Oxetane-terminated telechelic epoxy-functional polyesters as cationically polymerizable thermoset resins: Tuning the reactivity with structural design

Article abstract of DOI:10.1002/pola.27673

A series of epoxy-functional telechelic oligomers containing oxetane end groups have been synthesized. The precursor monomer, extracted from outer Birch bark, was first polymerized through enzyme-catalyzed esterification to form oligomers having epoxy and/or oxetane groups in the structures. The oligoesters were subsequently crosslinked through cationic polymerization either by epoxy or oxetane homopolymerization or copolymerization when both functionalities were present. A study of the polymerizations of the resins was performed "in situ" using real-time Fourier transform infrared spectroscopy revealing a preferred copolymerization when compared with the homopolymerization. By tailoring the different structures, it was possible to control the final mechanical properties of the networks.

Full text of DOI:10.1002/pola.27673

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Oxetane-Terminated Telechelic Epoxy-Functional Polyesters as
Cationically Polymerizable Thermoset Resins: Tuning the Reactivity with
Structural Design
Susana Torron, Mats Johansson
Department of Fibre and Polymer Technology, Division of Coating Technology, KTH Royal Institute of Technology, School of
Chemical Science and Engineering, SE-10044 Stockholm, Sweden
Correspondence to: M. Johansson (E-mail: matskg@kth.se)
Received 11 March 2015; accepted 10 April 2015; published online 00 Month 2015
DOI: 10.1002/pola.27673
ABSTRACT: A series of epoxy-functional telechelic oligomers
containing oxetane end groups have been synthesized. The
precursor monomer, extracted from outer Birch bark, was first
polymerized through enzyme-catalyzed esterification to form
oligomers having epoxy and/or oxetane groups in the struc-
tures. The oligoesters were subsequently crosslinked through
cationic polymerization either by epoxy or oxetane homopoly-
merization or copolymerization when both functionalities were
present. A study of the polymerizations of the resins was per-
formed “in situ” using real-time Fourier transform infrared
spectroscopy revealing a preferred copolymerization when
compared with the homopolymerization. By tailoring the differ-
ent structures, it was possible to control the final mechanical
properties of the networks. V
2015 Wiley Periodicals, Inc. J.
C
Polym. Sci., Part A: Polym. Chem. 2015, 00, 000–000
KEYWORDS: biobased monomer; cationic polymerization; enzy-
matic catalysis; enzymes; epoxide; oxetane; telechelic
polymers
5
INTRODUCTION The production of materials derived from
renewable resources is a field in continued development.
The substitution of fossil-based raw materials for other more
environmentally friendly alternatives is of great interest in
terms of costs, degradability, and recyclability. Moreover, the
large variety of compounds readily available in nature with
from adhesives to composites. Cationic polymerization of
ring-opening monomers, such as epoxides or oxetanes, is
compared with free radical polymerization of acrylates not
affected by oxygen inhibition. Furthermore, ring-opening
monomers exhibit a lower amount of cure shrinkage. Both
these features are of importance in several applications rang-
ing from thin organic coatings (oxygen inhibition issues) to
dental composites (cure shrinkage).
“
inbuilt” functionalities encourages its use as building blocks.
One group readily present in nature that can undergo a large
variety of chemistries are epoxides. Epoxides are present in
vegetable oils or cork. One example of an epoxy-functional
monomer present in nature that has been used as building
block for network formation is the 9,10-epoxy-18-hydroxyoc-
tadecanoic acid (EFA; Fig. 1) that can be retrieved from the
To form polymer networks through cationic polymerization
using epoxy-derived monomers, it is necessary that those
monomers possess at least two functionalities as in the case
of telechelic polymers. By definition, telechelic polymers are
“
prepolymers capable of entering into further polymerization
1
–3
outer bark of the Birch tree.
This monomer is capable to
6
or other reactions through its reactive end groups.” The
introduction of end functionalities will also increase the
options for network formation as different chemistries can
be applied to each group.
undergo a large variety of chemical reactions, for example,
4
through condensation reaction of the end groups or through
reaction of the epoxide.
