Thermally stable polybenzoxazines via tetrahydrophthalimide-functional monobenzoxazines: Synthesis, characterization and thermally activated polymerization kinetics

Article abstract of DOI:10.1016/j.tca.2019.03.006

Two benzoxazine monomers with attached tetrahydrophthalimide at para and ortho position have been synthesized via Mannich condensation. The chemical structures of obtained monomers have been confirmed by ¹H and ¹³C nuclear magnetic resonance spectroscopy (NMR) and Fourier transform infrared spectroscopy (FT-IR). The ring-opening polymerization has been examined by differential scanning calorimetry (DSC) analysis, which reveals that the ortho-tetrahydrophthalimide benzoxazine monomer exhibits lower polymerization temperature compared with its para-counterpart. Besides, the kinetic parameters have been determined by non-isothermal DSC at various heating rates. The apparent activation energy values of para- and ortho-tetrahydrophthalimide functional benzoxazine monomers are determined to be 107.12 and 91.00 kJ/mol, respectively, according to the Starink method. In addition, the polymerization processes of both monomers can be well described by autocatalytic kinetic models. The predicted curves based on the developed models from this study fit well with the experimental DSC thermograms. Moreover, the thermal properties of polybenzoxazines have also been evaluated by dynamic mechanical analysis (DMA) and thermogravimetric analysis (TGA).

Full text of DOI:10.1016/j.tca.2019.03.006

Accepted Manuscript
Title: Thermally stable polybenzoxazines via
tetrahydrophthalimide-functional monobenzoxazines:
Synthesis, characterization and thermally activated
polymerization kinetics
Authors: Xinye Yu, Zhikun Shang, Kan Zhang
PII:
DOI:
Reference:
S0040-6031(19)30025-5
https://doi.org/10.1016/j.tca.2019.03.006
TCA 78238
To appear in:
Thermochimica Acta
Received date:
Revised date:
Accepted date:
8 January 2019
5 March 2019
6 March 2019
Please cite this article as: Yu X, Shang Z, Zhang K, Thermally stable polybenzoxazines
via tetrahydrophthalimide-functional monobenzoxazines: Synthesis, characterization
and thermally activated polymerization kinetics, Thermochimica Acta (2019),
https://doi.org/10.1016/j.tca.2019.03.006
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Thermally
stable
polybenzoxazines
via
tetrahydrophthalimide-functional
monobenzoxazines:
Synthesis,
characterization and thermally activated polymerization kinetics
Xinye Yu, Zhikun Shang, Kan Zhang*
School of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, China.
*Corresponding author. E-mail address: zhangkan@ujs.edu.cn (K. Zhang).
Highlights
1. Two tetrahydrophthalimide-functional benzoxazines were successfully
synthesized.
2. Polymerization behavior and thermal properties were investigated.
3. Polymerization activation energy was calculated by Kissinger and
Starink methods.
4. The predicated autocatalytic models were in excellent agreement with
experimental results.
ABSTRACT: Two benzoxazine monomers with attached tetrahydrophthalimide at para
and ortho position have been synthesized via Mannich condensation. The chemical
1
13
structures of obtained monomers have been confirmed by H and C nuclear magnetic
resonance spectroscopy (NMR) and Fourier transform infrared spectroscopy (FT-IR). The
ring-opening polymerization has been examined by differential scanning calorimetry
(DSC) analysis, which reveals that the ortho-tetrahydrophthalimide benzoxazine
monomer exhibits lower polymerization temperature compared with its para-counterpart.
Besides, the kinetic parameters have been determined by non-isothermal DSC at various

heating rates. The apparent activation energy values of para- and
ortho-tetrahydrophthalimide functional benzoxazine monomers are determined to be
107.12 and 91.00 kJ/mol, respectively, according to the Starink method. In addition, the
polymerization processes of both monomers can be well described by autocatalytic
kinetic models. The predicted curves based on the developed models from this study fit
well with the experimental DSC thermograms. Moreover, the thermal properties of
polybenzoxazines have also been evaluated by dynamic mechanical analysis (DMA) and
thermogravimetric analysis (TGA).
Keyword: Benzoxazine; tetrahydrophthalimide; polymerization kinetics; thermal
stability.

1. Introduction
Benzoxazine resin is a relatively new thermoset which can be synthesized by Mannich
condensation from phenols, amines, and formaldehyde [1,2]. The corresponding
polymerized derivative of benzoxazine resin, namely polybenzoxazine, has been
extensively reviewed due to its outstanding advantages [3-6]. The major advantages of
polybenzoxazine include their low water absorption [7], high char yield [8], near-zero
shrinkage upon polymerization [9], low surface free energy [10,11], and excellent
mechanical and thermal properties [12,13]. These various characteristics of this class of
polymers are mainly attributed to the existence of inter/intra molecular hydrogen bonding
in the cross-linked network.
The molecular design flexibility offers myriad of modification in the structures simply
by replacing functionalized phenols or amines to achieve polybenzoxazines with desired
properties. To further improve the thermal stability of polybenzoxazines, benzoxazine
monomers with different functionalities have been synthesized. These various
preparations include the incorporation of functional groups such as maleimide [14],
norbornene [15], benzoxazole [16], phthalonitrile [17], allyloxy [18] and amide [19] into
the benzoxazine structures. In particular, polybenzoxazines derived from benzoxazines
consisting of phthalimide components have been shown to possess significant advantages
over the traditional benzoxazine resins [20-22]. However, the trade-off generally exists
between the thermal resistance and processability. The rigidity, originating from the
combination of imide structure with an aromatic structure, results in polybenzoxazines
with a very high glass transition temperature (Tg). Meanwhile, the high rigid chemical
structure also brings the difficulties in material processing due to the high melting
temperature and the narrow processing window [21]. Thus, developing new approaches
to improve the processability of imide functional benzoxazines is highly desired.
In this study, two benzoxazine monomers with tetrahydrophthalimide functionality
have been synthesized via Mannich condensation. The tetrahydrophthalimide

