Resorcinol-based benzoxazine with low polymerization temperature

Article abstract of DOI:10.1002/pola.27169

The industrial applications of benzoxazines are limited due to their high curing temperatures. This drawback can be overcome by more reactive precursor compared to conventional benzoxazines or by application of efficient initiators. We report the synthesis of a new resorcinol-based benzoxazine and its cationic polymerization with thermolatent super acids, namely organic sulfonium hexafluoroantimonates. This combination of a reactive precursor and an efficient initiator results in a curing temperature below 100 °C (differential scanning calorimetry onset) which is up to now one of the lowest polymerization temperatures for benzoxazine systems. Furthermore, the thermal stability of the formed polybenzoxazine has not been influenced by the applied initiators.

Full text of DOI:10.1002/pola.27169

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Resorcinol-Based Benzoxazine with Low Polymerization Temperature
Andre Arnebold,1 Oliver Schorsch,1 Johannes Stelten,2 Andreas Hartwig1
1
€
Fraunhofer-Institut fur Fertigungstechnik und Angewandte Materialforschung, Wiener Straße 12, D-28359 Bremen, Germany
2
€
€
Institut fur Organische und Analytische Chemie, Universitat Bremen, Leobener Straße NW2C, D-28359 Bremen, Germany
Correspondence to: A. Hartwig (E-mail: andreas.hartwig@ifam.fraunhofer.de)
Received 4 February 2014; accepted 13 March 2014; published online 29 March 2014
DOI: 10.1002/pola.27169
ABSTRACT: The industrial applications of benzoxazines are lim-
ited due to their high curing temperatures. This drawback can
be overcome by more reactive precursor compared to conven-
tional benzoxazines or by application of efficient initiators. We
report the synthesis of a new resorcinol-based benzoxazine
and its cationic polymerization with thermolatent super acids,
namely organic sulfonium hexafluoroantimonates. This combi-
nation of a reactive precursor and an efficient initiator results
in a curing temperature below 100 ꢀC (differential scanning cal-
orimetry onset) which is up to now one of the lowest polymer-
ization temperatures for benzoxazine systems. Furthermore,
the thermal stability of the formed polybenzoxazine has not
C
V
been influenced by the applied initiators. 2014 Wiley Periodi-
cals, Inc. J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 1693–
1699
KEYWORDS: benzoxazine; cationic polymerization; initiators;
latent initiator; resorcinol-based benzoxazine; ring-opening
polymerization; thermosets
of the benzoxazine units themselves, even though they react
to prepolymers at lower temperatures by polymerizable non-
oxazine groups. Some examples for known benzoxazines and
their self-polymerization temperatures Tc (measured by dif-
ferential scanning calorimetry, DSC) are phenol/aniline-based
benzoxazine with Tc of ꢀ263 ꢀC, p-cresol/aniline-based ben-
zoxazine with Tc of 2ꢀ69 C, 1-naphthol/aniline-based benzox-
azine with Tc of 156 C (broad DSC signal), and bisphenol-A/
INTRODUCTION Polybenzoxazines are phenolic-type resins
resulting from polymerization of 1,3-benzoxazine precur-
sors.1 These thermosets combine the advantages of epoxy
resins like design flexibility and mechanical behavior and
both thermal and flame retarded properties of phenolic res-
ins.2 Furthermore, polybenzoxazines show nearly no volume
shrinkage while curing, they do not release any byproducts
during reaction, exhibit high glass transition temperatures Tg
and high char yield. Additionally, benzoxazines do adsorb
water only in a low amount and at last they do not need any
initiators for polymerization reaction.1,3,4 On the other hand,
polybenzoxazines exhibit some shortcomings like brittleness
and high curing temperatures.1,5 In order to overcome the
drawback of high polymerization temperature, two ways
were examined in the past: First, the development of a more
reactive benzoxazine precursor and, second, the catalytic
acceleration of the polymerization reaction.
1,17
ꢀ
aniline-based benzoxazine with Tc of 249 C.
Except from
the 1-naphthol-based benzoxazine, the curing temperatures
of most benzoxazines are very high.
