Poly-2-oxazoline-derived polyurethanes: A versatile synthetic approach to renewable polyurethane thermosets

Article abstract of DOI:10.1002/pola.24744

2-Nonyl-2-oxazoline and 2-(9-decenyl)-2-oxazoline have been copolymerized in different proportions by cationic ring-opening polymerization to obtain a series of random linear copolymers with tailored molecular weight and double bond functionality in the side chains. Thiol-ene addition of 2-mercaptoethanol has been used to produce a set of polyoxazoline-polyols under mild conditions and with quantitative double bond transformation. The polyols obtained in this way were reacted with methylene-bis(phenylisocyanate) to yield a series of amorphous and semicrystalline polyurethane networks. The thermal stability and the thermomechanical properties of these thermosets have been studied and related with the structure of the parent polyols.

Full text of DOI:10.1002/pola.24744

Poly-2-oxazoline-Derived Polyurethanes: A Versatile Synthetic Approach
to Renewable Polyurethane Thermosets
`
´
ENRIQUE DEL RIO, GERARD LLIGADAS, JUAN CARLOS RONDA, MARINA GALIA, VIRGINIA CADIZ
Departament de Qu´ımica Anal´ıtica i Qu´ımica Orga`nica, Universitat Rovira i Virgili, Campus Sescelades,
Marcel.l´ı Domingo, s/n, 43007 Tarragona, Spain
Received 2 March 2011; accepted 22 April 2011
DOI: 10.1002/pola.24744
Published online 18 May 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: 2-Nonyl-2-oxazoline and 2-(9-decenyl)-2-oxazoline
have been copolymerized in different proportions by cationic
ring-opening polymerization to obtain a series of random lin-
ear copolymers with tailored molecular weight and double
bond functionality in the side chains. Thiol-ene addition of
2-mercaptoethanol has been used to produce a set of poly-
oxazoline–polyols under mild conditions and with quantita-
tive double bond transformation. The polyols obtained in
this way were reacted with methylene-bis(phenylisocyanate)
to yield a series of amorphous and semicrystalline polyur-
ethane networks. The thermal stability and the thermome-
chanical properties of these thermosets have been studied
C
V
and related with the structure of the parent polyols.
2011
Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 49:
3069–3079, 2011
KEYWORDS: cationic polymerization; polyoxazolines; polyure-
thanes; renewable resources
INTRODUCTION The depletion of world oil pool, rising price
of crude oil, and increased environmental concerns are pres-
surizing the scientists for the use of renewable natural
resources in different fields of applications as they are eco-
friendly and cost-effective materials.1 In this regard, vegeta-
ble oils are one of the most promising renewable raw mate-
rials because of their availability, structural variety, and
chemical versatility. Vegetable oils containing unsaturated
fatty acids can be used in polymerizations to make biobased
polymers.2–5 Moreover, numerous fatty acids are now avail-
able with a high purity that makes them attractive for syn-
thesis and they are also used as raw materials for the chemi-
cal industry.6
and further oxirane ring opening with alcohols and other
nucleophiles,9,10 transesterification with multifunctional alco-
hols,11,12 and hydroformylation or ozonolysis and subsequent
reduction of the carbonyl groups.13,14 Unfortunately, these
strategies usually lead to complex mixtures of polyols with
different hydroxyl content, molecular architecture, and mo-
lecular size. In a recent work, we prepared a set of polyether
polyols for synthesis of PUs by polymerization of the methyl
9,10-epoxyoleate and subsequent controlled reduction of the
pendant side ester groups.15–17 This new strategy allowed
an excellent control of the polyol hydroxyl content but failed
in the control of the polyol molecular weight and molecular
architecture due to the nature of the cationic and ionic-coor-
dinative polymerizations used. On the contrary, the living na-
ture of the cationic ring-opening polymerization (CROP) of
2-substituted-2-oxazolines allows the facile preparation of
well-defined homopolymers and copolymers with an excel-
lent control of their molecular weight, end group functional-
ities, and molecular architecture allowing an accurate control
of the final properties.18–24 Moreover, 2-oxazoline monomers
can be readily prepared directly from the carboxylic acids or
their nitrile derivatives following various synthetic path-
ways.25–30 In this way, 2-oxazolines are excellent candidates
to transform unsaturated carboxylic fatty acids in polymers
with tailored structure and functionality. Copolymerization of
saturated and unsaturated 2-oxazoline monomers allows the
preparation of linear copolymers with prefixed molecular
weights and the desired amount of pendant double bonds in
a random or block arrangement.31–33 Side chain double bond
Polyurethanes (PUs) are one of the most important and ver-
satile classes of polymers and can vary from thermosetting
to thermoplastic materials. PUs exhibit better abrasion re-
sistance, toughness, chemical resistance, and mechanical
strength compared to other polymers such as polyesters and
polyesteramides. Moreover, their mechanical, thermal, and
chemical properties can be tailored by reactions with various
isocyanates and polyols. Usually, both isocyanates and poly-
ols are petroleum based, but in recent, years, vegetable oils,
fatty acids and their derivatives have attracted significant
attention as raw materials for the preparation of polyur-
ethanes.7 For natural oils and derivatives to be used as raw
materials for polyol production, multiple hydroxyl functional-
ity is required.8 This can be accomplished by various strat-
egies like the epoxidation of carbon–carbon double bonds
Correspondence to: J. C. Ronda (E-mail: juancarlos.ronda@urv.cat)
C
V
Journal of Polymer Science Part A: Polymer Chemistry, Vol. 49, 3069–3079 (2011) 2011 Wiley Periodicals, Inc.
