Enhanced thermal properties of phthalonitrile networks by cooperating phenyl-s-triazine moieties in backbones

Article abstract of DOI:10.1016/j.polymer.2015.09.035

The development of phthalonitrile resins is of particular interest for aerospace applications because of their excellent heat resistance. Here, heteroaromatic phenyl-s-triazine moieties have been introduced into the architecture of the phthalonitrile resi

Full text of DOI:10.1016/j.polymer.2015.09.035

Polymer 77 (2015) 177e188
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Enhanced thermal properties of phthalonitrile networks by
cooperating phenyl-s-triazine moieties in backbones
Lishuai Zong, Cheng Liu, Shouhai Zhang, Jinyan Wang*, Xigao Jian
State Key Laboratory of Fine Chemicals, Department of Polymer Science and Materials, School of Chemical Engineering, Dalian University of Technology,
Dalian 116024, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The development of phthalonitrile resins is of particular interest for aerospace applications because of
their excellent heat resistance. Here, heteroaromatic phenyl-s-triazine moieties have been introduced
into the architecture of the phthalonitrile resin as thermally stable segments via a two-step, one-pot
reaction. Bis(4-[4-aminophenoxy]phenyl)sulfone was selected to promote the curing reaction, and the
gelation time could be readily controlled by the diamine concentration and the curing temperature as
evidenced by rheometric measurements. The curing procedure was optimized to give high cross-linking
density networks. The resulting networks exhibit high glass transition temperatures at around 400 ꢀC,
outstanding thermo-oxidative stability with weight retention of 95% ranging from 539 ꢀC to 552 ꢀC,
suggesting an improvement of the thermal properties imposed by phenyl-s-triazine units. Additionally,
the CF laminate of the resin possesses high flexural strength of 1722 MPa and interlaminar shear strength
of 71 MPa at ambient temperature, and these values retain 255 MPa and 36 MPa at 450 ꢀC, respectively.
© 2015 Elsevier Ltd. All rights reserved.
Received 23 April 2015
Received in revised form
7 September 2015
Accepted 15 September 2015
Available online 16 September 2015
Keywords:
Phenyl-s-triazine
Phthalonitrile resins
Composite laminates
1. Introduction
moderately elevated conditions. They are all associated with high
5% mass loss temperature and char yield at 1000 ꢀC [7], and could
Great efforts have recently been devoted to high-performance
materials facing aerospace industry mainly due to their high
modulus and strength to weight ratios [1]. One of the increasingly
demanding requirements for these materials is used in heat-
resistance components of aerospace vehicles such as the space
shuttle and reentry module. Phthalonitrile resins, as an emerging
high-performance material [2], have opened up a new possibility to
meet this requirement because of possessing a combination of
promising properties including excellent thermal properties, su-
perior flame and chemical resistance, less flaming smog, and facile
preparation [3].
Pioneering study on phthalonitriles for high-temperature ma-
terials was disclosed by Marvel and coworkers [4]. Unfortunately,
few attempts were performed on phthalonitrile resins until 1970s
[5]. Keller and coworkers have prepared a number of phthalonitrile
monomers and polymers, featuring random lengths of oligomeric
aromatic ether spacer units between curable terminal groups [6].
The networks were obtained by thermally induced curing under
be used as the competitive high-performance materials for marine
and aerospace applications [8]. Since the condition of curing neat
phthalonitriles was extremely sluggish [9], curing additives were
utilized to accelerate the cross-linking reaction. Among kinds of
curing additives, e.g., metals [10], metallic salts [11], organic acids
[3c], phenols [12], amines [13], and bispropargyl ether [14], di-
amines are more extensively used [6a,15]. Two factors have been
taken into account. First, amine-cured polymers are considered
more thermally stable than other cured counterparts. This can be
ascribed to the robust interaction between amines and phthaloni-
triles, which favorably leads to higher degree of polymerization.
Second, amines could also moderately react with phthalonitriles.
Thus, tailoring the type of diamines as well as their concentration
and the curing temperature could readily control the melt viscosity
of the curing system. This undoubtedly facilitates the processing of
the resins.
On the other hand, a tremendous amount of efforts has been
currently focused on a rational design of the structure of phthalo-
nitrile polymers. For instance, sterically rotatable meta, ortho-po-
sition bisphenoxyl moieties [16] along with flexible alkyl-center-
trisphenoxyl [13b] and silane [17] structures were separately
introduced to bind with phthalonitrile groups to generate
* Corresponding author. Tel.: þ86 41184986109.
E-mail address: wangjinyan@dlut.edu.cn (J. Wang).
http://dx.doi.org/10.1016/j.polymer.2015.09.035
0032-3861/© 2015 Elsevier Ltd. All rights reserved.