Wang and Ishida published investigations on different cati-
onic ring-opening mechanisms, which result in oxonium and
iminium ions as intermediates because of the basicity of the
oxazin-oxygen and -nitrogen atoms.18,19 The driving force of
benzoxazine ring-opening reaction is proposed to be the ring
stress of the distorted oxazine ring structure.4 Wang and Ish-
ida demonstrated that different initiators lead to two differ-
ent structures as a consequence of different mechanisms.19
These structures are known as Mannich base phenoxy-type
polybenzoxazine (type I) and as Mannich base phenolic-type
polybenzoxazine (type II) which are able to rearrange.20 Sev-
eral effective and even efficient catalysts and initiators were
developed to reduce the curing temperature of benzoxazines.
Initiators are usually Lewis acids like phosphorus penta-
chloride, titanium chloride, and aluminum chloride,18 carbox-
ylic acids like trifluoroacetic acid and sebacic acid,21 alcohols
In 1944, Holly and Cope synthesized the first benzoxazines
by condensation reaction of phenol, formaldehyde and a pri-
mary amine.6 Five years later, Burke et al. demonstrated that
benzoxazines react preferentially by their free ortho-posi-
tion.7 Furthermore, benzoxazines have been functionalized
by reactive groups like acetylene,8 allyl group,9 propagyl
group,10 nitrile,11 maleimide,12 coumarin,13 epoxy group,14
cardanol,15 hydroxymethyl group,16 or norbonane.12 Such
functional groups modify the mechanical and thermal behav-
ior but they do not decrease the polymerization temperature
C
V
2014 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699
1693

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
like p-cresol and 2-methylresorcinol,21,22 thiols like benzylth-
iol,23 metal complexes like acetylacetonato-complexes with
iron, manganese, or cobalt,24 sulfonates like p-toluenesulfo-
nate,17 salts like lithium iodide,5,25 and photolatent onium
salts like diphenyliodonium and triphenylsulfonium salts.26,27
Examples for really efficient known benzoxazine/initiator
systems and their respective curing temperatures (measured
equipped with a Golden Gate cell with a resolution of 4
cm21 (32 scans). 1H NMR, 13C NMR, HH-COSY, HSQC, and
HMBC spectra were recorded with a Bruker AVANCE NB 360
spectrometer. Tetramethylsilane was used as external stand-
ard. The spectra were measured in CDCl3 at room tempera-
ture. Mass spectra were measured with a Finnigan MAT 95
using electron-induced ionization at 200 ꢀC with an ioniza-
tion energy of 70 eV. The samples were injected with an
indirect inlet system. DSC was carried out in a sealed pan
with DSC 2920 Modulated from TA Instruments in a temper-
ꢀ
by DꢀSC) are BA-a/PCl5 with Tc of 122ꢀ C, pC-a/LiI with Tc of
197 C, pC-a/Zn(OTf)2 with Tc of 199 C, and pC-a/p-toluene-
sulfonates with Tc of 100–120 ꢀC (not measured by
DSC).5,17,18
ꢀ
ature range from 20 to 300 C with a heating rate of 10 K/
min. Thermogravimetric analysis (TGA) was carried out with
a Q5000 from TA Instruments. The measurements ꢀwere per-
formed under nitrogen flow, between 20 and 600 C, with a
heating rate of 10 K min21. Elementary analysis (EA) was
measured by Microanalytical Laboratory Pascher (Remagen,
Germany).
To the best of our knowledge, onium salts have not been
applied for thermal benzoxazine polymerization up to now.
Photo- or thermolatent initiators that are known for epoxide
polymerization react by a cationic mechanism and, for this
reason, they seem to be feasible for initiating benzoxazine
polymerization.28–33 Besides, such initiators can be applied in
formulations with long shelf-life. Thermolatent initiators like
p-methoxybenzyl tetrahydrothiophenium hexafluoroantimo-
nate (3) and benzyl tetrahydrothiophenium hexafluoroanti-
monat (4) form a super acid or a carbocation during thermal
or photochemical decomposition, respectively. These two ini-
tiators (3) and (4) show different activation temperatures for
divinyl ether and in other words different reactivity according
to their structures.34 They may also show different reactivity
toward benzoxazines. Recently, Oie et al. demonstrated a new
approach to form polybenzoxazines due to ring-opening
addition.22,35 This reaction of a benzoxazine ring with
2-methylresorcinol occurs at ambient conditions and is there-
fore a promising concept. In this case, the 2-methylresorcinol
is both initiator (protonation of benzoxazine by phenolic
hydroxyl group) and also monomer for addition reaction.