functionality is in the para and ortho position with respect to the oxygen group in the
oxazine ring of benzoxazines. As a hydrogenated product of phthalimide group,
tetrahydrophthalimide can significantly reduce the rigidity of benzene end-cap in
phthalimide benzoxazines, resulting in the easy synthesis and processability of
benzoxazine resins without sacrificing too much in thermal properties of the
corresponding polybenzoxazines. Besides, the polymerization kinetics of
tetrahydrophthalimide functional benzoxazines have been systematically investigated via
the non-isothermal DSC technique. As one of the most utilized methods to determine the
polymerization kinetic parameters, non-isothermal DSC method has been carried out at
various heating rates to evaluate the polymerization kinetics for benzoxazine
reins.[17,23,24] The predicted curves based on our developed models in this study fit
well with the experimental DSC thermograms. Notably, the highly purified monomers
were used in the current kinetic study in order to minimize the impurity-related variation
of the data within the same compound. The detailed synthetic strategy, characterization,
polymerization kinetics and thermal properties of tetrahydrophthalimide functional
benzoxazines are discussed in this article.
2. Experimental
2.1 Materials
Cis-1,2,3,6-tetrahydrophthalic
anhydride
(99%),
p-aminophenol
(99%),
paraformaldehyde (99%), and o-aminophenol (99%) were purchased from Sigma-Aldrich.
Acetic acid, xylenes, aniline were received from Sahn Chemical Technologies.
2-(3-phenyl-3,4-dihydro-2H-benzo[e][1,3]oxazin-8-yl)-3a,4,7,7a-tetrahydro-1H-isoindole
-1,3(2H)-dione (abbreviated as oHTI-a), was synthesized following our recent reported
method [22].
2.2 Methods
Synthesis of 2-(4-hydroxyphenyl)-3a,4,7,7a-tetrahydro-1H-isoindole-1,3(2H)-dione

(abbreviated as pHTI). Into
a
250 mL round flask were placed
cis-1,2,3,6-tetrahydrophthalic anhydride (4.56 g, 0.03 mol), p-aminophenol (3.27 g, 0.03
mol), and 80 ml of acetic acid. The mixture was stirred and refluxed for 6 h. After cooling
to room temperature, the precipitate was filtered and washed with 500 mL of deionized
water. Removal of water by evaporation afforded a white powder. The resulting white
1
product was recrystallized from isopropanol. (yield ca. 91%). H NMR (400 MHz,
DMSO-d6), ppm: δ = 2.22-2.27 (m, 2H, -CH2-), 2.40-2.49 (m, 2H, -CH2-), 3.22 (t, 2H),
5.95 (t, 2H, =C-H), 6.86-6.99 (4H, Ar), 9.76 (s, 1H, -OH).
Synthesis
of
2-(3-phenyl-3,4-dihydro-2H-benzo[e][1,3]oxazin-6-yl)-3a,4,7,7a-tetrahydro-1H-isoin
dole-1,3(2H)-dione (abbreviated as pHTI-a). Into a 100 mL round flask were added 30
mL of xylenes, aniline (0.93 g, 0.01 mol), pHTI (2.43 g, 0.01 mol), and
o
paraformaldehyde (0.61 g, 0.02 mol). The mixture was stirred at 120 C for 6 h. The
mixture was cooled to room temperature and yellow crystal was obtained. (yield ca. 86%).
1H NMR (CDCl3), ppm: δ = 2.29-2.33 (m, 2H, -CH2-), 2.70-2.75 (m, 2H, -CH2-), 3.26 (t,
2H), 4.67 (s, 2H, Ar-CH2-N, oxazine), 5.38 (s, 2H, O-CH2-N, oxazine), 6.00 (t, 2H,
13
=C-H), 6.77-7.37 (8H, Ar). C NMR (CDCl3), ppm: δ =50.48 (1C, Ar-CH2-N-), 79.69
(1C, -O-CH2-N-). FT-IR spectra (KBr), cm-1: 1771, 1706 (imide I), 1497 (stretching of
trisubstituted benzene ring), 1388 (imide II), 1239 (C-O-C asymmetric stretching), 942
(out-of-plane C-H), 697 (out-of-plane =C-H).
Anal. Calcd. for C22H20N2O3: C, 73.32; H, 5.59; N, 7.77. Found: C, 73.31%; H, 5.60%; N,
7.74%.
Polymerization of tetrahydrophthalimide functional benzoxazine monomers. Both
tetrahydrophthalimide functional benzoxazine monomers were thermally activated
polymerized at different temperature cycles without adding any initiators or catalysts. The
benzoxazine reins were polymerized stepwise at 120, 140, 160, 180, 200, 220 and 240 oC
for 1 h each. Afterwards, the samples were slowly cooled to room temperature, and the

