Synthesis and Thermal Curing of Benzoxazine Functionalized Polyurethanes

Article abstract of DOI:10.1021/acs.macromol.5b00954

Benzoxazine (BOX) functionalized polyurethanes (PU) are introduced to provide a conceptually new thermal curing mechanism for polyurethanes. 3,4-Dihydro-3-methyl-2H-1,3-benzoxazine (P-m) was carefully oligomerized through thermal treatment. In a straightf

Full text of DOI:10.1021/acs.macromol.5b00954

Article
pubs.acs.org/Macromolecules
Synthesis and Thermal Curing of Benzoxazine Functionalized
Polyurethanes
Stefan Kirschbaum,†,‡ Katharina Landfester,† and Andreas Taden*,†,‡
†Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany
‡Henkel AG & Co. KGaA, Adhesive Research, Henkelstrasse 67, 40191 Dusseldorf, Germany
̈
S
* Supporting Information
ABSTRACT: Benzoxazine (BOX) functionalized polyurethanes (PU)
are introduced to provide a conceptually new thermal curing mechanism
for polyurethanes. 3,4-Dihydro-3-methyl-2H-1,3-benzoxazine (P-m) was
carefully oligomerized through thermal treatment. In a straightforward
synthesis the newly formed hydroxyl groups are used for end-capping
reactions with isocyanate-terminated polyurethane prepolymers. The
isocyanate reactive hydroxyl content (IRH) of the benzoxazine oligomer
1
was investigated in detail via H NMR spectroscopy, HPLC-MS, indirect
potentiometric titration in various solvents, and comparison with model
substances and found to be strongly influenced by hydrogen bonding. The
corresponding polyurethane/benzoxazine hybrid materials (PU/BOX) can cross-link at elevated temperatures and do not suffer
from shelf-life issues or outgassing of blocked isocyanates. The thermally activated curing reaction was investigated via rheology
and DSC. Significant improvements over state-of-the-art systems based on phenol-capped PU prepolymers are shorter curing
times, increased moduli, and drastically increased glass transition temperatures.
benzoxazines can be used to introduce thermal curability to
polyurethane prepolymers. Consequently, a clever combination
of these very different polymer classes is expected to lead to
hybrid materials with a beneficial set of properties. In fact, it
should be able to address a broad range of applications by
tailoring hybrid materials with either high elasticity or high
modulus, or anything in between.
INTRODUCTION
■
Polybenzoxazines are a class of thermosetting phenolic resins
that recently gained increased interest due to their unusual set
of competitive material properties.1−3 They combine typical
properties of phenolic resins like high temperature resistance
and high char yield with unique properties like low water
uptake, high glass transition values (Tg) exceeding the curing
temperatures, and extremely strong mechanical performance.
Benzoxazines are synthesized from abundant raw materials such
as phenols, primary amines, and formaldehyde proceeding
water as a condensation product and can be polymerized
through thermally activated ring-opening reaction.4 The
polymerization proceeds without producing reaction by-
products, resulting in near zero shrinkage, and yields Mannich
bridged phenol resins. This structure comprising phenolic
hydroxyl groups and tertiary amines allows the formation of
various hydrogen bonds, including extremely stable intra-
molecular six-membered hydrogen bonds, which is the key
feature toward many unexpected properties that are character-
istic for benzoxazines.5−7
Polyurethanes, on the other hand, are among the most
important tough engineering polymers and can be adjusted to
serve a wide spectrum of properties.8,9 This versatility is based
on the segmented structure consisting of urethane-rich hard
segments and urethane-poor soft segments that are easily
generated through the polyaddition reaction of isocyanates and
higher molecular weight polyols. In general, polyurethanes
possess good adhesion and abrasion resistance as well as
excellent flexibility and are therefore principally suited to
improve the toughness of benzoxazine resins.10−12 Conversely,
The simplest approach for such a hybrid material is blending
of benzoxazines with reactive, isocyanate-terminated polyur-
ethane prepolymers and subsequent thermal induction of the
ring-opening polymerization of the benzoxazine.12−15 The
arising free hydroxyl groups of the benzoxazine polymer are
reactive toward isocyanates and yield aromatic urethanes. These
phenol-based urethanes are thermally labile and may reversibly
dissociate at the curing temperature, and regenerated isocyanate
functions additionally can react with urethane or urea units to
reversibly form allophanates or biurets moieties. Upon cooling,
finally isocyanate free resins are formed at room temperature.