Synthesis
Resorcinol-Based Benzoxazine Isomer Mixture (R-a)
Consisting of 3,9-Diphenyl-3,4,9,10-tetrahydro-2H,8H-
[1,3]oxazino-[6,5-f][1,3]benzoxazine (I) and 3,7-Diphenyl-
3,4,7,8-tetrahydro-2H,6H-[1,3]oxazino-[5,6-g]
[1,3]benzoxazine (II) (1)
Formaldehyde [3 mL (40 mmol)] were dissolved in 5 mL of
1,4-dioxane. Afterward, 1.85 mL (20 mmol) of aniline was
slowly dropped into the solution under stirring. The mixture
was stirred for 30 min until a white solid precipitated.
Thereafter, the mixture was started to reflux to dissolve the
precipitate. Subsequently, a solution of 1.10 g (10 mmol) res-
orcinol in 2.5 mL of 1,4-dioxane was slowly added. Then, the
mixture was refluxed for additional 60 min. The solvent was
removed under vacuum (1022 mbar) with a successive tem-
perature rise up to 80 ꢀC (the viscosity increases rapidly).
The obtained yellow, transparent solid was further purified
by refluxing with a mixture of 20 mL diethyl ether and 4 mL
toluene. After cooling to 20 ꢀC the liquid phase was sepa-
rated from the precipitated residue. The neat product was
dissolved in the liquid phase. After the removal of the sol-
vent and drying under vacuum (1022 mbar), we obtained a
white crystalline solid with a yield of 13%, which consisted
of the two isomers (I) and (II) in a molar ratio of 1:0.14 as
determined by different NMR experiments.
In this work, we synthesized and characterized a new reac-
tive resorcinol-based benzoxazine R-a (1). We investigated
the polymerization reaction of this new benzoxazine (1) in
comparison to an established bisphenol A-based benzoxazine
BA-a (2) as well as the influence of thermolatent initiators
(3) and (4) on both benzoxazines (1) and (2).
EXPERIMENTAL
Materials
All chemicals were used as received from commercial suppli-
ers. Toluene (99.8%), petroleum ether (p.a., 40–60 C), form-
ꢀ
NMR (360 MHz, CDCl3)
See Table 1; EIMS [m/z (%)]: 344 [M1], 239, 105, 77. IR
(ATR): m 5 1595, 1579, and 1492 (s) mC5C (aromatic), 1252
and 1043 (s) maryl-O (Caromatic-O-C), 1151 (m) mC-N (C-N-C),
925 (s) dC-H (1,2,4,5-tetrasubstituted aromatics), 867 (m)
dC-H (1,2,4,5-tetrasubstituted aromatics), 809 cm21 (m) dC-H
aldehyde (37 wt
% in water, 10–15% methanol as
stabilizer), aniline, and resorcinol (99%) were purchased
from Sigma Aldrich (Schnelldorf, Germany). 6,6’-(Propane-
2,2-diyl)bis(3-phenyl-3,4-dihydro-2H-benzo[e][1,3]oxazine)
R
V
(Araldite MT 35600, BA-a (2)) was obtained from Hunts-
(1,2,3,4-tetrasubstituted
aromatics).
EA:
Calcd
for
man (Everberg, Belgium). The latent initiators p-methoxy-
benzyl tetrahydrothiophenium hexafluoroantimonate (3) and
benzyl tetrahydrothiophenium hexafluoroantimonat (4) were
prepared according to the literature.28,36
C22H20N2O2: C 76.72, H 5.85, N 8.13, O 9.29; found: C 76.46,
H 5.84, N 8.18, O 9.57.