dark brown polybenzoxazine films were obtained.
2.3 Characterization
1
The H (400 MHz) and 13C (100 MHz) nuclear magnetic resonance (NMR) spectra
were applied to characterize the structures of each products dissolved in CDCl3 using the
Bruker AVANCE II 400 NMR spectrometer with tetramethylsilane as internal standard.
The average numbers of transients for 1H NMR and 13C NMR measurements were setting
as 64 and 1024. FTIR spectra of each simple were obtained using the Nicolet Nexus 670
Fourier transform infrared (FTIR) spectrophotometer with a range of 4000-500 cm-1.
Differential scanning calorimetry (DSC) was carried out by NETZSCH DSC of model
204f1. To study the polymerization kinetics, benzoxazine monomers were analyzed at a
temperature increase rates of 2, 5, 10, 15, and 20 °C/min from room temperature (RT) to
400 °C at a constant nitrogen flow rate of 50 mL/min. The dynamic mechanical analysis
(DMA) was performed on a NETZSCH DMA/242E analyzer in tension mode. An
amplitude of 10 µm and a temperature ramp rate of 3oC/min at a frequency of 1Hz was
applied during DMA testing. The thermal decomposition temperatures and char yields of
polybenzoxazines were measured using a NETZSCH STA449-C Thermogravimetric
o
Analyzer. Thermogravimetric analysis (TGA) was performed from RT to 800 C at a
heating rate of 10 oC/min with a flow rate of 40 mL/min under N2 atmosphere.
2.4 Kinetic analysis
The
kinetics
of
non-isothermal
polymerization
behaviors
for
tetrahydrophthalimide-functional benzoxazine monomers are conducted according to the
following rate equation:
푑훼 = 푘(푇)푓(훼)
푑푡
(1)
Where, α is the conversion rate, t is time, k(T) is the temperature dependent reaction rate
constant, and f(α) is the differential conversation function depending on polymerization
mechanisms. Herein, the rate constant, k(T), can be determined according to the
Arrhenius equation as shown in follows:

−퐸푎
( )
푘 푇 = 퐴exp( 푅푇
)
(2)
In terms of non-isothermal reaction condition, various methods have been adopted to
calculate the apparent activation energy, such as Friedman method [25], Ozawa method
[26], Kissinger method [27] and Starink method [23,24]. Among those reported
approaches, the Kissinger method is widely used to calculate the apparent activation
energy of benzoxazine resins due to its accuracy [28,29]. The equation of Kissinger
method is as follows:
퐸
1
푎
2
( ⁄
)
ln 훽 푇푃 = −
∙
+ 퐶
푇
푝
푅
(3)
where, β is the heating rate, Tp is the peak polymerization temperature, Ea is the apparent
activation energy of the reaction, R is the gas constant, 퐶 = ln
(퐴퐸푅), and A is the
푎
frequency factor.
However, the Kissinger method process as a single-step kinetics thus always yields a
single Ea. The Starink method can be used to test how Ea varies with conversation to
study the multi-step kinetic process [23,24]. The Starink method equation is as follows:
훽
푙푛 (푇1.92) = 퐶 − 1.0008( 퐸
)
(4)
훼
푅푇
훼
훼
where, β is the heating rate, Tα is the temperature at conversion α.
Scheme 1. Synthesis of tetrahydrophthalimide-functional benzoxazine monomers.

2.5 Synthesis of tetrahydrophthalimide-functional benzoxazine monomers
The successful synthesis of tetrahydrophthalimide-functional monobenzoxazines has
been achieved using primary amine (aniline), paraformaldehyde, and
tetrahydrophthalimide functional phenols (pHTI and oHTI) as shown in Scheme 1. Prior
to this study, various para-phthalimide functional benzoxazines have been attempted
[30,31]. However, incorporation of high rigid phthalimide functionality into benzoxazine
structure at para position required long reaction time, leading the difficulties in
benzoxazine synthesis. Herein, the reaction for synthesizing para-tetrahydrophthalimide
o
functional benzoxazine was completed in just six hours at 120 C in xylenes. The
tetrahydrophthalimide functionality has less rigidity compared with phthalimide, which
significantly increases the solubility of tetrahydrophthalimide-related compounds in
ordinary organic solvents, such as chloroform, dichloromethane, ethyl acetate and
dimethylacetamide (3.0%: w/v). As a result, both pHTI-a and oHTI-a exhibit great
advantages in synthesis in terms of imide-functional benzoxazines.

Fig. 1. 1H NMR spectra of pHTI-a (a) and oHTI-a (b).
1
The chemical structures of pHTI-a and oHTI-a were confirmed using H NMR, 13C
NMR and FTIR analyses. As depicted in Fig. 1, the typical resonances attributed to the
benzoxazine structure, Ar-CH2-N- and -O-CH2-N- for pHTI-a are observed at 4.67 and
5.38 ppm, respectively. However, two sets of doublets at around 4.65 and 5.32 ppm for
the typical resonances for oHTI-a can be observed, which is due to the existence of
atropisomerism in oHTI-a [22]. Besides, the protons of the vinylene group in both pHTI-a
and oHTI-a appear at 6.00 ppm. In addition, the 13C NMR spectra shown in Fig. 2 are

