a TMPTP/DAHP-BPA
70/30
60/40
d
peak was increased by 10 ꢁC. Regardless of
50/50 (mol/mol)
PEGDP/TMPTP
system, the tan
structure, all systems studied were very exothermic,
between ꢀ200 kJ/mol and ꢀ265 kJ/mol, with the potential to allow
for higher crosslink density prior to vitrification. Through the use of
various combinations of these alkynes, a variety of systems were
obtained with diverse reactivates, and thermal and physical
properties.
Acknowledgment
The authors of this paper would like to thank The University of
Southern Mississippi Composites Center, funded by The Office of
Naval Research and Northrop Grumman, Award No. N00014-07-1-
1057, for financial support of this research.
40
60
80
100 120 140 160 180 200
Temperature (°C)
Fig. 10. Tan d vs. temperature of DAHP-BPA cured using various mixtures of PEGDP and
TMPTP. Samples were equilibrated and held for 5 min at 35 ꢁC, and then heated at
a rate of 2 ꢁC/min to 200 ꢁC.
References
[1] Le Baut N, Díaz DD, Punna S, Finn MG, Brown HR. Polymer 2007;48:239e44.
[2] Rostovtsev VV, Green LG, Fokin VV, Sharpless KB. Angew Chem Int Ed 2002;
41:2596e9.
[3] Tornoe CW, Christensen C, Meldal M. J Org Chem 2002;67:3057e64.
[4] Hein JE, Fokin VV. Chem Soc Rev 2010;39:1302e15.
[5] Tomé AC. Five membered hetarenes with three or more heteroatoms. In:
Storr RC, Gilchrist TL, editors. Science of synthesis: HoubeneWeyl methods of
molecular transformations, vol. 13. Stuttgart: Thieme; 2004. p. 415e601.
[6] Gorman IE, Willer RL, Storey RF. ACS Div Polym Chem Polym Preprs 2010;
51(2):209e10.
composition. The data show that as the proportion of trifunctional
alkyne is increased, the crosslink density and the resulting glass
transition temperature increases. For the system containing 100%
PEGDP the tan
taining a 50/50 mixture of the two alkynes was found to have
a tan
peak at 90 ꢁC.
d
peak was found at 56 ꢁC; while the system con-
d
[7] Gorman IE, Kemp LK, Willer RL, Storey RF. SAMPE technical conference
proceedings: advanced materials and processes: enabling the future; Salt Lake
City, UT, October 11e14, 2010; Society for the Advancement of material and
Process Engineering; CD-ROM-14 pp.
4. Conclusion
[8] Gorman IE, Magenau AJD, Willer RL, Storey RF. ACS Div Polym Chem Polym
Preprs 2009;50(1):139e40.
[9] Gorman IE, Wiggins JS, Storey RF. ACS Div Polym Chem Polym Preprs 2011;
52(2):58e9.
[10] Liu Y, Diaz DD, Accurso AA, Sharpless KB, Fokin VV, Finn MG. J Polym Sci Part A
Polym Chem 2007;45:5182e9.
[11] Sheng X, Mauldin TC, Kessler MR. J Polym Sci Part A Polym Chem 2010;48:
4093e102.
[12] Willer RL, Lin W, Baum K. Final reports for Contracts N00014-02-M-0150
(ONR), N68936-02-P-0468 (NAWC), and N00014-03-C-0384 (ONR).
[13] Miller LA. US Patent 3,082,245, 1963.
[14] Moses JE, Ritson DJ, Zhang F, Lombardo CM, Haider S, Oldham N, et al. Org
Biomol Chem 2010;8:2926e30.
[15] Wang L, Song Y, Gyanda R, Sakhuja R, Meher NK, Hanci S, et al. J Appl Polym
Sci 2010;117:2612e21.
[16] Reppe W, et al. Ann 1955;596:25e38.
With the goal of developing a low-temperature-cure composite
resin system, we have developed a modified epoxy resin, derived
directly from commercial DGEBA-type resins, e.g., EPON 825, that
cures via triazole ring formation (Huisgen 1,3-dipolar cycloaddition
reaction) rather than the traditional oxirane/amine reaction.
Commercial DGEBA was easily converted to an azide-modified
resin through reaction with NaN3 in the presence of ammonium
chloride. A series of di-, tri-, and tetra-functional alkynes, of various
types, including propargyl esters, ethers, and amines, and propio-
late esters, were synthesized to serve as chain extenders/cross-
linkers for the resin; two of the synthesized alkynes, TPEDEA and
TMBTP, were completely novel and not previously reported. Taking
inspiration from azide/alkyne propellant binders, propargyl ester-
[17] Mekni N, Baklouti A. J Soc Chim Tunisie 2005;7:205e7.

2558
I.E. Gorman et al. / Polymer 53 (2012) 2548e2558
[22] Ingham JD, Petty WL, Nichols PL. J Org Chem 1956;21:373e5.
[18] Wang X, Feng Z, Ling J. Acta Polym Sin 2000;4:397e401.
[19] Yamada M, Aoki K. US Patent 5,196,583, 1993.
[23] Doyle CD. J Appl Polym Sci 1962;6:639e42.
[24] Barton JM. Brit Polym J 1979;11:115e9.
[25] Xue L, Wan L, Hu Y, Shen X, Huang F, Du L. Thermochim Acta 2006;448:
147e53.
[20] Standard test method for Arrehenius kinetic constants for thermally unstable
materials, ANSI/ASTM E698-79. Philadelphia: ASTM; 1979.
[21] Caron M, Sharpless KB. J Org Chem 1985;50:1557e60.