Tailor-made and chemically designed synthesis of coumarin-containing benzoxazines and their reactivity study toward their thermosets

Article abstract of DOI:10.1002/pola.27994

Coumarins are used as a natural renewable resource to synthesize coumarin-containing benzoxazine resins. The coumarin-containing benzoxazines are fully characterized in terms of their chemical structure by Fourier-transform infrared spectroscopy and proto

Full text of DOI:10.1002/pola.27994

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Tailor-Made and Chemically Designed Synthesis of Coumarin-Containing
Benzoxazines and Their Reactivity Study Toward Their Thermosets
Pablo Froimowicz,1,2 Carlos Rodriguez Arza,2 Seishi Ohashi,2 Hatsuo Ishida2
1Design and Chemistry of Macromolecules Group, Institute of Technology in Polymers and Nanotechnologies (ITPN), UBA-
CONICET, School of Engineering, University of Buenos Aires, Las Heras 2214 (CP 1127AAR), Buenos Aires, Argentina
2Department of Macromolecular Science and Engineering, Case Western Reserve University, Cleveland, Ohio 44106 7202
Correspondence to: P. Froimowicz (E-mail: pxf106@case.edu) or H. Ishida (E-mail: hxi3@case.edu)
Received 7 October 2015; accepted 11 November 2015; published online 17 December 2015
DOI: 10.1002/pola.27994
ABSTRACT: Coumarins are used as a natural renewable resource
to synthesize coumarin-containing benzoxazine resins. The
coumarin-containing benzoxazines are fully characterized in
terms of their chemical structure by Fourier-transform infrared
spectroscopy and proton nuclear magnetic resonance spectros-
copy. The influence of electronic effects caused by the substitu-
ents on the polymerization temperature is also evaluated.
Thermal properties of the resulting thermosets are characterized
by differential scanning calorimetry and thermogravimetric anal-
ysis, showing good stability and char yields higher than 50%.
The coumarin-containing polybenzoxazine thermosets show Tg
values in the range between 160 and 190 8C. Thus, the herein
presented coumarin-containing benzoxazine resins are proven
to be competitive monomers when compared with other
petroleum-based benzoxazine resins toward the generation of
C
V
high-performance thermoset.
2015 Wiley Periodicals, Inc. J.
Polym. Sci., Part A: Polym. Chem. 2016, 54, 1428–1435
KEYWORDS: chemical design; polybenzoxazines; renewable
resources; sustainable chemistry; thermosets
to polymerize at exact programed temperatures with high
precision without affecting the final properties of the
polybenzoxazines.