applied to identify the characteristic signal of oxazine ring and tetrahydrophthalimide
group. The characteristic carbon resonances of the oxazine ring in pHTI-a and oHTI-a are
assigned at 50.48 and 79.69 ppm, and 50.70 and 80.00 ppm, respectively.
Fig. 2. 13C NMR spectra of pHTI-a (a) and oHTI-a (b).
To further verify the chemical structures of pHTI-a and oHTI-a, FT-IR spectra of
benzoxazine monomers have been performed, and the results are shown in Fig. 3. A
number of infrared absorption bands are highlighted in Fig. 3 that are used to verify the
existence of oxazine ring and tetrahydrophthalimide in each monomer. The characteristic
doublets at 1771 and 1782 cm-1, and 1706 and 1709 cm-1 are the typical bands for imide
group, which are attributed to the imide C-C(=O)-C antisymmetric and symmetric
stretching for pHTI-a and oHTI-a, respectively [32]. Additionally, the presence of imide

is also determined by the characteristic bands at 1388 and 1389 cm-1 for pHTI-a and
oHTI-a, respectively, which are due to the axial stretching of C-N bonding [33,34]. The
characteristic bands of antisymmetric trisubstituted benzene that appear at 1497 and 1488
cm-1, confirming the incorporation of imide group into benzoxazine structure in both
benzoxazines. Moreover, the presence of the benzoxazine ring aromatic ether in the
monomers is indicated by band centered at 1239 and 1236 cm-1 due to the C-O-C
antisymmetric stretching modes [35]. Furthermore, the characteristic benzoxazine related
band can be observed at 942 and 939 cm-1 for pHTI-a and oHTI-a, respectively [36].
Fig. 3. FT-IR spectra of benzoxazine monomers.
3. Results and discussion
3.1 Polymerization behaviors of benzoxazine monomers
The thermal behaviors of benzoxazine monomers were studied by DSC at the heating
o
rate of 10 C/min under a nitrogen atmosphere. The thermograms of pHTI-a and oHTI-a
are shown in Fig. 4a. Very sharp melting peaks at 180 and 154 oC can be observed for
pHTI-a and oHTI-a, respectively, further indicating the excellent purity of both
monomers. Besides, two endothermic peaks can be found for oHTI-a during the melting
behavior, which is caused by the phase change of crystal-to-crystal transition [22]. As can

o
be seen in the figure, there is one exothermic peak with its maximum at 267 C and an
onset temperature of about 249 oC for pHTI-a. oHTI-a exhibits a lower temperature (252
oC) for the maximum of exothermic peak compared with pHTI-a. The DSC results in the
current study give another sample to indicate the ortho-functional benzoxazines possess
significant advantages over the para-counterparts in polymerization process [21,29].
Additionally, the temperatures of processing window from the DSC thermograms for
o
pHTI-a and oHTI-a are 87 and 98 C, which are much higher than that of reported
phthalimide functional benzoxazines [21]. Moreover, the completion for the curing of
both monomers can be determined by the compete disappearance of exothermic peaks as
shown in Figure 4b. Both polybenzoxazines show the characteristic of brittleness, which
is similar as other polybenzoxazines dervied from monofunctional benzoxazine resins.
Fig. 4. (a) DSC polymerization behavior of mono-functional benzoxazine monomers. (b)
DSC thermograms of polybenzoxazines.
3.2 Polymerization kinetics
Fig. 5 shows the DSC thermograms of pHTI-a and oHTI-a at heating rates of 2, 5, 10,
o
15 and 20 C/min, respectively. It can be observed that the initial polymerization
temperature (Ti) and exothermic peak temperature (Tp) shift to higher temperatures as
increasing the heating rate. The polymerization parameters, including Ti, Tp and terminal

polymerization temperature (Tf) of pHTI-a and oHTI-a are summarized as shown in
Tables 1 and 2.
Fig. 5. DSC curves of pHTI-a (a) and oHTI-a (b) at different heating rates.
Table 1 The characteristic temperatures of pHTI-a at different heating rates.
Heating rate β (oC/min)
Ti/ oC
212.8
225.3
232.4
241.4
245.2
Tp/ oC
234.5
253.9
267.5
275.2
281.3
Tf/ oC
256.8
284.0
289.9
295.4
307.2
2
5
10
15
20
Table 2 The characteristic temperatures of oHTI-a at different heating rates.
Heating rate β (oC/min)
Ti/ oC
190.5
199.3
212.7
221.6
229.0
Tp/ oC
216.4
235.8
252.2
260.8
267.0
Tf/ oC
238.0
256.8
271.6
277.7
284.0
2
5
10
15
20

The apparent activation energy (Ea) values of benzoxazine monomers were then
calculated by Kissinger method [25]. Basing on the data in Tables 1 and 2, the plots of
2
ln(β/Tp ) against 1000/Tp are produced as shown in Fig. 6 for both pHTI-a and oHTI-a. As
a result, the apparent Ea values of pHTI-a and oHTI-a are estimated to be 106.83 and
90.78 kJ/mol, respectively, according to Kissinger method. The estimated apparent
activation energy values are lower than that of some reported benzoxazines based on
Kissinger method, such as 247 KJ/mol for imide functional benzoxazine and 169 KJ/mol
for amide functional benzoxazine [37]. These relatively lower values of activation energy
indicate that both tetrahydrophthalimide functional benzoxazine monomers are easy to be
activated to polymerize. In addition, oHTI-a also shows a relatively lower Ea value
compared with pHTI-a. The ortho-attached electro-withdrawing group of
tetrahydrophthalimide may lead a more acidic phenol species, and therefore result in a
lower apparent activation energy during the polymerization of oHTI-a.
Fig. 6. Plots generated following the Kissinger method for determination the apparent
activation energy values of pHTI-a and oHTI-a.