However, simple blends of benzoxazines and reactive
isocyanates are moisture sensitive and exhibit limited stability
and shelf-life issues. Hence, a different approach using blocked
isocyanates was described. The isocyanate functionality was
converted to thermally labile urethanes by use of bisphenol A,16
nitrophenol,17 or butanone oxime.18 This approach renders the
BOX/PU blend more stable, and depending on the actual
thermal stability of these urethanes and the applied polymer-
Received: May 5, 2015
Revised: May 29, 2015
Published: June 11, 2015
© 2015 American Chemical Society
3811
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816

Macromolecules
Article
ization temperature the urethane bonds can dissociate to
regenerate free isocyanates. Unfortunately, the blocking agents,
which are often volatile organic compound (VOC), may be
released from the material. Among these blocking agents,
phenol derivatives show special properties as free phenols are
catalytically active for the polymerization reaction of benzox-
azines.19 Furthermore, they can be attacked by iminium ion
intermediates to be incorporated in the network structure and
hence covalently connect the PU prepolymers to the
benzoxazine, even if dissociation did not occur.16 However,
these incorporated phenols dilute the benzoxazine network as
they do not polymerize themselves and hence reduce the
material performance. Another approach is the precuring of the
polyurethane through multifunctional diols20 or moisture.21
The benzoxazine swollen polymer network then provides an
improved dimensional stability at elevated temperatures for the
benzoxazine curing reaction. The generated interpenetrating
network (IPN) was shown to have a synergistic effect on the
glass transition temperature due to intimate hydrogen bonding
between both materials. Furthermore, it was observed that the
presence of thermally labile allophanate or biureh groups
regenerate low quantities of free isocyanate and allow the
formation of covalent connections between both networks.
Recently, benzoxazines were covalently attached to a polyur-
ethane structure previous to the benzoxazine curing reaction.
Namely aliphatic hydroxyl group containing benzoxazines were
synthesized for the direct incorporation in the polyurethane,22
and furan functional benzoxazines were synthesized for the
Diels−Alder addition reaction with maleimide functionalized
polyurethanes.23 However, both approaches suffer from high
synthetic effort.
In this contribution we introduce a straightforward synthetic
approach that makes use of the inherent hydroxyl groups in
benzoxazine oligomers. These oligomers can be generated
through thermal treatment but are also obtained as byproduct
in crude benzoxazine resins. Their content is dependent on the
actual synthetic conditions like temperature, reaction time, or
the dielectric constant of the solvent and hence can be adjusted
without further synthetic effort.24 The oligomer containing
benzoxazine is then covalently attached to the polyurethane.
Aliphatic isocyanate-based polyurethanes renders the resulting
phenolic urethane linkage relatively stable with approximate
dissociation temperatures of >180 °C.25 The use of methyl-
amine-based benzoxazines that polymerize at comparatively low
temperatures26 and the presence of catalytically active phenolic
hydroxyl groups in the benzoxazine oligomer allow the curing
reaction to proceed below the deblocking temperature of the
benzoxazine functionalization. Nevertheless, if dissociation
occurs to some extent, the free isocyanate may additionally
catalyze the benzoxazine curing via stabilization of the iminium
ion intermediate27 and is likely to yield a similar phenolic
urethane upon addition due to the great excess of free hydroxyl
groups.
benzoxazines and provides novel insights into the reactivity of
phenolic hydroxyl groups in benzoxazine oligomers and
polymers toward isocyanates and the importance of hydrogen
bonding in benzoxazine-based systems.
EXPERIMENTAL SECTION
■
Materials. Phenol (99%, Acros Organics), dimethylphenol (DMP)
(98%, Alfa Aesar), paraformaldehyde (96%, Acros Organics), methyl-
amine 40% in water (Acros Organics), dimethyltin dineodecanoate
(Fomrez UL-28, Momentive), cyclohexyl isocyanate (98%, Alfa
Aesar), dibutylamine (99%, Merck), isophorone diisocyanate (IPDI)
(Merck, 99%), and poly(tetramethylene oxide) (PTMO, Mn = 650 g/
mol) (PolyTHF 650 S, BASF) were used as received.
