Facile, one-pot synthesis of aromatic diamine-based phosphinated benzoxazines and their flame-retardant thermosets

Article abstract of DOI:10.1002/pola.24247

Three aromatic diamine-based, phosphinated benzoxazines (79) were prepared from three typical aromatic diamines4,4′-diamino diphenyl methane (1), 4,4′-diamino diphenyl sulfone (2), and 4,4′-diamino diphenyl ether (3) by a one-pot procedure. To clarify the

Full text of DOI:10.1002/pola.24247

Facile, One-Pot Synthesis of Aromatic Diamine-Based Phosphinated
Benzoxazines and Their Flame-Retardant Thermosets
CHING HSUAN LIN,1 HUNG TSE LIN,1 JHAO WEI SIE,1 KUEN YUAN HWANG,2 AN PANG TU2
1Department of Chemical Engineering, National Chung Hsing University, Taichung, Taiwan, Province of China
2Chang Chun Plastics, ShinChu, Taiwan, Province of China
Received 19 April 2010; accepted 4 July 2010
DOI: 10.1002/pola.24247
Published online in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Three aromatic diamine-based, phosphinated ben-
zoxazines (7–9) were prepared from three typical aromatic
diamines—4,40-diamino diphenyl methane (1), 4,40-diamino di-
phenyl sulfone (2), and 4,40-diamino diphenyl ether (3) by a
one-pot procedure. To clarify the reaction mechanism, a two-
pot procedure was applied, in which the reaction intermedi-
ates (4–6) were isolated for characterization. The structures of
intermediates and benzoxazines were confirmed by high reso-
lution mass, IR, and 1D and 2D-NMR spectra. In addition to
self-polymerization, (7–9) were copolymerized with cresol
novolac epoxy (CNE). After curing, the homopolymers of P(7–9)
are brittle while the copolymers of (7–9)/CNE are tough. Dynamic
mechanical analysis shows the Tgs of (7–9)/CNE copolymers are
187, 190, and 171 ꢀC, respectively. Thermal mechanical analysis
shows the CTEs of
(7–9)/CNE copolymers are 46, 38, and 46
ppm, respectively. All the (7–9)/CNE copolymers belong to an UL-
C
V
94 V-0 grade, demonstrating good flame retardancy. 2010 Wiley
Periodicals, Inc. J Polym Sci Part A: Polym Chem 48: 4555–4566,
2010
KEYWORDS: flame retardance; ring-opening polymerization;
thermosets
INTRODUCTION Benzoxazines are resins that can be poly-
merized to thermosets via thermally activated cationic ring-
opening reactions. Thermosets with low water absorption,
superior electrical properties, and low surface energy can be
obtained after curing.1–13 Benzoxazines have great potential
to replace epoxy and phenolic resins as material for copper
clad laminate and encapsulation. Although carbon fiber rein-
forced polybenzoxazines have shown improved flame retard-
ancy, neat polybenzoxazines are still flammable, and may not
meet the requirements of certain electronic applications, in
which a UL-94 V-0 grade is required. Thus, research on
flame-retardant polybenzoxazines is attractive to the elec-
tronic industries. Flame-retardant benzoxazines based on
phenylphosphine oxide-containing biphenol,14 phosphinate
triphenol,15 and phosphinate triamine16 have been reported.
Recently, Ulrich et al.,17 Lin et al.,18 and Ca´diz and coworkers 19
reported a novel method of preparing phosphorus-containing
benzoxazines, and the flame retardancy of resulting thermo-
sets has been proven.
based on aromatic diamines with either electro-withdrawing
or electron-donating linkage were successfully prepared.
Almost at the same time, Ronda and coworkers prepared
deuterated benzoxazines to confirm the reaction mechanism
of the procedure.26 Very recently, Ishida and coworkers suc-
cessfully prepared 4,40-diaminodiphenyl sulfone-based ben-
zoxazine using the traditional one-step procedure.27 How-
ever, to the best of our knowledge, there are no reports on
the preparation of flame-retardant benzoxazines using aro-
matic diamines as starting materials. In this study, we report
a facile procedure to prepare aromatic diamine-based flame-
retardant benzoxazines. To prove the widely-useful charac-
teristic of this approach, diamines with a weak electron-
donating methylene group, a strong electron-withdrawing
sulfonyl group, and a strong electron-donating ether group
were chosen as starting materials. Thus, three phosphorus-
containing benzoxazines (7–9), derived from three typical ar-
omatic diamines: 4,40-diamino diphenyl methane (1), 4,40-
diaminodiphenyl sulfone (2), and 4,40-diaminodiphenyl ether
(3), were prepared via a one-pot procedure. In addition, to
clarify the reaction mechanism, a two-pot procedure was
applied, in which the reaction intermediates (4–6) were iso-
lated for characterization. The synthesized benzoxazines
were self-polymerized or copolymerized with cresol novolac
epoxy (CNE). The microstructure, thermal properties, and
the structure-property relationship of the resulting copoly-
mers were studied.