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3069

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 1 Synthesis of the oxazoline monomers I and II from 10-undecenoic and decanoic acids.
hydroxylation can be accomplished in different ways. Hydro-
boration/oxidation techniques have been applied successfully
to polymers like polybutadiene,34 poly(3-hydroxyundecen-
10-enoate),35 poly(4-allyloxyphenyloxazoline),36 and poly[2-
(9-decenyl)-2-oxazoline37 but this approach needs low tem-
perature conditions, a complex work up, and uses expensive
reagents and strongly basic and oxidizing media what is
undesirable in oxazoline-based polymers. Thiol-ene chemis-
try, which implies the free-radical addition of thiols onto
double bonds, is a highly efficient tool that has been applied
to polymer synthesis.38–40 Click chemistry reactions have
also been used in polyoxazolidines41 and recently efficient
thiol-ene modification of poly[2-(3-butenyl)-2-oxazoline],42
poly[2-isopropyl/3-butenyl)-2-oxazoline,43 and 2-oxazoline
monomers44 has been described.
ture.31,47 Purity was estimated as more than 98% by gas
chromatography (GC).
Random Copolymerization
The copolymerization of I and II was carried out with benzyl
bromide in anhydrous ACN at 100 ꢀC for 24 h. Polymeriza-
tions and work-up were carried out following a similar pro-
cedure previously described.48 In a typical copolymerization
procedure, the appropriate amounts of I and II (see Table 1
for composition) (30 mmol) were placed in a flame-dried
screw-capped shlenck flask under argon. Next, the necessary
amount of anhydrous ACN was added to reach a 2 M solu-
tion. The mixture was stirred at room temperature and the
necessary amount of anhydrous benzyl bromide was added
by means a precision syringe. The polymerization mixture
ꢀ
was heated and kept at 100 C for 24 h. After cooling to 50
ꢀC, the polymerization was stopped by adding few drops of
10% KOH solution in water and stirred for 1 h. The resulting
solution was concentrated under vacuum; the residue was
dissolved in ethyl acetate and the extracts were washed sev-
eral times with brine, dried with anhydrous MgSO4, and con-
centrated. The isolated polymer was dried at 40 ꢀC under
vacuum for 24 h. Results are summarized in Table 1.
This work describes the synthesis of random copolymers of
2-nonyl-2-oxazoline and 2-(9-decenyl)-2-oxazoline by CROP
and its thiol-ene functionalization with 2-mercaptoethanol to
produce a set of linear polyoxazoline–polyols under mild
conditions, quantitative double bond transformation, and
with control of the molecular weight and the hydroxyl con-
tent. The polyols obtained in this way were reacted with
methylene-bis(phenylisocyanate) (MDI) to yield a series of
polyurethane networks that were characterized by fluores-
cence infrared spectroscopy (FTIR), differential scanning cal-
orimetry (DSC), thermogravimetric analysis (TGA), and ther-
modynamic mechanical analysis to establish the relationship
of polyol structure–PU properties.
1H nuclear magnetic resonance (NMR) (CDCl3, tetramethylsi-
lane (TMS), d in ppm): 7.40–7.10 (m, ArCH2), 5.85–5.70 (m,
CH¼¼CH2), 5.03–4.87 (dd, Jcis 10.0 Hz, Jtrans 17.2 Hz
ACH¼¼CH2), 4.65 (m, ACH2Ar), 4.35 (NHACH2CH2OCOA),
3.72 (NACH2CH2OH), 3.61 (NACH2CH2OH), 3.50–3.24 (m,
CH2AN),
2.80
(NHACH2CH2OCOA),
2.42–2.10
(m,
ACH2ACONA), 2.05–1.95 (m, CH2ACH¼¼CH2), 1.65–1.45 (m,
EXPERIMENTAL
ACH2ACH2ACOA), 1.43–1.20 (m, aliphatic backbone), 0.86
(t, ACH3). 13C NMR (CDCl3, TMS,
d in ppm): 173.8
Materials
(ACOANA), 139.2 (ACH¼¼CH2), 136.4, 129.1, 128.7, 126.3
(ArACH2A), 114.2 (ACH¼¼CH2), 58.6 (NACH2CH2OH), 51.7
(NACH2CH2OH), 45.3 (NACH2Ar), 33.8 (ACH2CH¼¼CHA),
33.4 (ACH2CONA), 31.9 (ACH2ACH2ACH3), 29.6–29.0 (ali-
phatic backbone), 25.4 (ACH2CH2CONA), 22.7 (ACH2ACH3),
14.4 (ACH3).
2,2-Dimethoxy-2-phenylacetophenone (DMPA), 10-undece-
noic acid, 2-mercaptoethanol, and MDI were purchased from
Aldrich. Benzyl bromide and thionyl chloride were obtained
from Fluka, and decanoic acid and ethanolamine were from
Merck. Tetrahydrofuran (THF) and acetonitrile (ACN) from
Aldrich were dried with sodium benzophenone and P2O5,
respectively, and distilled immediately before use.
Thiol-ene Polymer Modification
In a shlenck flask under argon, 2.1 g of polyoxazoline (1.5
mmol of 1a, 1b, or 1c; 2.0 mmol of 2a, 2b, or 2c; and 3.0
mmol of 3a, 3b, or 3c) was dissolved in 5 mL of anhydrous
THF. DMPA (1% respect to the C¼¼C) as initiator and mer-
captoethanol (2.5 equiv. respect to the C¼¼C) were added.
The mixture was stirred and irradiated at 365 nm (18 W)
for 2 h at room temperature. Excess reagent was removed
by washingꢀwith brine several times, and the polymer was
dried at 40 C under vacuum for 48 h.