INTRODUCTION Polybenzoxazine is a relatively new class of
thermoset material attracting researchers’ attention from
both scientific and industrial communities. For instance,
while much work and effort are still being invested in fur-
ther developing deeper understanding and new formulations
for this kind of polymers, benzoxazines and polybenzoxa-
zines are already commercially available. Polybenzoxazines
are synthesized by cationic ring-opening polymerization of
1,3-benzoxazine resins.1 As in other polymers, the properties
exhibited by polybenzoxazines strongly depend on the nature
of the benzoxazine monomer and the manner, in which the
polymerization is carried out. Polybenzoxazines can be poly-
merized without added initiator or catalyst requiring in gen-
eral temperatures in the range of 150–230 8C, although
some benzoxazines have been shown to polymerize at lower
temperature.2 Thus, the use of catalysts aiming at reducing
the polymerization temperature has also been investi-
gated.3–5 However, these complex systems in which addi-
tional compounds must be added might not be fully
beneficial from an economic nor practical standpoint since
extra effort must be put on the processing and this could
also affect the very desirable properties of the final prod-
uct.6,7 On the contrary, it would certainly be advantageous if
the monomeric benzoxazines themselves could be tailor-
made and chemically designed in such a manner to be able
It would be also beneficial if new industrial polymeric sys-
tems can be prepared from natural renewable materials.8–10
As known, the synthesis of benzoxazines involves phenolic
compounds. They are usually derived from the petrochemis-
try but could easily be replaced by other biomass-based phe-
nolic compounds.11–14 The feasibility of using biomass
renewable resources as raw materials for the synthesis of
benzoxazine resins has been reported15 by not only substi-
tuting the phenolic moiety but also the amine portion16 of
the resins. Expectedly, attempts were also made to replace
both the phenolic and the amine moieties,17 placing the ben-
zoxazines as pending groups in existing natural occurring
polymers,18 and copolymerizing benzoxazines with natural
renewable oligomers.19
In this regard, coumarins can be considered as a viable alter-
native for the synthesis of benzoxazines12 since they are in
general natural and renewable compounds in themselves, or
they can be directly obtained from other derivatives from the
biomass such as resorcinol.20 Of particular interest are cou-
marins bearing a hydroxyl group in position 7, such as umbel-
liferone and 4-methylumbelliferone. These two compounds
C
V
2015 Wiley Periodicals, Inc.
1428
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
present their ortho positions unsubstituted with respect to
the hydroxyl group, thus making them good phenolic reac-
tants toward the synthesis of 1,3-benzoxazines. Additionally, if
the focus is addressed to the benzoxazine nuclei the coumarin
moiety (e.g., umbelliferone) could be viewed as a substituent,
which will provide specific properties and reactivity to the
resulting benzoxazine. Then, substituted coumarin moieties
(for instance, 4-methylumbelliferone) should in consequence
be viewed as substituted substituents, which will modify the
specific properties and reactivity of the benzoxazine. Knowing
the features of each moiety forming the entire molecule, one
may be able to understand its behavior. Nevertheless, the
knowledge might be used to chemically design compounds
with tailored properties and reactivity.
4-methyl-9-phenyl-9,10-dihydro-2H,8H-Chromeno[8,7-
E][1,3]oxazin-2-One (MU-a)
MU-a was obtained with a yield of 70%. 1H NMR (600 MHz,
CDCl3, 20 8C) d, ppm: 7.35 (d, 1H, H7), 7.25 (d, 2H, Hb), 7.12 (d,
2H, Ha), 6.94 (t, 1H, Hc), 6.76 (d, 1H, H8), 6.11 (s, 1H, -C(CH3)@CH-
COO-), 5.42 (s, 2H, –O-CH2-N–), 4.81 (s, 2H, Ar-CH2-N-), and 2.36
(2, 3H, Ar-C(CH3)5CH-). FT-IR m (cm21): 1732 (C@O str.), 1433
(CH2 sciss.), 1356 (aromatic C–N str.), 1313 (–CH2– twist), 1265
(aromatic C–O str.), 1215 (aliphatic C–N str.), 1057 (aliphatic C–O
str.), and 908 (C–H out-of-plane benzoxazine bend.).
Preparation of the Polybenzoxazines
Polymerizations were carried out by heating at a rate of
10 8C/min until the onset of the polymerization tempera-
ture for each monomer, followed by isothermal heating
for 1 h at this temperature. Next, samples were heated at
the same 10 8C/min rate until the exotherm peak temper-
ature, followed by 1 h of isothermal heating. Both onset
and exotherm peak temperature in each system were
obtained from DSC experiments. Thus, U-a was heated at
210 and 220 8C, MU-a at 220 and 232 8C, while P-a at
250 and 260 8C.
Having this motivation, the present work aims to uncover
the contribution of each substituent on coumarin-based ben-
zoxazines and correlate them to the thermal properties. This
article supplements our earlier report on the synthesis and
characterization of a coumarin-functional benzoxazine mono-
mer and its cross-linked material.21
EXPERIMENTAL
Characterization
Proton nuclear magnetic resonance (1H NMR) spectra were
acquired on a Varian Oxford AS600 at a proton frequency of
600 MHz. The average number of transients was 64. A relax-
ation time of 10 s was used for the integrated intensity
Material
Aniline, 4-methylumbelliferone (ꢀ98%), and paraformalde-
hyde (96%) were used as received from Sigma-Aldrich.
Umbelliferone (>98%) was purchased from TCI America. 3-
phenyl-3,4-dihydro-2H-benzo[e][1,3]oxazine (PH-a) was syn-
thesized and purified following a procedure described else-
where.22 Toluene and basic alumina were purchased from
Fischer scientific and used as received.