Due to the triazine networks that result from the condensa-
tion of aromatic diamines and formaldehyde,20 gelation occur
during the preparation of some aromatic diamine-based ben-
zoxazines. As
a result, benzoxazines based on difunc-
tional,21,22 multifunctional aromatic amines or their deriva-
tives23 have seldom been reported. Recently, we proposed a
widely-useful, three-step procedure to prepare aromatic dia-
mine-based benzoxazines.24,25 In that study, benzoxazines
Additional Supporting Information may be found in the online version of this article. Correspondence to: C. H. Lin (E-mail: linch@nchu.edu.tw)
C
V
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 4555–4566 (2010) 2010 Wiley Periodicals, Inc.
AROMATIC DIAMINE-BASED PHOSPHINATED BENZOXAZINES, LIN ET AL.
4555

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 1 One-pot synthesis of
(7–9).
EXPERIMENTAL
removed and the time at which the polymer self-extin-
guished (t1) and the dripping characteristics were recorded.
Cotton ignition was noted if polymer dripping occurred dur-
ing the test. After cooling, the second ignition was performed
on the same sample and the time at which the polymer self-
extinguished (t2) and the dripping characteristics were
recorded. Five specimens were performed, and the self-extin-
guish time was averaged. If the average t1 plus t2 was less
than 10 s without any dripping, the polymer is considered to
be a V-0 material. If t1 plus t2 was in the range of 10–30 s
without any dripping, the polymer is considered to be a V-1
material. IR spectra were obtained by Perkin–Elmer RX1
infrared spectrophotometer. High resolution mass spectra
were obtained by a Finnigan/Thermo Quest MAT 95XL mass
spectrometer.
Materials
2-Hydroxybenzaldehyde was purchased from Showa. 4,40-
Diamino diphenyl methane (1), 4,40-diaminodiphenyl sulfone
(2), and 4,40-diaminodiphenyl ether (3) were purchased
from ChrisKev and used as-received. 9, 10-Dihydro-9-oxa-10-
phosphaphenanthrene 10-oxide (DOPO) and was purchased
from TCI. N,N-dimethylacetamide (DMAc) and N,N-dimethyl-
formamide (DMF) were purchased from TEDIA and purified
by distillation under reduced pressure over calcium hydride
(Acros) and stored over molecular sieves. 1,4-Dioxane, chlo-
roform, and ethanol (99.5%) with a HPLC or ACSgrades
were purchased from various commercial sources and used
without further purification.
Characterization
Differential scanning calorimetry (DSC) was performed with
a Perkin–Elmer DSC 7 in a nitrogen atmosphere at a heating
rate of 10 C/min. Tg was taken as the midpoint of the heat
One-Pot Synthesis of (7) from (1)
(1) 19.82 g (0.1 mole), 2-hydroxybenzaldehyde 24.42 g (0.2
mole), DOPO 43.26 g (0.2 mole), and 300-mL DMAc were
introduced into a round 500-mL glass flask equipped with a
nitrogen inlet, a condenser, and a magnetic stirrer (Scheme
1). The reaction mixture was stirred at room temperature
for 18 h. Then, 24.3 g (0.3 mole) of 37% formaldehyde was
added. The mixture was stirred at room temperature for 6 h,
ꢀ
capacity transition between the upper and lower points of
deviation from the extrapolated liquids and glass lines. Ther-
mal gravimetric analysis (TGA) was performed using a Per-
kin–Elmer TGA 7 at a heating rate of 20 ꢀC/min in a nitro-
gen atmosphere from 100 to 800 ꢀC. Dynamic mechanical
analysis (DMA) was performed with a Perkin–Elmer Pyris
Diamond DMA with a sample size of 5.0 ꢁ 1.0 ꢁ 0.2 cm3.