Synthesis of 3,4-Dihydro-3-methyl-2H-1,3-benzoxazine (P-
m). 94.11 g (1 mol) of phenol, 66.06 g of paraformaldehyde
(equivalent to 2.2 mol of formaldehyde), 190 g (50 wt %) of toluene,
and 31.06 g (1 mol) of methylamine 40% in water were placed in a
flask equipped with a magnetic stirrer, a water separator, and a reflux
condenser. The mixture was heated to reflux until water separation
stopped. The product was washed with an aqueous solution of sodium
hydroxide (2 mol/L) and water several times. Solvent was evaporated
under reduced pressure to yield the crude P-m.
Oligomerization of P-m. A 250 mL flask was charged with 22 g of
crude P-m, heated to 70 °C, and purged with nitrogen. The flask was
then placed in an oil bath of 160 °C for 6 min under vigorous stirring.
The oligomerization was visually monitored by the increasing viscosity
and stopped through abrupt cooling in an ice bath. Indirect
potentiometric titration indicates 10.9% isocyanate reactive phenol
structures and 93.0% of the theoretical basic nitrogen content.
Vacuum Distillation of P-m. 72 g of the crude P-m was vacuum
distilled to obtain the pure, low viscous, and colorless P-m.
Condensate was collected at a reduced pressure of 0.8 mbar and a
vapor temperature of 85−87 °C. In the course of distillation the
temperature of the residue needs to be increased to ensure a constant
condensate flow due to enrichment of oligomers and by products and
at some point fast and exothermic curing of the residue occurs. Yield:
83%. Elemental analysis: C, 72.5; H, 7.5; N, 9.4; O, 10.5 (calculated:
1
C, 72.46; H, 7.43; N, 9.39; O, 10.72 for C9H11NO). H NMR (400
MHz, CDCl3, 298 K) δ: 2.60 ppm (s, 3H, N−CH3), 3.95 (s, 2H, Ar−
CH2−N), 4.79 (s, 2H, O−CH2−N), 6.79−7.12 (m, 4H, Ar−H). The
spectrum is shown in Figure S1 of the Supporting Information.
Polymerization Study of P-m. Samples of 0.3 g of distilled P-m
each were flushed with nitrogen, sealed, and placed in an oven of 160
°C. Every 5 min a sample was removed from the oven and
immediately cooled in a water bath. The samples were subsequently
1
subjected to H NMR (400 MHz, CDCl3, 298 K) spectroscopy and
indirect potentiometric titration.
Synthesis of N,N-Bis(3,5-dimethyl-2-hydroxybenzyl)-
methylamine (DMP2-m). DMP2-m was synthesized referring to
the procedure to this described by Dunkers.29 36.65 g of 2,4-
dimethylphenol (0.3 mol), 9.91 g of paraformaldehyde (equivalent to
0.33 mol of formaldehyde), 57 g (50 wt %) of toluene, and 11.65 g
(0.15 mol) of methylamine 40% in water were placed in a flask
equipped with a magnetic stirrer, a water separator, and a reflux
condenser. The mixture was heated to reflux until water separation
stopped. Solvent was evaporated under reduced pressure. The bulk
material was heated to 150 °C for 1 h. The product was then
recrystallized from diethyl ether until white crystals were obtained.
Elemental analysis: C, 75.9; H, 8.5; N, 4.7; O, 10.9 (calculated: C,
76.22; H, 8.42; N, 4.68; O, 10.69 for C19H25NO2). MS (APCI): m/z =
Besides phosphorus NMR spectroscopy of derivatized,
oligomerized benzoxazines,28 so far surprisingly little attention
was paid to benzoxazine oligomers. Consequently, there is a
lack of characterization and quantification. Therefore, we
developed an indirect potentiometric titration method that is
capable to quantify the isocyanate reactive hydroxyl content
(IRH) as well as the basic amine content in benzoxazines. This
method was successfully applied to a series of benzoxazine
oligomers and N,N-bis(3,5-dimethyl-2-hydroxybenzyl)-
methylamine (DMP2-m) as model substance for polymerized
1
300.4 [M + H]+ (calculated: 300.4 for C19H26NO2). H NMR (400
MHz, CDCl3, 298 K) δ: 2.21 ppm (s, 6H, Ar−CH3), 2.22 ppm (s, 3H,
N−CH3), 2.23 ppm (s, 6H, Ar−CH3) 3.64 (s, 4H, Ar−CH2−N), 6.73
(s, 2H, Ar−H), 6.87 (s, 2H, Ar−H) 5−9 ppm (br, 2H, OH). The
spectrum is shown in Figure S2 of the Supporting Information.