1
determination of H NMR spectra. Fourier transform infrared
(FT-IR) spectra were recorded using a Bomem Michelson
MB100 FT-IR spectrometer, which was equipped with a deu-
terated triglycine sulfate (DTGS) detector and a dry air purge
unit. Absorption spectra were obtained employing KBr
plates, and using 64 scans at a resolution of 4 cm21. Differ-
ential scanning calorimetry (DSC) measurements were car-
ried out in a TA Instruments DSC Model 2920 with a
nitrogen flow rate of 60 mL/min. Thermograms of the mono-
mers were obtained using a heating rate of 10 8C/min,
whereas the glass transition temperature (Tg) of the poly-
benzoxazines was determined employing a sample mass of
around 10 mg and a heating rate of 20 8C/min. All samples
were sealed in hermetic aluminum pans. Thermal decompo-
sition of the polybenzoxazines was determined by thermo-
gravimetric analysis using a TA Instrument Model Q500 TGA.
TGA analysis was performed in a single heating run from
room temperature to 900 8C (ca 3 mg) at a heating rate of
10 8C/min, with a nitrogen flow rate of 60 mL/min. All ther-
mal analysis and polymerizations were carried out under
nitrogen atmosphere.
Synthesis of Coumarin-Containing Benzoxazines
General procedure:
Aniline (15 mmol), 4-methylumbelliferone (15 mmol), para-
formaldehyde (33 mmol), and toluene (20 mL) were placed
into a 50 mL round-bottomed flask equipped with magnetic
stirring and refluxed for 12 h. The crude product was diluted
with toluene (25 mL) and passed through a basic alumina
column. After solvent removal, the resulting yellow–white
crystals were purified by recrystallizing in toluene at low
temperatures.
9-phenyl-9,10-dihydro-2H,8H-Chromeno[8,7-E][1,3]oxazin-
2-One (U-a)
U-a was obtained with a yield of 65%. 1H NMR (600 MHz,
CDCl3, 20 8C) d, ppm: 7.59 (d, 1H, Ar-CH@CHR (HIV)), 7.27
(t, 2H, (Hb)), 7.22 (d, 1H, (H7)), 7.14 (d, 2H, (Ha)), 6.95 (t,
1H, (Hc)), 6.75 (d, 1H, (H8)), 6.23 (d, 1H, Ar-CH@CH-COOR
(HIII)), 5.43 (s, 2H, Ar-O-CH2-NR (H2)), and 4.81 (s, 2H, Ar-
CH2-NR (H4)). FT-IR m (cm21): 1716 (C@O str.), 1433 (CH2
sciss.), 1360 (aromatic C–N str.), 1317 (–CH2– twist), 1234
(aromatic C–O str.), 1211 (aliphatic C–N str.), 1059 (aliphatic
C–O str.), and 918 (C–H out-of-plane benzoxazine bend.). All
spectral details are reported in Ref. [21].
RESULTS AND DISCUSSION
As described in the Introduction section, the interest of this
work was on designing coumarin-containing benzoxazines
with programmable polymerization temperatures. It is then
helpful to first identify not only each component nuclei, ben-
zoxazine, and coumarin, but also every position on the
resulting coumarin-based benzoxazine molecule. Figure 1
shows the numbering on each individual class of molecule
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435
1429

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
asymmetric stretching bands are found at 1265 and
1057 cm21. Finally, the characteristic band assigned to C–H
out-of-plane bending in the aromatic ring fused to the oxa-
zine ring is observed at 908 cm21. These results indicate the
coexistence of the two nuclei, benzoxazine and coumarin,
which suggest the success synthesis of MU-a.