Polymerization of R-a (1)
Physicochemical Characterization
Infrared spectra were obtained in attenuated total reflection
The polymerization of the resorcinol-based benzoxazine (1)
was investigated both (a) with one weight percent of the ini-
tiators (3) and (4) and (b) without initiator. In case of (a),
(ATR) with
a
Bruker Equinox 55 FTIR-spectrometer
1694
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 1 1H and 13C NMR Signals of Isomeric R-a (1) and Its Structure Elements
Asymmetric Isomer (I)
Symmetric Isomer (II)
Numbering/
Numbering/Structure
13C, d (ppm)
1H, d (ppm)
Structure
13C, d (ppm)
1H, d (ppm)
2
O-CH2
N-CH2
5Caryl
5CH
79.5
5.41
4.59
–
2
O-CH2
N-CH2
5Caryl
5CH
79.2
49.9
5.35
4.58
–
4
49.9
4
4a
112.1
4a
5
113.5
124.2
104.5
5
125.0
6.81
6.44
5.33
4.56
–
6.67
6.33
–
6
5CH
109.0
10
10a
11
12
13
14
5CH
8
O-CH2
N-CH2
5Caryl
5Caryl-O
N-Caryl
5CH
78.9
5Caryl-O
N-Caryl
5CH
153.5
148.1
10
46.0
–
10a
10b
11
109.2
ꢁ118
ꢁ129
ꢁ121
ꢁ7.14
ꢁ7.31
ꢁ6.98
150.8
–
5CH
148.1
–
5CH
12/18
13/19
14/20
17
118.1/117.8
129.0/128.9
121.2/121.0
148.3
7.14
ꢁ7.3
6.98/6.96
–
5CH
5CH
N-Caryl
the R-a (1) was mixed together with the initiator by a pestle.
The resorcinol-based benzoxazine R-a (1) was polymerized
for 24 h in a preheated oven at 120 ꢀC (heating profile A).
R-a (1) was also polymerized in anotheꢀr approach for 2 h at
ization was investigated in a preheated oven. We obtained a
deep red transparent solid in case of polymerization accord-
ing to heating profile B.
ꢀ
180 C and additionally for 2 h at 200 C (heating profile B)
RESULTS AND DISCUSSION
for comparability to the commercial benzoxazine (2). We
Synthesis of Resorcinol-Based Benzoxazine R-a (1)
obtained a red brown transparent solid in all six cases.
A new resorcinol-based benzoxazine R-a (1) was prepared
by reaction of resorcinol with aniline and formaldehyde in a
molar ratio of 1:2:4 as can be seen in Scheme 1. A separa-
tion of the asymmetric isomer (I) and the symmetric isomer
(II) is not necessary for polymerization experiments; thus, it
has not been carried out yet. Nevertheless, the isomeric mix-
ture of R-a (1) was completely characterized by standard
methods like infrared spectroscopy, NMR spectroscopy, EA,
and mass spectrometry. The IR spectrum shows all
Polymerization of BA-a (2)
The polymerization of BA-a (2) was investigated both (a)
with the initiators (3) and (4) and (b) without initiator. In
case of (a), the benzoxazine (2) was first pestled together
with one weight percent initiator. The polymerization was
carried out for both procedures (a) and (b) on the one hand
ꢀ
for 24 h at 120 C (heating profile A) and on the other hand
according to Huntsman (heating profile B).37 The polymer-
SCHEME 1 Synthesis of the resorcinol-based benzoxazine R-a (1) by reaction of resorcinol with aniline and formaldehyde in a
molar ratio of 1:2:4. The numbers (I) and (II) of R-a (1) denote the asymmetric and symmetric isomers, respectively. The number-
ings of the C-atoms in the structure of both isomers refer to the NMR signals, which are listed in Table 1.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699
1695

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
lithium iodide,5,25 zinc(II) triflate,5 p-toluenesulfonates,17 or
phosphorus pentachloride18 were investigated which lower
the polymerization temperature dramatically. Unfortunately,
these initiators also decrease the shelf-life of benzoxazine/
initiator mixture. The synthesis of a potential reactive
resorcinol-based benzoxazine has not been published yet.
The question arises whether this structure of two electron
donating aryl ether functions may be more reactive than
conventional benzoxazines like the commercial bisphenol A-
based one. Furthermore, it is conceivable that the application
of thermolatent initiators like p-methoxybenzyl tetrahydro-
thiophenium hexafluoroantimonate (3) and benzyl tetrahy-
drothiophenium hexafluoroantimonat (4), which are known
for effective polymerization of epoxy resins,29,30 reduces the
curing temperature of benzoxazines.
FIGURE 1 DSC thermogram of the polymerization of
resorcinol-based benzoxazine R-a (1) without initiator and also
in presence of both latent initiators (3) and (4).