Synthesis of Monomers
2-(9-Decenyl)-2-oxazoline (I) and 2-nonyl-2-oxazoline (II)
(Scheme 1) were synthesized starting from the correspond-
ing acids and ethanolamine following general reported pro-
cedures.45–48 The monomers were purified by fractional dis-
tillation over barium oxide under vacuum and stored under
argon. The physical properties and spectral characteristics
were in accordance with those described in the litera-
3070
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 1 Monomer Composition in the Feed and Molecular Weight of the Resulting Copolyoxazolines
Ia
IIa
I þ II/BnBr ratio
Mn (ꢁ 10ꢂ3 b
)
Mn (ꢁ10ꢂ3 c
)
Mn (ꢁ10ꢂ3 d
)
DPe
Mn/Mw
d
Sample
1a
1b
1c
2a
2b
2c
3a
3b
3c
a
0.15
0.15
0.15
0.20
0.20
0.20
0.30
0.30
0.30
0.85
0.85
0.85
0.80
0.80
0.80
0.70
0.70
0.70
8/1
1.70
2.89
4.09
1.70
2.90
4.10
1.71
2.92
4.12
2.09
3.20
4.12
2.16
3.25
4.20
1.97
3.20
4.40
2.24
3.23
4.09
2.24
3.28
4.17
2.21
3.25
4.20
11.2
16.2
20.5
11.2
16.4
20.9
11.0
16.1
20.9
1.08
1.08
1.11
1.08
1.08
1.15
1.08
1.09
1.13
14/1
20/1
8/1
14/1
20/1
8/1
14/1
20/1
d
e
Molar percentage in the feed.
Theoretical from the IþII/BnBr ratio.
Determined by 1H NMR.
Determined by SEC using polystrene standards.
Calculated from SEC results.
b
c
1H NMR (CDCl3, TMS, d in ppm): 7.40–7.10 (m, ArCH2), 4.65
(m, ACH2Ar), 3.70 (t, CH2AOH), 3.50–3.24 (m, CH2AN), 2.71
(t, SACH2ACH2OHA), 2.50 (t, ACH2ASACH2CH2OH), 2.40–
2.10 (m, ACH2CONA, 1.65–1.45 (m, ACH2ACH2ACONA),
1.43–1.20 (m, aliphatic backbone), 0.86 (t, ACH3). 13C NMR
(CDCl3, TMS, d in ppm): 173.9 (ACOANA), 136.4, 129.1,
128.7, 126.3 (ArACH2A), 60.5 (SACH2CH2OH), 58.6
(NACH2CH2OH), 51.7 (NACH2CH2OH), 45.3 (NACH2Ar), 35.1
(SACH2CH2OH), 33.4 (ACH2CONA), 31.7 (ACH2ACH2ACH3),
29.6–29.0 (aliphatic backbone), 25.5 (ACH2CH2CONA), 22.7
(ACH2ACH3), 14.2 (ACH3).
value of the internal standard, and gPOH is the amount of
polyol in grams.
Characterization
GC-mass spectrometry (GC-MS) analysis was done in a
6890N gas chromatograph and a 5973 mass spectrometer
(Agilent Techologies, Palo Alto, CA), using a HP-5MS capillary
column of 5% phenyl polydimethyl siloxane (30 m, 0.25 mm,
0.25 lm), from Agilent Technologies, and 99.999% pure he-
lium gas as the carrier at a flow rate of 1.5 mL minꢂ1. Sam-
ples were injected in a split/splitless injector at a split ratio
ꢀ
of 20:1, at a temperature of 280 C. The oven temperature of
Polyurethane Synthesis
ꢀ
ꢀ
GC was initially held at 60 C, raised to 260 C at a rate of 6
ꢀC minꢂ1, and then held for 50 min. The mass spectrometer
acquired data in scan mode with an m/z interval from 35 to
800. The compounds were identified by comparison with
NIST08 MS library.
Around 2–3 g of polyol was dissolved under argon in the
same weight of anhydrous toluene. The solution was heated
at 50 ꢀC and added to a 50% (w/w) dispersion of MDI in
toluene. In all cases, a ratio of 1.02 equivalents of isocyanate
group per hydroxyl was used. The mixture was stirred under
argon until it become homogeneous and casted over silan-
1H NMR (400 MHz) and 13C NMR (100.6 MHz) spectra were
recorded in CDCl3 using a Varian Gemini 400 spectrometer.
Chemical shifts were reported in ppm relative to TMS as in-
ternal standard.
ꢀ
ized glass plaquesꢀpreheated at 90 C. The polyurethane was
ꢀ
maintained at 90 C for 2 h and at 130 C for 3 h producing
films with 0.03–0.04 mm thickness.
FTIR spectra were recorded on a Bomem Michelson MB 100
FTIR spectrophotometer with a resolution of 4 cmꢂ1 in the
absorbance mode. An attenuated total reflection (ATR) de-
vice with thermal control and a diamond crystal (Golden
Gate heated single-reflection diamond ATR, Specac-Tknok-
roma) was used.
Determination of Hydroxyl Content
Quantification of the hydroxyl content was carried out using
1H NMR spectroscopy and 2-phenylethanol as internal stand-
ard using the signal at 2.85 ppm as reference. Accurate
weights of polyol and the internal standard were introduced
in a NMR tube. The intensity of the partially overlapped sig-
nals at 3.79–3.70 ppm (side chain ACH2OH and chain-end
ACH2OH groups) and the signal at 2.71 ppm
(ASACH2ACH2OH side chain groups) was compared with
that of the intensity of the signals at 3.86 and 2.85 ppm
(PhACH2ACH2AOH) of internal standard. The hydroxyl con-
tent in millimole of hydroxyl groups per gram of polyol was
calculated using eq 1.
MALDI-TOF MS measurements were performed with a Voy-
ager DE-RP mass spectrometer (Applied Biosystems, Fra-
mingham, MA) equipped with a nitrogen laser delivering 3
ns laser pulses at 337 nm. 1,8,9-Antracenetriol was used as
matrix and potassium trifluoroacetate as dopant. Samples
were prepared from a THF solution containing 40 mg/mL of
matrix, 3 mg/mL of polymer, and 0.1 mg/mL of dopant. The
homogenized mixture was casted in the MALDI plate.
value OH ¼ ðmmolIS ꢁ IntPOHÞ=ðIntIS ꢁ gPOHÞ
(1)
Size exclusion chromatography (SEC) analysis was carried
out with an Agilent 1200 series system with PLgel 3 lm
MIXED-E, PLgel 5 lm MIXED-D, and PLgel 20 lm MIXED-A
where mmolIS is the internal standard amount in mmol,
IntPOH is the integral value of the polyol, IntIS is the integral
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3071

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
SCHEME 2 Cationic ring-opening copolymerization of oxazoline monomers and possible chain termination mechanisms.
columns in series and equipped with an Agilent 1100 series
refractive-index detector. Calibration curves were based on
polystyrene standards having low polydispersities. THF was
used as an eluent at a flow rate of 1.0 mL/min; the sample
concentrations were 5–10 mg/mL, and injection volumes of
100 lL were used.
for I and 76% for II). Physical properties and structural
characteristics were in accordance with those described in
the literature.31,47
I was copolymerized with the necessary molar amount of II
to obtain copolymers with a 15% (1 series), 20% (2 series),
and 30% (3 series) of unsaturated comonomer (Table 1).
