DSpace at VNU: Synthesis of Allyl End-Block Functionalized Poly(epsilon-Caprolactone)s and Their Facile Post-Functionalization via Thiol-Ene Reaction

ARTICLE

WWW.POLYMERCHEMISTRY.ORG

JOURNAL OF
POLYMER SCIENCE

Synthesis of Allyl End-Block Functionalized Poly(e-Caprolactone)s
and Their Facile Post-Functionalization via Thiol–Ene Reaction
Thuy Thu Truong,1 Son Hong Thai,1 Ha Tran Nguyen,1,2 Vinh-Dat Vuong,2 Le-Thu T. Nguyen1
1

Faculty of Materials Technology, Ho Chi Minh city University of Technology, Vietnam National University (VNU–HCM),
268 Ly Thuong Kiet, District 10, Ho Chi Minh City, Vietnam
2
Materials Technology Key Laboratory (Mtlab), Ho Chi Minh City University of Technology, Vietnam National University
(VNU–HCM), 268 Ly Thuong Kiet, District 10, Ho Chi Minh City, Vietnam
Correspondence to: L.-T.T. Nguyen (E-mail: )
Received 13 July 2016; accepted 9 November 2016; published online in Wiley Online Library
DOI: 10.1002/pola.28454

ABSTRACT: A simple and facile strategy for the functionalization of
commercial poly(e-caprolactone) diols (PCLs) with pendant functionalities at the polymer chain termini is described. Well-defined
allyl-functionalized PCLs with varying numbers of allyl end-block
side-groups were synthesized by cationic ring-opening polymerization of allyl glycidyl ether using PCL diols as macroinitiators. Further functionalization of the allyl-functionalized PCLs was realized
via the UV-initiated radical addition of a furan-functionalized thiol

to the pendant allyl functional groups, showing the suitability for
post-modification of the PCL materials. Changes in polymer structure as a result of varying the number of pendant functional units at
C 2017 Wiley Periodicals,
the PCL chain termini were demonstrated. V

Inc. J. Polym. Sci., Part A: Polym. Chem. 2017, 55, 928–939

INTRODUCTION Poly(e-caprolactone)s (PCLs) are an essential
class of synthetic polymers widely used as biomaterials for
medical devices, drug delivery and tissue engineering, as
shape-memory polymers, adhesives, and soft segments for
polyurethane synthesis as well as in packaging and microelectronics.1 Such extensive application of PCLs is possible as
a result of their tailorable mechanical properties, miscibility
with a large range of other polymers and biodegradability.2
PCLs are hydrophobic, semi-crystalline and have low melting
points (typically 60 8C). The typical synthesis route to PCLs
is either free radical polymerization of 2-methylene-1-3dioxepane3 or ionic ring-opening polymerization of ecaprolactone (CL)4 using a variety of enzymatic,5–7 organic,8–12 and metal catalyst systems.13–15 They are usually
obtained by metal-alcoholate initiators, producing hydroxyl
end-functionalized polymers.16–20 The use of diols affords
a,x–hydroxytelechelic PCLs.4,16,17 Alternatively, the use of
metal carboxylate initiators for CL polymerization has also
been well studied.20–22

along PCL chains consists of the synthesis and polymerization of functional CL substituted at the a- or c-position by a
large series of functional groups such as halogen, acid, acyloxy, alcohol, amine, allyl, or an ATRP initiating group.2,23–32
Usually a multi-step process involving protection and deprotection of functional groups before and after polymerization was necessary, since the functional groups may coordinate with metal catalysts.27 Alternatively, CL has been ringopening copolymerized with other functionalized monomers.33–36 A different post-polymerization route was the
grafting of functional groups or macromolecular chains onto
PCLs through the a-position of the carbonyl, although chain
degradations were hardly avoidable.37–40

Functional groups have often been added to render PCLs
more hydrophilic, adhesive, or biocompatible to diversify
their applications, or for subsequent coupling or crosslinking
reactions toward different macromolecular structures. The
methodology for introduction of pendant functional groups