178
L. Zong et al. / Polymer 77 (2015) 177e188
monomers with low-melting point, wide processing window and
excellent thermal properties. These monomers were competent to
face more cost-effective processing methods such as resin transfer
molding (RTM), resin infusion molding (RIM), and vacuum-assisted
resin transfer molding (VARTM). Combined all the merits displayed
by phthalonitriles, incorporating curable phthalonitrile units is
widely deemed as a possible approach for tailoring the compre-
hensive properties of the known commercial resins (e.g. Epoxy [18],
Benzoxazine [19], Novolac [20], Poly(aryl ether)s [20d,21], Poly-
amide [22], and Polyimide [23]) by many researchers. This strategy
has been already accomplished through either chemical modifica-
tion or physical blending method, and the resulting thermosetting
polymers are proved attractive for advanced technological appli-
cations including composites, adhesives, electronic conductors,
magnetic and microwave absorption materials [24]. However, these
works neglect to incorporate thermally stable and high-stiff groups
into phthalonitrile-based polymer backbones. Such groups may
enhance the stability and stiffness of the polymer main chain, in
turn to bring about the increase for the heat resistance of the cross-
linked product.
Aluminum(III) chloride (AlCl3, A.R.) and Ammonium chloride
(NH4Cl, A.R.) were purchased from Bodi Chemical Reagent Co., Ltd.,
Tianjin, China. All chemicals and solvents above were used without
further purification. Anhydrous potassium carbonate (K2CO3, 99%,
Beijing Chemical Co., China.) was ground and dried in vacuum at
100 ꢀC for 24 h before used. N-methyl pyrrolidone (NMP, A.R.,
Kermel Chemical Reagent Co., Ltd., Tianjin, China.) was refluxed
with CaH2 for 2 h, and then vacuum distilled. The 125e127 ꢀC
boiling range fraction was collected and stored over molecular
sieves (type 4 Å). T300 CF fabric was provided by Aviation Industry
Corporation of China (AVIC). T700 continuous CF was purchased
from TORAY and disposed at 350 ꢀC for 5e10 min before used.
Bis[4-(4-aminophenoxy)phenyl]sulfone (BAPS, 6) was prepared
according to described work by Barikani [31]. The product was
recrystallized from toluene, and brown flaky crystal was obtained.
Purity: 99 wt %. APCI/MS (M þ Calcd. as C21H13N2F2: m/z ¼ 432.11):
m/z ¼ 432.00 (Mþ).
2.2. Methods and equipment
Inspired by this concept, our previous work demonstrated the
preparation of the phthalazinone-bearing monomers [25], poly(-
arylene ether amide)s [26], poly(arylene ether imide)s [27] as well
as poly(arylene ether nitrile)s [28] terminated with cyano groups,
subsequently reporting their trimerization to high-performance s-
triazine resins. The twisted, non-coplanar phthalazinone structures
in the polymer's main chain endow the polymers high thermal
stability and good solubility.
In this paper, we furthered this concept for pursuing ultrahigh
performance phthalonitrile polymers with the aid of phenyl-s-
triazine units. s-Triazine units, one of the most thermally stable
heteroatomic ring reported, were manifested having strong charge
transfer interaction between s-triazine ring and aromatic ring in-
side [29]. This would undoubtedly enhance the rigidity of the
polymer backbone. According to previous reports, the synthesis of
the monomers containing phenyl-s-triazine segment suffer from
toxic and irritant smelling SO3 or NH3 and low yield [30]. Alter-
natively, we presented herein a facile one-step synthetic strategy to
generate phenyl-s-triazine-containing monomer, namely 2-
High performance liquid chromatography (HPLC) was per-
formed on a HewlettePackard (HP) 1100 liquid chromatograph
using a mixture of acetic acid (0.1 wt%) and methanol (v/v ¼ 90:10)
as eluting solvent and a 2.0 ꢁ 150 mm Microbore column (Waters
Spherisorb® S5 ODS2) as column. Inherent viscosities (hinh) of the
polymers were measured using an Ubbelohde capillary viscometer
at a concentration of 0.5 gdLꢂ1 in NMP at 25 ꢀC. Fourier transform
infrared (FT-IR) measurements were performed with a Thermo
Nicolet Nexus 470 FT-IR spectrometer. 1H NMR (400 MHz), 1He1H
gCOSY NMR (400 MHz) and 13C NMR (100 MHz) spectra were
recorded with a Varian Unity Inova 400 spectrometer using CDCl3
or CDCl3 and CF3COOD mixed solvents and listed in parts per
million downfield from tetramethylsilane (TMS). Matrix-assisted
laser desorption ionization time-of-flight mass spectroscopy
(MALDI-TOF-MS) analyses were performed on a Micromass GC-TOF
CA 156 MALDI-TOF/MS. Elemental analyses were conducted on a
Vario ELIII CHNOS Elementaranalysator from Elementar-
analysesyteme GmbH for C, H, N, and S determinations. Gel
permeation chromatography (GPC) analysis was carried out on an
Agilent PL-GPC 50 Intergrated GPC system equipped with two PLgel
phenyl-4,6-bis(4-fluorophenyl)-1,3,5-triazine (1).
A
phenyl-s-
triazine-containing phthalonitrile oligomer was subsequently
designed and prepared (named as 5). The kinetics of curing reac-
tion, melting behaviors and curing procedure of 5 cured with bis[4-
(4-aminophenoxy)phenyl]sulfone (named as 6) were investigated
in details. The thermal stabilities of the resultant resins (named as
7) are significantly modified by the introduction of phenyl-s-
triazine moieties and increase on the addition of 6 for 2e10 wt %
loading. Simultaneously, the resins exhibit limited water absorp-
tion capability, which is also closely related to the % 6 inclusion. The
carbon fiber (CF) reinforced composite of the 5 wt % hybrid was
fabricated. The result of the mechanical tests indicated that this
type of phthalonitrile resin might be potentially used as structural
matrix resin for high-performance applications.
5
m
m MIXED-C columns (300 ꢁ 7.5 mm) arranged in series with
NMP as solvent calibrated with polystyrene standards. The solu-
bility test was performed by dissolving 0.04 g oligomer in 1 mL
solvent (4%, w/v) at different temperatures. The thermal properties
of the obtained materials were determined using a modulated TA
Q20 instrument at a heating rate of 10 ꢀC/min under a nitrogen flow
of 50 mL/min. The kinetics study of the reaction between 5 and 6
(5 wt %) was conducted by the non-isothermal DSC method at the
heating rate of 5, 10, 15, and 20 ꢀC/min. The samples were prepared
by thoroughly grounding for three times to ensure homogeneity.
The DSC tests of the cured 7s was evaluated at the heating rate of
20 ꢀC/min. Rheological behavior was investigated using a TA
AR2000 instrument under a frequency of 1 Hz and a strain of
0.02 N. The sample was compacted into a flaky cylinder with the
2. Experimental
dimension of
F
25 ꢁ 1 mm3 in advance. Dynamic mechanical
analysis (DMA) measurements were carried out with a TA Q800
instrument at 1 Hz and a heating rate of 3 ꢀC/min from 25 ꢀC to
400 ꢀC under air atmosphere using a 30 ꢁ 10 ꢁ 2 mm3 T300
composite sample, in order to characterize the dynamic mechanical
2.1. Materials
4,40-Biphenol (BP, 99%, 2), bis(4-chlorophenyl)sulphone (BCS,
99%) and 4-aminophenol (APO, 99%) were purchased from Haiqu
chemical Co., shanghai, China. 4-Fluoro-benzonitrile (FBN, 99%) and
4-nitro-phthalonitrile (NPh, 99%, 4) were purchased from Jiakai-
long chemical Co., Wuhan, China. Benzaldehyde (BA, A.R.), chloro-
benzene (CB, A.R.), toluene (A.R.) and other solvents were
purchased from Kermel Chemical Reagent Co., Ltd., Tianjin, China.
spectra including mechanical damping tan d
, storage modulus E0
and loss modulus E00. Thermal Gravimetric Analysis (TGA) was oꢂb1-
tained from a TA Q500 instrument at a heating rate of 20 ꢀC/min
in N2 or air. The gel contents of the cured samples were studied
according to ASTM D2765-11 standard. NMP was selected as the
extracting solvent. The oxidative stability was performed via