Indirect Potentiometric Titration. An aliquot of the sample
containing approximately 0.4−2.0 mmol of phenolic hydroxyl groups
was placed in a 40 mL screw-cap glass. The sample was dissolved in 5
mL of dry xylene before 0.02 g of dimethyltin dineodecanoate and 5
3812
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816

Macromolecules
Article
Figure 1. Synthesis and polymerization of P-m. Additionally to the idealized chemical structures, the corresponding monomeric and oligomeric
species are schematically depicted. Aromatic moieties are represented by a black circle and ring-closed polymerizable units by a blue circle. The OH
groups are drawn explicitly, as they can serve as reaction sites for isocyanate groups.
1
Figure 2. (a) H NMR spectra upon polymerization of P-m. (b−e) Peak intensities over the reaction time.
mL of cyclohexyl isocyanate solution (0.5 mol/L in dry xylene) were
added. The glass was flushed with nitrogen and sealed tight. The
samples were stored in an oven at 80 °C for 60 min. At room
temperature 7 mL of dibutylamine solution (0.5 mol/L in xylene) was
added. The sample was then transferred to a 400 mL beaker using 15
mL of xylene and 50 mL of isopropyl alcohol and was subjected to a
potentiometric titration with aqueous hydrochloric acid (0.5 mol/L),
while the potential of the solution was recorded using a solvotrode
(Metrohm 6.0229.010 or Mettler Toledo DGI113-SC). For some
titrations also different reaction conditions such as other catalysts,
solvents, or isocyanates were used as indicated.
Synthesis of PU/BOX-Hybrids. 47.08 g (72.4 mmol) of PTMO
was dried at 75 °C under reduced pressure. 24.00 g (108 mmol) of
IPDI was added at 40 °C. A sample was taken for IR spectroscopy.
Then 0.09 g (125 mg/100 g sample) of dimethyltin dineodecanoate
was added, and the mixture was heated to 80 °C until the desired
intensity of the NCO stretching vibration band at about 2250
cm−1 was achieved. Subsequently aliquots of the sample were taken for
end-capping reactions. 5.83 g (70 wt %) was reacted with 2.50 (30 wt
%) of the oligomerized P-m. Another aliquot of 10.00 (70 wt %) g was
reacted with 0.94 g (6.6 wt %) of phenol to end-cap the reactive
isocyanate groups. After almost complete conversion of the isocyanate
was confirmed by IR spectroscopy, 3.34 g (23.4 wt %) of the
oligomerized P-m was added to the phenol-capped PU.
Curing of PU/BOX Hybrids and Rheological Investigation.
The thermally induced curing reaction of the polyurethane/
benzoxazine hybrids was performed and monitored in a rheological
plate/plate setup using an Anton Paar MCR 302 rheometer.
Disposable aluminum cups with a diameter of 53 mm were charged
with 1.5 g of the sample and used as bottom plate. Disposable
aluminum top plates with a diameter of 25 mm were used at a gap
distance of 0.35 mm. The starting values were monitored for 1 min at
25 °C, at a sinusoidal strain of 10% and a frequency of 1 Hz. The
temperature was then increased with a rate of 10 °C/min to 170 °C,
and the sample was cured for 120 min at that temperature, whereas the
sample deformation was ramped linearly from 10% to 0.2% to remain
within the instrument limitation and to maintain linear viscoelastic
behavior as the moduli increase over several orders of magnitude upon
curing. Subsequently, a normal force of zero was applied to allow
automatic gap control and minimize stress from thermal contraction or
expansion of the solidified sample. The temperature was then
decreased with a rate of 10 °C/min to 25 °C. After 60 min postcuring
time at room temperature the thermal properties of the cured sample
were investigated. Therefore, the sample was heated with a rate of 2
°C/min to 150 °C, kept at that temperature for 5 min, and cooled with
a rate of 2 °C/min.