KEYWORDS: activated monomer; allyl; functionalization; poly(e-

caprolactone); thiol–ene

A strategy of choice to obtain PCLs functionalized at one end
is the polymerization of CL initiated by functional alcohols,
such as 2-hydroxyethyl methacrylate,41 propargyl alcohol,42,43 2-mercaptoethanol,44 or a-(2,4-dinitrophenylthio)ethanol.45 Hedfors et al.44 has also employed functional
terminators in an enzyme-catalyzed polymerization process
of CL to end–cap PCLs with the thiol functionality. On the
other hand, end-functionalization of pre-formed PCLs represents a convenient alternative. The hydroxyl end groups of
PCLs can be chemically modified by coupling reactions of
hydroxyl groups with carboxylic acid, anhydride, acyl halide,

Additional Supporting Information may be found in the online version of this article.
C 2017 Wiley Periodicals, Inc.
V

928

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939


JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

isocyanate, or carboimidazole.46–51 As such, functional groups,
such as methacrylate,42,43,52,53 acrylate,46,54,55 isocyanate,49

furan and maleimide,47 amine,56 bromide,57 carboimidazole,50
allyl,51 and thiol45,58 have been incorporated in PCLs as terminal groups. Alternatively, polymers bearing end-block triplebond, benzaldehyde, naphthoyl chloride, benzyl chloroformate,
and iodomethane groups, can be obtained using macroinitiators with functional end groups to initiate either cationic
activated monomer (AM)59–63 or anionic polymerization64,65
of functional cyclic monomers.
Aliphatic PCLs containing double bond substituents are of
particular interest owing to the derivatization of the double
bond into a large variety of functional groups and the exploitation of its reactivity to obtain polymeric networks. The
synthesis of PCLs end-capped with acrylate and methacrylate
groups via chemical modification of hydroxyl end groups, followed by further derivatization of these functional groups or
subsequent free-radical crosslinking of polymer chains, have
been reported.42,43,46,52–55,66–68 Recently, Boire et al.32 copolymerized CL with a-allyl carboxylate e-caprolactone to synthesize PCLs bearing pendant allyl groups, which were
photo-crosslinked via free-radical polymerization of allyl
groups. Taha and co-workers51 obtained multi-branched
poly(ester urethane)s bearing PCL segments and allyl end
functions via two steps, including the synthesis of isocyanate
terminated prepolymers by reacting a hydroxyl-terminated
PCL and glycerol with an excess of a diisocyanate and the
subsequent reaction of the isocyanate terminated prepolymers with allyl amine. These multiallyl-functionalized PCLbased polymers were then free-radical photo-copolymerized
with 2-hydroxyethylmethacrylate.
During the last decades, the polymer functionalization strategy
via “click” and coupling reactions has already proved to be an
efficient and straightforward pathway.69–72 Among a wide
range of metal-free “click” reactions,73 the photoinduced and
thermally induced thiol–ene radical addition reactions have
been widely recognized as a fast and robust tool for polymer
synthesis and modification.74–77 Although in many cases,
radical-mediated thiol–ene reactions do not fulfill the “click”
criteria and can only be considered as coupling reactions, they
have still proved to be a promising approach to readily attach

functional groups, particularly biomolecules like proteins and
peptides commonly bearing free thiols.71,78,79
In this respect, there are few examples where PCLs containing
either thiol side groups or (meth)acrylate end groups have been
post-modified via thiol–ene coupling reactions.38,42,54,55,80 Nottelet et al.38 synthesized a PCL grafted with pendant thiol
groups and crosslinked the obtained thiol-functionalized polymer via the thermally initiated thiol–ene coupling reaction with
a tri-allyl compound. The group of Mather and co-workers54,55
conducted the photo-crosslinking of terminally di-acrylated
PCLs by the thiol–ene reaction with a tetrathiol compound. The
radical-initiated thiol–methacrylate addition reaction has been
applied on methacrylated PCLs to attach an amino acid or
cholesterol moiety for imparting desired bioactivity.42,80