The storage modulus E0 and tan d were determined as the
sample was subjected to the temperature scan mode at a
programmed heating rate of 5 ꢀC/min at a frequency of
1 Hz. The test was performed by a bending mode with an
amplitude of 5 lm. Thermal mechanical analysis (TMA) was
performed with a Seiko TMA/SS6100 at a heating rate of
ꢀ
and then further stirred at 80 C for 12 h. After that, the so-
lution was pourꢀed into water. The precipitate was filtered
and dried at 80 C in a vacuum oven. About 81.6 g of brown
powder (94% yield) with an exothermic peak temperature
ꢀ
of 261 C (DSC) was obtained. Since the phosphorus and the
adjacent aliphatic carbon are both chiral centers, two dia-
stereomers coexist in (7). 1H-NMR (DMSO-d6, a comma is
added for the other diastereomer), d ¼ 3.52–3.56 (2H, H1,
0
0
0
5
ꢀC/min from 25 to 250 ꢀC. The coefficient of thermal
H1 ), d ¼ 5.04–5.53 (6H, H6, H6 , H25, and H25 ), d ¼ 6.52–
0
0
0
0
expansion (CTE) was taken in the range from 40 to 120 ꢀC.
NMR measurements were performed using a Varian Inova
600 NMR in DMSO-d6, and the chemical shift was calibrated
by setting the chemical shift of DMSO-d6 as 2.49 ppm. The
UL-94 vertical test was performed according to the testing
procedure of FMVSS 302/ZSO 3975 with a test specimen bar
of 127 mm in length, 12.7 mm in width and about 1.27 mm
in thickness. In the test, the polymer specimen was subjected
to two 10-s ignitions. After the first ignition, the flame was
6.59 (4H, H4, H4 ), 6.80–6.85 (7H, H3, H3 , H8, H8 , H10 ), 6.18
0 0
(1H, H10), 7.02 (1H, H11 ), 7.17 (1H, H9), 6.21 (1H, H9 ),
0
7.23–7.28 (2H, H11 and H22 ), 7.291 (1H, H21), 7.33–7.35
0
0
(2H, H21 and H22), 7.39 (1H, H23), 7.49 (1H, H230), 7.52–7.57
0
(2H, H15 and H15 ), 7.79–7.85 (2H, H16 and H16 ), 7.86 (1H,
0
H140), 7.94 (1H, H14), 8.21 (1H, H20), 8.24–8.27 (2H, H17 and
0
H20 ), 8.29 (1H, H17 ). FTIR (KBr): 960 cmꢂ1 (NACAO
stretch), 1037 cmꢂ1 (ArAOAC symmetric stretch), 1234
cmꢂ1 (ArAOAC asymmetric stretch), 1368 cmꢂ1 (CAN
4556
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 1 1H-NMR spectrum of (7).
stretch). HR-MS (FAB) m/z: calcd. 862.2362 for
53H40O6N2P2; anal., 862.2360 for C53H40O6N2P2 (Mþ).
z: calcd. 912.1824 for C52H38O8N2P2S; anal., 913.1889 for
C
C
52H39O8N2P2S (M þ 1þ).
One-Pot Synthesis of (9) from (3)
One-Pot Synthesis of (8) from (2)
Compounds (9) was synthesized with a preparation proce-
dure similar to (7) using diamine (3) as the starting material
(Scheme 1). Yellow powder (94% yield) with an exothermic
peak temperature of 242 ꢀC (DSC) was obtained. Since the
phosphorus and the adjacent aliphatic carbon are both chiral
centers, two diastereomers coexist in (9). 1H-NMR (ppm,
DMSO-d6, a comma is added for the other diastereomer), d
Compounds (8) was synthesized with a preparation proce-
dure similar to (7) using diamine (2) as the starting material
(Scheme 1). White powder (97% yield) with an exothermic
peak temperature of 288 ꢀC (DSC) was obtained. Since the
phosphorus and the adjacent aliphatic carbon are both chiral
centers, two diastereomers coexist in (8). 1H-NMR (ppm,
DMSO-d6, a comma is added for the other diastereomer): d
0
0
¼ 5.07–5.63 (6H, H5, H5 , H24, and H24 ), 6.62–6.67 (4H, H3,
0
0
0
0
0
¼ 05.31–5.81 (6H, H5, H5 , H24, and H24 ), 6.81–6.90 (6H, H3,
H3 ), 6.68–6.73 (4H, H2, H2 ), 6.86–6.92 (2H, H10, H100 ), 7.0–
0 0
0
0
H3 , H9, and H10 ), 6.98–07.0 0 (2H, H7, H7 ), 7.04–7.07 (1H,
7.06 (2H, H8, H8 ),0 7.14–7.26 (3H, H9, H9 , and H22 ), 7.28–
0 0
0
H100), 7.16–7.21 (2H, H9 , H8), 7.25–7.36 (3H, H8 ,H20, and
7.34 (2H, H20, H20 ), 7.35–7.46 (4H, H7, H7 , H201 , and H22),
7.47–7.51 (1H, H21), 7.52–7.60 (2H, H14, H14 ), 7.76–7.84
0
0
H200), 7.44–7.053 (6H, H2, H2 , H14, H14 ), 7.70–7.77 (4H, H15
,
,
0
0
0
H150, H22, H22 ), 7.92–7.95 (2H, H13, H13 ), 8.12–8.14 (2H, H201
(2H, H15, H15 ), 7.92–7.98 (1H, H13), 7.98–8.06 (1H, H13 ),
0 0
0
H21 ), 8.20–8.21 (2H, H19, H19 ), 8.26–8.28 (2H, H16, H16 ).