One of the strategies most frequently used for the prepara-
tion of thermosets derived from epoxy compounds is cationic
polymerization. One of the key features of cationic polymer-
ization is the possibility of triggering the reaction with UV
light. The so-called cationic UV curing is extensively used
nowadays for the preparation of diverse materials ranging
However, the synthesis of telechelic polyesters is not always
easy to accomplish by traditional acid catalysis when dealing
with epoxy derivatives. One way to circumvent this issue is
to make use of selective catalysts such as enzymes. With the
aid of the enzymes, it is possible to synthesize epoxy-
functional polyesters directly from epoxy-functional x-
Additional Supporting Information may be found in the online version of this article.
2015 Wiley Periodicals, Inc.
V
C
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
EXPERIMENTAL
Materials
EFA was extracted from outer birch bark (Betula pendula)
FIGURE 1 9,10-Epoxy-18-hydroxyoctadecanoic acid (EFA)
supplied by Holmen AB (c/o Holmen Energi, SE-89810) as
monomer.
1
4–16
previously reported in the literature by us and others.
4
Dimethyl adipate (DA), hexanol, hexadecanedioic acid (99%),
ethylene glycol, and immobilized CalB (Novozyme 435) were
purchased from Sigma-Aldrich. Trimethylpropyl oxetane
hydroxy fatty acids without altering the epoxide or to incor-
porate in the same polymer backbone functional groups that
by other synthetic routes would be more difficult to have
simultaneously due to self-reaction of those as a thiol and an
(
TMPO) was used as received from Perstorp AB. The
7
“photoinitiator” Uvacure 1600 was supplied by CYTEC.
alkene.
Synthesis and Characterization of Telechelic Oligomers
For the synthesis of telechelic polymers containing an epox-
ide in the backbone, that is, using the EFA monomer as pre-
cursor, the synthetic route used is similar to the one
The control of the mechanical properties of the formed ther-
mosets will rely on the possibility of selectively reacting
with the functional groups present in the polymer. Having
functional groups in the main chain and at the ends, that is,
in the telechelic positions, increase dramatically the options
of network formation by either reaction of those groups with
themselves separately or at the same time. These options
will increase when the end groups and the ones present in
the main chain are also capable to react with each other,
that is, to copolymerize with each other. This is the case of
epoxides and oxetanes, for example, as shown by Sasaki
1
4,17
previously reported in the literature.
The enzyme-
catalyzed polymerization is done in bulk and in a “one-pot
synthesis” using the reactants in the adequate stoichiometric
ratios depending on the desired degree of polymerization
(
DP). The quantities in all polymerizations correspond to an
aimed total yield of 250 mg. The formed oligomers were
characterized with Fourier transform infrared spectroscopy
1
8
9,10
(FTIR), H NMR, MALDI-TOF MS, and size exclusion chroma-
et al. and Crivello et al.,
when a small amount of epoxy
tography (SEC). Detailed assignments can be found in Sup-
porting Information.
monomer is present in the cationic polymerization of an
oxetane-derived monomer, the polymerization rates highly
increase as a consequence of the fast generation of an oxo-
nium ion from the epoxy ring (eq 1 and Fig. 2) that gives a
Synthesis of a,x-Oxetane End-Capped Epoxy-Functional
Polyester DP3 (EODP3) and DP6 (EODP6)
“
2
kick-start” to the cationic polymerization of the oxetane (eq
and Fig. 2).
EFA, DA, and TMPO were put into a 25-mL round-bottomed
flask in a 3:1:2 ratio for DP3 and 6:1:2 for DP6 and heated
up to melting temperature at around 85 8C. Once the reac-
tants had reached the molten state, the immobilized enzyme
was added (10% of the total amount of monomers), and the
reaction was allowed to proceed at open atmosphere for 2 h.
After that time, the flask was connected to a vacuum line,
and the reaction was run overnight (Scheme 1).
1
1
Other factors such as the differences in reaction enthalpies
also affect the polymerization rates. This and other studies
mainly dealing with primary epoxides and oxetane mono-
mers clearly show that epoxy and oxirane monomers readily
copolymerize to give polyether polymers with varying ether
1
2,13
segments due to the specific copolymerization rates.