L. Zong et al. / Polymer 77 (2015) 177e188
179
handling a 0.5 g sample in a muffle furnace at various temperatures
(250e450 ꢀC) each for 8-h intervals under static air atmosphere.
The retention weights for every temperature step were recorded.
The long-term water absorptions were studied based on ASTM
D570-98 standard. The flexural strength and interlaminar shear
strength were determined using an Instron-5869 machine with a
capacity of 500 N on T700 composite samples according to ASTM
D790-10 and ISO 14130 standards, respectively. The high-
temperature (450 ꢀC) mechanical properties were achieved on-
line via a peripheral heating oven. The resin content of the com-
posite (T700) was measured according to ASTM D3171-15 standard.
Annex A7 procedure was used to digest the matrices at 600 ꢀC, and a
paralleled sample of carbon fiber was heated simultaneously for
calculating the original mass of the carbon fiber in the composite.
SEM studies were performed on the cracked section of the com-
posite with a FEI QUANTA 450 instrument at 15 kV.
NaOH solution. The filtered solid was rinsed continuously with the
5% aqueous NaOH solution until the filtrate turned to colorless. Then
the solid was washed with distilled water until neutral. The yield of
dried crude product was 92%. Subsequently, the crude product was
purified by reprecipitation from NMP into ethanol, and then dried at
120 ꢀC in a vacuum oven for 20 h. The constitutional repeating unit
of the target product (n) was theoretically designed to be 1 by
controlling the feed ratio. The value for n of the obtained product
was calculated via the terminal group analysis method based on the
1H NMR spectrum (Fig. S6) [32]. It is approximately 1.8. Total yield:
78%.1H NMR (400M, CDCl3 and CF3COOD):
d
8.64 (d, J ¼ 13.7 Hz, 2H),
8.54 (d, J ¼ 13.4 Hz,1H), 7.93e7.79 (m,1.1H), 7.72 (dd, J ¼ 16.4, 8.1 Hz,
4H), 7.44e7.34 (ss, 1.14H), 7.27 (t, J ¼ 8.8 Hz, 4H), 7.21 (d, J ¼ 8.5 Hz,
1.04H). 13C NMR (101 MHz, CDCl3 þ CF3COOD):
d 167.59 (s), 167.20
(s), 166.03 (d, J ¼ 7.7 Hz), 162.53 (d, J ¼ 6.8 Hz), 160.79 (s), 160.37 (s),
159.96 (s), 159.54 (s), 153.95 (d, J ¼ 17.2 Hz), 153.02 (d, J ¼ 16.4 Hz),
137.76 (s),137.44 (s),136.79 (s),175.83e103.05 (m),136.00 (s),133.44
(s), 130.12 (s), 129.76 (s), 129.59e128.61 (m), 123.35 (s), 121.97 (d,
J ¼ 12.1 Hz), 121.13 (d, J ¼ 11.8 Hz), 118.84 (s), 118.05 (s), 116.89 (s),
116.01 (s), 114.83 (s), 114.36 (s), 113.17 (s), 110.33 (s), 107.54 (d,
J ¼ 4.7 Hz).
2.3. Synthesis of 2-phenyl-4,6-bis(4-fluorophenyl)-1,3,5-triazine
(BFPT, 1)
A facile one-step procedure is shown as follows (Scheme 1): To a
stirred solution of FBN (0.10 mol, 12.111 g), BA (0.05 mol, 5.1 mL)
and NH4Cl (0.05 mol, 2.675 g) in CB (100 mL) was added AlCl3
(0.05 mol, 6.667 g) progressively at an ice-bath temperature (ꢂ5 to
2.5. Curing procedure for the formation of the networks (7)
ꢀC). Then the ice bath was removed. The yellow solution was
The formation of the networks (7) was typically achieved by a
thermally induced procedure as follows (Scheme 2). To 1.0 g olig-
omer 5 was added 0.02e0.1 g 6 (2e10 wt %). The mixture was
thoroughly grounded, and then the well-mixed sample was com-
0
gradually heated to 130 ꢀC and maintained at this temperature for
8e12 h. The product was precipitated by slowly pouring the
mixture into an ice-containing 5% aqueous HCl solution. CB was
removed via steam distillation. The material was collected using
Büchner funnel and rinsed with distilled water until neutral. Then,
the obtained white solid was added into a 250 mL of three-necked
flask, and stirred with 200 mL methanol. The suspension was
filtered, and the crude product was dried at 100 ꢀC under vacuum
for 24 h. Finally, the product was recrystallized from toluene, white
long thin needle crystal was obtained, which should be ground
before used. Yield: 82%, m.p.: 260.2e260.8 ꢀC; purity: 99 wt %.
pressed to a void-free flaky cylinder (
F
25 ꢁ 2 mm3) under elevated
pressure. The sample was thermally cured by heating in an oven at
250 ꢀC/3 h þ 285 ꢀC/1 h þ 325 ꢀC/3 h þ 350 ꢀC/2 h þ 375 ꢀC/8 h to
afford the cross-linked polymer. Then, a portion of 7 was pulverized
and used for DSC and TGA analysis. The residual bulky sample was
used for long-term oxidative stability and water-uptake studies.
2.6. Fabricating process of the 7/T700 composite
MALDI-TOF/MS (M
þ
Calcd. as C21H13N2F2 345.1078): m/
z ¼ 345.1067(Mþ). 1H NMR (400M, CDCl3):
7.74e7.50 (m, 3H), 7.39e7.12 (m, 4H).
d 8.91e8.61 (m, 6H),
The network 7 containing 5 wt % 6 was selected to fabricate the
composite (T700). To a three-necked flask was charged 50 g 5, and
then the flask was heated to 250 ꢀC. 2.5 g 6 was added to the melted
5 with vigorous stirring. The mixture was maintained at 250 ꢀC for
7 min, then cooled down. The obtained pre-polymer (B-stage resin)
was dissolved in 100 mL NMP. The T700 carbon filament was dip-
ped into the solution, and winded continuously to an iron frame of
300 ꢁ 210 mm2 (Fig. 1) to afford the unidirectional prepreg. After
dried under the infrared lamp, the prepreg was transferred to a
vacuum oven, dried at 180 ꢀC for 36 h, 200 ꢀC for 1 h. The solvents
have been thoroughly removed as evidenced by TGA (Fig. S3). Then,
the prepreg was tailored into 100 ꢁ 60 mm2 dimension. Twelve
tailored prepregs were compacted into a tight steel mold with a
pressure of 2.5 MPa at 250 ꢀC for 2 h. Then the steel mold was
removed, and the obtained laminate was cured at 250 ꢀC for 1 h,
285 ꢀC for 1 h, 325 ꢀC for 3 h, 350 ꢀC for 2 h, and 375 ꢀC for 8 h in a
muffle furnace. The thickness of the finished laminate was
2.4. Synthesis of oligomer 5
As Scheme 2 shows, to a 250 mL three-necked flask equipped
with a Dean-Stark trap with a nitrogen inlet was added 2 (0.06 mol,
11.189 g), K2CO3 (0.08 mol, 11.057 g), 30 mL NMP and 30 mL toluene.
The mixture was stirred at room temperature for 5 min, and then the
temperature was raised gradually to 140 ꢀC under a nitrogen at-
mosphere. After the generated water was completely azeotroped off
with toluene, the reaction was allowed to cool to room temperature.
1 (0.03 mol, 10.369 g) along with 20 mL NMP was charged into the
flask. The system was heated stepwise to 190 ꢀC, and held for 3 h.
The obtained solution was then cooled to 50 ꢀC. NPh (0.07 mol,
12.119 g) and 30 mL NMP were added into the flask. The reaction
system was heated to 80 ꢀC for 8e10 h. After cooling to room tem-
perature, the resultant brown mixture was poured to a 5% aqueous
approximately
2 mm. It should be sawed and sanded into
80 ꢁ 10 ꢁ 2 mm3 size and 20 ꢁ 10 ꢁ 2 mm3 size for flexural
strength and interlaminar shear strength tests, respectively. The
weight percentage of the fiber of the laminate was 62.7% based on
the ASTM D3171-15 standard.
3. Results and discussion
3.1. Synthesis of oligomer 5
Scheme 1. Synthetic route to 1.
A phthalonitrile-terminated oligomeric poly(aryl ether) (5) was

180
L. Zong et al. / Polymer 77 (2015) 177e188
Scheme 2. Synthetic route of oligomer 5 and subsequent network 7 formation.
successfully prepared by a two-step, one-pot reaction, as Scheme 2
shows. The first reaction step is a typical nucleophilic substitution
reaction. An s-triazine-activated dihalo compound 1 first reacted
with biphenol 2 through a Meisenheimer complex at 190 ꢀC. NMP/
toluene solvent system was favorably chosen, because NMP would
provide sufficient solubility for the product to ensure the reaction
proceeded steadily towards the target product. Toluene would help
to azeotrope off the by-product water. Then the toluene was
removed from the system before polymerization. This is contrary to
the reported procedure [7b] and attributed to the limited solubility
of intermediate 3 imposed by the steric planarity and the electron
affinity of the phenyl-s-triazine groups. The presence of toluene
may induce 3 precipitated from the reaction solution. Conse-
quently, the resulting product exhibits a broader polydispersity
index (Fig. S3). In the second step, the reaction underwent a milder
nitro-substituted reaction at 80 ꢀC for 12 h. An excess amount (5%)
of 4 was used to convert the phenolates into the cross-linkable
phthalonitrile moieties completely. The relatively synthetic de-
tails are summarized in Table 1. As expected, solubility study re-
veals that 5 is readily soluble in NMP, and also in some less efficient
solvents such as tetrachlorethane and hot pyridine (Table S1). The
rational solubility may be resulted from the strongly polar phtha-
lonitrile terminals. These units would be easily polarized by the
solvent molecule, and in turn to solubilize the oligomer efficiently.
This would substantially make the oligomer 5 competent for
composite applications through solution impregnating technology.
3.2. Characterization of oligomer 5
The MOLDI-TOF/MS spectrum (Fig. S4b) has provided primary
Fig. 1. The schematic diagram for the fabrication of the 7/T700 composite.
evidence to certify the structure of the resultant 5, due to the