FTIR (KBr): 970 cmꢂ1 (NACAO stretch), 1037 cmꢂ1
(ArAOAC symmetric stretch), 1240 cmꢂ1 (ArAOAC asym-
metric stretch), 1379 cmꢂ1 (CAN stretch). HR-MS (FAB) m/
8.18–8.24 (2H, H19, H19 ), 8.25–8.30 (2H, H16, H16 ). FTIR
(KBr): 962 cmꢂ1 (NACAO stretch), 1039 cmꢂ1 (ArAOAC
symmetric stretch), 1235 cmꢂ1 (ArAOAC asymmetric
stretch), 1367 cmꢂ1 (CAN stretch). HR-MS (FAB) m/z: calcd.
AROMATIC DIAMINE-BASED PHOSPHINATED BENZOXAZINES, LIN ET AL.
4557

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 2 1H-1H COSY NMR spectrum of (7).
864.2154
C
for C52H38O7N2P2;
anal.,
865.2234
for
samples were allowed to cool slowly to room temperature to
52H39O7N2P2 (M þ 1þ).
prevent cracking. In the (7–9)/CNE systems, the equivalent
number of oxirane (epoxy linkage) and phenolic OH (after
ring opening of benzoxazine) is set to be equal. Since each
benzoxazine resulted in two phenolic OH after ring opening,
the molar ratio of oxirane to (7–9) is controlled to 2.
RESULTS AND DISCUSSION
One-Pot Synthesis of (7–9)
As in Scheme 1, benzoxazines (7–9) were successfully pre-
pared by a one-pot procedure from aromatic diamine (1–3)
without any extraction procedures (the reaction mechanism
will be discussed later). Figure 1 shows the 1H-NMR spectrum
of (7). The characteristic peaks of oxazine linkage, OCH2N and
Synthesis of (4–9)
Syntheses and the characterization of (4–9) are listed in sup-
plementary information.
Curing Procedure
phCHN, at around 5.1–5.5 ppm support the formation of ben-
0
Benzoxazine (7–9) and (7–9)/CNE were melted in an alumi-
num mold, and then cumulative curing at 160, 180, 200, and
220 ꢀC for 2 h each in an air-circulating oven. Thereafter,
zoxazines. Two sets of 0signals for methyl peaks (H1 an0d H1 ),
NACHAph (H6 and H6 ), and NACH2AO (H25 and H25 ) con-
firm the coexistence of two pairs of diastereomers since the
4558
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

FIGURE 3 13C-NMR spectrum of (7).
FIGURE 4 1H-13C HETCOR NMR spectrum of (7).

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 Two-pot synthesis of
(7–9).
FIGURE 5 1H-NMR spectrum of (4).
4560
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 6 13C-NMR spectrum of (4).
phosphorus and its adjacent carbon (C6) are both chiral. The
Figure 3 shows the 13C-NMR spectrum of (7). The characteris-
tic peaks at 57.4–58.6 and 76.3–76.7 ppm, standing for the
NACH2AO and NACHAph of oxazine linkage, support the for-
0
peak pattern of H25 and H25 was a little complex because the
H6 is a chiral center, which makes the two hydrogens of H6
magnetically nonequivalent, and splits each other with a
geminal coupling. The peaks of ArAH, assigned by the corre-
1
mation of benzoxazines. Because of the JP-C coupling and the
coexistence of two diastereomers, four peaks were observed
for NACHAph. The peaks of ArAC, assigned by the correla-
tions in the 1H-13C HETCOR NMR spectrum (Fig. 4), were
1
lations in the H-1H COSY spectrum (Fig. 2), were marked on
Figure 1, further confirming the structure of (7).
FIGURE 7 FTIR spectra of (4) and
(7).
AROMATIC DIAMINE-BASED PHOSPHINATED BENZOXAZINES, LIN ET AL.
4561

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 3 Proposed ring-open-
ing mechanism of benzoxazine.