HPLC-MS Analysis. 0.6455 g of DMP2-m (2.16 mmol), 0.5400 g
of cyclohexyl isocyanate (4.31 mmol), and 0.02 g of dimethyltin
dineodecanoate were dissolved in 10 mL of dry xylene. The glass was
flushed with nitrogen and sealed tight. The sample was stored in an
oven at 80 °C for 60 min. Subsequently, 0.2770 g of dibutylamine
(2.14 mmol) was added at room temperature. A sample of this mixture
was subjected to HPLC-MS analysis. The HPLC-MS measurement
was performed using a Waters Acquity UPLC system coupled to a
Thermo Scientific LTQ XL linear ion trap mass spectrometer.
Chromatographic separation was archived using an Acquity UPLC
BEH C18 column. Mobile phase A was composed of water containing
0.1% formic acid, while mobile phase B consisted of acetonitrile and
0.1% formic acid. The gradient elution program employed was 100−
0% A from 0 to 10 min at a flow rate of 0.2 mL/min. The UV detector
was set to measure in the range of 200−400 nm. Atmospheric pressure
chemical ionization (APCI) was applied, and the mass spectrometer
was operated in positive ion mode scanning from 100 to 2000 m/z.
RESULTS AND DISCUSSION
■
Synthesis and Oligomerization of the P-m. The
synthesis and polymerization reactions of P-m are displayed
in Figure 1. Additionally to the idealized linear polymerization
reaction of unsubstituted, monofunctional benzoxazines, the
free para position leads to branching through substitution in the
less favored para position.30 Hence, comparatively high
3813
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816

Macromolecules
Article
molecular weight species can be generated and partially
insoluble material is generated at high conversion.
conversions of the benzoxazine. At the same time the basic
nitrogen content is decreased by about 28% of the initial value.
Additional results of indirect potentiometric investigations are
provided and discussed in detail in the Supporting Information
(Figures S4−S8). This results strongly suggest that most of the
hydroxyl groups formed are trapped in very stable hydrogen
bondings, for example six-membered intramolecular hydrogen
bonding between the hydroxyl moiety and the Mannich bridge,
and are therefore not accessible for the reaction with
isocyanates. In fact, the pronounced hydrogen-bonding
capabilities are widely recognized to be responsible for many
unusual properties of polybenzoxazines.5−7 Hence, the
comparably low IRH values seem plausible and may also be
seen as measure to quantify the strength of hydrogen bonding
in a given benzoxazine system. Consequently, the IRH is a
crucial parameter for the tailored design of PU/BOX-hybrids,
albeit the strong dependence on the reaction conditions.
Naturally, solvent type (dielectricity constant), catalyst type,
and isocyanate structure do affect the IRH value, and more
comprehensive information is given below and in the
Supporting Information. The main article however predom-
inantly does focus on the characterization of the PU/BOX
hybrid materials.
Benzoxazine Functionalization of the Polyurethane. A
polyurethane prepolymer with a theoretical average molecular
weight of Mn = 2000 g/mol was synthesized from PTMO and
IPDI. The molecular weight was confirmed via IR-spectro-
scopic end group analysis (see Figure S10 in the Supporting
Information). The PTMO backbone was chosen due to its high
compatibility with benzoxazines. Other polyols like polypropy-
lenoxide (PPO, Mn = 2000 g/mol) caused undesired
macrophase separation upon polymerization of the hybrid
material. However, in a previous publication our group has
shown that the morphology of polyether containing benzox-
azines can be adjusted by varying the amount and ratio of
PTMO and PPG segments and that tailored microphase
separation can be very beneficial for the toughness of resin
materials.11 The reactive, isocyanate-terminated prepolymer
was devided into two portions. One was directly charged with
the oligomerized P-m and as a reference, and the other one was
terminated with phenol and subsequently charged with the
oligomerized P-m to obtain the same final content of
polyurethane and benzoxazine resin (including phenol). The
oligomerized P-m was determined to have a relative, isocyanate
reactive hydroxyl content of 10.9% determined in xylene. Based
on this result, a formulation of 42 wt % PU and 58 wt % BOX
would be necessary for a stoichiometric functionalization, and
an even greater excess of benzoxazine was expected to be
necessary to avoid cross-linking due to oligomers containing
multiple hydroxyl groups. The synthetic pathway is schemati-
cally depicted in Figure 4.