WWW.MATERIALSVIEWS.COM

ARTICLE

In this study, we present a straightforward approach to
obtain PCLs decorated with a variable number of functional
side groups situated at the chain termini. This strategy
involves consecutive additions of protonated allyl glycidyl
ether ring-molecules to hydroxyl-ended PCL macromolecules,
proceeding according to the AM mechanism.59,61,62,81 While
PCL-based polymers consisting of either pendant allyl functions randomly distributed along the backbone or multibranches terminated by allyl groups have been created,32,51
this is the first study of linear PCL materials fitted with multiple pendant non–activated double bonds at the same chain
ends, which can be facilely prepared via a one-stage reaction.
Finally, the use of allyl side groups as modification sites for
further functionalization by a photoinduced radical thiol–ene
coupling reaction with furfuryl mercaptan has been demonstrated. The use of the “food-grade” furfuryl mercaptan compound as a model thiol has the advantages of non-toxicity
and that the 1H NMR furan signals are clearly separated

from all others and thus can be integrated without interference. Although not within the scope of this study, we believe
that the possibility to localize as well as vary the number of
functional groups may result in final materials with diversified thermal and physical properties, and that the resulting
furan-functionalized PCLs with variable content of furan
pendants are expected to find broad applications, such as in
the fabrication of drug carrier materials,82,83 self-healing and
thermoreversible polymeric materials exploiting the diene
role of furan in Diels–Alder reactions,84 or biocompatible
materials with enhanced fire-resistance imparted by furan
entities.85–89
EXPERIMENTAL

Materials
Poly(e-caprolactone) diol (Mn 5 2000 g mol21, manufactured
with diethylene glycol as initiator, Mn 1H NMR 5 2112 g mol21,
Acros) and poly(e-caprolactone) diol (CAPA 2403D, manufactured with butanediol as initiator, Mn 5 4000 g mol21, Mn 1H
21
, Perstorp) were azeotropically dried with
NMR 5 4090 g mol
toluene before use. Allyl glycidyl ether (AGE, 991%, TCI Chemicals–Japan) was dried over molecular sieves and distilled
under vacuum before use. Dichloromethane (99.9%, Fisher
Chemicals) was dried over CaH2 and distilled. Tetrafluoroboric
acid diethyl ether complex (HBF4ÁEt2O, 901%, Sigma-Aldrich),
furfuryl mercaptan (FM, 971%, Sigma-Aldrich), 2,2-dimethoxy-2-phenylacetophenone (DMPA, 99%, Sigma-Aldrich),
calcium oxide (96%, Fisher Chemicals), triphenylphosphine
(TPP, 99%, Sigma-Aldrich), thioglycerol (TG, 98%, Evans
Chemetics/Bruno Bock), 1,10 -(methylenedi-4,1-phenylene)
bismaleimide (95%, Sigma-Aldrich), ethyl acetate (99%, Fisher
Chemicals), methanol (99%, Fisher Chemicals), and n-heptane
(99%, Fisher Chemicals) were used as received.

Measurements
H NMR spectra were recorded in deuterated chloroform (CDCl3)
with TMS as an internal reference, on a Bruker Avance 300 at
300 MHz. Transmission Fourier transform infrared (FT-IR) spectra, collected as the average of 128 scans with a resolution of
1

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

929


ARTICLE

JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

TABLE 1 Characterization of Allyl-Functionalized PCLs
Starting PCL-Diol

Structurea

Mn
(g mol21)b

HOA(CL)18AOH

2112

HOA(CL)35AOH

4090

Allyl-Functionalized PCL

Mn
(g mol21)e

intensity Ratio of
Observed-to-Theoretical
Terminal AGE (a0 ) Signal

Structurea

Entry

HOA(AGE)yA(CL)18A
(AGE)zAOH

1

5.8

5.0

2683

0.96

2

3.7

3.6

2523

1.20

3

3.1

3.0

2455

1.12

4

5.0

4.0

4547

1.09

5

10.0

8.8

5094

1.23

6

20.0

16.1

5928

1.05

7

30.0

17.7

6110

1.09

HOA(AGE)yA(CL)35A
(AGE)zAOH

a

Additional unit corresponding to initiator used in the synthesis of the
commercial PCL-diol present in PCL is not shown.
b
As estimated from 1H NMR spectra of the starting PCL-diol (Figs. S1
and S2, Supporting Information).
c
Calculated on the basis of the Mn (NMR) of the starting PCL-diol and
the feeding amount of AGE (for complete AGE conversion).

d
Calculated on the basis of 1H NMR analysis of the purified product
[Fig. 1(a) and Figs. S4–S9, Supporting Information].
e
On the basis of 1H NMR analysis, Mn 5 Mn (starting PCL) 1 Mn (AGE)
(y 1 z).