Note that the electron-withdraw-
ing O¼P bond destabilized the
iminium ion.
marked on Figure 3, also confirming the structure of (7).1H
and 13C-NMR were performed to confirm the structure of
benzoxazines (8–9), and the detailed assignment of each peak
is shown in the supplementary information (Figs. S1–S4).
equilibrium to the right, making it easy for the reaction to
complete. Therefore, despite the low nucleophility of dia-
mine (2), compound (5) can be synthesized with high yield
and purity at room temperature without any catalyst. Figure
5 shows the 1H-NMR spectrum of (4). Two phenolic OH
peaks at 9.4 and 9.5 ppm, two secondary amino peaks at
Two-Pot Synthesis of (7–9)
To clarify the reaction mechanism, a two-pot procedure was
applied, in which the reaction intermediate (4–6) was iso-
lated for characterization. As in Scheme 2, compounds (4–6)
were synthesized from the reaction of aromatic diamines
(1–3), 2-hydroxybenzaldehyde and DOPO. In the first step,
2-hydroxybenzaldehyde reacted with the amino groups of
(1–3), resulting in an intermediate with an o-hydroxy phe-
nylimine structure. DOPO then carried out an addition reac-
tion on the resulting imine linkage, forming compounds (4–
6) with a secondary amine linkage. In this procedure, the
reaction temperature was kept low to prevent the DOPO-
aldehyde addition, which occurs at high temperature.28
Although the formation of the imine linkage is a reversible
reaction, the yield and purity of compounds (4–6) were
high. It is thought that the DOPO-imine addition shifted the
6.4 and 6.1 ppm, and two aliphatic hydrogen peaks (H6 and
0
H6 ) at 5.4 and 5.3 ppm confirm the coexistence of two dia-
stereomers. Figure 6 shows the 13C-NMR spectrum of (4).
Four aliphatic carbons signals at 47–50 ppm appeared. The
1
coexistence of two diastereomers, along with the strong JP-C
coupling that further splits the peaks, explains the origin of
four aliphatic carbon peaks. In the 31P-NMR spectrum, two
peaks at 37.5 and 34.2 were observed, also confirming the
coexistence of two diastereomers. 1H and 13C-NMR were
performed to confirm the structure of benzoxazines (5–6),
and the detailed assignment of each peak is shown in the
supporting information (Figs. S5–S8). In the second step,
formaldehyde was added to the chloroform solution of (4–
6) to react with the secondary amine, producing an interme-
diate with hydroxymethylamine (NCH2OH) structure.
SCHEME 4 Ideal structure of P(DDM-Bz) and P(7).
4562
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 8 DSC thermograms of benzoxazines (7–9) and polybenzoxazines P(7–9). [Color figure can be viewed in the online issue,
which is available at wileyonlinelibrary.com.]
Thereafter, the reaction temperature was raised to prompt
the condensation reaction between hydroxymethylamine
and hydroxy, leading to benzoxazines (7–9). Figure 7 shows
the FTIR spectra of (4) and (7). The disappearance of strong
absorptions of NH and OH at around 3200 cmꢂ1, the
appearance of the ArAOAC absorption at 1037 cmꢂ1 (sym-
metric stretch) and 1234 cmꢂ1 (asymmetric stretch), and
the oxazine absorptions at 960 and 1368 cmꢂ1 support the
ring closure of oxazine linkage. According to other spectro-
scopic data, benzoxazines (7–9) prepared by this method
had the same patterns as those prepared by the one-pot
procedure. The total yields of (7–9) by two-step procedure
are 72, 63, and 73%, respectively, while the yields of (7–9)
by one-pot reaction are 94, 97, and 94%, respectively. The
result demonstrates the high yield characteristic of one-pot
procedure.
FIGURE 9 IR spectra of (7) after ac-
cumulative curing at each stage
for 20 min. [Color figure can be
viewed in the online issue, which
is available at wileyonlinelibrary.
com.]
AROMATIC DIAMINE-BASED PHOSPHINATED BENZOXAZINES, LIN ET AL.
4563

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
opening of the oxazine linkage. Based on the IR spectra, the
structure of P(7) is proposed in Scheme 4.
Thermal Properties of Homopolymers
Tgs of P(7–9) homopolymers were measured by DSC, and
ꢀ
the Tg values are 184, 187, and 169 C, respectively, (Fig. 8).