In addition, the Starink method was also carried out to see how Ea varies with
conversation during the polymerization process. From the equation (4), a straight line
1.92
( ⁄
)
vs. 1000/Tα
with a slope of -1.0008Eα/R can be obtained by taking a plot of 푙푛 훽 푇훼
in the same conversion range for a series of dynamic DSC experiments at different
1.92
( ⁄
)
vs. 1000/Tα with a
heating rates for both pHTI-a and oHTI-a. The plots of 푙푛 훽 푇훼
conversion rate between 0.1 and 0.9 are shown in the Fig. 7. The plots of both pHTI-a
and oHTI-a are obviously in good linear relationship.
1.92
⁄
Fig. 7. 푙푛 훽 푇훼
vs. 1000/Tα at different conversion rates of pHTI-a (a) and oHTI-a (b).
Therefore, the Eα values at different conversion rates can be obtained through slopes at
different conversion rates as shown in Fig. 8. The average apparent activation energy
values for pHTI-a and oHTI-a were calculated as 107.12 and 91.00 kJ/mol, respectively.
This results for Eα values of both benzoxazine monomers are very close to the values
obtained from the Kissinger method.

Fig. 8. Variation of Eα vs. α by Starink method for pHTI-a and oHTI-a.
Generally, the thermally activated polymerization mechanism of thermosetting resin
can be divided into two types: one is the nth-order reaction, and the other one is the
autocatalytic reaction [38,39]. In case of nth-order reaction:
푛
( )
푓 훼 = (1 − 훼)
(5)
And in the case of autocatalytic reaction:
푚
푛
( ) )
(
푓 훼 = 훼 1 − 훼
(6)
where, n is the number of reaction stages, m is the number of autocatalytic series.
According to the basic kinetic theory, the reaction model determines the shape of the
kinetic curve, that is, the shape of the kinetic curve can be used to distinguish the
polymerization reaction model of benzoxazine reins. Fig. 9 shows the variation of the
conversion degree 훼 as a function of temperature for pHTI-a and oHTI-a at different
heating rates.

Fig. 9. Conversion⁡훼 as a function of temperature at different heating rates of pHTI-a
(a) and oHTI-a (b).
To further determine the polymerization kinetic models of oHTI-a and pHTI-a, the
relationship between the reaction rate dα/dt and the conversion rate α was obtained at a
heating rate of 15 oC/min as shown in Fig. 10. It can be observed in Fig. 10 that the
maximum reaction rates for both oHTI-a and pHTI-a occur at an intermediate conversion
degree, which suggests that the polymerization processes of pHTI-a and oHTI-a follow
the autocatalytic reaction mechanism. Ordinarily, the nth-order kinetic model shows the
maximum reaction rate at the beginning of polymerization. Therefore, pHTI-a and
oHTI-a reach maximum reaction rate at the conversion values of 0.53 and 0.57,
respectively.

Fig. 10. dα/dt vs. α at the heating rate of 15 oC/min for pHTI-a and oHTI-a.
Although the reaction mechanism of pHTI-a and oHTI-a was preliminarily judged
according to the relationships of conversion vs. temperature and reaction rate vs.
conversion, a further more detailed investigation was carried out by Friedman method
[40,41]. The Friedman equation is as follows:
푑훼
푑훼
퐸
푎
[
( )]
ln 푑푡 = ln훽 푑푇 = ln 퐴푓 훼 − 푅푇
(7)
The Friedman equation turns into Equation (8) by the combination of the nth-order
reaction equation as aforementioned:
푑훼
퐸
푎
[
( )]
ln 퐴푓 훼 = ln훽 푑푇
+
푅푇 = 푙푛퐴 + 푛푙푛(1 − 훼)
(8)
where, each character represents the same meaning as above. Notably, the Ea in Equation
(8) was the values obtained from above apparent activation energy calculated by
Kissinger method.

푑훼 = 훽퐴 exp
(− 푅푇)훼푚(1 − 훼)푛
According to the Equation (8), it should form a linear relationship between ln[Af(α)]
and ln(1-α) if the polymerization mechanisms of both pHTI-a and oHTI-a follow the
nth-order reaction. Fig. 11 shows the plots of ln[Af(α)] against ln(1-α) for pHTI-a (Fig.
11a) and oHTI-a (Fig. 11b). Clearly as seen in Fig. 11, the ln[Af(α)] of the two monomers
do not possess the simple linear relationship with ln(1-α). The value of ln[Af(α)] reaches
the maximum and then decreases as the value of ln(1-α) increases for both monomers.
The results clearly indicate that the polymerization processes of pHTI-a and oHTI-a
follow an autocatalytic reaction model rather than the nth-order reaction [38].
Fig. 11. ln[Af(α)] vs. ln(1-α) at different heating rates of pHTI-a (a) and oHTI-a (b).
Afterwards, by substituting Equation (6) in Equation (8), a new Equation (9) can be
obtained as follows:
퐸
푎
푑푇
(9)
Equation (9) can be turned into Equation (10) by taking logarithm transformation:
푑훼
퐸
푎
(
)
( )
ln (훽 푑푇) = − 푅푇 + 푛푙푛 1 − 훼 + 푚푙푛 훼 + 푙푛퐴
(10)
Equation (10) can be solved by multiple linear regression, in which the dependence
푑훼
(
)
( )
variable is ln (푑푇) , and the independent variables are 1/T, 푙푛 1 − 훼 ⁡and 푙푛 훼 .Then
the values of n, m, and A can be obtained using the average apparent activation energy