The chemical structure of MU-a was further studied and
confirmed by 1H NMR spectroscopy. Figure 3 shows the 1H
NMR spectrum. As in FT-IR analysis, chemical shift values of
each signal generated by protons belonging to MU-a show
high similarities to other coumarin-based benzoxazines given
the nature of their chemical structures. Thus, the characteris-
tic resonances belonging to the methylene groups in the oxa-
zine ring are found at 5.42 and 4.81 ppm in the spectrum of
MU-a. Unlike U-a, which bears one proton in position III and
one in position IV observed as doublets at 6.23 and 7.59
ppm (see Experimental section), MU-a contains a single pro-
ton in position III observed as a singlet at 6.11 ppm and a
methyl group at position IV observed as a singlet at 2.36
ppm. As in other reported coumarins substituted in position
VII,23 the absence of a singlet in the entire aromatic region
on the spectrum of MU-a indicates that only one isomer
shown in Scheme 1 and Figure 3 was obtained following the
synthesis procedure herein described.
FIGURE 1 Full numbering in benzoxazine and coumarin nuclei
showing complete identification of each position in each class
of molecule individually. Briefly, Arabic numbers are used in
the benzoxazine nuclei and Latin letters are used in the aniline
portion. Roman numbers describe the different position in the
coumarin nuclei. An apostrophe (’) is used to show that two
positions apparently chemically and/or magnetically equivalent
could, in fact, be different.
Polymerization behavior of both coumarin-containing ben-
zoxazine resins and the unsubstituted PH-a as well as the
thermal properties of their corresponding thermosets were
investigated by DSC and TGA and are discussed next.
nuclei, benzoxazine, and coumarin, as well the resulting
coumarin-containing benzoxazines. It can be seen in the fig-
ure that the conventional numbering system in each nuclei is
respected, using Arabic numbers on the benzoxazine nuclei
and Roman ones on the coumarin. To avoid confusions, in
the resulting coumarin-containing benzoxazines, each num-
ber system is respected.
Figure 4 shows the DSC thermograms for the three monomers
and a summary of the thermal properties is presented in Table 1.
It can be seen from Figure 4 that each of the three thermo-
grams corresponding to each monomeric resins presented
Using two different coumarins bearing one functionalizable
hydroxyl group, such as umbelliferone and 4-methylumbelliferone,
it is possible to design and synthesize two different coumarin-
based benzoxazines. Thus, Scheme 1 shows how umbelliferone-
aniline benzoxazine (U-a) and 4-methylumbelliferone-aniline ben-
zoxazine (MU-a) were successfully prepared via the Mannich con-
densation reaction using aniline as the amine, umbelliferone, and
4-methylumbelliferone as the phenol source and paraformaldehyde
in refluxing toluene.
Formation of the structures of MU-a was studied by FT-IR
spectroscopy (Fig. 2). The spectrum shows expected similar-
ities to other coumarin-containing benzoxazines, except for
the particular methyl group present in position IV. Thus, a
C@O stretching bands is observed at 1732 cm21 for MU-a.
The shift of this C@O stretch band to a higher frequency
compared to the one from U-a (observed at 1716 cm21, see
Experimental section) is due to the methyl group of MU-a
since forces the carbonyl group to be slightly out of plane,
thus interfering with the resonance. Typical benzoxazine’s C–
N–C symmetric and asymmetric stretching are observed at
1356 and 1215 cm21. Similarly, the C–O–C symmetric and
SCHEME
1 Synthesis
of
phenol-aniline
(PH-a),
4-
methylumbelliferone-aniline (MU-a), and umbelliferone-aniline
(U-a) benzoxazine resins from phenol, 4-methylumbelliferone,
umbelliferone, aniline, and paraformaldehyde.
1430
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
This double bond is substituted with a weak electron-
donating methyl group in its very activated position IV,
and also substituted in position 5 (meta position) with the
oxygen from the ester group which is a moderately
electron-donating group.
ꢁ When comparing U-a with MU-a in terms of electronic
contribution, the effects generated by the oxygen in posi-
tion 5 can be overlooked since it should be comparable
for both compounds.
ꢁ C@C double bond substituents are usually very weak
electron-donating groups, whereas activated C@C double
bond via conjugation such as Michael acceptors behaves
as strong electron-withdrawing groups.
As shown in Figure 4, U-a is the most reactive resin since it
has the lowest polymerization temperature followed by MU-a,
and finally PH-a which has the highest polymerization tem-
perature thus being the least reactive resin. Compound PH-a
is known to have the four positions in the aromatic ring only
slightly activated, being position 8 the most reactive one since
it is in the ortho position to its most activating group, the oxy-
gen in the oxazine ring.