The Tgs are 45–63 ꢀC higher than that of mono-functional
polybenzoxzine with similar structure,18 demonstrating the
advantage of the difunctional characteristic of (7–9). The Tg
depends on the structure of the polybenzoxazines. P(8), with
the polar sulfonyl group, shows the highest Tg. In contrast,
P(9), with a flexible ether linkage, shows the lowest Tg. In
our previous study,24 we reported the Tg of a diamine (1)-
based polybenzoxazines, P(DDM-Bz), is 208 ꢀC, which is 24
ꢀC higher than that of P(7). As in Scheme 3, DDM-Bz and (7)
have similar structure except for the bulky phosphinate
pendant. The lower Tg of P(7) than that of P(DDM-Bz) might
be attributed to the increased free volume caused by the
bulky phosphinate pendant. In addition, as evidenced by
smaller exothermic enthalpy, the lower crosslinking density
resulting from steric hindrance might also be responsible for
the lower Tg of P(7). As to the thermal stability, the 5%
decomposition temperatures oꢀf P(7–9) are in the range of
FIGURE 10 DSC thermogram of (7)/CNE mixture. [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
DSC Thermosgrams and IR Analysis
DSC thermograms of benzoxazines (7–9) are shown in Fig-
ure 8. No melting endotherms were observed in the thermo-
grams. The coexistence of four stereoisomers may be respon-
sible for the amorphous characteristic of (7–9). However, as
conventional benzoxazines do, benzoxazines (7–9) show an
apparent exotherm in the DSC thermograms, demonstrating
the thermal curing characteristic. The values of exothermic
enthalpy for (7–8) are 87.9 and 88.4 kJ/mol, respectively,
which are relatively small compared with those (158 and
114 kJ/mol) of diamines (1) and (2)-based benzoxazines.24
The bulky phosphinate pendant, which hindered the curing
reaction due to steric hindrance, might be responsible for
the smaller exothermic enthalpy. The exothermic peak tem-
peratures are 261, 288, and 242 ꢀC for benzoxazines (7–9),
respectively. The polymerization temperature increased with
the electron-withdrawing/donating character of ‘‘X’’ shown
in Scheme 1. Benzoxazine with the electro-withdrawing sul-
fonyl group shows the highest exothermic temperature, while
benzoxazine with electro-donating ether linkage showed the
lowest exothermic temperature. This trend was also
observed by Ronda and coworkers29 and Lin et al.24 It has
been reported that the ring-opening of benzoxazine is
related with the iminium ion.30,31 Ronda and coworkers
reported that the intermediate during benzoxazine curing,
the iminium ion, is strongly destabilizes by the electron-
withdrawing substituents on the para-position of the imi-
nium ion.29 In this case, the electron-withdrawing sulfonyl
group destabilized the iminium ion the most, increasing the
activation energy of curing the most, so (8) exhibits the high-
est polymerization temperature. In contrast, the electron-
donating ether linkage stabilizes the iminium ion the most,
so (9) exhibit the lowest polymerization temperature. In
addition, the electron-withdrawing O¼¼P bond destabilized
the iminium ion (Scheme 3), hindering the ring-opening of
benzoxazine, and thus explaining the high exothermic tem-
perature of (7–9). Figure 9 shows IR spectra of (7) after ac-
cumulative curing at each stage for 20 min. After curing at
ꢀ
338–364 C, which are 30–50 C higher than that of bisphe-
nol F and aniline-based polybenzoxazine, P(F-a), demonstrat-
ing moderate thermal stability.
Properties of (7–9)/CNE Copolymers
In addition to self-curing, (7–9) were also copolymerized
with CNE. Figure 10 shows the DSC thermogram of (7)/CNE
mixture. Two exothermic peaks were observed in the heating
scan. The first peak is thought to be associated with the
ring-opening of (7), forming polybenzoxazine with phenolic
OH linkage. The second is thought to be associated with the
phenolic OH-oxirane reaction. IR spectra were applied to
confirm the speculation. Figure 11 shows FTIR spectra of the
(7)/CNE system after accumulative curing at each stage for
20 min. The ArAOAC absorption at 1035 and 1234 cmꢂ1and
the oxazine absorptions at 960 and 1368 cmꢂ1 reduced
ꢀ
gradually after curing at 180 C, supporting the ring-opening
of benzoxazine. The absorption of oxirane (914 cmꢂ1) did
ꢀ
220 C, the ArAOAC absorption at 1035 (symmetric stretch)
and 1234 cmꢂ1(asymmetric stretch) and the oxazine absorp-
FIGURE 11 IR spectra of the (7)/CNE system after accumulative
tions at 960 and 1368 cmꢂ1 disappeared, indicating the ring
curing at each stage for 20 min.
4564
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
SCHEME 5 Proposed polymerization mechanism of (7)/CNE copolymer.