The free hydroxyl groups generated upon polymerization
provide the opportunity of functionalization reactions. To gain
more insights into the evolution of hydroxyl groups and their
reactivity toward isocyanates, samples of the purified P-m were
smoothly polymerized at 160 °C, while a sample was removed
from the oven every 5 min. Up to 45 min reaction time all
samples were completely soluble in xylene and chloroform and
were analyzed via 1H NMR spectroscopy to observe the overall
progress of the reaction. Figure 2 shows the spectra and
structural assignments as well as plots of their intensities over
the reaction time.
All aliphatic CH2 signals from educts and products appear in
the spectral region of 3.0−5.0 ppm. This region (yellow
background, Figure 2) was used as internal standard for the
integration. The normalized peak intensities of signals, assigned
to the educt, indicate a similar reaction profile with an overall
conversion of about 90% after 45 min. Samples that have been
treated for a longer time were partially insoluble in common
organic solvents. The Mannich bridged structure of the main
product was confirmed by HBMC-NMR spectroscopy (see
Figure S3 in the Supporting Information). The same samples
have also been investigated using an indirect potentiometric
titration method. This method was shown to quantify the
hydroxyl content of electron-rich phenols with good accuracy
(see Table S1 in the Supporting Information), and in contrast
to standard anhydride-based titration methods, the procedure
does not cause ring-opening or ring extension31 of the
benzoxazine. Also, the reactivity of the aliphatic cyclohexyl
isocyanate used for quantification is similar to that of IPDI and,
hence, was expected be transferable to the functionalization
reaction. Additionally, a second equivalent point in the
potential curve allows the quantification of basic amine due
to different basicity of dibutylamine and the amine structures in
benzoxazines. The relative, isocyanate reactive hydroxyl content
(IRH) is an important parameter for the tailored design of PU/
BOX-hybrids and is plotted in Figure 3 in conjunction with the
relative, basic nitrogen content.
The isocyanate reactive hydroxyl content evolves similarly to
the overall reaction profile but remains low in comparison to
the theoretical values. Only up to 21% of the theoretically
possible hydroxyl content could be determined through the
reaction with cyclohexyl isocyanate in xylene, even at high
In fact, IR spectroscopy indicated that a formulation of 70 wt
% PU and 30 wt % oligomerized P-m was sufficient to
completely convert the isocyanate groups without gelation (see
Figure S10 in the Supporting Information). This formulation
(PU/BOX 70/30) corresponds to a stoichiometry of 1/0.3
NCO/OHdetermined and a stoichiometry of 1/2.6 NCO/
OHtheoretically. The reaction behavior shows that under synthetic
conditions in bulk more hydroxyl groups are accessible for the
reaction with isocyanates. In contrast to the unpolar xylene
solvent utilized as reaction medium for the IRH determination,
the nondiluted polyurethane system provides a more viscous
and polar, protic reaction environment that weakens the
Figure 3. Isocyanate reactive hydroxyl content (IRH) and relative
basic nitrogen content of P-m upon polymerization determined by
indirect potentiometric titration with cyclohexyl isocyanate in xylene.
3814
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816

Macromolecules
Article
Figure 4. Synthesis and curing of PU/BOX hybrid formulations.
Figure 5. (a) Curing of the PU/BOX-hybrids at 170 °C. (b) Temperature dependence of the storage modulus upon heating and cooling. The PU/
BOX 70/30 cures faster and shows significantly improved mechanical properties in comparison to the state-of-the-art PU-Phenol/BOX 70/30.
intramolecular hydrogen bonds and obviously increase the
reactivity of phenolic hydroxyl groups in polybenzoxazines. To
substantiate this hypothesis, the hydroxyl content of DMP2-m
as model substance for polybenzoxazines was investigated
under variation of different reaction parameters. Indeed, it was
observed that more polar solvents, basic catalysts, or more
reactive aromatic isocyanates are capable to increase the
hydroxyl content contributing to the urethane formation
reaction (see Table S2). The reference hybrid material (PU-
Phenol/BOX 70/30) was composed of 70 wt % polyurethane,
6.6 wt % of phenol, and 23.4 wt % of oligomerized P-m. In
contrast to the PU/BOX 70/30, the use of phenol as end-
capping agent prevents a covalent binding between the
polyurethane and the oligomerized P-m upon formulation
and provides a state-of-the-art material that allows covalent
attachment upon the thermally induced attack of benzoxazines
to the terminal phenol moieties.