FIGURE 2 FT-IR spectrum of 4-methylumbelliferone-aniline
benzoxazine resin (MU-a).
two thermal processes clearly defined, an endothermic peak
attributed to the melting processes of the resin as well as an
exothermic one assigned to their polymerizations.
A
straightforward comparison between PH-a and both
coumarin-containing resins reveals that, in U-a and MU-a,
positions 5 and 6 are occupied by the coumarin moiety,
while positions 7 and 8 are unsubstituted. Interestingly, posi-
tion 8 in the coumarin-containing resins is being simulta-
neously activated by the oxygen atom in the oxazine ring
and the –OC(O)R group, while deactivated by the conjugated
C@C double bond which still directs any reaction to position
8. Thus, it is reasonable to accept that both coumarin-
containing resins are more reactive than PH-a toward the
polymerization reaction. By comparing U-a to MU-a, it can
The benzoxazine resins exhibited melting temperatures of 54
8C for PH-a, 147 8C for U-a, and 153 8C for MU-a, while their
polymerization temperatures presented the following pat-
tern: PH-a polymerizes at the highest temperature (261 8C),
followed by MU-a (232 8C), and U-a at the lowest tempera-
ture (220 8C).
The polymerization temperature pattern can be rationalized
straightforwardly by doing structure–property analysis. Com-
pared to the unsubstituted PH-a, benzoxazines with the cou-
marin substituent have significantly reduced the
polymerization temperature. Motivated by this fact, we have
developed an interest in understanding and correlating the
influence of electronic effects on the reactivity of these ben-
zoxazine resins. To achieve this goal, we will first interpret
the influence of these electronic effects induced by the sub-
stituents on the polymerization mechanism by considering
each substituent separately. Next, the manner in which each
effect may affect the others will also be taken into account.
And finally, the thermal behavior of the entire molecule will
be interpreted. To begin, a description of the unique struc-
tural parameters of each monomer is presented as follows:
ꢁ PH-a is considered as reference since no substituents are
present in this resin.
ꢁ U-a is substituted in position 6 (para position with
respect to the oxygen in the oxazine ring) with a strong
electron-withdrawing conjugated C@C double bond, and in
position 5 (meta position) with the oxygen from the ester
group which is a moderately electron-donating group.
ꢁ MU-a is substituted in position 6 (para position with
respect to the oxygen in the oxazine ring) with a relatively
strong electron-withdrawing conjugated C@C double bond.
FIGURE 3 1H NMR spectra of 4-methylumbelliferone-aniline
benzoxazine resins (MU-a). [Color figure can be viewed in the
online issue, which is available at wileyonlinelibrary.com.]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435
1431

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Thermal Properties of Benzoxazine Resins
Polymerization
temperature (8C)
Melting
Monomer
temperature (8C)
Onset
Max.
PH-a
U-a
54
255
215
229
261
220
232
147
153
MU-a
methyl group. Clearly, this methyl group is not a direct sub-
stituent on the benzoxazine nuclei, and even so it still influen-
ces its reactivity. Nevertheless, its effect can straightforward
be understood when one considers this methyl group as a
substituent of the activated electrophilic olefins which is the
actual substituent on the benzoxazine nuclei. Certainly, differ-
ent activated electrophilic olefins will affect the reactivity of
the benzoxazine resins in different manners. This is in fact the
case in our systems. Therefore, the two coumarin-containing
benzoxazines can be viewed as the same resin bearing two
different substituents in position 6. The electrophilicity of this
isolated model system is here straightforwardly related to the
capacity to act as an electron-withdrawing group when the
same motif is a substituent group. Thus, it can be inferred
that the substituent in position 6 in U-a is a stronger
electron-withdrawing group than the one present in MU-a at
the same position. This results in U-a being more reactive
than MU-a with respect to their polymerization reactions.