ꢀ
not reduce after curing at 180 C, but reduced gradually after
shows the TMA thermogram of (7)/CNE copolymer. Only one
transition temperature at 162 C was observed, supporting the
ꢀ
ꢀ
curing at temperature higher than 180 C, supporting the phe-
nolic OH-oxirane reaction. The appearance of an aromatic
ether (ArAO) absorption at 1208 cmꢂ1 and an aliphatic ether
(RAO) absorption at 1054 cmꢂ1 after curing at temperature
higher than 180 ꢀC also supports the oxirane-phenolic OH
reaction. According to IR and DSC analyses, we proposes a po-
lymerization mechanism of (7)/CNE copolymer in Scheme 5.
The first stage is the ring-opening of oxazine, and the second
stage is the nucleophilic addition of phenolic OH on oxirane.
homogeneous structure. As shown in Figure 13, a CTE of 46
ppm/ꢀC was obtained, which is relatively small for an epoxy
thermoset. The CTEs of (8–9)/CNE copolymers are 38 and 46
ppm, respectively, also showing good thermal stability. As
listed in Table 1, the 5% decomposition temperatures of (7–
9)/CNE copolymers are in the range of 345–356 ꢀC, showing
moderate thermal stability. As to the flame retardancy, all the
(7–9)/CNE copolymers belong to an UL-94 V-0 grade, demon-
strating that good flame reatardancy was achieved via the
incorporation of phosphinated benzoxazines. In addition to the
phosphorus element, the nitrogen-phosphorus synergistic
effect32,33 might also contribute to the excellent flame retard-
ancy of this system.
DMA and TMA were applied to measure their thermal proper-
ties of phenolic OH-oxirane reaction. Figure 12 shows the
DMA thermograms of (7–9)/CNE thermosets. The Tgs of (7–
9)/CNE copolymers are 187, 190, and 171 ꢀC, respectively.
The trend is the same as those of P(7–9). As shown in Figure
12, only one tan d peak was observed, indicating that a homo-
geneous copolymer was obtained. The homogeneous structure
can also be confirmed by TMA measurements. Figure 13
FIGURE 12 DMA thermograms of (7–9)/CNE thermosets.
FIGURE 13 TMA curve of (7)/CNE copolymer.
AROMATIC DIAMINE-BASED PHOSPHINATED BENZOXAZINES, LIN ET AL.
4565

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
TABLE 1 Thermal Properties of (7–9)/CNE Thermosets
Thermoset ID
E0 (GPa)a
Tg (ꢀC)b
Tg (ꢀC)c
CTEd
Td (ꢀC)e
Char Yieldf
P (wt %)g
N (wt %)h
UL-94 Grade
(7)/CNE
(8)/CNE
(9)/CNE
a
2.06
2.18
1.79
187
190
171
162
167
156
46
38
46
356
353
345
e
32
42
48
4.91
4.72
4.89
2.21
2.13
2.21
V-0
V-0
V-0
Modulus at 50 ꢀC, measured by DMA.
5% decomposition temperature in nitrogen atmosphere (ꢀC).
Residual weight percentage at 800 ꢀC in nitrogen atmosphere (wt %).
Phosphorus content.
b
c
f
Peak temperature of tan d, measured by DMA (ꢀC).
Measured by TMA (ꢀC).
g
h
d
Coefficient of thermal expansion before Tg (ppm/ꢀC).
Nitrogen content.
CONCLUSIONS
12 Andreu, R.; Reina, J. A.; Ronda, J. C. J Polym Sci Polym
Chem 2008, 46, 6091–6101.
Three diamine-based, phosphinated benzoxazines (7–9) were
successfully synthesized by a one-pot procedure, and the reac-
tion mechanism was confirmed by analyzing the intermediate
prepared by a two-pot procedure. According to this proce-
dure, the molecule-design flexibility of benzoxazines can be
increased. According to DSC thermograms and IR analysis, the
phosphinated benzoxazines can be thermally polymerized.
Depending on the structure, moderate to high Tgs can be
achieved. In addition to self-curing, (7–9) were also applied
as curing agents of CNE, and the reaction sequence can be
monitored by IR analysis. After curing, homogeneous and
tough copolymers with good flame retardancy can be
achieved.
13 Iha, R. K.; Wooley, K. L.; Nystrom, A. M.; Burke, D. J.; Kade,
¨
M. J.; Hawker, C. J. Chem Rev 2009, 109, 5620–5686.
14 Choi, S. W.; Ohba, S.; Brunovska, Z.; Hemvichian, K.; Ishida,
H. Polym Degrad Stab 2006, 91, 1166–1178.
15 Lin, C. H.; Cai, S. X.; Leu, T. S.; Hwang, T. Y.; Lee, H. H.
J Polym Sci Polym Chem 2006, 44, 3454–3468.