The copolymerizations were carried out at 100 ꢀC in anhy-
drous ACN using benzyl bromide as initiator (Scheme 2).
Calorimetric studies were carried out on a Mettler DSC822
differential scanning calorimeter using N2 as a purge gas (20
mL/min). Dynamic mechanical thermal analysis (DMTA) and
tensile tests were performed using a TA DMA 2928 in the
controlled force–tension film mode with a preload force of
0.1 N, an amplitude of 10 lm, and at a fixed frequency of 1
Each mixture of monomer was polymerized using three dif-
ferent monomer/initiator ratios 8/1 (a series), 14/1 (b se-
ries), and 20/1 (c series). These initiator ratios were
selected to obtain copolymers with a theoretical polymeriza-
tion degree (DP) of 8, 14, and 20, respectively. In this way,
nine different copolymers with varying compositions and
molecular weights were obtained essentially with quantita-
tive conversions. Results determined by SEC and 1H NMR
are collected in Table 1, together with the theoretical values
expected from the monomer–initiator ratio used.
ꢀ
ꢀ
Hz in the ꢂ80 to 170 C range and at a heating rate of 3 C/
min. The tensile assays were performed in triplicates on rec-
tangular samples (10 ꢁ 5 ꢁ 0.5 mm3) measuring the strain
while applying a ramp of 0.5 N/min at 25 ꢀC. A preload
force of 0.05 N and a soak time of 3 min were used. Thermal
stability studies were carried out on
a Mettler TGA/
SDTA851e/LF/1100 with N2 as purge gas. The studies were
performedꢀin the 30–800 ꢀC temperature range at a scan
rate of 10 C/min.
All copolymers were obtained in quantitative conversion and
so the comonomer composition should match with the feed.
The 1H NMR spectra [(Fig. 1(a)] show the expected signals
for copolymers obtained via a CROP initiated by benzyl bro-
mide. In addition to the main and side chain proton signals
of both comononer units, some other low intensity signals
can be observed. These signals do not disappear or change
their intensity by precipitating the polymer and so it can be
assumed that they are due to end groups instead of polymer
impurities. Signals between 7.40–7.12 ppm and 4.65–4.47
ppm can be assigned to the starting benzyl moieties whereas
the signals at 3.77 and 3.63 ppm can be assigned to the final
hydroxyethyl groups formed in the termination step by reac-
tion with water. This was confirmed by ‘‘in situ’’ derivatiza-
tion of the polymer with trichloroacetylisocyanate, which
produces a downfield shift of 0.48 ppm for the signal at 3.77
ppm in the 1H NMR spectrum. The relative intensity of sig-
nals at 4.65–4.47 ppm was used to estimate the molecular
weight by comparing with the main chain methylene signal
at 3.42 ppm. The signals at 4.15 and 2.90–2.60 ppm could
be assigned to an ester end group, which can be formed
RESULTS AND DISCUSSION
As commented, the main purpose of this work is preparing
tailor-made thermosetting PUs from comb-like copoly-2-oxa-
zolines functionalized with hydroxyl groups in the side chain.
To prepare the starting 2-oxazoline monomers, we chose 10-
undecenoic acid and decanoic acid as unsaturated and satu-
rated precursors, respectively. Both starting materials are
from renewable origin. 10-Undecenoic acid is the main prod-
uct of castor oil caustic fusion or pyrolisis,49 and decanoic
acic occurs naturally in coconut and palm kernel oils and is
the major component of eml seed oil (60–70%).50 The syn-
thesis of the 2-oxazoline monomers was accomplished using
the standard two-step procedure consisting of the reaction
of the acid with ethanolamine followed by treatment with
thionyl chloride and basic hydrolysis45–47 (Scheme 1).
The resulting 2-oxazolines were purified by fractional distil-
lation under vacuum giving colorless oils in good yield (82%
3072
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
FIGURE 1 1H NMR spectra of (a) 2c and (b) its thiol-ene addition product 2cOH
.
alternatively by attack of water to the position 5 of the
growing oxazolinium group as it has been established by
Schubert and coworkers51 (Scheme 2).
The main distribution fits with the expected structure for a
random copolymer with benzyl and hydroxyl end groups
and presents molecular weight distributions close to that
observed by SEC. Each peak in the main distribution
presents a fine structure which can be associated to the dif-
ferent comonomer composition as is shown in the expanded
zone of the peak corresponding to DP 14 in Figure 3(b). As
can be seen, the peaks of II14, II13/I1, II12/I2 and up to II7/
I7 can be observed, by varying its relative intensity with the
amount of II comonomer in the 3b. The same MALDI-TOF-
MS patterns are observed for 1 and 2 copolymer series. In
all cases, a secondary peak distribution is observed. The
mass of these secondary peaks fits with a random copolymer
with hydrogen and hydroxyl end groups (difference of 91 Da
with the main series peaks). These peaks can be explained
on the basis of known chain transfer reactions that take
place during the polymerization of oxazolines52 leading to
hydrogen-initiated and enamine chains. Alternatively, these
species can be formed as result of chain-transfer processes
with traces of water or H-nucleophiles present in the poly-
merization media.53 This secondary set of peaks follows the
In all cases, SEC chromatograms of polymers (Fig. 2) show
narrow monomodal distributions with polydispersities
around 1.08 with the exception of the polymers obtained
with 20/1 monomer/initiator ratio that presents a slightly
higher polydispersity (1.11–1.15; Table 1). The Mn values
obtained by SEC were somewhat higher than the theoretical
ones taking into account the monomer/initiator ratio used.