The cationic polymerization of resorcinol-based benzoxazine
(1) was carried out (A) at 120 ꢀC for 24 h and (B) in
another approach, according to the established polymeriza-
tion protocol of the commercial bisphenol A-based benzoxa-
zine (2).37 The reactivity of both the thermally induced and
also the latent initiator started (in the following named
super acid initiated) polymerization was investigated by DSC
which is illustrated in Figure 1. The temperatures of maxi-
mum heat flow Tpeak are listed in Table 2. R-a (1) has an
significant oxazine ring signals at 1043 cm21(aryl-O), 1252
cm21 (aryl-O), and 1151 cm21 (C-N-C). In addition, the infra-
red signals at 925 and 867 cm21 belong to a 1,2,4,5-tetra-
substituted aromatic ring whereas the signal at 809 cm21
indicates a 1,2,3,4-tetrasubstitution which are both charac-
teristic for the isomers (I) and (II). Two-dimensional NMR
confirms clearly the formation of isomeric R-a (1) as can be
seen in Table 1. All C- and H-atoms are assigned to a NMR-
ꢀ
exothermic Tpeak value of 179 C, which accords to the poly-
merization temperature and is much lower than 245 ꢀC of
BA-a (2). Thermolatent cationic initiators like (3) and (4),
which are known for cationic polymerization of epoxy res-
ins,29,30 are predestinated for benzoxazine formulations with
long shelf-life. The applicability of similar photoactive onium
compounds has already been shown by Kasapoglu et al. for
benzoxazine photopolymerization in solution.26 The latent
super acid (3) exhibits the lowest curing temperature with a
Tpeak of 107 ꢀC. This drastically decreased polymerization
temperature was verified by three repeated measurements.
Initiator (4) is not as reactive as (3) because the p-methoxy
group of (3) leads to an increased resonance stability of the
formed carbocation. Further, an electron donating effect of
the methoxy group activates the latent initiator which results
in a lower initiation temperature as known from the poly-
merization of divinyl ethers.34 All samples the pure R-a (1)
and the mixtures with initiators (3) and (4) show a second
exotherm or a shoulder, respectively. This observation may
the result of multiple polymerization according to Ishida
et al.18 or probably the consequence of rearrangements
which are also described in the literature.5 The bisphenol A-
1
signal by measuring H NMR, 13C NMR, HH-COSY, HSQC, and
HMBC spectra. The resulting two isomeric structures (I) and
(II) have a ratio of approximately 7:1 as analyzed by the aro-
matic ring proton integrals. Furthermore, mass spectrometry
reveals typical fragments at m/z 77, 105, and 239. A molecu-
lar radical cation at m/z of 344 corresponds to the theoreti-
cal mass of R-a (1). At last, EA shows that the resorcinol-
based benzoxazine R-a (1) is pure and that its molecular
composition corresponds to the theoretical one. DSC meas-
urements of R-a (1) exhibits two endothermic signals, which
indicate crystallinity (Fig. 1). Each endothermic may belong
to one of the two isomers.
Polymerization of R-a (1)
It is well known that one main drawback in benzoxazine
chemistry is the high curing temperature.3 To overcome such
a disadvantage, there are promising concepts like developing
new reactive benzoxazines or the application of highly active
initiators. In the last decades, several effective initiators like
TABLE 2 DSC Peak Temperatures Tpeak in ꢀC for the Polymerization Reaction of the Commercial Benzoxazine (2) and the
Resorcinol-Based Benzoxazine (1) With and Without Initiators (3) and (4)
R-a (1)
BA-a (2)
Initiator
–
(3)
73
(4)
122
140
157
–
–
(3)
40
–
(4)
42
–
Endothermic signal
111
146
179
229
41
–
–
Exothermic signal
107
143
245
–
225
–
223
–
1696
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
to that of BA-a (2). The combination of R-a (1) and initiator
(3) leads to one of the lowest known polymerization temper-
atures for benzoxazine systems with a DSC onset tempera-
ꢀ
ture below 100 C.