FIGURE 4 DSC thermograms of the benzoxazine resins stud-
ied: PH-a (-.-), MU-a (–), and U-a (-). [Color figure can be
viewed in the online issue, which is available at wileyonline
library.com.]
be seen that they have a similar activation pattern produced
by the coumarin moiety over the benzoxazine nuclei, being
the methyl group in position IV the only difference. To better
understand the influence of this methyl group, and therefore,
the difference in reactivity between U-a and MU-a, localized
electronic effects must be considered and explained sepa-
rately first and then combined together.
As previously mentioned, the C@C double bond from the
coumarin moiety can be considered as a single substituent
in position 6 in the benzoxazine nuclei. However, this C@C
double bond is not isolated in the structure, but instead is
conjugated with the carbonyl group of the ester. When a
C@C double bond is conjugated with very strong electron-
withdrawing groups, such as esters, this double bond
becomes part of a more complex system, known as an acti-
vated electrophilic olefin, also referred to as a Michael
acceptor. Figure 5 depicts an example for the selective activa-
tion of this kind of system, where two chemical structures of
the same compound are in resonance. This is in fact the rea-
son for this C@C double bond to be considered as a strong
electron-withdrawing group substituent in the coumarin-
containing benzoxazine resins.24
Finally, the unsubstituted PH-a is the least reactive of all
three resins since this benzoxazine is not activated by any
means. An idealized polymerization mechanism is proposed
for MU-a and U-a, showing the electronic influence in each
monomer (Scheme 2). The classic cationic polymerization of
PH-a is also presented for comparison.
With the interest of studying the coumarin-containing poly-
benzoxazines toward potential applications and based on the
DSC results, each coumarin-containing resin was polymerized
according to the following cycle, tailored for each resin. Each
resin was heated at a rate of 10 8C/min until the onset of
the polymerization, followed by isothermal heating for 1 h at
this temperature. Then, the sample was heated at the same
10 8C/min rate until the exotherm peak temperature, fol-
lowed by 1 h of isothermal heating. DSC and TGA analyses
were carried out on the three polybenzoxazines.
The system in resonance on the right-hand side is less elec-
trophilic than the one at the left due to the weakening effect
of the positive charge by the neighboring electron-donating
FIGURE 5 Schematic depiction showing the resonance of a compound and the electronic implications, evidencing the origin of
the activated electrophilic olefins, also referred to as Michael acceptors.
1432
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
SCHEME 2 Idealized proposed polymerization mechanism of PH-a, MU-a, and U-a showing the electronic influence in each mono-
mer. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Figure 6 shows the DSC thermograms of the three polyben-
zoxazines, poly(PH-a), poly(MU-a), and poly(U-a), which
were obtained from the monomeric resins, PH-a, MU-a, and
U-a, respectively.
The second result observed in Figure 6 for these fully poly-
merized polybenzoxazines is that all three systems showed to
be very stable in the entire range of temperatures analyzed
since no degradation process was detected. This result is of
particular importance since the two coumarin-containing poly-
benzoxazines herein presented are more stable than any other
similar or comparable systems derived from natural renew-
able resources found in the literature.11,12
The DSC thermograms indicate that poly(MU-a) has a Tg of
189 8C which is the highest, followed by poly(U-a) at 183
8C, while poly(PH-a) exhibits the lowest one at 163 8C.
The relatively low Tg values for these coumarin-containing
polybenzoxazines were in a way expected. A reason for this
can be rationalized by the fact that two of the usually active
positions for polymerization and cross-linking reactions in the
aromatic ring, adjacent to the oxazine moiety, are occupied by
the coumarin moiety. These results might suggest that
poly(MU-a) is the most cross-linked of the three polybenzoxa-
zines obtained after polymerization, followed by poly(U-a)
and finally poly(PH-a) which is the least cross-linked. How-
ever, another reason for the higher Tg of poly(MU-a) com-
pared to that of poly(U-a) might be due to the presence of
the methyl group in the repeating units, which increases the
steric constraints toward molecular rearrangement and conse-
quently impairing chain mobility resulting in a higher value of
Tg. Poly(PH-a) shows the lowest value of the Tg obtained in
this study and is consistent with results previously reported
in the literature.11 It has been documented that a higher
cross-linking in methyl cinnamate-containing resins, with
respect to the unsubstituted PH-a, is due to additional cross-
linking sites, such as the extra C@C double bond, as well as
more possible cross-linking reactions as the reported transes-
terification between ester groups and the –OH generated dur-
ing polymerization of the benzoxazines.11,21,25
FIGURE 6 DSC thermograms of poly(PH-a), poly(MU-a), and
poly(U-a). Tg was defined as the midpoint of the inflection.