This behaviour is especially remarkable in the case of the
low-molecular weight polymers.
Molecular weights estimated by 1H NMR using the benzyl
group signal were also higher than the theoretical ones
because other existing initiations were not taken into
account in the measurements. To gain insight about the
structure of these polymers, MALDI-TOF-MS analysis was
performed. In Figure 3(a), the representative spectrogram
obtained for 2b is shown. In all samples, at least two differ-
ent Gaussian peak distributions can be observed.
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3073

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
The hydroxyl number of all polyoxazoline–polyols was deter-
mined by 1H NMR measuring the amount of hydroxyl end
groups and side chain hydroxyl groups. The intensity of the
signals at 3.80–3.70 ppm corresponding to side chain
ACH2OH and chain-end ACH2OH groups were compared
with the intensity of the signals at 3.86 and 2.85 ppm
(PhACH2ACH2AOH) with a known amount of phenylethanol
as internal standard. Also, the intensity of the signal at 2.71
ppm, corresponding to the ASACH2CH2OH, was measured to
calculate the hydroxyl content due exclusively to side chain
groups. Both hydroxyl contents expressed in mmol of OH
per gram of polymer are collected in Table 2. The calculated
side chain hydroxyl values are in good agreement with the
theoretical ones taking into account the copolymer composi-
tion thus confirming the efficiency of the functionalization
under the mild conditions used. For each 3PU, 3PU, and 3PU
FIGURE 2 SEC chromatograms of 3a and 3c (continuous line)
and the corresponding thiol-ene addition products 3aOH and
3cOH (dashed line).
same comonomer composition as that of the main peak se-
ries but not a Gaussian distribution. As expected, for species
formed in chain-transfer processes, the peaks are more
intense in the low-molecular weight region and its intensity
decreases progressively as the DP increases. According to the
relative peak intensity, the amount of chain transfer species
is more important in 1c, 2c, and 3c compounds, which is in
agreement with their wider molecular weight distribution
observed by SEC. The presence of copolymer chains not initi-
1
ated by benzyl groups also explains why the H NMR molec-
ular weight calculation gave higher values than the theoreti-
cal ones.
We used photoinitiated thiol-ene addition of 2-mercaptoetha-
nol to functionalize the pendant copolyoxazoline double
bonds (Scheme 3 and Table 3).
After 2 h of irradiation, complete double bond conversion was
achieved and the polyoxazoline-based poliols were obtained
1
with simple work-up in quantitative yields. H NMR analysis
[Fig. 1(b)] show the absence of double bond signals and
the appearance of three new signals at 2.51 ppm
(ACH2ASACH2CH2AOH), 2.71 ppm (ACH2ASACH2CH2AOH),
and 3.71 ppm (ACH2ASACH2CH2AOH) corresponding to the
2-hydroxyethylthioether group.
SEC analysis of polyoxazoline polyols (Fig. 2 and Table 2)
show similar molecular weight distributions and polydisper-
sities when compared with the starting unfunctionalized
polyoxazolidine, thus indicating the absence of degradation
and side chain processes.
FIGURE 3 (a) MALDI-TOF-MS spectra of 3b, numbers indicate
the DP. (b) Amplification of the peak corresponding to DP ¼
14. Numbers indicate the copolymer composition: the first indi-
cate the number of saturated monomeric units II and the sec-
MALDI-TOF-MS spectra show a complex pattern with several
series of peaks separated by 78 mass units, which corre-
sponds to the ACH2ASACH2CH2AOH moiety. These series of
peaks could be assigned to the copolymers initiated with H
and Bn groups and cationized either by sodium or
potassium.
ond the number of unsaturated
I ones. The series with
apostrophe correspond to the polymers chains not initiated
with benzyl groups.
3074
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
SCHEME 3 Functionalization of the copolyoxazolines by thiol-ene addition of 2-mercaptoethanol.
polyol series, the total hydroxyl content increases as the mo-
lecular weight decreases due to the contribution of the chain
end hydroxyl groups.
MDI toluene suspension leading to homogeneous systems
before the curing cycle.
FTIR analysis (Fig. 4) demonstrated the completion of the
reaction by the disappearance of the band at 2240 cmꢂ1 cor-
responding to the AN¼¼C¼¼O stretching of the isocyanate
group and the appearance of the characteristic absorbances
of the urethane links at 1728 cmꢂ1, which appears clearly
differentiated from the band at 1641 cmꢂ1 corresponding to
the C¼¼O stretching of the polyoxazoline amide groups. The
C¼¼O of these urethane linkages appears as a broad band
between 1745 and 1690 cmꢂ1 with a maximum at 1728
cmꢂ1 and a shoulder at 1710 cmꢂ1, which are attributable
to the associated and nonassociated C¼¼O urethane groups.
No additional shoulders at about 1660 cmꢂ1 attributable to
urea linkages were observed. Moreover, the broad band cen-
tred at 3500 cmꢂ1, corresponding to the OAH stretching,
shifts to lower frequencies and shows a maximum at 3315
cmꢂ1, characteristic of the NAH stretching of urethanes.
The resulting polyoxazolidine–polyols were yellowish clear
waxy solids. DSC analysis shows its semicrystalline nature
ꢀ
with glass transition temperature (Tg) values below 0 ꢀ C and
multiple melting endotherms in the range of 50–100 C (Ta-
ble 2). Both, first- and second-order transitions increase with
molecular weight for all copolymer series. Low-molecular
weight copolyols showed a single melting endotherm at
ꢀ
about 50 C whereas the rest of polyols show complex endo-
therms. According to results, the different thermal behavior
of these samples seems to be related to the differences in
molecular weight rather that to the hydroxyl group content.