L. Zong et al. / Polymer 77 (2015) 177e188
181
Table 1
Synthetic data of Oligomer 5.
Temperature (ꢀ C)
190
Time (h)
3
Mn
930
Mn
1316
Mn
1011
PDc
Yield(%)
92
hinhd (gdLꢂ1
)
Color
Gray
a
b
c
Sample
Solvent
NMP
Charge ratio (2:1:4)
2:1:2.1
5
3.69
0.013
a
Designed molar mass controlled by the charge ratio of 2 versus 1.
b
c
Number-average molar mass of 5 calculated by the integration area of characteristic peaks through 1H NMR spectrum in CDCl3 and CF3COOD.
Number-average molar mass and polydispersity of 5 measured by GPC in NMP.
d
Inherent viscosity of 5 determined at a concentration of 0.5 g/dL in NMP at 25 ꢀC.
correlation between the detected molar mass and the designed
values. No trace of hydroxy-terminated oligomer peaks indicates
sufficient displacement of the nitro groups by the ether linkages.
Detailed analyses of the chemical structure of oligomeric 5 were
further performed using instrumental techniques including FT-IR,
NMR and 1He1H gCOSY. The FT-IR spectra of oligomer 3 and 5
(Fig. S5) demonstrate the characteristic bands of s-triazine group at
1488 cmꢂ1 and 1363 cmꢂ1. The cyano group is located by the
stretching band at 2230 cmꢂ1 in the spectrum of oligomer 5.
Compared to the spectrum of 3, the appearance of the cyano group
indicates successful introduction of this cross-linkable moiety into
the molecular structure. The exact structure of oligomer 5 was
certified by the NMR spectra (Fig. S6eS8). Each discernible shifting
peak was accurately assigned with the assistance offered by the
1He1H gCOSY spectrum of oligomeric 5 (Fig. S7). The set of peaks
shifting downfield at 8.49e8.72 ppm are attributed to the protons
(H5, 6), which are adjacent to s-triazine groups (Fig. S6bc). Note
that the H5 and H6 shift at different fields with the integral in-
tensity ratio of 2:1 in mixed solvents of CDCl3 and CF3COOD
(Fig. S6c). These signals can be used as the referenced signals to
position the protons (H1-4, 7, 8). The rest protons resonating at
8.32e8.40 ppm are associated with H9, 10 of phthalonitrile moiety.
The disappearance of active halogen (OeH) signal (5.23e5.27 ppm,
Fig. S6a) confirms the complete transformation of the hydroxyl
groups to ether-linked phthalonitrile moieties. Since all the area
peaks can be well ascertained, the value of n, which represents the
number of polymer repeating units, could be calculated via termi-
nal analysis method [32]. The Mn of oligomer 5 was subsequently
calculated as 1316 g/mol. The value, higher than the designed
molecular mass (930 g/mol), might be caused by the loss of low
molar product during the purification procedure, and is in agree-
ment with previous reports [33]. Thus, the characterization study
certifies that the proposed phthalonitrile oligomer was successfully
synthesized.
Fig. 2. DSC curves of oligomer 5 under N2 atmosphere: (a) First run at the heating rate
of 10 ꢀC/min; (b) Cooling curve at the cooling rate of 10 ꢀC/min; (c) Second run at the
heating rate of 10 ꢀC/min.
Kissinger method [35] was employed to calculate the cure kinetics
parameters of the reaction between 5 and 6 (See the analysis details
in Supplementary data) [36]. The results are listed in Table S2. The
activation energy of the reaction between 5 and 6 (107.7 kJ/mol) is
significantly higher than the self-promoted phthalonitrile resins
disclosed by Zhou [16a] as well as the well-known Epoxy [37],
bismaleimide (BMI) [38], and benzoxazine (BA) resins [1e],
implying more energy is entailed to overwhelm the extremely high
3.3. Thermal cure behaviors
The thermal property of the neat oligomer 5 was investigated by
DSC up to 400 ꢀC (Fig. 2). In the first heating curve, a slight glass
transition appeared at 97 ꢀC, implying a rather low molecular
weight (the reported Tg of polymer PAEP is 241 ꢀC) [30a]. Then an
endothermic peak was recorded at 211 ꢀC. Such a wide peak is
associated with the broad molecular weight distribution
(PD ¼ 3.69). When the sample was quenched and reheated
immediately, the oligomer showed an increased Tg of 137 ꢀC,
accompanied by vanishing of the melting peak. It suggests that 5
experienced a somewhat self-promoting reaction under elevated
temperature, which agrees with Keller's analysis [34].
For promoting the curing rate, a small amount (5 wt %) of
typically used diamine 6 was combined with 5 during the thermally
induced curing process. The non-isothermal DSC tests of 5/6
mixture (Fig. 3) were carried out, followed by elucidating the ki-
netics of the curing reaction. A significant exothermic peak appears
in each DSC trace with the peak temperature of 242, 254, 264,
269 ꢀC for the heating rate of 5, 10, 15, 20 ꢀC/min, respectively.
Fig. 3. DSC curves of oligomer 5 with 5 wt % 6 in N2 atmosphere: (a, b, c, d) first run
with the heating rate of 5, 10, 15, 20 ꢀC/min, respectively.

182
L. Zong et al. / Polymer 77 (2015) 177e188
Table 2
Three cure schedules upon 5 with 5 wt % 6 under environmental atmosphere.
Index
Specific procedure
a
b
c
250 ꢀC/3h þ 285 ꢀC/1h þ 325 ꢀC/8h
250 ꢀC/3h þ 285 ꢀC/1h þ 325 ꢀC/3h þ 350 ꢀC/8h
250 ꢀC/3h þ 285 ꢀC/1h þ 325 ꢀC/3h þ 350 ꢀC/2h þ 375 ꢀC/8h
potential barrier, which intuitively behaves as requiring much
higher temperature. Additionally, the high frequency factor (A,
2.0 ꢁ 1010) along with the untidy reaction order (n, less than 1)
demonstrates a complex curing reaction.
In this context, for pursuing highly cross-linked thermoset
resins, three curing schedules were empirically designed to poly-
merize the oligomer 5 with 5 wt % 6 loading (Table 2). The reaction
is initiated by the nucleophilic attack of the amine group to
the carbon atom of the nitrile group to generate an
iminoisoindolenine-containing compound. Then it reacts with
another phthalonitrile group to give chain extension. And after
curing at even high temperature, isoindoline, triazine, dehy-
drophthalocyanine, and phthalocyanine was produced [6a,12a,13a]
(Scheme S1). As we know, the well-cured polymers are always
associated with attractive properties including high glass transition
temperature, excellent thermal stability, and commendable me-
chanical property. Thus, three set of tests (i.e., DSC, TGA, and flex-
ural strength measurement) for the cured 7 or its CF-reinforced
composite were carried out to give the most desirable curing cycle.
DSC study (Fig. 4) obviously reveals that the products cured under a
and b schedule soften at 179 ꢀC and 212 ꢀC, respectively, due to their
low degree of polymerization. Notably, no distinct transition was
detected in c-schedule curve, indicating 375 ꢀC curing step is
beneficial to the polymerization of phthalonitrile-terminaled resin/
diamine system, which is consistent with previous research [3a].
Likewise, TGA and DTG tests (Fig. 5) afford similar conclusion. Both
the initial decomposition temperatures and the residue mass of
thermosetting resin 7s significantly increase when the curing cycle
was used from a to c. DTG data disclose that the temperatures of
thermo-gravimetric rate of all the networks peak at 540 ꢀC and
590 ꢀC. The 540 ꢀC peak decreases gradually and eventually dis-
appears, while the 590 ꢀC peak behaves conversely as the final
curing temperature was elevated. However, the slight residue of
540 ꢀC peak in c-schedule trace has not escaped our observation,
which strongly suggests that an elevated temperature such as
425 ꢀC for curing is necessary. Thus, by post heated at 425 ꢀC for 8 h
under static air atmosphere, the network shows 5% mass loss
Fig. 5. TGA and DTG curves of networks (7) separately cured under a, b, c schedule
with 5 wt % 6 with the heating rate of 20 ꢀC/min detected in N2.
temperature rising from 537 ꢀC to 565 ꢀC, along with the char yield
at 900 ꢀC increasing for 2.5% and the disappearance of the 540 ꢀC
peak in DTG curve (Fig. S10). Unfortunately, the high-temperature
(425 ꢀC) curing schedule caused a non-negligible mass loss for
6.3%, suggesting that this cross-linking procedure should be
cautiously utilized. The flexural strength test of the continuous CF-
reinforced laminates of 7 (Fig. 6) also provided effective evidence
for choosing the optimal curing cycle. The c-schedule-cured lami-
nate exhibits the highest flexural strength of 1722 MPa as compared
to other laminates. As a result, c schedule was convinced as the
most suitable procedure for curing the phenyl-s-triazine-contain-
ing phthalonitrile oligomer/diamine system.
The structure of cured 7 were Characterized by FT-IR (Fig. S11),
the peak intensity of the nitrile group weakened, and the absorp-
tion at 1502, 1360, and 1004 cmꢂ1 appeared, both of which indicate
the curing of phthalonitrile groups and the formation of phthalo-
cyanine and s-triazine structures.
3.4. Meltability and processability
The rheological behavior of the resins is a key factor in deter-
mining their processablities. Therefore, the variation of the
Fig. 6. Flexural strength of the continuous CF-reinforced laminates of 7 separately
cured by a, b, c procedures with 5 wt % 6 under air atmosphere according to ASTM
D790-10 Standard.
Fig. 4. DSC curves of networks (7) separately cured under a, b, c schedule with 5 wt % 6
with the heating rate of 5 ꢀC/min.