16 Chang, C. W.; Lin, C. H.; Lin, H. T.; Huang, H. J.; Hwang, K.
Y.; Tu, A. P. Eur Polym J 2009, 45, 680–689.
17 Ulrich, W.; Franck, M. Flame-Proofing Agents, WO02057279
A1.
18 Lin, C. H.; Lin, H. T.; Chang, S. L.; Huang, H. J.; Hu, Y. M.;
Taso, Y. R.; Su, W. C. Polymer 2009, 50, 2264–2272.
The authors thank the National Science Council of the Republic
of China for financial support. Partially sponsorship by the
Green Chemistry Project (NCHU), as funded by the Ministry of
Education is also gratefully acknowledged.
19 Sponto´ n, M.; Lligadas, G.; Ronda, J. C.; Galia`, M.; Ca´diz, V.
Polym Degrad Stab 2009, 94, 1693–1699.
20 Brunovska, Z.; Liu, J. P.; Ishida, H. Macromol Chem Phys
1999, 200, 1745–1752.
REFERENCES AND NOTES
21 Shen, S. B.; Ishida, H.
J Appl Polym Sci 1996, 61,
1 Endo, T.; Sudo, A.
J Polym Sci Polym Chem 2009, 47,
1595–1605.
4847–4858.
22 Takeichi, T.; Kano, T.; Agag, T. Polymer 2005, 46,
2 Ghosh, N. N.; Kiskan, B.; Yagci, Y. Prog Polym Sci 2007, 32,
12172–12180.
1344–1391.
23 Subrayan, R. P.; Jones, F. N. Chem Mater 1998, 10,
3 Nagai, A.; Kamei, Y.; Wang, X. S.; Omura, M.; Sudo, A.; Nishida,
H.; Kawamoto, E.; Endo, T. J Polym Sci Polym Chem 2009, 46,
2316–2325.
3506–3512.
24 Lin, C. H.; Chang, S. L.; Hsieh, C. W.; Lee, H. H. Polymer
2008, 49, 1220–1229.
4 Kiskan, B.; Aydogan, B.; Yagci, Y. J Polym Sci Polym Chem
25 Lin, C. H.; Chang, S. L.; Lee, H. H.; Chang, H. C.; Hwang, K.
Y.; Tu, A. P.; Su, W. C. J Polym Sci Polym Chem 2008, 46,
4970–4983.
2009, 47, 804–811.
5 Herrera, P. V.; Ishida, H. J Polym Sci Polym Chem 2009, 47,
5871–5881.
26 Andreu, R.; Ronda, J. C. Synth Commun 2008, 38,
6 Chernykh, A.; Agag, T.; Ishida, H. Macromolecules 2009, 42,
2316–2329.
5121–5127.
27 Agag, T.; Jin, L.; Ishida, H. Polymer 2009, 50, 5940–5944.
7 Wang, C. F.; Su, Y. C.; Kuo, S. W.; Huang, C. F.; Sheen, Y. C.;
28 Liu, Y. L.; Wu, C. S.; Hsu, K. Y.; Chang, T. C. J Polym Sci
Polym Chem 2002, 40, 2329–2339.
Chang, F. C. Angew Chem Int Ed 2006, 45, 2248–2251.
8 Yagci, Y.; Kiskan, B.; Ghosh, N. N. J Polym Sci Polym Chem
29 Andreu, R.; Reina, J. A.; Ronda, J. C. J Polym Sci Polym
2008, 47, 5565–5576.
Chem 2008, 46, 3353–3366.
9 Sponto´ n, M.; Larrechi, M. S.; Ronda, J. C.; Galia`, M.; Ca´diz,
30 McDonagh, A. F.; Smith, H. E. J Org Chem 1968, 33, 1–8.
31 McDonagh, A. F.; Smith, H. E. J Org Chem 1968, 33, 8–12.
V. J Polym Sci Polym Chem 2008, 46, 7162–7172.
10 Kiskan, B.; Demirel, A. L.; Kamer, O.; Yagci, Y. J Polym Sci
32 Wu, C. S.; Liu, Y. L.; Chiu, Y. S. Polymer 2002, 43,
Polym Chem 2008, 46, 6780–6788.
4277–4284.
11 Chou, C. I.; Liu, Y. L. J Polym Sci Polym Chem 2008, 46,
6509–6517.
33 Zhu, S. W.; Shi, W. F. Polym Degrad Stab 2002, 75, 543–547.
4566
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA