Substituent effects of N-alkyl groups on thermally induced polymerization behavior of 1,3-benzoxazines

Article abstract of DOI:10.1002/pola.24026

Thermally induced polymerizations of a series of 1,3-benzoxazines with a variety of substituents on the nitrogen atom were investigated in detail, particularly in the following three aspects of the polymerization: (1) N-alkyl1,3-benzoxazines are much more reactive than N-phenyl1,3-benzoxazine. (2) The polymerization rate depended on the bulkiness of the N-substituent The bulkier the substituent was, the slower the polymerization was. (3) The poly-merizations accompanied weight loss due to the elimination of the corresponding imine (R-N = CH₂), and its extent became larger when R was more bulky.

Full text of DOI:10.1002/pola.24026

Substituent Effects of N-Alkyl Groups on Thermally Induced
Polymerization Behavior of 1,3-Benzoxazines
ATSUSHI SUDO, LONG-CHAO DU, SHOJI HIRAYAMA, TAKESHI ENDO
Molecular Engineering Institute, Kinki University, Kayanomori 11-6, Iizuka, Fukuoka 820-8555, Japan
Received 16 November 2009; accepted 13 March 2010
DOI: 10.1002/pola.24026
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Thermally induced polymerizations of a series of
,3-benzoxazines with a variety of substituents on the nitro-
merizations accompanied weight loss due to the
elimination of the corresponding imine (R-N ¼ CH ), and its
extent became larger when
was more bulky. VC 2010
1
2
gen atom were investigated in detail, particularly in the
R
following three aspects of the polymerization: (1) N-alkyl-
Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 48:
2777–2782, 2010
1
1
,3-benzoxazines are much more reactive than N-phenyl-
,3-benzoxazine. (2) The polymerization rate depended on
the bulkiness of the N-substituent. The bulkier the substitu-
KEYWORDS: benzoxazine; crosslinking; kinetics (polym.); ring-
ent was, the slower the polymerization was. (3) The poly-
opening polymerization
INTRODUCTION Benzoxazines are a class of heterocyclic
N-alkyl group on the polymerization behavior have been not
monomers that can be easily synthesized from the corre-
fully clarified yet.
1
–5
sponding phenols, amines, and formaldehyde.
The combi-
On the other hand, the current applications of benzoxazines
have encountered a serious elimination of volatiles from
their polymerization systems. So far, thermal degradation of
poly(benzoxazine) and the corresponding evolution of vola-
tiles have been investigated; however, evolution of volatiles
during polymerization of benzoxazine has not been studied
in detail.
nations of these precursors are almost unlimited, permitting
diverse molecular design of benzoxazines so that their corre-
sponding polymers would have excellent properties suitable
for specific applications. Intrinsically, poly(benzoxazine)s
27
6
7
exhibit high mechanical strength, thermal stability, and
8
durability under humid environment, which have attracted
considerable attention of polymer scientists and engineers
and prompted their tremendous efforts for the development
and applications of a variety of benzoxazine-type monomers
With these backgrounds in our mind, we prepared a series
of monofunctional benzoxazines 1 and that of bifunctional
benzoxazines 2 having different substitutents on nitrogen
atom (Scheme 1). Rates of thermally induced polymeriza-
tions of 1, heat evolution behaviors during thermally induced
polymerizations of 2 were studied with focusing on their
dependences on the bulkiness of N-alkyl groups. Another
major content of this article is our investigation on the
weight loss by evolution of volatiles during the polymeriza-
tions of 2, which showed clear dependence on the bulkiness
of N-alkyl group.
9
–19
having various functional groups.
Due to the current expansion of various application fields of
benzoxazine-type monomers and the concurrent develop-
ment of a wide structural variety of them, more fundamental
aspects such as the correlation between their structures and
reactivity have become more important. Recently, Ronda and
coworkers. reported their investigation on the polymeriza-
tion behavior of N-aryl benzoxazines, which clarified the
2
0
influence of electronic nature of the N-aryl group. Ishida
and coworkers. reported the boron trifluoride-catalyzed pol-
ymerizations of several N-alkylbenzoxazines by FTIR.
2
1
EXPERIMENTAL
Besides, preparation of a series of N-alkyl benzoxazines and
Synthesis of Benzoxazines
2
8
23
29,30
31
32
33
their detailed spectroscopic analyses have been also
1a, 1b, 1c,
1d, 1e, and 2a-2e were synthesized
2
2,23
34
reported.
In addition, polymerizations of various bifunc-
by a general procedure for synthesis of benzoxazines. All
of these monomers were purified by column chromatography
tional N-alkyl benzoxazines and the corresponding volume
2
4
changes have been reported.
There have been a few
(SiO , eluent ¼ hexane/ethyl acetate (9/1-2/1 in v/v). 1b
2
articles on the polymerization mechanisms of benzoxa-
and 1c were obtained as viscous oils. The other benzoxa-
2
5,26
zines.
However, effects of steric and electronic nature of
zines were obtained as solids, which were further purified
Correspondence to: T. Endo (E-mail: tendo@mol-eng.fuk.kindai.ac.jp)
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 48, 2777–2782 (2010) V
C
2010 Wiley Periodicals, Inc.
SUBSTITUENT EFFECTS OF N-ALKYL GROUPS, SUDO ET AL.
2777

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 Thermally induced polymerizations of monofunc-
tional benzoxazines 1.
(
DSC) were performed with a SEIKO EXSTAR6000 (Seiko
SCHEME 1 Monofunctional and bifunctional benzoxazines.
Instruments).
Thermally Induced Polymerizations of Monofunctional
Benzoxazines 1
by recrystallization from ethyl acetate. The purity was
checked by H NMR. As typical examples, the spectra of 1a
1
General Procedure: Benzoxazine 1 (43.8 mmol) was divided
into six portions and each of them was placed in a test tube.
Argon inlets were attached to the resulting six test tubes and
then they were heated in an oil bath at 150 C. From time to
time, these test tubes were taken away from the oil bath one-
and 2a are shown in Figure 1.
ꢀ
Measurements
1
H NMR spectra were recorded in chloroform-d (CDCl ) on a
3
by-one, and each of the mixture was dissolved in CDCl
analyzed by H NMR to determine conversions of 1.
3
and
1
JEOL EX-300 ( H, 300 MHz) spectrometer, with tetramethyl-
1
silane (TMS) as an internal standard, acquisition period ¼
5
.45 s; pulse delay ¼ 1.55 s; accumulation times ¼ 32.
Thermally Induced Polymerizations of Bifunctional
Benzoxazines 2
General procedures: (1) 10 mg of 2 was put into an alumi-
Thermogravimetric analysis (TG), differential thermal analy-
sis (DTA), and differential scanning calorimetric analysis
num pan, placed in the DSC instrument, and heated from
ꢀ
ꢀ
3
0 to 300 C at a rate of 10 C/min under nitrogen flow to
study the corresponding temperature-heat flow relationship.
(
2) 500 mg of 2 was placed on a glass dish and heated in an
ꢀ
oven at 180 or 200 C for 3 h. The resulting weight was
measured to determine the weight loss during the thermally
induced polymerization. (3) 5 mg of 2 was put into an alu-
minum pan, placed in the TG/DTA instrument, and heated
from 30 to 300 C at a rate of 10 C/min under nitrogen
flow to examine the corresponding temperature-weight loss
relationship.
ꢀ
ꢀ
RESULTS AND DISCUSSION
Thermally Induced Polymerization Behavior of
Monofunctional Benzoxazine 1
The series of N-alkyl benzoxazines 1a-1d were heated at
ꢀ
1
50 C to examine their thermally induced ring-opening
polymerizations (Scheme 2). The progress of the polymeriza-
1
tions was monitored by H NMR, and the corresponding
time-conversion relationships are shown in Figure 2. The
most reactive monomer was N-methyl benzoxazine 1a, and
the second most reactive one was N-n-propyl benzoxazine
1
1
b. N-t-butyl benzoxazine 1c and N-cyclohexyl benzoxazine
d were less reactive, implying that polymerization rate sig-
nificantly decreased as the bulkiness of N-alkyl group
ꢀ
became larger. In contrast, at 150 C, N-phenyl benzoxazine
1e did not undergo the polymerization at all, even though
the bulkiness of phenyl group is similar to or less than that
of cyclohexyl group. This significant difference in reactivity
would be caused by the difference in electronic nature
1
FIGURE 1 H NMR spectra of monofunctional benzoxazine 1a
and bifunctional benzoxazine 2a.
2778
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 2 Time-conversion relationships for the thermally induced
ꢀ
polymerizations of monofunctional benzoxazines 1 at 150 C.
FIGURE 3 DSC-heat evolution profiles for the thermally
induced polymerizations of bifunctional benzoxazines 2.
between alkyl and aromatic groups. Another remarkable dif-
ference between the polymerizations of N-alkyl benzoxazines
and N-phenyl benzoxazine was found in structure of the
obtained polymers: Previously, we reported that the poly-
merization of 1e afforded the corresponding polymer having
N,O-acetal structure in the main chain, and its thermally
induced rearrangement gave polymer 3 having phenolic
structure. In contrast, formation of N,O-acetal structure
was not detected during the polymerizations of N-alkylben-
zoxazines, implying that these polymerizations afforded the
phenolic polymers 3 directly.
the behavior of N-cyclohexyl benzoxazine 2c: In spite of the
higher bulkiness of cyclohexyl group than n-propyl group,
the peak top temperature observed in the DSC analysis of
the polymerization of 2c was slightly lower than that of 2b.
It is noteworthy that the present study with using DSC
clearly revealed that N-phenyl benzoxazine was much less
reactive than N-alkyl benzoxazines. The heat evolution by
the polymerization of N-phenyl benzoxazine 2e started from
3
5
ꢀ
ꢀ
around 250 C, implying its reactivity below 200 C would
be negligible. In fact, the monofunctional analogue 1e did
ꢀ
not undergo the polymerization at 150 C at all, and our pre-
Thermally Induced Polymerizations of Bifunctional
Benzoxazine 2
vious study has already revealed that more than 3 h was
necessary for the complete consumption of 1e at 180 C.
ꢀ
35
Thermally induced polymerizations of the series of bifunc-
tional benzoxazines 2 were carried out with monitoring the
corresponding heat evolution by DSC (Scheme 3). The
Weight Loss During Thermally Induced Polymerization
of Bifunctional Benzoxazine 2
ꢀ
benzoxazines were heated from 30 to 300 C in a rate of 10
ꢀ
C/min under nitrogen flow, and the resulting heat evolution
As was described in the introduction, one of the drawbacks
of benzoxazine is the out-gassing issue during the polymer-
ization process, which causes serious foaming of the prod-
ucts, formation of voids in the cured materials, and pollution
of working environment by volatile organic compounds.
However, the extent of this issue and its correlation with
profiles are shown in Figure 3. Among them, that of
N-methyl benzoxazine 2a indicated its peak top at 191 C,
ꢀ
which was much lower than those for the other benzoxa-
zines to imply the highest reactivity of 2a. These profiles
exhibited a clear tendency, that is, the bulkier the N-alkyl
group was, the higher the peak top temperature was. This
tendency was in good agreement with the dependence of the
reactivity of monofunctional benzoxazines 1 on the bulkiness
of the substituent on nitrogen atom. Only the exception was
TABLE 1 Substituent Effects on Weight Loss During Thermally
a
Induced Polymerization of 2
ꢀ
ꢀ
R
Weight Loss/% at 180 C
Weight Loss/% at 200 C
Me
1.6
2.0
4.2
7.7
8.5
1.9
1.9
4.9
12
n-Pr
cyc-Hex
t-Bu
Ph
8.7
SCHEME 3 Thermally induced polymerizations of bifunctional
benzoxazines 2.
a
500 mg of 2 was heated on a glass dish in an oven for 3 h.
SUBSTITUENT EFFECTS OF N-ALKYL GROUPS, SUDO ET AL.
2779

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
ꢀ
ꢀ
FIGURE 4 Time-weight loss relationships for the polymerizations of 2 under isothermal conditions: (a) At 180 C and (b) at 200 C.
molecular structure of benzoxazine have been not clarified
yet.
the weight loss occurred mainly in the early stages. For
example, when N-methyl benzoxazine 2a was heated at 180
ꢀ
C, 1.8% weight loss occurred within 20 min, and after that,
Table 1 shows the weight losses that occurred by heating
there was no increase in the weight loss. Similarly, in the
ꢀ
bifunctional benzoxazines 2 at 180 or 200 C under air, on
ꢀ
case of N-n-propyl benzoxazine 2b, its weight loss at 180 C
an uncovered glass dish in an oven. When N-methyl benzoxa-
occurred within 40 min to reach the plateau at 2.0%. When
N-cyclohexyl benzoxazine 2c was heated, its weight loss con-
tinued for a longer period with approaching the plateau at
zine 2a and N-n-propyl benzoxazine 2b were heated at
ꢀ
1
80 C for 3 h, nearly 2% of the initial weights were lost.
ꢀ
The weight losses of the other monomers at 180 C were
more significant, and were further increased at 200 C. The
4.3% more slowly. In the case of N-t-butyl benzoxazine 2d,
ꢀ
its weight loss continued throughout the heating process to
reach 14% at 180 min; however the major part of it was
observed before 60 min. In addition, a similar tendency was
most remarkable feature observed herein was the strong
dependence of weight loss on bulkiness of R on the nitrogen
atom, that is, the higher bulkiness of R lead to the larger
weight loss.
ꢀ
observed in the isothermal experiments at 200 C.
These results can be summarized as follows: (1) The
dependence of the weight loss on the bulkiness of R on
nitrogen atom studied with TG was in good accordance with
Although these experiments demonstrated the significance of
the weight loss caused by evolution of volatiles that leads to
the out-gassing issue, they did not clarify the mechanism for
it. The important point is whether the weight loss occurred
‘
‘during’’ or ‘‘after’’ the polymerization. To clarify this point,
ꢀ
benzoxazines were heated isothermally at 180 or 200 C for
h in a TG instrument. Figure 4 shows the corresponding
3
time-dependences of the weight losses, which clarified that
FIGURE 5 Schematic illustration of the apparatus to trap the
SCHEME 4 Possible reaction pathways involved in the ther-
volatiles evolved during the thermally induced polymerization
ꢀ
mally induced polymerization process of benzoxazine.
of bifunctional benzoxazine 2b at 180 C.
2780
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA

ARTICLE
ꢀ
volatiles as a condensate at ꢁ78 C (Fig. 5). Upon heating
.0 g of 2b, roughly 20 mg of the condensate was obtained,
3
which was dissolved in CDCl immediately and then analyzed
3
1
by H NMR. In parallel, we prepared an authentic sample of
1,3,5-tri(n-propyl)hexahydro-1,3,5-triazine (TPHT) 5b (just
by mixing n-propylamine and paraformaldehyde in dioxane
at rt), which is the trimer of N-n-propylimine 4b. Figure 6
1
shows the H NMR spectrum of the volatile fraction and that
of authentic 5b, which clearly revealed the presence of 5b in
the condensed volatile fraction. These results supported the
validity of the postulated mechanism for the weight loss
based on the elimination of imine from the zwitter-ionic
intermediate.
SUMMARY
1
ꢀ
FIGURE 6 H NMR spectra (measured at 25 C) of CDCl
3
solu-
A series of N-alkyl benzoxazines were compared in terms of
polymerization rate and weight loss during the polymeriza-
tion. These two parameters were closely related to each
other and were strongly dependent on bulkiness of alkyl
group on the nitrogen atom: The increased bulkiness of N-
alkyl group decelerate the polymerization to permit another
reaction pathway leading to the release of a larger amount
of N-alkylimine as a volatile. As a result, N-methyl-benzoxa-
zine, which had the smallest alkyl group, underwent the fast-
est polymerization with accompanying the smallest weight
loss.
tions of (a) 1,3,5-tri(n-propyl) hexahydro-1,3,5-triazine (TPHT)
and (b) the condensed volatiles that were evolved during the
ꢀ
thermally induced polymerization of 2b at 180 C.
that observed in the polymerizations on an uncovered glass
plate. In both the cases, the bulkier R was, the larger the
weight loss was. (2) Weight loss of benzoxazine upon heat-
ing occurs mainly in the early stages, implying that a certain
side reaction proceeded to form volatiles in parallel with the
ring-opening polymerization of benzoxazine. (3) The period
required for reaching the plateau of weight loss becomes
longer when the bulkiness of R on the nitrogen atom
becomes larger. This is coincidentally in good agreement
with the tendency that the polymerization became slower
when the bulkiness of R on the nitrogen atom becomes
higher.
REFERENCES AND NOTES
1
2
Burke, W. J. J Am Chem Soc 1949, 71, 609–612.
Schreiber, H. German Offen 2,255,504, (1973).
These tendencies can be interpreted into the following possi-
ble mechanisms illustrated in Scheme 4. Step A is a ther-
mally reversible ring-opening reaction of benzoxazine into
the corresponding zwitter-ionic intermediate (I). This inter-
3 Schreiber, H. German Offen 2,323,936, (1973).
4
Reiss, G.; Schwob, J. M.; Guth, G.; Roche M.; Laude, B.
Polym Prepr (Am Chem Soc Div Polym Chem) 1984, 25, 41–42.
3
6,37
5 Reiss, G.; Schwob, J. M.; Guth, G.; Roche M.; Laude, B.
Advances in Polymer Synthesis; Culbertson, B. M.; McGrath,
J. E., Eds.; Plenum Press: New York, 1984, pp 27–49.
mediate was directly observed by NMR
or successfully
3
8
trapped by its chemical transformation, supporting our
postulation that it would be the key species in the polymer-
ization of benzoxazine. Step B would involve (1) polyaddition
of the zwitter-ionic intermediate to safford poly(benzoxa-
zine) and/or (2) ring-opening polymerization that proceeds
via nucleophilic attack of benzoxazine monomer to the immi-
nium moiety at the propagating chain end. When this poly-
merization is hindered by bulkiness of R, another reaction
pathway, that is, dissociation of the zwitter-ionic intermedi-
ate (I) to give another zwitter-ionic intermediate (II) would
become significant. This dissociation is accompanied by
release of imine 5, which can vaporize from the polymeriza-
tion system to cause the weight loss. Imine 5 would be im-
mediately trimerize to give hexahydrotriazine 6 and/or
hydrolyzed by moisture to give the corresponding amine and
formaldehyde. To confirm whether this postulated mecha-
nism for the weight loss based on liberation of imine 5 is
6
Ishida, H.; Sanders, D. P. J Polym Sci Part B: Polym Phys
000, 38, 3289–3301.
2
7 Ning, X.; Ishida, H. J Polym Sci Part A: Polym Chem 1994,
32, 1121–1129.
8
Ishida, H.; Allen, D. J. J Polym Sci Part B: Polym Phys 1996,
4, 1019–1030.
3
9
Ghosh, N. N.; Kiskan, B.; Yagci, Y. Prog Polym Sci 2007, 32,
344–1391.
1
10 Agag, T.; Takeichi, T. Macromolecules 2003, 36, 6010–6017.
1 Ergin, M.; Kiskan, B.; Gacal, B.; Yagci, Y. Macromolecules
2007, 40, 4724–4727.
1
1
2 Agag, T.; Takeichi, T. Macromolecules 2001, 34, 7257–7263.
1
3 Chernykh, A.; Agag, T.; Ishida, H. Polymer 2009, 50,
3
153–3157.
reasonable or not, N-n-propyl benzoxazine 2b was heated at
ꢀ
1
80 C under a gentle flow of argon gas with collecting the
14 Brunovska, Z.; Ishida, H. J Polym Sci 1999, 73, 2937–2949.
SUBSTITUENT EFFECTS OF N-ALKYL GROUPS, SUDO ET AL.
2781

JOURNAL OF POLYMER SCIENCE: PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
15 Qi, H.; Ren, H.; Pan, G.; Zhuang, Y.; Huang, F.; Du, L. Polym
27 Hemvichian, K.; Kim, H. D.; Ishida, H. Polym Degrad Stabil
Adv Technol 2009, 20, 268–272.
2005, 87, 213–214.
16 Liu, Y.-L.; Chou, C.-I. J Polym Sci Part A: Polym Chem 2005,
28 Wang, Y.-X.; Ishida, H. Macromolecules 2000, 33,
43, 5267–5282.
2839–2847.
17 Reina, J. A. J Polym Sci Part A: Polym Chem 2006, 44,
29 Chylinska, J. B.; Urbanski, T.; Mordarski, M. J Med Chem
1529–1540.
1963, 6, 484–487.
18 Yei, D.-R.; Fu, H.-K.; Chen, W.-Y.; Chang, F.-C. J Polym Sci
30 Phongtamrug, S.; Chirachanchai, S.; Tashiro, K. Macromol
Part A: Polym Chem 2006, 44, 347–358.
Symp 2006, 242, 40–48.
19 Liu, Y.; Zhang, W.; Chen, Y.; Zheng, S. J Appl Polym Sci
31 Papachristos, M. J.; Evans, J. S. U.K. Pat. GB 2,308,849 A (1997).
2006, 99, 927–936.
3
2 Kimura, H.; Matsumoto, A.; Hasegawa, K.; Ohtsuka, K.;
20 Andreu, R.; Reina, J. A.; Ronda, J. C. J Polym Sci Part A:
Fukuda, A. J Appl Polym Sci 1998, 68, 1903–1919.
Polym Chem 2008, 46, 3353–3366.
3
3 Kim H.-D.; Ishida, H. Macromolecules 2003, 36,
21 Cid, J. A.; Wang, Y.-X.; Ishida, H. Polym Comp 1999, 7,
8320–8329.
409–420.
3
4 Dunkers, J.; Ishida, H. Spectrochim Acta 1995, 51A,
2
2 Neuvonen, K.; Pihlaja, K. Mag Res Chem 1990, 28, 239–245.
3 Dunkers, J.; Ishida, H. Spectrochim Acta Part A: Mol Biomol
855.
2
35 Sudo, A.; Kudoh, R.; Nakayama, H.; Arima, K.; Endo, T. Mac-
Spec 1995, 51A, 1061–1071.
romolecules 2008, 41, 9030–9034.
24 Ishida, H.; Low, H. Y. Macromolecules 1997, 30, 1099–1106.
36 McDonagh, A. F.; Smith H. E. J Org Chem 1968, 33, 8–12.
25 Dunkers, J.; Zarate, E. A.; Ishida, H. J Phys Chem 1996, 100,
37 P e´ rez, S.; L o´ pez, C.; Caubet, A.; Roig, A.; Molins, E. J Org
13514–13520.
Chem 2005, 70, 4857–4860.
26 Hayakawa, T.; Osanai, Y.; Niizeki, K.; Haba, O.; Ueda, M.
38 Kudoh, R.; Sudo, A.; Endo, T. Macromolecules 2009, 42,
High Perform Polym 2000, 12, 237–246.
2327–2329.
2782
INTERSCIENCE.WILEY.COM/JOURNAL/JPOLA