L. Zong et al. / Polymer 77 (2015) 177e188
183
[19b], 3b-c [39], and 4M-P [3b], but still lower than those of the
RTM-processable resins like RPh [7a], N ¼ 2 [33], TDPE [13b], and
HSiPN-ViSiPN [17] (see Table S5, column Process window). After
gelation, the viscosity of the mixture increased abruptly as soon as
the temperature was elevated to 275 ꢀC. Intriguingly, the viscosity
then exhibited a steady trend from 300 ꢀC to 350 ꢀC, possibly
because the increasing temperature generally drives the viscosity
declined, simultaneously suppresses the curing rate (see Fig. 8a,
blue curve). This steady trend may be interpreted as a compromise
between the two factors. When the temperature exceeded 350 ꢀC,
the viscosity rose significantly again, implying the curing rate
renewedly increased.
Subsequently, the complex viscosity as a function of phthalo-
nitrile polymerization reaction was also investigated by conducting
isothermal rheometric measurements at different temperatures on
oligomer 5 with various concentrations of catalyst 6 (Fig. 8). All the
prepolymers initially exhibited a melt viscosity of less than 20 Pa s.
The gelation occurred ranging from 7.2 min to 40.9 min. This step
would be favorable for their meltability and processability.
Specially, as displayed in Fig. 8a, after gelation, the slope of the
increase of the viscosity fell as the maintained temperature, which
exceeded 250 ꢀC, increased. It indicates that those temperature
(>280 ꢀC) would be disadvantageous to the cross-linking of 5/6
mixture, probably caused by the volatility or degradation of
diamine 6 [15a]. Fig. 8b, c, d depict the complex viscosity correlated
with time at 235, 250, 280 ꢀC distinguished by the diamine loading,
respectively. Interestingly, the gelation time was negatively
dependent on the ratio of the diamine at 235 ꢀC and 250 ꢀC
(Fig. 8bc). To the best of our knowledge, this phenomenon has not
appeared in any other works. A possible explanation would be that
the bulky volume of the phenyl-s-triazine-containing prepolymer
has hindered the further polymerization occurred between the
prepolymer molecules; rather, the more mass ratio of diamine was
added, the more bulky chains were generated. The interaction be-
tween the bulky chains should leap over the high potential energy,
thus visibly resulting in delayed gelation (as Fig. 9 depicts). When
Fig. 7. Storage Modulus (G'), Loss Modulus (G'') and Complex viscosity (h*) of oligomer
5 as a function of temperature at a heating rate of 3 ꢀC/min with 2 wt % (a) and 5 wt %
(b) 6.
complex viscosity of oligomer 5 as a function of temperature was
measured from 175 ꢀC to 400 ꢀC, and the storage modulus and loss
modulus were simultaneously recorded. The results (Fig. S12)
indicate that the complex viscosity of neat 5 decreased rapidly after
melting at about 215 ꢀC, and then reached a minimum for 4.2 Pa s at
283 ꢀC. The gelation, which is determined from the crossover point
of the storage modulus (G') and the loss modulus (G''), occurred at
303 ꢀC. After this, both of G' and G'' leveled off, indicating that the
self-curing reaction proceeded sluggishly. However, the curing
process would be promoted in the presence of the additive 6. When
the loading ratio of 6 was 2 and 5 wt %, the gel points migrated
slightly to lower temperature at 303 ꢀC and 285 ꢀC, respectively
(Fig. 7). This means the process window of the 5/6 mixture is about
70e88 ꢀC relying on the diamine inclusion. The values are com-
parable to those of many phthalonitrile monomers such as BZBPh
Fig. 8. Complex viscosity (
h*) of oligomer 5 with different mass ratios of diamine 6 as a function of time at various temperature: (a) 2 wt % 6 at various temperature; (b) with
different mass ratios of 6 at 235 ꢀC; (c) with different mass ratios of 6 at 250 ꢀC; (d) with different mass ratios of 6 at 280 ꢀC.

184
L. Zong et al. / Polymer 77 (2015) 177e188
Fig. 9. The proposed graphical mechanism of the formation of the prepolymer.
the temperature was maintained at 280 ꢀC (Fig. 8d), the reaction
potential barrier was overwhelmed, the gelation time eventually
increased linearly with diamine mass fraction. The rheological be-
haviors of 5/6 system reveal that the rate of their polymerization
could be controlled by varying the diamine concentration or tem-
perature. This would be a key factor that advantageously affects
their processability through molding or lamination technology.
the polymers possess excellent thermal and thermo-oxidative
stabilities, which increase on the addition of 6. The fact is
ascribed to the corresponding increase of the cross-linking density
driven by the increasing 6 inclusion. Furthermore, owing to the
presence of the thermally stable phenyl-s-triazine units in the
polymer backbone, these thermal data are higher than those of
many cured poly(aryl ether)-based phthalonitrile networks (e.g.,
PAEK-CN [40], PEN-t-BAPh [21a], 2CN-o-PEEK [21c], and also
several oligomers disclosed by Keller [3f,9,33,41]), the ether-linked
monomers (e.g., BDS [42], BPh, BAPh, 6FPH [15b], RPh [7a], and
monomers 1e3 [16b]), and the self-curable monomers (e.g., 2OeP
[16a], 3a-b [39], and 4O-M [3b]), but slightly lower than those of the
cured TDPE [13b] and silane-bearing HSiPN-ViSiPN [17] (The cor-
responding thermal data of the reported networks are summarized
in Table S5). However, a certain length of the molecular chain of 5
lowers the cross-linking density of 7 as compared to the above
mentioned phthalonitrile monomers, which weakens the role of
phenyl-s-triazine groups, in turn to lower the thermal stabilities of
3.5. Thermal and thermo-oxidative stability
The thermal and thermo-oxidative stabilities of the networks
(7) prepared by schedule c were investigated by TGA analysis. The
relevant thermograms are presented in Figs. 10 and 11, respectively.
The data are summarized in Table S3. The samples distinguished by
diamine loading from 2 to 10 wt % showed the weight retention of
95% at 530e555 ꢀC under N2 atmosphere, and at 539e552 ꢀC under
a flow of air. The residue mass of the polymers ranged from 64 to
74%, when they are heated to 900 ꢀC in N2. The relative aerobic char
yields for 7 are from 3 to 31% at 800 ꢀC. These values show that all
Fig. 10. TGA and DTG thermograms of the networks (7) after cured with different mass
ratios of 6 under N2 atmosphere.
Fig. 11. TGA thermograms of the networks 7 after cured with different mass ratios of 6
under air atmosphere.