from Kissinger method. Herein, the interval for calculation was selected from the initial
stage to the conversion α reaching the maximum reaction rate (α=0.1-0.5). The data of
kinetic parameters for pHTI-a and oHTI-a are listed in Table 3 and Table 4, respectively.
Table 3 Polymerization kinetic parameters of pHTI-a.
β
Eα
Correlation
lnA
Standard
Error
Standard
Error
Standard
Error
n
m
(℃/min) (kJ/mol) Coefficient(R2) (s-1)
2
0.9643
0.9716
0.9623
0.9749
0.9833
0.9713
24.4100 0.1179
24.4610 0.1202
24.4944 0.2113
24.6530 0.1248
24.6360 0.0870
0.9524 0.0286
1.0533 0.0229
0.8746 0.0407
1.0037 0.0214
0.9734 0.0165
0.7022 0.0259
0.7385 0.0230
0.7099 0.0492
0.7950 0.0306
0.7618 0.0167
5
10
15
20
107.12
Average 107.12
24.53
0.1322
0.98
0.0260
0.74
0.0291
Table 4 Polymerization kinetic parameters of oHTI-a.
β
Eα
Correlation
lnA
Standard
Error
Standard
Error
Standard
Error
n
m
(℃/min) (kJ/mol) Coefficient(R2) (s-1)
2
0.9707
0.9738
0.9368
0.9558
0.9547
0.9584
21.6170 0.1640
21.5785 0.1548
21.3972 0.2227
21.4909 0.1806
21.5499 0.1821
1.0331 0.0450
0.9613 0.0414
0.8467 0.0534
0.8797 0.0544
0.8293 0.0487
0.7329 0.0363
0.7677 0.0275
0.7061 0.0482
0.7639 0.0328
0.7516 0.0366
5
10
15
20
91.00
Average 91.00
21.53
0.1808
0.91
0.0486
0.74
0.0363
From the parameters in above tables, the kinetic equations for pHTI-a and oHTI-a can
be obtained as follows, respectively.
⁡⁡푑훼 = 훽 = 푒24.53exp
(−107120/푅푇)훼0.74(1 − 훼)0.98
푑훼
푑푡
푑푇
(11)

푑훼 = 훽 = 푒21.53exp
(−91000/푅푇)훼0.74(1 − 훼)0.91
푑훼
푑푡
푑푇
(12)
In order to further verify the mathematical models shown in Equations (11) and (12), a
comparison between the predicted and experimental results for pHTI-a and oHTI-a was
carried out. As shown in Fig. 10, the calculated data from the model equations are in good
agreement with the experimental results.
Fig. 12. Calculated (symbol) and experimental (solid line) DSC curves corresponding to
the polymerization processes of pHTI-a (a) and oHTI-a (b).
3.3 Thermal properties of polybenzoxazines
Dynamic mechanical analysis (DMA) has been performed to study the
thermomechanical properties of both polybenzoxazines obtained in this study. The DMA
results of for the storage modulus (E’), loss modulus (E”), and tan δ are shown in Fig. 13.
As seen from Fig. 13a, The glass temperature (Tg) determined as the peak temperature of
tan δ for poly(pHTI-a) is around 215 oC. In addition, the polybenzoxazine derived from
the ortho-benzoxazine isomer has a higher Tg temperature of 241 oC as seem from Fig.
13b.

Fig. 13. Dynamic mechanical spectra of poly(pHTI-a) (a) and poly(oHTI-a) (b).
TGA was carried out to evaluate the thermal stability of polybenzoxazines from room
temperature to 800 °C under nitrogen atmosphere. Thermal stability was determined by
initial decomposition temperatures, the temperature at 5% (Td5) and 10% (Td10) weight
o
loss and char yield (Yc) at 800 C. As shown in Fig. 14, the Td5 values of poly(pHTI-a)
and poly(oHTI-a) are 334 and 305 °C, respectively. Besides, the Td10 values of
o
poly(pHTI-a) and poly(oHTI-a) are 377 and 346 C, respectively. The polybenzoxazine
derived from para-functional isomer shows higher Td5 and Td10 values than ortho-isomer.
o
The weight loss difference in the range from 300 to 400 C is possible caused by the
benzoxazole formation from ortho-hydroxyimide in poly(oHTI-a) as other reported

ortho-imide functional benzoxazines [21]. Moreover, poly(pHTI-a) also exhibits
improved thermal stability as compared to typical bisphenol A based benzoxazine (Td5
o
326 and Td10 344 C) [42]. However, the derivative weight loss curves in Fig. 14 show a
o
high weight-loss stage in the range of 400 to 550 C for poly(pHTI-a). On the contrary, a
much lower weight-loss stage can be observed for poly(oHTI-a) at the same temperature
range. As a result, the Yc value of polybenzoxazine based on ortho-isomer (47%) is much
higher than the polybenzoxazine derived from para-isomer (28%). Furthermore, the TGA
results suggest the possibility to form thermally stable thermosets from low molecular
weight monobenzoxazines.
Fig. 14. Thermogravimetric analysis of polybenzoxazines.
4. Conclusions
Two benzoxazine monomers with attached tetrahydrophthalimide at para- and
ortho-position have been synthesized in this study. The structure of new achieved
1
para-tetrahydrophthalimide functional benzoxazine monomer was confirmed by H and
13C NMR and FT-IR spectra. The DSC results showed the high purity and excellent
processability of these benzoxazine monomers. Tetrahydrophthalimide can significantly
reduce the rigidity of phthalimide group, which results an easier approach for