The reaction was stopped by dissolving the product in chlo-
roform and filtering off the immobilized enzyme. The prod-
uct was characterized and used without further purification
The incorporation of epoxy- and oxetane-functional groups
in the same polyfunctional resin would be an extension of
this concept of copolymerization and give a new platform in
designing thermoset network structures.
(yield of EODP3: 94%; yield of EODP6: 65%).
Synthesis of a,x-Hexane End-Capped Epoxy-Functional
Polyester DP3 (EDP3)
EFA, DA, and hexanol were put into a round-bottomed flask
in a 3:1:2 ratio and heated up to 85 8C. After the reaction
This study aims to describe the concept of incorporation of
two different functional groups capable of copolymerizing
into telechelic oligomers as a tool of tailoring the final ther-
moset properties.
FIGURE 2 Initiation of the cationic polymerization of oxetanes
SCHEME 1 Reaction scheme for the formation of oligomers
by reaction with an oxonium ion (AX, photoinitiator).
EODP3 and EODP6.
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
tate was filtered out and dried under vacuum to yield a
white powder.
1
Yield: 69%. H NMR (400 MHz, CDCl , d): 1.23–1.26 (20H,
3
m, ACH A), 1.60 (4H, m, ACH ACH ACOA, J 5 7.2), 2.28
2
2
2
(
4H, t, ACH ACOA, J 5 7.6), 3.64 (6H, s, CH OA).
2
3
SCHEME 2 Reaction scheme for the formation of oligomer
Network Formation
EDP3.
The cationically crosslinked networks were prepared in all
cases from a mixture containing 25 mg of each oligomer, 4%
(
w/w) of the photoinitiator (Uvacure 1600), and 100 mg of
mixture had reached that temperature, the enzyme was
added to the flask (10% of the total amount of monomers),
and the reaction was allowed to proceed at open atmosphere
for 2 h and afterward under vacuum overnight (Scheme 2).
chloroform. For the photo real-time infrared (photo-RTIR)
measurements, the networks were formed in situ on the sur-
face of the attenuated total reflection (ATR) crystal of the
FTIR after evaporation of the solvent and continuous expo-
2
sure to the UV light with an intensity of 10 mW/cm for a
The reaction was stopped by filtering out the enzyme, and
the product was characterized and used without further
purification (yield: 65%).
period of 1 h. The rest of the films were prepared by solvent
casting of the mixture on a glass substrate with the aid of a
pipette. After evaporation of the solvent, the mixtures were
covered with a quartz microscope slide to obtain homogene-
ously cured films. Free-standing films were obtained after 15
min of exposure to UV light with an intensity of 20 mW/cm
(
the results obtained after the real-time FTIR (RT-FTIR)
studies.
Synthesis of a,x-Oxetane End-Capped Aliphatic Polyester
DP3 (ODP3)
2
To prepare the aliphatic homologous of the polyester with
DP3, end capped with oxetane, hexadecanedioic acid was
chosen as monomer. Because of the nonsolubility of this
monomer in any suitable solvent for the enzymatic polymer-
ization, the equivalent methyl ester was prepared prior to
polymer formation (Scheme 3).
2
dose: 18 J/cm ). The time of exposure was selected from
Analytical Methods
1
H NMR
The spectra were recorded at 400 MHz on a Bruker AM 400
Dimethyl hexadecanedioate, ethylene glycol, TMPO, and
molecular sieves of 4 Å (25 wt % of the total amount of
monomers) were put into a round-bottomed flask in a 3:2:2
ratio and heated up to 85 8C. After the reaction mixture had
reached that temperature, the enzyme was added to the flask
using deuterated chloroform (CDCl ) as solvent. The solvent
3
signal was used as reference.
MALDI-TOF MS
MALDI-TOF MS spectrum acquisitions were conducted on a
Bruker UltraFlex MALDI-TOF MS with SCOUT-MTP Ion
(
10% of the total amount of monomers), and the reaction
was allowed to proceed at open atmosphere for 2 h and
afterward under vacuum overnight. The reaction was
stopped by filtering out the enzyme. The product was puri-
fied by column chromatography using silica as stationary
phase and a gradient of eluents, starting with a mixture of
ethyl acetate:heptane 30:70 and increasing to 100% ethyl
acetate (yield: 11%).
Source (Bruker Daltonics, Bremen) equipped with a N
2
laser
(
337 nm), a gridless ion source, and reflector design. All
spectra were acquired using a reflector-positive method with
an acceleration voltage of 25 kV and a reflector voltage of
26.3 kV. The detector mass range was set to 200–2500 m/z.
A THF solution of 2,5-dihydroxybenzoic acid was used as
matrix. The obtained spectra were analyzed with FlexAnaly-
sis Bruker Daltonics, Bremen, version 2.
Synthesis of Dimethyl Hexadecanedioate
Size Exclusion Chromatography
SEC using DMF (0.2 mL/min) with 0.01 M LiBr as the mobile
phase was performed at 50 8C using a TOSOH EcoSEC HLC-
Hexadecanedioic acid (3.5 mmol, 1 equiv) was suspended in
methanol (20 mL). Afterward, HCl (2.2 equiv) was added
dropwise at room temperature. After 4 h, the formed precipi-
8320GPC system equipped with an EcoSEC RI detector and
three columns (PSS PFG 5 lm; microguard, 100 Å, and 300
Å; MW resolving range: 100–300,000 g/mol) from PSS
GmbH. A calibration method was created using narrow linear
polystyrene standards. Corrections for the flow rate fluctua-
tions were made using toluene as an internal standard.
Fourier Transform Infrared Spectroscopy
FTIR analysis was carried out using a Perkin-Elmer Spec-
trum 2000 FTIR instrument (Norwalk, CT) equipped with a
single reflection (ATR) accessory unit (Golden Gate) from
Graseby Specac LTD (Kent, England) and a TGS detector
SCHEME 3 Reaction scheme for the formation of oligomer
ODP3.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
using the Golden Gate setup. Each spectrum collected was
2
1
based on 16 scans averaged at 4.0 cm resolution range of
2
1
6
00–4000 cm . Data were recorded and processed using
the software Spectrum from Perkin-Elmer UV light sources.
The light source used for the crosslinking of the films was a
Black Ray B-100AP (100 W, k 5 365 nm) Hg UV lamp with
2
an irradiance of ꢀ21 mW/cm as determined with an UVI-
CURE Plus High Energy UV integrating radiometer (EIT,
USA), measuring UVA at 320 nm ꢁ k ꢁ 390 nm. The light
source used for the photo-RTIR measurements was a Hama-
matsu L5662 equipped with a standard medium pressure
200-W L6722-01 Hg–Xe lamp and provided with optical
fibers. The UV intensity was measured using a Hamamatsu
UV light power meter (model C6080-03) calibrated for the
main emission line centered at 365 nm. The intensity of the
2
lamp was 7 mW/cm .
Real-Time FTIR Analysis
The RTIR analysis was made using a Perkin-Elmer Spectrum
2
000 FTIR instrument (Norwalk, CT) equipped with a single
FIGURE 3 a,x-Functionalized telechelic polyesters.
reflection (ATR) accessory unit (Golden Gate) from Graseby
Specac LTD (Kent). RT-FTIR continuously recorded the chem-
ical changes over the range 4000–600 cm . Spectroscopic
data were collected at an optimized scanning rate of 1 scan
per 1.67 s with a spectral resolution of 4.0 cm using Time
BaseVR software from Perkin-Elmer.
2
1
Characterization of Telechelic Oligomers
To confirm the formation of the telechelic polyesters,
1
H
2
1
NMR spectroscopy, MALDI-TOF, and SEC analysis were per-
formed. As exemplified in Figure 4 (see also Supporting
Information Figs. SI3–SI6), the success in the synthesis of
the EODP3 polymer is confirmed by the appearance of new
signals: p, q, and r, at around 2.3, 4, and 4.2 ppm, respec-
tively, as a result of the formation of two new ester bonds
between the EFA monomer and the DA, and the EFA mono-
mer and the TMPO. At the same time, the peak correspond-
ing to the methyl group of the DA, peak k, disappears as
methanol is formed and eliminated as a byproduct of the
esterification. It is also noteworthy that the epoxide is kept
intact during the synthesis, as confirmed by peak i.
Dynamic Mechanical Thermal Analysis
The study of the physical properties of the polymerized resins
was performed on a Q800 dynamic mechanical thermal ana-
lyzer (TA Instruments), equipped with a film fixture for tensile
testing. The measurements were done between 260 8C and
150 8C, with a heating rate of 5 8C/min. The tests were per-
formed in controlled strain mode with a frequency of 1 Hz, an
oscillating amplitude of 5 lm, and a force track of 125%.
Although the a,x-difunctional oligomers are obtained in
majority, a fraction of monosubstituted oligomers is also
RESULTS AND DISCUSSION
obtained as confirmed by MALDI-TOF MS (Supporting Infor-
Synthesis of Telechelic Oligomers
1
mation Figs. SI7–SI10). The DP was calculated from the
H
To understand how the polymer network is formed when
the two cyclic ethers capable to undergo cationic polymeriza-
tion in similar manners are present in the system, telechelic
oligomers with and without an epoxide built in the backbone
and/or end capped with an oxetane were synthesized. In Fig-
ure 3, all the oligomers used in this work are depicted.
NMR spectra as the relationship between the peaks of the
repeating unit and the end capper. The results are summar-
ized in Table 1.
The values obtained for the degrees of polymerization are
coherent with the molecular weight distributions obtained with
MALDI-TOF analysis (Supporting Information Figs. SI7–SI10).
The utilization of immobilized enzymes as catalysts not only
afford very mild conditions to be used but also make the
product purification very simple as the catalyst is easily
removed by filtering it off. The final structure can also be
easily adjusted by appropriate choice of the in-going stoichi-
ometry of the reactants. The variation in degree of di-end-
functionalization is within the normal range for this type of
systems. This study did not include efforts in optimizing the
synthesis toward 100% di-end-functionalization as the
obtained percentages were considered to be high enough for
the purpose of the study.
Network Formation and Final Properties
In general terms, when two or more functional groups are
present in the polymer chain, the reaction between them-
selves or of those with a crosslinker yields thermoset forma-
tion. In the case of the networks formed through cationic
polymerization, a strong nucleophile is first needed to form
the oxonium ion that will subsequently initiate the propaga-
tion reactions through ring opening of the cyclic ethers
(Fig. 2). The most common initiators used in cationic ring
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
1
FIGURE 4 H NMR spectra of the EODP3 oligomer before (lower spectrum) and after reaction (upper spectrum).
1
8
opening of cyclic ethers consist of diaryliodonium or triar-
Polymerization Performance
1
9
ylsulfonium salts. For the oligomers with both epoxide and
oxetane groups in the main chain, three possibilities for ther-
moset formation were contemplated: cationic homopolymeri-
zation of the epoxy group with itself, of the oxetane with
itself, and copolymerization by reaction of the epoxide with
oxetane or vice versa. Although the epoxide is slightly more
ring strained than the oxetane and will ring open faster
when attacked by a nucleophile, the oxetane is slightly more
basic which will make it more reactive toward cationic poly-
As aforementioned, for the better understanding of the pro-
gress of the reaction, four different oligomers were synthe-
sized prior crosslinking. For the polymers containing both
2
0
merization; therefore, the three contemplated possibilities
were considered equally probable. Most previous studies on
the copolymerization of oxetanes and epoxides have further-
more been performed on primary epoxides being more reac-
1
1
tive than the secondary epoxides used in the current study.
However, because of the different nature of the system con-
sidered in this study, this assumption cannot be directly
applied. RT-FTIR studies and dynamic mechanical analysis
(
DMA) were performed to further understand how the net-
works were formed.
TABLE 1 Degrees of Polymerization and Percentage of End
Functionalization for the Different Di-End-Functionalized
1
Oligomers Calculated from Their H NMR Spectrum
Oligomer
DPNMR
% Di-End-Functionalized
a
e
EODP3
EODP6
EDP3
2.2
71
b
f
5.1
89
c
g
2.7
99
FIGURE 5 Schematic representation of the formation of the
polymer thermoset by reaction of the epoxides, and FTIR spec-
tra of the EDP3 oligomer taken every 10 min with focus in the
d
ODP3
2.3
NA
a
b
Calculated from the relationship between peaks: i 3 2 5 r 3 DP; i 3
5 q 3 DP; i 3 2 5 (q 2 i) 3 DP; h 3 2 5 i 3 DP.
Calculated from the relationship between peaks: (m, m ) and q; (m,
m ) and p; (q 1 p) and i. NA, not applicable due to peak overlapping.
21
c
d
region between 1120 and 1000 cm (PI, photoinitiator). [Color
figure can be viewed in the online issue, which is available at
wileyonlinelibrary.com.]
2
e
0
f
0
g
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
epoxides and oxetanes, two possibilities were contemplated:
homopolymerization and copolymerization.
To be able to discriminate the peaks corresponding to the
homopolymerization, homologous oligomers only containing
epoxides and only containing oxetanes were synthesized.
Homopolymerization of secondary epoxides giving a substi-
tuted polyethylene glycol structure give rise to a new
21
absorption band at 1073 cm (Fig. 5), whereas homopoly-
merization of the oxetane on the other hand will yield a new
2
1
band at 1062 cm (Fig. 6) corresponding to a polypropyl-
ene glycol structure. The polymerization of resins EODP3
and EODP6, however, gave rise to a much more complex
absorption pattern in the ether region, which strongly indi-
cates that a copolymerization is preferred (Fig. 8).
As can be seen in Figures 5 and 6, the cationic polymeriza-
tion of the oligomer only containing epoxy groups (EDP3) is
faster than the one of the oligomer only containing oxetanes
(
ODP3). Although these results are qualitative and no quanti-
tative kinetic studies have been performed, it is possible to
see that after 30 min all the epoxides have reacted while the
oxetane is still reacting as seen by a continued increase in
2
1
the peak at 1062 cm
.
It is also noteworthy that in the first 10 min of the polymer-
2
1
ization of the ODP3 polymer, the peak at 1117 cm appears
2
1
at the same time as the one at 1107 cm disappears (Fig.
2
1
6
, squared). After that time, the peak at 1107 cm
varies
2
1
less, whereas the rest of the peaks at around 1062 cm
seemingly increase to a larger extent. This can be explained
as in the first 10 min, there is a continuous generation of ali-
FIGURE 6 Schematic representation of the formation of the poly-
mer thermoset by reaction of the telechelic oxetanes, and FTIR
spectra of the ODP3 oligomer taken every 10 min with focus in
21
phatic ethers (peak at 1117 cm ) originated from the ring
21
opening of the cyclic oxetanes (peak at 1107 cm ).
21
the region between 1120 and 1000 cm . Squared: spectra taken
every 10 s during the first 10 min for the region between 1120
As can be seen in Figure 7, after this time the generation of
aliphatic ethers continues at a slower rate indicating that
other factors such as molecular mobility and dilution effects
starts to have increased influence.
21
and 1080 cm (PI, photoinitiator). [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]
FIGURE 7 Absorbance versus time for the two different peaks of ODP3 enhanced at the right side. [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 8 Schematic representation of the formation of the polymer thermoset by copolymerization between epoxides and oxe-
tanes, and FTIR spectra of the EODP3 and EODP6 oligomers taken every 10 min with focus in the region between 1120 and
21
1
000 cm (PI, photoinitiator). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
2
1
Because of the overlapping peaks, it is difficult to assess the
suppressed as the peak at 1073 cm is no longer the one
governing this region of the spectra.
2
1
different peaks in the area between 1300 and 1000 cm for
the EODP3 oligomer. However, it is possible to observe the
success of the cationic polymerization as new peaks appear
already after some minutes of irradiation. These new absorp-
tion suggest that there is a complex mixture of different
reaction sequences present in the formation of the network
structure and that the homopolymerization of the epoxide is
It is also worth highlighting that when the variation in
absorbance as a function of the reaction time of the peak at
is compared with the different oligomers con-
taining oxetanes (Fig. 9, left), there is a clear increase in the
reaction rate for the oligomers containing both oxetane and
2
1
1062 cm
21
FIGURE 9 Absorbance versus time for the 1062 and 1073 cm peaks. [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 2015, 00, 000–000
7

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Summary of the Mechanical Properties of the
Oligomers
0
00
0
a
0
b
Oligomer
T
g
(8C)
tan d (E /E )
E (GPa)
E rp (GPa)
EDP3
5
0.51
1.03
0.79
2
2
2
0.3
1
EODP3
EODP6
15
12
0.2
a
0
E : storage modulus at 260 8C.
rp: storage modulus in the rubbery plateau.
b
0
E
opposite trend. The most interesting feature is that the tan d
curve narrows and increases in height with increased cross-
link density. The normal trend is that the transition is broad-
ened and that the peak height decreases. This implies that
the network formation routes are different for the different
resins, that is, the polymerization mechanism differs from
homopolymerization of epoxides (EDP3) to mixed epoxy
homopolymerization/copolymerization (EODP6), and mainly
epoxy–oxetane copolymerization (EODP3). A more narrow
transition indicates a more homogeneous network structure
with high regularity in the network. This type of change is
often seen when changing the reaction mechanism going
from a chainwise to a stepwise mechanism as in the case of
comparing free radical acrylate polymerization (chainwise)
FIGURE 10 Tan d versus temperature for cured resins EDP3,
EODP3, and EODP6. [Color figure can be viewed in the online
issue, which is available at wileyonlinelibrary.com.]
epoxide. The increase is much higher for the oligomer with
DP3 than for the one with DP6 (Fig. 9). This can be
explained as the epoxy:oxetane ratio for the polymer with
DP3 is higher than for the polymer with DP6 (3:2 and 3:1,
respectively) and would also confirm the hypothesis of the
increase in reaction rate due to copolymerization between
those polymers. The same trend is observed for the peak at
2
1
to thiol-ene systems (stepwise).
2
1
1
073 cm (Fig. 9, right), although the difference in the reac-
The crosslink density is also directly related to the rubbery
modulus, and it is clearly seen that EODP3 is more crosslinked
than the others (Table 2 and Supporting Information Fig. SI16).
The difference between EODP6 and EDP3 is not significant
enough to be validated by the rubbery modulus level because
of the experimental uncertainties, that is, the level of uneven-
ness of the films was too large. More studies are, however,
needed to further elaborate on details of the copolymerization
mechanism, and this is out of scope of the current study.
tion rate between EODP3 and EDP3 is not as big as for the
previous case. It should, however, be noted that the very
broad and overlapping peaks only allow for a qualitative
evaluation of the rate differences. The FTIR spectra before
and after polymerization of the oligomers can be found in
the Supporting Information.
Mechanical Properties of the Networks
To support the conclusion that the telechelic end groups
(
oxetanes) do connect to the main chain in the telechelic,
CONCLUSIONS
DMA was performed. For all cases except of the ODP3 poly-
mer, free-standing films were obtained allowing mechanical
analysis to be performed. The initial resin structure as well
as the different ways the networks are formed, that is, homo-
polymerization or copolymerization, will result in different
mechanical performances of the films. The main factors
affecting the mechanical performance of a thermoset are the
rigidity of the backbone, the polarity, the crosslink density,
and the presence of loose chain ends. The homogeneity of
In this work, the synthesis of a series of telechelic polyesters
and their consecutive incursion into a polymer network is
presented. The intrinsic epoxy functionality of the precursor
monomer (EFA), first extracted from outer Birch bark, is pre-
served with the aid of enzymatic catalysis.
By understanding how the different groups present in the
polymer chain react with each other, it will be able to predict
the final properties of the network.
the network is furthermore reflected in the width of the T
g
transition. All the present resins are aliphatic polyesters
crosslinked with ether linkages, and hence, there is a similar-
ity in polarity and rigidity of the backbone. When analyzing
the tan d, a clear trend is seen as an increased peak temper-
ature value going from EDP3 to EODP6 to EODP3 is
observed (see Fig. 10 and Table 2).
Recently, our group reported the lipase-catalyzed synthesis of
methacrylate end-capped EFA oligoesters and their subsequent
1
4
polymerization for network formation. The groups were
reacted either separately or in parallel using different polymer-
ization techniques, that is, radical polymerization or cationic
polymerization. The different ways these groups were cova-
lently linked to each other had direct repercussion on the prop-
erties of the formed networks, which could be seen as a
variation on the mechanical properties of the network.
This is normal for polymers with a similar backbone struc-
ture and associated with an increased crosslink density. An
increase of unreacted chain ends will, however, lead to an
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
In this work, we extend this concept by increasing the possi-
bilities for network formation. In this case, an oxetane was
introduced as end capper of the enzyme-catalyzed esterifica-
tion. As previously reported in the literature, oxetanes and
epoxides are capable to copolymerize with each other. How-
ever, to our knowledge, this concept had never been applied
for polymers containing these compounds in the same back-
bone. As proven in this work, these two compounds not only
do copolymerize but also this reaction is preferred, seen as
an increase in the reaction rates in comparison with the
homopolymerizations.
6 IUPAC. Compendium of Chemical Terminology, 2nd ed. In
IUPAC (the “Gold Book”). Compiled by A. D. McNaught and A.
Wilkinson. Blackwell Scientific Publications, Oxford, UK 1997.
7
M. Takwa, K. Hult, M. Martinelle, Macromolecules (Washing-
ton, DC) 2008, 41, 5230–5236.
H. Sasaki, J. M. Rudzinski, T. Kakuchi, J. Polym. Sci. Part A:
Polym. Chem. 1995, 33, 1807–1816.
8
9
2
U. Bulut, J. V. Crivello, J. Polym. Sci. Part A: Polym. Chem.
005, 43, 3205–3220.
1
0 J. V. Crivello, B. Falk, M. R. Zonca, Jr., J. Polym. Sci. Part A:
Polym. Chem. 2004, 42, 1630–1646.
1
1 J. V. Crivello, J. Polym. Sci. Part A: Polym. Chem. 2014, 52,
2
934–2946.
1
2 S.-C. Yang, J.-H. Jin, S.-Y. Kwak, B.-S. Bae, J. Appl. Polym.
Sci. 2012, 126, E380–E386.
ACKNOWLEDGEMENTS
1
3 J. V. Crivello (Rensselaer Polytechnic Institute). U.S. Patent
0060041032-A1, 2006.
2
This work received funding from the People Programme
1
4 S. Torron, S. Semlitsch, M. Martinelle, M. Johansson, Mac-
(
Marie Curie Actions) of the European Union’s Seventh
romol. Chem. Phys. 2014, 215, 2198–2206.
5 A. R u€ diger, P. Hendil-Forssell, C. Hedfors, M. Martinelle, S.
Trey, M. Johansson, J. Renew. Mater. 2013, 1, 124–140.
6 T. Iversen, H. Nilsson, A. Olsson (Innventia AB, Sweden).
WO Patent 2010093320-A1, 2010.
7 S. Semlitsch, S. Torron, M. Johansson, M. Martinelle,
unpublished work.
Framework Programme FP7/2007 2013/ under REA grants
agreement number 289253. The authors thank Stefan Sem-
litsch and Mats Martinelle for fruitful discussions about
enzymatic catalysis.
1
1
1
REFERENCES AND NOTES
1
1
8 J. V. Crivello, J. L. Lee, J. Polym. Sci. Part A: Polym. Chem.
989, 27, 3951–3968.
1
2
R. Ekman, Holzforschung 1983, 37, 205–211.
A. Gandini, C. Pascoal Neto, A. J. D. Silvestre, Prog. Polym.
1
4
9 S. R. Akhtar, J. V. Crivello, J. L. Lee, J. Org. Chem. 1990, 55,
222–4225.
Sci. 2006, 31, 878–892.
3
P. A. Krasutsky, Nat. Prod. Rep. 2006, 23, 919–942.
A. Olsson, M. Lindstr o€ m, T. Iversen, Biomacromolecules
007, 8, 757–760.
2
0 S. Penczek, P. Kubisa, K. Matyjaszewski, Adv. Polym. Sci.
4
2
1985, 68, 2–32; 52–64; 66–77.
2
1 A. F. Senyurt, H. Wei, B. Phillips, M. Cole, S. Nazarenko, C.
5
M. Sangermano, N. Razza, J. V. Crivello, Macromol. Mater.
E. Hoyle, S. G. Piland, T. E. Gould, Macromolecules 2006, 39,
Eng. 2014, 299, 775–793.
6315–6317.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2015, 00, 000–000
9