These polyoxazoline-based polyols were used to prepare
polyurethane thermosets by reaction with MDI in solution of
toluene. In all cases, a slight excess of isocyanate group over
the theoretical amount based on the total hydroxyl content
of polyol was used. The PU prꢀepolymers were casted onto
glass plaques and cured at 130 C for 3 h to produce flexible
to hard yellowish transparent materials as films with 0.03–
0.04 mm thickness. The synthesized polyols having primary
hydroxyl groups showed an excellent compatibility with the
The thermal properties of polyurethanes were investigated
by DSC (Fig. 5 and Table 3). The most significantꢀfeature is
the appearance of a melting endotherm at 57–58 C for PUs
obtained from polyols of PUb with medium molecular weight
(2900 Da) and PUc with high molecular weight (4100 Da).
TABLE 2 Molecular Weight, Hydroxyl Content, and Thermal Properties of the Polyoxazoline Polyols
Sample
Mn ꢁ 10ꢂ3 (Mn/Mw)a
mmolOH/g polyolb
mmolOH/g polyolc
Tg (ꢀC)d
Tm (ꢀC)d
DH (J/g)d
1aOH
1bOH
1cOH
2aOH
2bOH
2cOH
3aOH
3bOH
3cOH
a
2.34 (1.11)
3.41 (1.11)
4.31 (1.11)
2.42 (1.11)
3.57 (1.11)
4.56 (1.16)
2.50 (1.10)
3.62 (1.11)
4.58 (1.14)
0.79
0.80
0.80
1.08
1.09
1.09
1.20
1.23
1.21
1.22
1.08
0.89
1.45
1.34
1.18
1.66
1.47
1.32
c
ꢂ11.6
ꢂ9.0
ꢂ2.5
ꢂ10.9
ꢂ7.4
ꢂ3.6
ꢂ11.7
ꢂ6.6
ꢂ3.9
50.7
14.1
40.7
41.3
22.9
40.6
41.9
21.1
40.1
41.8
58.3, 81.0, 88.3
58.3, 87.3, 101.3
56.3
59.0, 84.3
58.3, 82.7, 93.7
54.7
57.7, 78,7
60.3, 84.7
Determined by SEC using polystrene standards.
Side chain OH content determined by 1H NMR.
Side chain plus chain-end OH content determined by 1H NMR.
b
d
Determined by DSC.
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3075

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
FIGURE 4 FTIR-ATR spectrum of 2cPU
.
In these cases, the crystallinity is higher for the PUs obtained
from the higher molecular weight polyoxazoline polyols.
When comparing PUs obtained from polyols of the same mo-
three main stages for most of the PUs [Fig. 6(a)]. The first
ꢀ
ꢀ
has a maximum that increases fromꢀ219 C to 272 C in the
ꢀ
2PU series and from 233 C to 252 C in the 3PU series in a
lecular weight but increasing hydroxyl content (1bPU, 2bPU
,
parallel way with the increase of molecular weight of the
starting polyol. The associated weight loss (5–10% of the
total mass) decreases with the same parameter and so this
decomposition process can be associated in part to the par-
ent polyol moiety fragmentation. This can be inferred when
comparing the TGA curves of a PU, its parent polyol and the
original polyoxazoline [(Fig. 6(b)]. As can be seen, all poly-
mers have a main degradation process with a maximum
3bPU and 1cPU, 2cPU, 3cPU series), a decrease in its relative
crystallinity degree is observed which can be related to the
increasing difficulty in the chain packing as the crosslinking
degree increases.
ꢀ
The Tg for all PUs is around 20 C showing a slight increase
with the starting polyol molecular weight in series 1PU and
2PU but not in the case of series 3PU. This complex behavior
can be related with the occurrence of two different effects.
Lower molecular weight polyols have higher hydroxyl con-
tent and consequently leads to PUs with higher number of
urethane links; however, PUs from high-molecular weight
polyols have crystalline regions that act as physical crosslink-
ing restricting mobility. In the absence of other effects, PUs
made from polyols of similar molecular weight but increas-
ing hydroxyl content, should lead to an increase in the num-
ber of urethane links and consequently in the Tg values. This
behavior is effectively observed in PUs obtained from similar
comb-like amorphous polyetherpolyols.15 In this case, it
must be considered that the semicrystalline nature of these
PUs and the fact that the strong polar interactions between
the amide groups can be disrupted as the amount of ure-
thane links increases. This consideration is supported by the
fact that polyoxazoline–polyols with the similar molecular
weight but different hydroxyl content (i.e., 1cOH, 2cOH, and
3cOH) have very close Tg values (Table 2).
ꢀ
around 400 C, but the polyoxazoline polyol and the polyur-
ethane present an additional low temperature degradation
process that is not observed in the polyoxazolidine polymer.
This seems to indicate that this degradation process is
related in part with the cleavage of the ASACH2ACH2AOA
moieties.
The thermal stability of polyurethanes was studied in N2
atmosphere (Table 3). The shapes of the weight-loss curves
were very similar for all PUs and differences in the thermal
stability seem to be negligible. The decomposition shows
FIGURE 5 Second heating DSC traces of 2aPU, 2bPU, and 2cPU
.
3076
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
TABLE 3 Thermal and Mechanical Properties of Pus
Tg (ꢀC)
Tm
TGA (N2)
DMTA
Tmax (ꢀC)a
DH (J/g)a
Tmax (ꢀC)
Young’s Modulus (MPa)
Stress (MPa)
Strain (%)
a
b
Sample
1= DC
tan dmax
2
p
1aPU
1bPU
1cPU
2aPU
2bPU
2cPU
3aPU
3bPU
3cPU
a
12
20
26
20
21
23
24
21
18
49
56
64
51
50
53
52
51
48
–
–
224, 335, 415
338, 413
6
2.2
9.3
14.3
3.1
6.2
7.9
4.5
5.3
8.0
177
76
58
58
–
8.3
9.7
–
85
254
6
338, 414
25
219, 334, 410
238, 333, 410
272, 337, 410
233, 338, 411
246, 340, 409
252, 336, 410
b
268
94
57
57
–
2.9
6.1
–
41
46
8
83
229
163
160
–
–
13
29
58
1.8
Determined by DSC.