synthesizing benzoxazines while still maintaining excellent thermal stability of
polybenzoxazines. In addition, the kinetic models of benzoxazine monomers were
evaluated by multi-heating-rate DSC methods, and two autocatalytic models for the
polymerization processes of both para and ortho benzoxazine isomers were confirmed.
Moreover, the calculated mathematical models were in excellent agreement with the
experimental data from non-isothermal DSC results.
Acknowledgments
The authors express their gratitude to the National Natural Science Foundation of
China (NSFC) for their financial support (No. 51603093). This project is also supported
by the Science and Technology Agency of Jiangsu Province (No. BK 20160515) and
China Postdoctoral Science Foundation (No. 2016M600369 and No. 2018T110451).
References
[1] F.W. Holly, A.C. Cope A, Condensation products of aldehydes and ketones with
o-aminobenzyl alcohol and o-hydroxybenzylamine, J. Am. Chem. Soc. 66 (11)
(1944) 1875-1879.
[2] X. Ning and H. Ishida, Phenolic materials via ring-opening polymerization:
Synthesis and characterization of bisphenol-A based benzoxazines and their
polymers, J. Polym. Sci., Part A: Polym. Chem. 32 (6) (1994) 1121-1129.
[3] H. Ishida, Chapter 1-Overview and historical background of polybenzoxazine
research, in: H. Ishida, T. Agag (Eds), Handbook of Benzoxazine Resins. Elsevier,
Amsterdam, 2011, pp. 3-81.
[4] C.P.R. Nai, Advances in addition-cure phenolic resins, Prog. Polym. Sci. 29 (5)
(2004) 401–498.
[5] N.N. Ghosh, B. Kiskan, Y. Yagci, Polybenzoxazines-new high performance

thermosetting resins: synthesis and properties, Prog. Polym. Sci. 32 (11) (2007)
1344–1391.
[6] T. Takeichi, T. Kawauchi, T. Agag, High performance polybenzoxazines as a novel
type of phenolic resin, Polym. J. 40 (12) (2008) 1121–1231.
[7] H. Ishida, D. Allan, Physical and mechanical characterization of near-zero shrinkage
polybenzoxazines, J. Polym. Sci., Part B: Polym. Phys. 34 (6) (2015) 1019-1030.
[8] H.J. Kim, Z. Brunovska, H. Ishida, Synthesis and thermal characterization of
polybenzoxazines based on acetylene-functional monomers, Polymer 40 (23) (1999)
6565-6573.
[9] H.I. And, H.Y. Low, A study on the volumetric expansion of benzoxazine-based
phenolic resin, Macromolecules 30 (4) (1997) 1099-1106.
[10] C.F. Wang, Y.C. Su, S.W. Kuo, C.F. Huang, Y.C. Sheen, F.C. Chang,
Low-Surface-Free-Energy Materials Based on Polybenzoxazines, Angew. Chem. Int.
Ed. 45 (14) (2006) 2248-2251.
[11] S.W. Kuo, Y.C. Wu, C.F. Wang, K.U. Jeong, Preparing low-surface-energy polymer
materials by minimizing intermolecular hydrogen-bonding interactions, J. Phys.
Chem. C. 113 (48) (2009) 20666-20673.
[12] X. Xiong, Z. Zhang, R. Ren, X. Cui, P. Chen, Alkynl-functionalized benzoxazine
containing phthalide side group: Synthesis, characterization and curing mechanism.
Polym. Test. 72 (2018) 232-237.
[13] K. Zhang, J. Liu, H. Ishida, An Ultrahigh Performance Cross-Linked
Polybenzoxazole via Thermal Conversion from Poly (benzoxazine amic acid) Based
on Smart o-Benzoxazine Chemistry, Macromolecules 47 (24) (2014) 8674-8681.
[14] L. Jin, T. Agag, H. Ishida, Bis(benzoxazine-maleimide)s as a novel class of high
performance resin: synthesis and properties, Eur. Polym. J. 46 (2) (2010) 354-363.

[15] K. Zhang, H. Ishida, Thermally stable polybenzoxazines via ortho-norbornene
functional benzoxazine monomers: Unique advantages in monomer synthesis,
processing and polymer properties, Polymer 66 (2015) 240-248.
[16] P. Yang, Y. Gu, Synthesis of a benzoxazine containing benzoxazole structure, Chin.
Chem. Lett. 21 (5) (2010) 558-562.
[17] M. Xu, Y. Luo, Y. Lei, X. Liu, Phthalonitrile-based resin for advanced composite
materials: Curing behavior studies, Polym. Test. 55 (2016) 38-43.
[18] Y. Tang, Q.L. Zhao, K. Zeng, P.K. Miao, K. Zhou, W.R. Tang, H.F. Zhou, T. Liu, Y.P.
Wang, G. Yang, Synthesis of a benzoxazine monomer containing maleimide and
allyloxy groups, Chin. Chem. Lett. 18 (8) (2007) 973-976.
[19]L. Han, K. Zhang, H. Ishida, P. Froimowicz, Study of the Effects of Intramolecular
and Intermolecular Hydrogen-Bonding Systems on the Polymerization of
Amide-Containing Benzoxazines, Macromol. Chem. Phys. 218 (18) (2017),
1600562.
[20] M.W. Wang, C.H. Lin, T.Y. Juang, Steric hindrance control synthesis of primary
amine-containing benzoxazines and properties of the resulting poly (benzoxazine
imide) thermosetting films, Macromolecules 46 (22) (2013) 8853-8863.
[21] K. Zhang, J. Liu, S. Ohashi, X. Liu, Z. Han, H. Ishida, Synthesis of high thermal
stability polybenzoxazoles via ortho-imide-functional benzoxazine monomers, J.
Polym. Sci., Part A: Polym. Chem. 53 (11) (2015) 1330-1338.
[22] K. Zhang, Z. K. Shang, C. J. Evans, L. Han, H. Ishida, S.F. Yang, Benzoxazine
Atropisomers: Intrinsic Atropisomerization Mechanism and Conversion to High
Performance Thermosets, Macromolecules 51 (2018) 7574-7585.
[23] T. Zhang, J. Wang, M. Derradji, N. Ramdani, H. Wang, Z. W. Lin, W. Liu, Synthesis,
curing kinetics and thermal properties of a novel self-promoted fluorene-based
bisphthalonitrile monomer, Thermochim. Acta. 602 (2015) 22-29.
[24] L. Sun, K. Zhang, C. Y. Min, Y. Q. Liu, Y. T. Wang, J. X. Zhang, S. J. Li, Synthesis,

characterization and structural thermally rearrangement of orthoamide functional
benzoxazine containing acetylene group, Thermochim. Acta. 668 (2018) 1-8.
[25] H. Friedman, Kinetics and gaseous products of thermal decomposition of
polymers, J. Polym. Sci., Part A: Polym. Chem. 1 (1) (1967) 57-79.
[26] T. Ozawa, Kinetic analysis of derivative curves in thermal analysis, J. Therm. Anal.
Calorim. 2 (3) (1970) 301-324.
[27] H. E. Kissinger, Reaction kinetics in differential thermal analysis, Anal. Chem. 29
(11) (1957) 1702-1706.
[28] S. Vyazovkin, A. K. Burnham, J. M. Criado, L. A. Pérez-Maqueda, C. Popescu, N.
Sbirrazzuoli. Ictac kinetics committee recommendations for performing kinetic
computations on thermal analysis data, Thermochim. Acta. 520 (1) (2011) 1-19.
[29] K. Zhang, H. Ishida, Thermally stable polybenzoxazines via ortho -norbornene
functional benzoxazine monomers: Unique advantages in monomer synthesis,
processing and polymer properties, Polymer 66 (2015) 240-248.
[30] K. Zhang, J. Liu, H. Ishida, High performance crosslinked polyimide based on
main-chain type polybenzoxazine RSC. Adv. 4 (2014) 62550-62556.
[31] Y. Gao, F. Huang, Q. Yuan, Y. Zhou, L. Du, Synthesis of imide-functionalized
fluorinated benzoxazines and properties of their thermosets, High. Perform. Polym.
25 (2013) 677-684.
[32] K. Zhang, Y. Liu, Z. Shang, C. J. Evans, S. Yang. Effects of end-caps on the
atropisomerism, polymerization, and the thermal properties of ortho-imide
functional benzoxazines. 11 (3) (2019) 339.
[33] C. Min, D. Liu, C. Shen, Q. Zhang, H. Song, S. Li, X. Shen, M. Zhu, K. Zhang.
Unique synergistic effects of graphene oxide and carbon nanotube hybrids on the
tribological properties of polyimide nanocomposites. Tribol. Int. 117 (2018)
217-224.
[34] B.T. Low, Y. Xiao, S. C. Tai, L. Ye, Simultaneous occurrence of chemical grafting,

cross-linking, and etching on the surface of polyimide membranes and their impact
on H2/CO2 separation, Macromolecules 41 (4) (2008) 1297-1309.
[35] T. Agag, T. Takeichi, Synthesis and characterization of benzoxazine monomers
containing allyl groups and their high performance thermosets, Macromolecules 36
(16) (2003) 6010-6017.
[36] L. Han, D. Iguchi, P. Gil, T.R. Heyl, V.M. Sedwick, C.R. Arza, S. Ohashi, D.J. Lacks,
H. Ishida, Oxazine ring-related vibrational modes of benzoxazine monomers using
fully aromatically substituted, deuterated, 15N isotope exchanged and
oxazine-ring-substituted compounds and theoretical calculation, J. Phys. Chem. A.
121 (2017) 6269-6282.
[37] K. Zhang, H. Ishida, An anomalous trade-off effect on the properties of smart
ortho-functional benzoxazines, Polym. Chem. 6 (2015) 2541−2550.
[38] C. Jubsilp, S. Damrongsakkul, T. Takeichi, S. Rimdusit, Curing kinetics of
arylamine-based polyfunctional benzoxazine resins by dynamic differential scanning
calorimetry, Thermochim. Acta. 447 (2) (2006) 131-140.
[39] Z. Lei, H. Xiao, Autocatalytic curing kinetics of thermosetting polymers: a new
model based on temperature dependent reaction orders, Polymer, 51 (16) (2010)
3814-3820.
[40] H.L. Friedman, Kinetics of thermal degradation of char-forming plastics from
thermogravimetry. application to a phenolic plastic. J. Polym. Sci. Part C: Polym.
Symp. 6 (1) (1964) 183–195.
[41] C. Jubsilp, K. Punson, T. Takeichi, S. Rimdusit, Curing kinetics of benzoxazine–
epoxy copolymer investigated by non-isothermal differential scanning calorimetry,
Polym. Degrad. Stabil. 95 (6) (2010) 918-924.
[42] N.K. Sini, M. Azechi, T. Endo, Synthesis and properties of spiro-centered
benzoxazines, Macromolecules 48 (20) (2015) 7466-7472.