[Color figure can be viewed in the online issue, which is avail-
able at wileyonlinelibrary.com.]
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435
1433

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
yield, defined as the residual weight at 800 8C under N2
atmosphere, are presented in Table 2.
Figure 7 shows the monotonous degradation profiles for both
polybenzoxazines, poly(U-a) and poly(MU-a). Both thermosets
underwent very little weight loss below 310 8C, showing their
high thermal stability. The main degradation occurred from 310
to 560 8C, where the main weight loss is observed in the ther-
mograms. A slow degradation profile takes place from there to
800 8C, where the high char yield obtained for each polybenzox-
azine evidenced once again the high thermal stability. As previ-
ously discussed for the DSC results, and once again
demonstrated by the relatively high Td5, Td10, and specially the
high char yields generated by our coumarin-containing poly-
benzoxazines (poly(U-a) and poly(MU-a)), a good thermal sta-
bility has been achieved for poly(MU-a) without significantly
affecting the polymerization temperature when compared to
poly(U-a). This is of special importance because it shows that
these two different thermosets with similar properties can be
designed and synthesized using different compounds obtained
from different natural renewable resources. Additionally, and to
the best of our knowledge, these materials exhibit better ther-
mal properties than any other similar structures11,12 and com-
parable benzoxazine resins from natural renewable resources
reported to date.27 The additional advantage of knowing and
understanding the chemistry behind the process let us manipu-
late by design the systems to obtain different materials exhibit-
ing similar desirable properties.
FIGURE 7 TGA thermograms of the coumarin-containing poly-
benzoxazines: poly(U-a), poly(MU-a), and PH-a. [Color figure
can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
Thermal degradation of conventional polybenzoxazines has
been reported to proceed following mainly three steps: (a)
evaporation of the amine from the chain ends and branches;
(b) evaporation of the amine from the main chain; and (c)
the simultaneous breakage of the phenolic linkage and deg-
radation of the Mannich base.26 The specific temperatures at
which these general features may occur and the profile of
the thermogram could certainly be altered depending on dif-
ferent possible chemical structures, degree of cross-linking
as well as quantity and nature of existing substituent groups.
For instance, while one coumarin-containing benzoxazine has
been reported to start its degradation at temperatures
around 275 8C generating a char yield of only 36%,12 a novel
coumarin-based benzoxazine was recently chemically
designed to be thermally more stable, starting its degrada-
tion at higher temperature and generating a char yield of
55%.21
CONCLUSIONS
Novel coumarin-containing benzoxazine resins were success-
fully synthesized using coumarins as phenolic raw materials,
which are a family of natural renewable resources. The
implications of the electronic effects caused not only by the
coumarin itself on the reactivity of benzoxazine toward
polymerization, but also the one induced by the methyl
group modifying the coumarin, which in turn ended to affect
the benzoxazine reactivity as well, were studied.
To further investigate the unexpected stability of the cross-
linked coumarin-containing polybenzoxazines herein pre-
sented, TGA measurements were carried out and the results
are shown in Figure 7. A summary of the main thermal prop-
erties, Td5 and Td10, defined as the temperatures at which
the weight loss is 5% and 10% respectively, and the char
Thermal properties of the resulting thermosets showed unex-
pected stabilities and high char yields, higher than 50%, for this
class of materials. Both coumarin-containing thermosets
showed Tg in the range between 160 and 190 8C, being these
values higher than the respective Tg for the unsubstituted PH-a.
The coumarin-containing benzoxazine resins reported in this
work have been demonstrated to be competitive starting
materials comparable to other petroleum-based benzoxazine
resins toward the generation of useful thermoset materials.