Curing of the Polyurethane/Benzoxazine Hybrids. The
curing reaction was conducted in the rheometer at 170 °C.
Rheometric plots are shown in Figure 5.
The storage modulus of the immediately benzoxazine
functionalized polyurethane PU/BOX 70/30 evolves faster
and reaches a final value that is about 1 order of magnitude
higher than that of the phenol functionalized start-of-the-art
3815
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816

Macromolecules
Article
PU-Phenol/BOX 70/30 hybrid material. Furthermore, the pure
PU/BOX 70/30 gels faster and the glass transition temperature
(Tg) of the cured materials is more narrow and about 45 °C
higher than that of the PU-Phenol/BOX 70/30. Especially the
difference in the Tg provides a tremendously improved thermal
resistance. Another advantage is the far more distinct softening
behavior at the glass transition, which is highly favorable and a
clear demand for the majority of engineering processes and
applications. The sharp decline of the storage modulus is
related to the breakup of hydrogen bonds near the Tg.
Generally, this effect is very strong for benzoxazines in
comparison to most polymers whose properties does not
depend as heavily on hydrogen bonding.1 The improved curing
profile as well as physical and thermal properties can be
attributed to an unperturbed benzoxazine curing reaction. In
contrast, if phenol is used for the PU-functionalization instead,
the inactive but Mannich bridge accepting phenol moieties
weaken and dilute the polybenzoxazine network structure. This
effect is expected to escalate with increasing phenol content, if
lower molecular weight polyurethanes or higher polyurethane
contents become applied. These results are substantiated via
DSC, whereas the diluted phenol-based sample shows a
significantly reduced reaction enthalpy and an increased peak
temperature for the curing reaction (see Figure S11 in the
Supporting Information).
ACKNOWLEDGMENTS
■
The authors greatly acknowledge the analytical department of
Henkel for the opportunity to use the potentiometric
equipment.
REFERENCES
■
(1) Ishida, H. In Handbook of Benzoxazine Resins; Agag, H. I., Ed.;
Elsevier: Amsterdam, 2011; Chapter 1, pp 3−81.
(2) Ghosh, N. N.; Kiskan, B.; Yagci, Y. Prog. Polym. Sci. 2007, 32
(11), 1344−1391.
(3) Yagci, Y.; Kiskan, B.; Ghosh, N. N. J. Polym. Sci., Part A: Polym.
Chem. 2009, 47 (21), 5565−5576.
(4) Dunkers, J.; Ishida, H. J. Polym. Sci., Part A: Polym. Chem. 1999,
37 (13), 1913−1921.
(5) Dunkers, J.; Zarate, E. A.; Ishida, H. J. Phys. Chem. 1996, 100
(32), 13514−13520.
(6) Kim, H.-D.; Ishida, H. J. Phys. Chem. A 2002, 106 (14), 3271−
3280.
(7) Wirasate, S.; Dhumrongvaraporn, S.; Allen, D. J.; Ishida, H. J.
Appl. Polym. Sci. 1998, 70 (7), 1299−1306.
(8) Delebecq, E.; Pascault, J.-P.; Boutevin, B.; Ganachaud, F. Chem.
Rev. 2012, 113 (1), 80−118.
(9) Chattopadhyay, D. K.; Raju, K. V. S. N. Prog. Polym. Sci. 2007, 32
(3), 352−418.
(10) Prisacariu, C. Polyurethane Elastomers: From Morphology to
Mechanical Aspects; Springer: Berlin, 2011; pp 1−250.
(11) Sawaryn, C.; Landfester, K.; Taden, A. Polymer 2011, 52 (15),
3277−3287.
CONCLUSION
(12) Sawaryn, C.; Kreiling, S.; Schonfeld, R.; Landfester, K.; Taden,
̈
■
A. In Handbook of Benzoxazine Resins; Hatsuo, I., Tarek, A., Eds.;
Elsevier: Amsterdam, 2011; Chapter 35, pp 605−620.
(13) Takeichi, T.; Guo, Y.; Agag, T. J. Polym. Sci., Part A: Polym.
Chem. 2000, 38 (22), 4165−4176.
(14) Takeichi, T.; Guo, Y. Polym. J. 2001, 33 (5), 437−443.
(15) Rimdusit, S.; Pirstpindvong, S.; Tanthapanichakoon, W.;
Damrongsakkul, S. Polym. Eng. Sci. 2005, 45 (3), 288−296.
(16) Jamshidi, S.; Yeganeh, H.; Mehdipour-Ataei, S. Polym. Adv.
Technol. 2011, 22 (11), 1502−1512.
(17) Jamshidi, S.; Yeganeh, H.; Mehdipour-Ataei, S. Polym. Int. 2011,
60 (1), 126−135.
(18) Liu, Y.; Zhang, X.; Gao, N. Polym. Eng. Sci. 2015, 55 (3), 604−
In summary, our PU/BOX-hybrid approach combines two
totally different material classes in a straightforward and
synergistic manner. We were able to achieve significantly
improved reaction profiles and material properties compared to
state-of-the-art polyurethane/benzoxazine hybrids. The materi-
al design is extremely versatile and allows the tailored
adjustment of product properties over a very broad range by
simply changing the composition. Our investigations on the
concentration and reactivity of hydroxyl groups clearly
demonstrate the importance of hydrogen bonding in
benzoxazine-based systems and contribute to a better analytical
understanding of the corresponding interaction strength.
Strong differences can not only be revealed and adjusted by
different chemical structures but also by choosing different
reaction media, which is of tremendous importance for the
postmodification of oligomerized and even polymerized
benzoxazines. Finally, we want to emphasize that the capability
of benzoxazine functionalized polyurethanes to attack electron-
rich phenolic structures and covalently bind to phenolic
precursors like lignin in wood or phenol resins can be highly
beneficial for many applications.
613.
(19) Ishida, H.; Rodriguez, Y. J. Appl. Polym. Sci. 1995, 58 (10),
1751−1760.
(20) Cui, Y.; Chen, Y.; Wang, X.; Tian, G.; Tang, X. Polym. Int. 2003,
52 (8), 1246−1248.
(21) Rimdusit, S.; Sudjidjune, M.; Jubsilp, C.; Tiptipakorn, S. J. Appl.
Polym. Sci. 2014, 131 (13), 40502.
(22) Baqar, M.; Agag, T.; Ishida, H.; Qutubuddin, S. Polymer 2011,
52 (2), 307−317.
(23) Gaina, C.; Ursache, O.; Gaina, V.; Varganici, C. D. J. Polym. Res.
2014, 21 (11), 1−11.
(24) Ishida, H.; Liu, J.-P. Benzoxazine Chemistry in Solution and
Melt. In Handbook of Benzoxazine Resins; Agag, H. I., Ed.; Elsevier:
Amsterdam, 2011; Chapter 2, pp 85−102.
ASSOCIATED CONTENT
■
(25) Wicks, Z. W., Jr. Prog. Org. Coat. 1975, 3 (1), 73−99.
(26) Sudo, A.; Du, L. C.; Hirayama, S.; Endo, T. J. Polym. Sci., Polym.
Chem. 2010, 48 (13), 2777−2782.
S
* Supporting Information
Figures S1−S11; Tables S1 and S2. The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/acs.macromol.5b00954.
(27) Sudo, A.; Mori, A.; Endo, T. J. Polym. Sci., Part A: Polym. Chem.
2011, 49 (10), 2183−2190.
(28) Chutayothin, P.; Ishida, H. Eur. Polym. J. 2009, 45 (5), 1493−
AUTHOR INFORMATION
1505.
■
(29) Dunkers, J.; Ishida, H. Spectrochim. Acta, Part A 1995, 51 (5),
855−867.
Corresponding Author
*E-mail andreas.taden@henkel.com; Ph 0049 211 7972744
(A.T.).
(30) Burke, W. J.; Bishop, J. L.; Glennie, E. L. M.; Bauer, W. N. J.
Org. Chem. 1965, 30 (10), 3423−3427.
(31) Kudoh, R.; Sudo, A.; Endo, T. Macromolecules 2009, 42 (7),
2327−2329.
Notes
The authors declare no competing financial interest.
3816
DOI: 10.1021/acs.macromol.5b00954
Macromolecules 2015, 48, 3811−3816