Determined by DMTA.
ꢀ
The second degradation step has a maximum at 335–340 C
for all samples and can be associated to the initial decompo-
sition of the urethane links. The third and main decomposi-
The dynamomechanical behavior of PUs was studied by
DMTA. Typical DMTA plots for crosslinked polymeric net-
works were observed in all cases. A glassy state with a high
modulus plateau E0 at lower temperatures and a rubbery
state with lower E0 at higher temperatures is observed. For
ꢀ
tion step has a maximum at 410–415 C for all PUs and sup-
poses the complete degradation of the polyoxazoline and
urethane moieties leading to a carbonaceous residue of
about 1% of the initial weight in all cases.
FIGURE 7 Variation of tan d peak of PUs with (a) the molecular
weight of the parent polyol (3aPU, 3bPU, and 3cPU) and (b) with
the hydroxyl content of the parent polyol (1cPU, 2cPU, and
3cPU).
FIGURE 6 (a) First derivative curve of the TGA plots of 3aPU
,
3bPU, and 3cPU. (b) TGA plots of 3c, 3cOH, and 3cPU
.
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3077

JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY DOI 10.1002/POLA
using the thiol-ene addition of mercaptoethanol, to produce
renewable polyoxazoline–polyols. These polyols have preset
characteristics that can be used to prepare PUs with tailored
characteristics ranging from hard-rigid to soft-elastic
materials.
The authors express their thanks to MICINN (Comisio´n Intermi-
nisterial de Ciencia y Tecnologı´a) (MAT2008-01412) for finan-
cial support for this work.
REFERENCES AND NOTES
1 Belgacem, M. N.; Gandini, A. Monomers, Polymers and Com-
posites from Renewable Resources; Elsevier: Oxford, 2008.
2 Guner, F. S.; Yagci, Y.; Erciyes, A. T. Prog Polym Sci 2006,
¨
31, 633–679.
FIGURE 8 Stress–strain plots of all PUs.
3 Meier, M. A. R.; Metzger, J. O.; Shubert, U. S. Chem Soc Rev
2007, 36, 1788–1802.
all PU series, the storage moduli at the rubbery plateau
slightly increases with the molecular weight of the starting
polyol. PUs behave like homogeneous networks with a
unique broad transition that according to DSC corresponds
to the glass transition in the amorphous PU and the glass
transition and melting in the semicrystalline PUs.
4 (a) Sharma, V.; Kundu, P. P. Prog Polym Sci 2008, 33,
1199–1215; (b) Sharma, V.; Kundu, P. P. Prog Polym Sci 2006,
31, 983–1008.
5 Xia, Y.; Larock, R. C. Green Chem 2010, 12, 1893–1909.
6 Biermann, U.; Friedt, W.; Lang, S.; Luhs, W.; Machmuller, G.;
¨
¨
Metzger, J. O.; Klaas, G.; Rusch, M.; Schafer, H. J.; Schneider,
¨
¨
Figure 7(a,b) shows the representative tan d of some synthe-
sized PUs. Consistent with DSC results, this peak shows a
slight shift of maxima when increasing the molecular weight
of the parent polyol for series 1PU and 2PU, which is consist-
ent with the increase in its crystallinity degree. In the case
of PU3 series that are practically amorphous, a slight
decrease of the maxima is observed [Fig. 7(a)]. The variation
of the tan d peak when comparing 1PU, 2PU, and 3PU series
[Fig. 7(b)] is also consistent with the decrease in the crystal-
line content as the functionality of the parent polyol
increases. These results seem to indicate that mechanical
properties of these PUs are mostly governed by their crystal-
line content, which is directly related to the molecular
weight of the parent polyol, instead of the amount of ure-
thane links introduced.
M. P. Angew Chem Int Ed Engl 2000, 39, 2206–2224.
7 Petrovic, Z. S. Polym Rev 2008, 48, 109–155.
8 Lligadas, G.; Ronda, J. C.; Galia`, M.; Ca´diz, V. Biomacromole-
cules 2010, 11, 2825–2835.
9 Harry-O’kuru, R. E.; Holster, R. A.; Abbott, T. P.; Weisleder, D.
Oil Ind Crop Prod 2002, 15, 51–58.
10 Guo, Y.; Hardesty, J. H.; Mannri, V. M.; Massingill, J. J. L.
J Am Oil Chem Soc 2007, 84, 928–935.
11 Gryglewicz, S.; Piechocki, W.; Gryglewicz, G. Bioresour
Technol 2003, 87, 35–39.
12 Campanella, A.; Bonnaillie, L. M.; Wool, R. P. J Appl Polym
Sci 2009, 112, 2567–2578.
13 Petrovic, Z. S.; Guo, A.; Javni, I. Biomacromolecules 2005,
Figure 8 shows the stress–strain curves, and Table 3 summa-
rizes the mechanical properties of obtained PUs. Young’s
modulus and tensile strength at break increase as the molec-
ular weight of the parent polyol increases for all series. As
expected, an enhancement of the modulus and tensile
strength can be observed when increasing the semicrystal-
line nature of the PUs. What is most notable is the high
modulus of 1cPU that is consistent with its high crystalline
degree and its maximum Tg and tan d values determined,
respectively, by DSC and DMTA.
6, 713–719.
14 Petrovic, Z. S.; Guo, A.; Javni, I.; Cvetkovic, I.; Hong, D. P.
Polym Int 2008, 57, 275–281.
15 Lligadas, G.; Ronda, J. C.; Galia`, M.; Biermann, U.; Metzger,
J. O. J Polym Sci, Part A: Polym Chem 2006, 44, 634–645.
16 del Rio, E.; Galia`, M.; Cadiz, V.; Lligadas, G.; Ronda, J. C.
J Polym Sci Part A: Polym Chem 2010, 48, 4995–5008.
17 del Rio, E.; Lligadas, G.; Ronda, J. C.; Galia`, M.; Ca`diz, V.