TABLE 2 Thermal Properties of Benzoxazine Resins
Thermoset
Tg (8C)
Td5 (8C)a
Td10 (8C)b
Char yield (%)c
Poly(U-a)
183
189
163
320
312
288
353
361
343
55
51
40
Poly(MU-a)
Poly(PH-a)
REFERENCES AND NOTES
1 H. Ishida, In Handbook of Benzoxazine Resins; H. Ishida and
T. Agag, Eds.; Elsevier: Amsterdam, 2011, pp 3–81.
a
Temperature at which the weight loss is 5%.
Temperature at which the weight loss is 10%.
b
c
2 A. Chernykh, T. Agag, H. Ishida, Polymer 2009, 50, 3153–
Char yield, defined as the remaining weight% at 800 8C, measured
under N2.
3157.
1434
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
3 J. Dunkers, H. Ishida, J. Polym. Sci., Part A: Polym. Chem.
1999, 37, 1913–1921.
17 N. K. Sini, J. Bijwe, I. K. Varma, J. Polym. Sci., Part A:
Polym. Chem. 2014, 52, 7–11.
4 A. Sudo, S. Hirayama, T. Endo, J. Polym. Sci., Part A: Polym.
Chem. 2010, 48, 479–484.
18 T. Agag, K. Vietmeier, A. Chernykh, H. Ishida, J. Appl.
Polym. Sci. 2012, 125, 1346–1351.
5 B. Kiskan, Y. Yagci, Polymer 2008, 49, 2455–2460.
19 K. Chiou, P. Froimowicz, K. Landfester, A. Taden, H. Ishida,
Macromolecules 2014, 47, 3297–3305.
6 C. Liu, D. M. Shen, R. M. Sebastian, J. Marquet, R.
Schonfeld, Polymer 2013, 54, 2873–2878.
20 S. R. Trenor, A. R. Shultz, B. J. Love, T. E. Long, Chem. Rev.
2004, 104, 3059–3077.
7 X. Liu, Y. Gu, J. Appl. Polym. Sci. 2002, 84, 1107–1113.
8 A. Gandini, Macromolecules 2008, 41, 9491–9504.
9 F. H. Isikgor, C. R. Becer, Polym. Chem. 2015, 6, 4497–4559.
10 A. Gandini, Green Chem. 2011, 13, 1061–1083.
21 C. Rodriguez Arza, P. Froimowicz, H. Ishida, RSC Adv. 2015,
5, 97855–97861.
22 T. Agag, J. Liu, R. Graf, H. W. Spiess, H. Ishida, Macromole-
cules 2012, 45, 8991–8997.
11 M. Comi, G. Lligadas, J. C. Ronda, M. Galia, V. Cadiz, J.
Polym. Sci., Part A: Polym. Chem. 2013, 51, 4894–4903.
23 M. E. Perelson, Y. N. Sheinker, G. P. Syrova, K. F. Turchin,
Khimiya Prir. Soedin. 1970, 6–11.
12 B. Kiskan, Y. Yagci, J. Polym. Sci., Part A: Polym. Chem.
2007, 45, 1670–1676.
24 B. D. Mather, K. Viswanathan, K. M. Miller, T. E. Long,
Progr. Polym. Sci. 2006, 31, 487–531.
13 A. Arnebold, O. Schorsch, J. Stelten, A. Hartwig, J. Polym.
Sci., Part A: Polym. Chem. 2014, 52, 1693–1699.
25 C. Zuniga, M. S. Larrechi, G. Lligadas, J. C. Ronda, M. Galia,
V. Cadiz, J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 1219–
1227.
14 E. Calo, A. Maffezzoli, G. Mele, F. Martina, S. E. Mazzetto, A.
Tarzia, C. Stifani, Green Chem. 2007, 9, 754–759.
15 K. Chiou, H. Ishida, Curr. Org. Chem. 2013, 17, 913–925.
26 J. Liu, H. Ishida, Macromolecules 2014, 47, 5682–5690.
27 A. Van, K. Chiou, H. Ishida, Polymer 2014, 55, 1443–1451.
16 Y. L. Liu, C. I. Chou, J. Polym. Sci., Part A: Polym. Chem.
2005, 43, 5267–5282.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1428–1435
1435