Infrared spectroscopy was applied for structure determina-
tion of poly(R-a). The polymerization of R-a (1) was per-
formed (a) with both initiators (3) and (4) and also (b)
without initiator. For both cases (a) and (b), two tempera-
ꢀ
ture profiles were carried out: (A) 24 ꢀh at 120 C and (B) 2
ꢀ
h at 180 C and additional 2 h at 200 C. The reaction of R-a
(1) without initiator and at temperature profile A) (24 h at
120 ꢀC) is not further discussed because it is uncompleted
under given conditions. The same counts for the polymeriza-
tion of BA-a (2) with and without initiator in case of heating
profile A). The IR spectra for complete conversion to poly(R-
a) of both the thermally induced and the super acid initiated
polymerization are identical. Poly(R-a) exhibits typical broad
IR signals compared to the monomer (1) as can be seen in
Figure 2. The aryl ether signals at 1252 and 1043 cm21 dis-
appeared after polymerization so that the presence of the
phenoxy-type structure can be excluded. The polymer spec-
trum shows an additional signal at 1290 cm21 designating
the formed phenolic-type structure. Furthermore, the signals
for 1,2,4,5-tetrasubstituted aromatics (isomer II) at 925 and
867 cm21 disappeared after polymerization. This observation
indicates a change in the degree of substitution of the aro-
matic ring. The broad signal at 810 cm21, which has the
same value as the signal of 1,2,3,4-tetrasubstituted aromatics
(isomer I), may the result of a pentasubstituted aromatic
system while the presence of residual isomer (I) in the
FIGURE 2 Infrared spectra of (a) R-a (1) and (b) poly(R-a).
based benzoxazine (2) exhibits only one exothermic peak for
all cases shown in Table 2. The polymerization of pure BA-a
ꢀ
(2) shows a Tpeak at 245 C but, in the presence of initiators
(3) and (4), which feature identical reactivity relating to BA-
a (2), Tpeak is 20 ꢀC lower. At such a high temperature, no
differences in reactivity for both initiators were observed.
Initiators (3) and (4) have already been both activated at
significantly lower temperature compared to the polymeriza-
tion temperature of the benzoxazine so that the super acid
or carbocation is already available in the matrix before the
polymerization temperature is reached. Although the activity
of (3) and (4) is the same, it is also possible to reduce the
polymerization temperature of commercial BA-a (2) with
both latent initiators. The effect of the latent super acid (4)
on the new resorcinol-based benzoxazine (1) is very similar
SCHEME 2 Cationic polymerization of R-a (1) to pentasubstituted poly(R-a) with phenolic-type structure.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699
1697

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 3 Data of Thermogravimetric Analysis for Poly(R-a) and Poly(BA-a)
Samples
T5% (ꢀC)
Char Yield (%)
Dm(1) (%)
Tmax(1) (ꢀC)
Dm(2) (%)
Tmax(2) (ꢀC)
Poly(R-a)
267
327
56
37
24
42
292
396
20
22
416
459
Poly(BA-a)
polymer can be excluded because of its missing aryl ether
signals. We propose the formation of a pentasubstituted
phenolic-type poly(R-a) by cationic polymerization as
depicted in Scheme 2. The structure of the thermally induced
and super acid-initiated bisphenol A-based polybenzoxazine
was also verified by IR. The infrared spectrum exhibits a
phenolic-type polybenzoxazine for thermally induced poly-
merization corresponding to the literature.18 As expected,
the structures of the super acid-initiated and thermally
induced polymerizations are the same.
pounds, which decompose preferentially to amines, phenolic
compounds, and benzene derivatives.38–40 In addition, the
homolytic decomposition of the hydroxyl group in phenolic
compounds to form radical benzene derivatives which are
able to recombine to biphenyls was described. In considera-
tion of Hemvichians observations poly(R-a) may be able to
form a diradical because of two hydroxyl groups in the res-
orcinol unit. This recombination of radicals may also result
in higher char formation after common pyrolysis reactions.
Furthermore, poly(R-a) has not got any weak iso-propyl link-
age which is known for its reduction of char yield.1 In sum-
mary poly(R-a) degrades at lower temperature but yields in
significant higher char formation than poly(BA-a). It is also
notable that thermal decomposition is the same independent
from the usage of latent initiators (3) and (4). This observa-
tion shows that the generated super acid does not influence
thermal stability negatively.
TGA was carried out under nitrogen atmosphere to charac-
terize the thermal stability of the polymers and to determine
ꢀ
the char yield at 600 C. The T5% value is chosen as measure
of thermal stability, which represents the temperature at five
percent mass loss. All following TGAꢀdata are summarized in
Table 3. Poly(R-a) has a T5% at 267 C which is considerably
lower compared to 327 ꢀC of poly(BA-a). That means pol-
y(BA-a) exhibits higher thermal stability than the resorcinol
based pendant. However, the thermogravimetric curves of
poly(R-a) and poly(BA-a) cross each other at around 400 ꢀC
so that, henceforth, the polymeric resorcinol-based benzoxa-
zine shows a higher thermal resistance as shown in Figure
3. Furthermore, poly(R-a) yields 56% char at 600 ꢀC which
is significantly higher compared to that one of the commer-
cial polymerized benzoxazine with a value of 37%. Both
polymers exhibit two main mass losses Dm differing in their
amount and in their temperature of maximum weight loss
CONCLUSIONS
A new resorcinol-based benzoxazine was synthesized show-
ing higher reactivity compared to common benzoxazines due
to its specific structure. This structure offers complex elec-
tronic effects because of two meta positioned aryl ethers in
the precursor and due to both the correspondent hydroxyl
groups in the ring-opened form and the formed alkyl amino
group in ortho- or para-position. Polymerization of benzoxa-
zines is sensitive to electronic effects resulted from their
structure as indicated for the polymerization of R-a and BA-
a. The complex influence and direction of electronic effects is
not yet understood completely and will be object of further
examinations in which the structure of R-a will be varied
systematically.
Tmax. Poly(R-a) has distinct lower Tmax values but also lower
amounts of mass losses which at last ends in higher char
yield as described before. Hemvichian et al. demonstrated
the degradation mechanism of polybenzoxazine model com-
A polymerization DSC onset temperature of about 100 ꢀC
can be obtained by application with a thermolatent initia-
tor. This is one of the lowest known curing temperatures
for a benzoxazine. A complete conversion of R-a (1) to
poly(R-a) is achieved by heating 24 h at 120 ꢀC with a
thermolatent cationic initiator. Polymerization of pure ben-
zoxazine does not only take place at lower temperature
compared to common benzoxazines but it can be also fur-
ther accelerated by application of latent cationic initiators
based on organic sulfonium hexafluoroantimonates. The
decreased polymerization temperature of
a commercial
benzoxazine was not as pronounced but with 20 ꢀC, it is
significant. The drawback of high curing temperature can
be overcome by the application of latent initiators prefera-
bly in combination with the resorcinol-based benzoxazine.
Although the curing temperature is drastically reduced by
the latent initiators, the thermal stability of formed poly-
mers is not influenced.
FIGURE 3 Thermogravimetric analysis of poly(R-a) and pol-
y(BA-a) measured under nitrogen atmosphere with a heating
rate of 10 K min21
.
1698
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
ACKNOWLEDGMENTS
21 J. Dunkers, H. Ishida, J. Polym. Sci. Part A: Polym. Chem.
1999, 37, 1913–1921.
Martina Osmers is gratefully acknowledged for supporting the
nomenclature of the isomers of resorcinol-based benzoxazine.
22 H. Oie, A. Mori, A. Sudo, T. Endo, J. Polym. Sci. Part A:
Polym. Chem. 2012, 50, 4756–4761.
23 Z. Beyazkilic, M. U. Kahveci, B. Aydogan, B. Kiskan, Y.
Yagci, J. Polym. Sci. Part A: Polym. Chem. 2012, 50, 4029–4036.
REFERENCES AND NOTES
24 A. Sudo, S. Hirayama, T. Endo, J. Polym. Sci. Part A: Polym.
Chem. 2010, 48, 479–484.
1 H. Ishida, T. Agag, Handbook of Benzoxazine Resins; Elsevier:
Amsterdam, Oxford, 2011.
€
25 C. Liu, D. Shen, R. M. Sebastian, J. Marquet, R. Schonfeld,
2 B. S. Rao, K. R. Reddy, S. K. Pathak, A. R. Pasala, Polym. Int.
2005, 54, 1371–1376.
Macromolecules 2011, 44, 4616–4622.
26 F. Kasapoglu, I. Cianga, Y. Yagci, T. Takeichi, J. Polym. Sci.
Part A: Polym. Chem. 2003, 41, 3320–3328.
3 Y. Yagci, B. Kiskan, N. N. Ghosh, J. Polym. Sci. Part A:
Polym. Chem. 2009, 47, 5565–5576.
27 M. A. Tasdelen, B. Kiskan, Y. Yagci, Macromol. Rapid Com-
mun. 2006, 27, 1539–1544.
4 N. N. Ghosh, B. Kiskan, Y. Yagci, Prog. Polym. Sci. 2007, 32,
1344–1391.
28 T. Endo, H. Uno, J. Polym. Sci. Polym. Lett. Ed. 1985, 23,
359–363.
€
5 C. Liu, D. Shen, R. M. Sebastian, J. Marquet, R. Schonfeld,
Polymer 2013, 54, 2873–2878.
29 K. Morio, H. Murase, H. Tsuchiya, J. Appl. Polym. Sci. 1986,
32, 5727–5732.
6 F. W. Holly, A. C. Cope, J. Am. Chem. Soc. 1944, 66, 1875–
1879.
€
30 A. Hartwig, K. Koschek, A. Luhring, O. Schorsch, Polymer
2003, 44, 2853–2858.
7 W. J. Burke, J. Am. Chem. Soc. 1949, 71, 609–612.
8 H. J. Kim, Z. Brunovska, H. Ishida, Polymer 1999, 40, 4365–
4376.
31 J. C. Crivello, In Ring-Opening Polymerization; D. J. Bru-
nelle, Ed.; Hanser Verlag: Munchen, 1995; pp 157–196. Chapter
€
9 T. Agag, T. Takeichi, Macromolecules 2003, 36, 6010–6017.
10 T. Agag, T. Takeichi, Macromolecules 2001, 34, 7257–7263.
5.
32 Y. Yagci, I. Reetz, Prog. Polym. Sci. 1998, 23, 1485–1538.
11 Z. Brunovska, H. Ishida, J. Appl. Polym. Sci. 1999, 73, 2937–
33 J. V. Crivello, K. Dietliker, Photoinitiators for Free Radical
Cationic & Anionic Photopolymerization, 2nd ed.; Wiley-VCH:
Weinheim, 1998.
2949.
12 H. Ishida, S. Ohba, Polymer 2005, 46, 5588–5595.
13 B. Kiskan, Y. Yagci, J. Polym. Sci. Part A: Polym. Chem.
2007, 45, 1670–1676.
€
34 P.-E. Sundell, S. Jonsson, A. Hult, J. Polym. Sci. Part A:
Polym. Chem. 1991, 29, 1535–1543.
14 R. Andreu, M. A. Espinosa, M. Galia, V. Cadiz, J. C. Ronda,
J. A. Reina, J. Polym. Sci. Part A: Polym. Chem. 2006, 44,
1529–1540.
35 H. Oie, A. Mori, A. Sudo, T. Endo, J. Polym. Sci. Part A:
Polym. Chem. 2013, 51, 3867–3872.
36 A. Kikkawa, T. Takata, T. Endo, J. Polym. Sci. Part A: Polym.
Chem. 1991, 29, 1089–1095.
15 B. S. Rao, A. Palanisamy, React. Funct. Polym. 2011, 71,
148–154.
R
V
37 Huntsman Selector Guide, Araldite Benzoxazine Thermoset
Resins, Huntsman Corporation, 2009.
16 M. Baqar, T. Agag, H. Ishida, S. Qutubuddin, J. Polym. Sci.
Part A: Polym. Chem. 2012, 50, 2275–2285.
17 A. Sudo, H. Yamashita, T. Endo, J. Polym. Sci. Part A:
Polym. Chem. 2011, 49, 3631–3636.
38 K. Hemvichian, H. D. Kim, H. Ishida, Polym. Degrad. Stab.
2005, 87, 213–224.
18 Y.-X. Wang, H. Ishida, Polymer 1999, 40, 4563–4570.
19 Y.-X. Wang, H. Ishida, Macromolecules 2000, 33, 2839–2847.
20 P. Chutayothin, H. Ishida, Macromolecules 2010, 43, 4562–4572.
39 K. Hemvichian, A. Laobuthee, S. Chirachanchain, H. Ishida,
Polym. Degrad. Stab. 2002, 76, 1–15.
40 K. Hemvichian, H. Ishida, Polymer 2002, 43, 4391–4402.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 1693–1699
1699