L. Zong et al. / Polymer 77 (2015) 177e188
185
Fig. S13), strongly implying sulfur-containing diamine units are
thermally weak segments in polymer components. We can further
envision to load diamine with high-temperature-resistance struc-
ture as additive to pursue phthalonitrile resins which could be used
at higher temperature (such as 450 ꢀC) for long term (>200 h).
3.6. Water absorption capability
The water absorption capability of 7s was conducted with a
sample soaked in distilled water at the ambient temperature
(Fig. 13). The hygroscopicity of all the networks reached
a
maximum amount of 1.72e1.93 wt % over the course of approxi-
mately 65 days. The water absorption decreases as the levels of
diamine concentrations increase toward 5 wt %. This is because the
increasing cross-linking density hampered the osmosis of hydrone
into polymer chains. Irregularly, the water absorption for the net-
works (8e10% diamine mass ratio) ultimately saturated at 1.89%
and 1.93% after a 56-day immersion, respectively. This is thought as
a result of the increasing amino groups in polymer backbone which
would incur increasing hydrophilic capacity of the networks.
Compared to other phthalonitrile counterparts [7b,42], the limited
water absorption will definitely benefit the polymers 7 for applying
at a high-humidity or aqueous environment.
Fig. 12. Oxidative aging of networks (7) cured with different mass ratios of 6 after an
approximately 60-h treatment at various temperatures.
the networks. The DTG curves disclose that these polymers un-
dergo catastrophic failure at around 590 ꢀC under N2 atmosphere,
while two stages of decomposition at temperatures about 590 ꢀC
and 660 ꢀC was found with an atmosphere of flowing air, implying
two different decomposition mechanisms. Research into the
decomposition mechanisms is already underway.
3.7. Mechanical properties of network 7s/CF laminates
Dynamic mechanical analysis (DMA) was used to investigate the
The oxidative stabilities of the polymers were evaluated at
various temperatures (to a maximum of 450 ꢀC) each for 8-h in-
tervals under ambient atmosphere (Fig. 12). The results showed
that the thermosets (7) retained 50e72% of their initial weight
varied by the 6 contents after the approximately 60-h period
treatment. It was noteworthy that no distinct mass loss (0.5e1.5 wt
%) could be observed during the 375 ꢀC-treatment. The results
could be roughly comparable to those of the famous PMR-Ⅱ-30/
Celion 6000 laminates [43] (aging at 371 ꢀC for 200 h, mass loss for
12.5 wt %). The oxidative stabilities of 7s increase with the increase
of diamine concentrations corresponding to the increasing cross-
link density, which is consistent with their thermal stabilities
detected by TGA. Note that the sulfur loss is higher than other el-
ements after separately treated at steplike temperatures (i.e., 375,
425, 450 ꢀC) as evidenced by the elemental analysis (Table S4,
Fig. 14. DMA traces of networks 7/CF fabric composites cured with various concen-
trations of diamine 6: (a) G', G'' and tan
d
of 5 wt % diamine containing 7 composites as
Fig. 13. Water absorption ratios of networks (7) cured with different mass ratios of 6
a function of temperature; (b) tan d as a function of temperature with respect to the 6
versus time at ambient temperature.
concentrations.

186
L. Zong et al. / Polymer 77 (2015) 177e188
temperature DSC tests were performed on the thermoset 7s
(Fig. S15). The results disclose that the Tgs of 7s were from 418 to
424 ꢀC. High diamine loading (6 wt%e10 wt%) would caused a slight
decrease of Tg. Compared to the poly(aryl ether)-based phthaloni-
trile oligomers like PAEK-CN [40], PEN-t-Ph [44], 2CN-o-PEEK [21c],
PEN-CN3 [45], and 6 [7b], the Tgs of 7s are higher than those of
phthalonitrile thermosets (Table S5, column Tg). This is resulted
from the contribution of the high-stiff phenyl-s-triazine moieties to
the rigidity of the polymer backbone. As expected, the overall
thermal properties of the networks (7) are significantly promoted
by cooperating phenyl-s-triazine moieties into the polymer
architecture.
Based on the DMA results, the network 7 comprising of 5 wt %
diamine was selected to reinforce with continuous T700 carbon
fiber. The mechanical property of the resulting composite laminate
was evaluated according to ASTM D790M standard and ISO 14130
standard (Fig. 15). It was observed that the composite exhibited
excellent mechanical property with the flexure strength of
1722 MPa and the interlaminar shear strength of 71 MPa at ambient
temperature. When detected on-line at 450 ꢀC, these values
retained 15% and 51% of their initial values for 255 MPa and 36 MPa,
respectively. The performance is a little disappointing, and prob-
ably results from a combined effect of superabundant ether linkage
encompassed in polymer framework and a certain length of the
polymer main chain, both of which would encourage the mobility
of the main chain of the networks. For some positive aspects, the
data at 450 ꢀC would be a reference in further research. Succes-
sively, scanning electron microscopy (SEM) studies were performed
on the surfaces of the cracked samples to assess the failure mech-
anism. Room-temperature failure (Fig. 16ab) underwent a co-break
process, accompanied by a little extraction of CF from resins (white
arrows). The fracture morphology of the resin phase is smooth (red
arrows), indicating a brittle cracked mechanism. In contrast, when
Fig. 15. Flexural strength (a) and interlaminar shear strength (b) of 7/CF laminate at
ambient temperature (1) or 450 ꢀC (2), respectively.
thermal properties of the composite laminates derived from the
networks 7 and T300 CF fabric (The details of fabricating composite
laminates are provided in Supplementary data). The corresponding
DMA profiles are given in Fig. 14. Taking the composites of 5 wt %
addition network 7 as representative, the storage modulus fell with
the temperature increased, and retained 14.5 GPa, 57% of its initial
storage modulus, at 400 ꢀC. Simultaneously, the tan
at around 400 ꢀC, related to the glass transition of the polymer.
Fig. 14b depicts the tan traces of all the composites versus tem-
perature. When the diamine inclusion was 5 wt %, the tan exhibits
d value peaked
d
d
the lowest height, indicating a minimum loss of its mechanical
property. However, not any complete tan delta peaks could be
obtained from Fig. 14. In order to determine the Tg of 7s, high
Fig. 16. The cross-sectional images for the network 7/CF laminate broken at ambient temperature (a, b) or 450 ꢀC (c, d) monitored by SEM instrument.

L. Zong et al. / Polymer 77 (2015) 177e188
187
laminates were broken at 450 ꢀC, the fracture surface was coarse,
and the resins covered the cross section of the fiber (Fig. 16cd, blue
arrows). However, the interfacial adhesion between 7 and CF is just
physical connection which could be easily broken; thus, enhancing
the interfacial strength via chemically modifying the surface of fi-
bers is necessary. The work is underway and will be reported in the
future.
[5] (a) T.M. Keller, J.R. Griffith, Abstr. Pap. Am. Chem. S 176 (1978) 105e106;
(b) T.M. Keller, J.R. Griffith, J. Fluor. Chem. 13 (4) (1979) 315e324.
[6] (a) T.M. Keller, T.R. Price, J. Macromol. Sci. A A18 (6) (1982) 931e937;
(b) T.M. Keller, Polymer 34 (5) (1993) 952e955;
(c) M.L. Warzel, T.M. Keller, Polymer 34 (3) (1993) 663e666;
(d) M. Laskoski, T.M. Keller, S.B. Qadri, Polymer 48 (26) (2007) 7484e7489.
[7] (a) T.M. Keller, D.D. Dominguez, Polymer 46 (13) (2005) 4614e4618;
(b) M. Laskoski, D.D. Dominguez, T.M. Keller, J. Polym. Sci. Part A Polym. Chem.
43 (18) (2005) 4136e4143.
[8] (a) S.B. Sastri, J.P. Armistead, T.M. Keller, Polym. Compos. 17 (6) (1996)
816e822;
4. Conclusion
(b) S.B. Sastri, J.P. Armistead, T.M. Keller, U. Sorathia, Polym. Compos. 18 (1)
(1997) 48e54;
(c) D.D. Dominguez, H.N. Jones, T.M. Keller, Polym. Compos. 25 (5) (2004)
554e561.
[9] T.M. Keller, Chem. Mater. 6 (1994) 302e305.
[10] (a) R.P. Linstead, A.R. Lowe, J. Chem. Soc. (1934) 1022e1027;
(b) T.R. Walton, J.R. Griffith, Abstr. Pap. Am. Chem. S 169 (1975) 98.
[11] Y.Z. Meng, I.A. Abu-Yousef, A.R. Hlil, A.S. Hay, Macromolecules 33 (25) (2000)
9185e9191.
[12] (a) A.W. Snow, J.R. Griffith, N.P. Marullo, Macromolecules 17 (8) (1984)
1614e1624;
(b) K. Zeng, K. Zhou, S. Zhou, H. Hong, H. Zhou, Y. Wang, P. Miao, G. Yang, Eur.
Polym. J. 45 (4) (2009) 1328e1335;
(c) Z. Chen, X. Yang, M. Xu, X. Liu, High. Perform. Polym. 26 (1) (2013) 3e11.
[13] (a) P.J. Burchill, J. Polym. Sci. Part A Polym. Chem. 32 (1) (1994) 1e8;
(b) H. Sheng, X. Peng, H. Guo, X. Yu, C. Tang, X. Qu, Q. Zhang, Mater. Chem.
Phys. 142 (2e3) (2013) 740e747.
[14] D. Augustine, D. Mathew, C.P. Reghunadhan Nair, Polymer 60 (0) (2015)
308e317.
[15] (a) S.B. Sastri, T.M. Keller, J. Polym. Sci. Part A Polym. Chem. 36 (11) (1998)
1885e1890;
This paper mainly discussed the synthesis, cross-linking, phys-
ical properties, and advanced applications of novel phenyl-s-
triazine-containing phthalonitrile resins. The preparation of oligo-
meric precursor was achieved by a two-step, one-pot reaction with
high yields. Diamine 6 was added to elevate the curing rate, and the
diamine concentration and the temperature could readily control
the gelation time. The curing schedule was subsequently optimized
to give high cross-linking density networks with excellent thermal
and mechanical properties. The resulting networks (7) all
possessed high glass transition temperature at around 400 ꢀC,
outstanding thermal and thermo-oxidative stability as well as
limited water absorption (1.72e1.93 wt %). The 7/CF composite
materials exhibited high flexural strength of 1722 MPa and inter-
laminar shear strength of 71 MPa at ambient temperature. The
retention of value at 450 ꢀC was 15% and 51%, respectively. All of the
merits displayed above obviously suggest that the phenyl-s-
triazine-bearing phthalonitrile resins are qualified for the applica-
tions as high temperature structural materials.
(b) S.B. Sastri, T.M. Keller, J. Polym. Sci. Part A Polym. Chem. 37 (13) (1999)
2105e2111.
[16] (a) H. Zhou, A. Badashah, Z. Luo, F. Liu, T. Zhao, Polym. Adv. Technol. 22 (10)
(2011) 1459e1465;
(b) A. Badshah, M.R. Kessler, H. Zhou, A. Hasan, Polym. Int. 63 (3) (2014)
465e469.
[17] Z.B. Zhang, Z. Li, H. Zhou, X.K. Lin, T. Zhao, M.Y. Zhang, C.H. Xu, J. Appl. Polym.
Sci. 131 (20) (2014).
Acknowledgments
[18] (a) D.D. Dominguez, T.M. Keller, J. Appl. Polym. Sci. 110 (4) (2008)
2504e2515;
We thank the National High Technology Research and Devel-
opment Program (“863”Program) of China (Nos. 2015AA033802)
and the National Science Foundation of China (Nos. 21074017).
(b) H. Guo, Y. Zou, Z. Chen, J. Zhang, Y. Zhan, J. Yang, X. Liu, High. Perform.
Polym. 24 (7) (2012) 571e579;
(c) X. Zhao, Y. Lei, R. Zhao, J. Zhong, X. Liu, J. Appl. Polym. Sci. 123 (6) (2012)
3580e3586;
(d) X. Zhao, H. Guo, Y. Lei, R. Zhao, J. Zhong, X. Liu, J. Appl. Polym. Sci. 127 (6)
(2013) 4873e4878.
Appendix A. Supplementary data
[19] (a) Z. Brunovska, R. Lyon, H. Ishida, Thermochim. Acta 357e358 (2000)
195e203;
(b) F. Zuo, X. Liu, J. Appl. Polym. Sci. 117 (3) (2010) 1469e1475.
[20] (a) M.J. Sumner, M. Sankarapandian, J.E. McGrath, J.S. Riffle, U. Sorathia,
Polymer 43 (19) (2002) 5069e5076;
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.polymer.2015.09.035.
References
(b) H. Guo, Y. Lei, X. Zhao, X. Yang, R. Zhao, X. Liu, J. Appl. Polym. Sci. 125 (1)
(2012) 649e656;
(c) B. Zhang, Z. Luo, H. Zhou, F. Liu, R. Yu, Y. Pan, Y. Wang, T. Zhao, High
Perform. Polym. 24 (5) (2012) 398e404;
[1] (a) T. Agag, J. Liu, R. Graf, H.W. Spiess, H. Ishida, Macromolecules 45 (22)
(2012) 8991e8997;
(d) D. Augustine, D. Mathew, C.P.R. Nair, Polym. Int. 64 (1) (2015) 146e153.
[21] (a) R. Du, W. Li, X. Liu, Polym. Degrad. Stab. 94 (12) (2009) 2178e2183;
(b) X. Yang, J. Zhang, Y. Lei, J. Zhong, X. Liu, J. Appl. Polym. Sci. 121 (4) (2011)
2331e2337;
(c) H. Zhang, T. Liu, W. Yan, Y. Su, H. Yu, Y. Yang, Z. Jiang, High Perform. Polym.
26 (8) (2014) 1007e1014.
[22] K. Zeng, H.B. Hong, S.H. Zhou, D.M. Wu, P.K. Miao, Z.F. Huang, G. Yang, Poly-
mer 50 (21) (2009) 5002e5006.
[23] (a) T.M. Keller, Polymer 34 (1993) 952e955;
(b) P. Selvakumar, M. Sarojadevi, P. Sundararajan, Mater. Sci. Eng. B ADV 168
(1e3) (2010) 214e218.
[24] (a) M. Xu, X. Shi, X. Zou, H. Pan, M. Liu, K. Jia, X. Liu, J. Magn. Magn. Mater. 371
(2014) 20e28;
(b) M. Naffakh, A.M. Diez-Pascual, C. Marco, G.J. Ellis, M.A. Gomez-Fatou, Prog.
Polym. Sci. 38 (8) (2013) 1163e1231;
(c) C. Tridech, H.A. Maples, P. Robinson, A. Bismarck, ACS Appl. Mat. Interfaces
5 (18) (2013) 9111e9119;
(d) M. Antunes, J.I. Velasco, Prog. Polym. Sci. 39 (3) (2014) 486e509;
(e) I. Hamerton, L.T. McNamara, B.J. Howlin, P.A. Smith, P. Cross, S. Ward,
Macromolecules 47 (6) (2014) 1935e1945.
[2] (a) D.M. Wu, Y.C. Zhao, K. Zeng, G. Yang, J. Polym. Sci. Part A Polym. Chem. 50
(23) (2012) 4977e4982;
(b) X.Q. Zou, M.Z. Xu, K. Jia, X.B. Liu, J. Appl. Polym. Sci. 131 (23) (2014)
41203e41209;
(c) K. Zeng, Y. Zou, G. Yang, Des. Monomers Polym. 17 (2) (2014) 186e193;
(d) J. Hu, Y. Liu, Y. Jiao, S. Ji, R. Sun, P. Yuan, K. Zeng, X. Pu, G. Yang, RSC Adv. 5
(21) (2015) 16199e16206;
(b) L.F. Tong, Z.J. Pu, Z.R. Chen, X. Huang, X.B. Liu, Polym. Compos. 35 (2)
(2014) 344e350;
(c) M. Liu, K. Jia, X. Liu, J. Appl. Polym. Sci. 132 (10) (2015) 41595e41602;
(d) T.M. Robert, D. Augustine, S.M. Chandran, D. Mathew, C.P.R. Nair, RSC Adv.
5 (2) (2015) 1198e1204;
(e) P. Yuan, Y. Liu, K. Zeng, G. Yang, Des. Monomers Polym. 18 (4) (2015)
343e349.
[3] (a) D.D. Dominguez, T.M. Keller, Polymer 48 (2007) 91e97;
(b) B. Amir, H. Zhou, F. Liu, H. Aurangzeb, J. Polym. Sci. Part A Polym. Chem. 48
(24) (2010) 5916e5920;
(e) K. Jia, P. Wang, L. Yuan, X. Zhou, W. Chen, X. Liu, J. Mater. Chem. C 3 (15)
(2015) 3522e3529;
(c) M. Laskoski, A. Neal, T.M. Keller, D. Dominguez, C.A. Klug, A.P. Saab,
J. Polym. Sci. Part A Polym. Chem. 52 (12) (2014) 1662e1668;
(d) Y. Zhan, X. Yang, H. Guo, J. Yang, F. Meng, X. Liu, J. Mater. Chem. 22 (12)
(2012) 5602e5608;
(f) F. Meng, X. Liu, RSC Adv. 5 (10) (2015) 7018e7022.
[25] G.P. Yu, C. Liu, X.P. Li, J.Y. Wang, X.G. Jian, C.Y. Pan, Polym. Chem. 3 (4) (2012)
1024e1032.
[26] G.P. Yu, J.Y. Wang, C. Liu, E.C. Lin, X.G. Jian, Polymer 50 (7) (2009) 1700e1708.
[27] C. Liu, J.Y. Wang, E.C. Lin, L.S. Zong, X.G. Jian, Polym. Degrad. Stab. 97 (3)
(2012) 460e468.
[28] (a) G.P. Yu, J.Y. Wang, C. Liu, E.C. Lin, X.G. Jian, Polymer 50 (2009) 1700e1708;
(b) G.P. Yu, C. Liu, J.Y. Wang, G.H. Li, Y.J. Han, X.G. Jian, Polymer 51 (1) (2010)
(e) H. Guo, Y. Zhan, Z. Chen, F. Meng, J. Wei, X. Liu, J. Mater. Chem. A 1 (6)
(2013) 2286e2296;
(f) M. Laskoski, D.D. Dominguez, T.M. Keller, J. Polym. Sci. Part A Polym. Chem.
51 (22) (2013) 4774e4778.
[4] C.S. Marvel, J.H. Rassweiler, J. Am. Chem. Soc. 80 (5) (1958) 1197e1199.

188
L. Zong et al. / Polymer 77 (2015) 177e188
100e109.
[36] Z.S. Guo, S.Y. Du, B.M. Zhang, Z.J. Wu, J. Appl. Polym. Sci. 92 (5) (2004)
3338e3342.
[37] J.K. Herman Teo, K.C. Teo, B. Pan, Y. Xiao, X. Lu, Polymer 48 (19) (2007)
5671e5680.
[38] Y.Y. Han, G.X. Liao, Y.J. Xu, G.P. Yu, X.G. Jian, Polym. Eng. Sci. 49 (12) (2009)
2301e2308.
[39] A. Badshah, M.R. Kessler, H. Zhou, J.H. Zaidi, S. Hameed, A. Hasan, Polym.
Chem. 4 (12) (2013) 3617e3622.
[29] (a) D.M. Tigelaar, A.E. Palker, C.M. Jackson, K.M. Anderson, J. Wainright,
R.F. Savinell, Macromolecules 42 (6) (2009) 1888e1896;
(b) D.M. Tigelaar, A.E. Palker, R.H. He, D.A. Scheiman, T. Petek, R. Savinell,
M. Yoonessi, J. Membr. Sci. 369 (1e2) (2011) 455e465.
[30] (a) S. Matsuo, J. Polym. Sci. Part A Polym. Chem. 32 (11) (1994) 2093e2098;
(b) R. Fink, C. Frenz, M. Thelakkat, H.W. Schmidt, Macromolecules 30 (26)
(1997) 8177e8181;
(c) R. Fink, Y. Heischkel, M. Thelakkat, H.W. Schmidt, C. Jonda, M. Huppauff,
Chem. Mater 10 (11) (1998) 3620e3625.
[40] T. Liu, Y. Yang, T. Wang, H. Wang, H. Zhang, Y. Su, Z. Jiang, Polym. Eng. Sci. 54
(7) (2014) 1695e1703.
[31] M. Barikani, S. MehdipoureAtaei, J. Polym. Sci. Part A Polym. Chem. 38 (9)
(2000) 1487e1492.
[32] R. Singh, A.S. Hay, Macromolecules 25 (3) (1992) 1025e1032.
[33] D.D. Dominguez, T.M. Keller, High Perform. Polym. 18 (3) (2006) 283e304.
[34] (a) T.M. Keller, T.R. Price, J.R. Griffith, Abstr. Pap. Am. Chem. S 180 (1980)
165e166. Aug;
[41] M. Laskoski, D.D. Dominguez, T.M. Keller, Polymer 48 (21) (2007) 6234e6240.
[42] X. Peng, H. Sheng, H. Guo, K. Naito, X. Yu, H. Ding, X. Qu, Q. Zhang, High
Perform. Polym. 26 (7) (2014) 837e845.
[43] R.D. Vannucci, Sampe Q. Soc. Adv. Mater. Process Eng. 19 (1) (1987) 31e36.
[44] Y. Zou, J. Yang, Y. Zhan, X. Yang, J. Zhong, R. Zhao, X. Liu, J. Appl. Polym. Sci.
125 (5) (2012) 3829e3835.
(b) T.M. Keller, J. Polym. Sci. Part A Polym. Chem. 26 (12) (1988) 3199e3212.
[35] H.E. Kissinger, Anal. Chem. 29 (11) (1957) 1702e1706.
[45] J. Yang, X.L. Yang, Y.K. Zou, Y.Q. Zhan, R. Zhao, X.B. Liu, J. Appl. Polym. Sci. 126
(3) (2012) 1129e1135.