J Polym Sci Part A: Polym Chem 2010, 48, 5009–5017.
18 Kobayashi, S.; Uyama, H. J Polym Sci Part A: Polym Chem
CONCLUSIONS
2002, 40, 192–209.
The transformation of saturated and unsaturated fatty acids
in oxazoline monomers is a straightforward and convenient
methodology to produce polymers with controlled molecular
weight, composition, and end group functionality using
CROP. The side chain unsaturated groups can be effectively
and quantitatively functionalized, under mild conditions
19 Hoogenboom, R. Macromol Chem Phys 2007, 208, 18–25.
20 Park, J.-S.; Kataoka, K. Macromolecules 2007, 40,
3599–3609.
21 Fijten, M. W. M.; Kranenburg, J. M.; Thijs, H. M. L.; Paulus,
R. M.; van Lankvelt, B. M.; de Hullu, J.; Springintveld, M.;
3078
WILEYONLINELIBRARY.COM/JOURNAL/JPOLA

ARTICLE
Thielen, D. J. G.; Tweedie, C. A.; Hoogenboom, R.; van Vliet, K.
J.; Schubert, U. Macromolecules 2007, 40, 5879–5886.
37 Cai, G.; Litt, M. H. J Polym Sci Part A: Polym Chem 1996,
34, 2689–2699.
22 Fijten, M. W. M.; Hoogenboom, R.; Schubert, U. J Polym
38 Hoyle, C. E.; Bowman, C. N. Angew Chem Int Ed Engl 2010,
Sci, Part A: Polym Chem 2008, 46, 4804–4816.
49, 1540–1573.
23 Weber, C.; Becer, C. R.; Baumgaertel, A.; Hoogenboom, R.;
39 Hoyle, C. E.; Lowe, A. B.; Bowman, C. N. Chem Soc Rev
Schubert, U. Design Monom Polym 2009, 12, 149–165.
2010, 39, 1355–1387.
24 Kempe, K.; Jacobs, S.; Lambermont-Thijs, H. M. L.; Fijten,
M. W. M.; Hoogenboom, R.; Schubert, U. Macromolecules
2010, 48, 5533–5540.
40 Lluch, C.; Ronda, J. C.; Galia`, M.; Lligadas, G.; Ca´diz, V. Bio-
macromolecules 2010, 11, 1646–1653.
41 Luxenhofer, R.; Jordan, R. Macromolecules 2006, 39,
25 Aoi, K.; Okada, M. Prog Polym Sci 1996, 21, 151–208.
3509–3516.
26 Zhou, P.; Blubaum, J. E.; Burns, C. T.; Natale, N. R. Tetrahe-
42 Gress, A.; Volkel, A.; Schlaad, H. Macromolecules 2007, 40,
¨
dron Lett 1997, 38, 7019–7020.
7928–7933.
27 Bandgar, B. P.; Pandit, S. S. Tetrahedron Lett 2003, 44,
43 Diehl, C.; Schlaad, H. Macromol Biosci 2009, 9, 157–161.
2331–2333.
44 Cortez, M. A.; Grayson, S. M. Macromolecules 2010, 43,
28 Stefano, C.; Abigail, Y.; Bruno, L. Tetrahedron Lett 2004, 45,
4081–4090.
9611–9615.
45 Culbertson, B. M. Prog Polym Sci 2002, 27, 579–626.
29 Hioki, K.; Takechi, Y.; Kimura, N.; Tanaka, H.; Kunishima, M.
46 Holerca, M. N.; Percec, V. Eur
J Org Chem 2000,
Chem Pharm Bull 2008, 56, 1735–1737.
2257–2263.
30 Kempe, K.; Lobert, M.; Hoogenboom, R.; Schubert, U. S.
47 Moaddel, H.; Ameri, R. Iranian Polym J 2005, 14, 849–854.
J Comb Chem 2009, 11, 274–280.
48 Jordan, R.; Martin, K.; Arder, H. J.; Unger, K. K. Macromole-
31 Cai, G.; Litt, M. H. J Polym Sci Part A: Polym Chem 1996,
cules 2001, 34, 8858–8865.
34, 2679–2688.
49 Van der Steen, M.; Stevens, C. V. ChemSusChem 2009, 2,
32 Cai, G.; Litt, M. H. J Polym Sci Part A: Polym Chem 1996,
692–713.
34, 2701–2709.
50 Badami, R. C.; Patil, K. B. Prog Lipid Res 1981, 19, 119–153.
33 Hoogenboom, R.; Wiesbrock, F.; Huang, H.; Leenen, M. A.
M.; Thijs, H. M. L.; van Nispen, S. F. G. M.; van der Loop, M.;
Fustin, C.-A.; Jonas, A. M.; Gohy, J.-F.; Schubert, U. S. Macro-
molecules 2006, 39, 4719–4725.
51 Baumgaertel, A.; Altuntas, E.; Kempe, K.; Crecelius, A.;
Schubert, U. S. J Polym Sci Part A Polym Chem 2010, 48,
5533–5540.
34 Chung, T. C.; Raate, M.; Berluche, E.; Schulz, D. N. Macro-
52 Wiesbrock, F.; Hoogenboom, R.; Leenen, M.; van Nispen,
S. F. G. M.; van der Loop, M.; Abeln, C. H. C. H.; van der
Berg, A. M. J.; Schubert, U. S. Macromolecules 2005, 38,
7957–7966.
molecules 1988, 21, 1903–1907.
35 Eroglu, M. S.; Hazer, B.; Ozturk, T.; Caykara, T. J Appl Polym
Sci 2005, 97, 2132–2139.
36 Cai, G.; Litt, M. H. J Polym Sci Part A: Polym Chem 1996,
53 Warakomski, J. M.; Thill, B. P. J Polym Sci Part A: Polym
34, 2711–2717.
Chem 1990, 28, 3551–3563.
POLY-2-OXAZOLINE-DERIVED PUs, RIO ET AL.
3079