4 cm21, were recorded from KBr disk on an FT–IR Bruker Tensor
27. Attenuated total reflectance (ATR) FT-IR spectra were collected as the average of 128 scans with a resolution of 4 cm21 on a
FT-IR Tensor 27 spectrometer equipped with a Pike MIRacle ATR
accessory with a diamond/ZnSe element. Gel permeation chromatography (GPC) measurements were performed on a Polymer
PL-GPC 50 gel permeation chromatograph system equipped with
an RI detector, with THF as the eluent at a flow rate of 1.0 mL/
min. Molecular weight and molecular weight distribution (Ð)
were calculated with reference to polyethylene glycol standards.
Thermogravimetric analysis (TGA) measurements were performed under nitrogen flow using a NETZSCH STA 409 PC Instruments with a heating rate of 10 8C/min from ambient

temperature to 800 8C. Differential scanning calorimetry (DSC)
measurements were carried out with a DSC Q20 V24.4 Build 116
calorimeter under nitrogen flow, from 240 to 170 8C at a heating
rate of 10 8C/min and cooling rate of 50 8C/min.

vacuum line. Yields: 78–85% for Entries 1–6, Table 1 and 56%
for Entry 7 Table 1.

Synthesis of Allyl-Functionalized PCLs
A typical reaction procedure for the synthesis of allylfunctionalized PCLs is described: Commercial PCL diol (CAPA
2403D) with Mn 5 4090 g mol21 (3.45 g, 1.69 mmol of –OH
groups) was dissolved in 17.5 mL of dichloromethane in a
round–bottom flask. To this solution, 0.017 mL (0.12 mmol) of
HBF4ÁEt2O was added. Then, a nitrogen flow was passed over
the mixture and the flask was closed with a rubber septum.
0.5 mL (4.22 mmol) of AGE was slowly introduced with a
syringe during 8 h. The reaction mixture was kept at room
temperature for 24 h, and, after that, the acid catalyst was neutralized with solid CaO. After filtration of CaO, the solution was
concentrated and the product was isolated by precipitation
three times in methanol (dichloromethane/methanol 5 1/5, v/
v), washed three times with methanol, filtered and dried on

930

y 1 z, 1H
NMRd

y 1 z,
Theor.c

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

Thiol–Ene Addition Reaction of Allyl-Functionalized
PCLs with Furfuryl Mercaptan
As an example, the reaction of an allyl-functionalized PCL (Entry
4, Table 1, PCL4000–4ene, HOA(AGE)yA(CL)35A(AGE)z–OH)
with y 1 z 5 4, Mn 5 4546.6 g mol21, containing in average 4
allyl groups per chain) with furfuryl mercaptan is described. In
a flask containing a stirring bar and closed with a rubber septum, 2.82 g (2.48 mmol of allyl groups) of PCL4000–4ene, in a
round-bottom flasked dipped in an oil bath, was melted at 55 8C
under stirring. After stopping heating, PCL4000–4ene remained
as a clear liquid and a minimum amount of tetrahydrofuran was
added to maintain the polymer in the liquid state at room temperature. Then, 64 mg (10 mol% with respect to allyl groups) of
DMPA and 0.5 mL (4.96 mmol of thiol groups) of furfuryl mercaptan were added in succession. In the case of using TPP, TPP
was added at the same time as thiol. The reaction mixture was
degassed and was purged with nitrogen through a needle using
vacuum/nitrogen line. After an overnight exposure to UV light
(wavelength of 365 nm, with twelve lamps of 9 W circularly oriented), the product was collected by precipitation three times in
diethyl ether and was further dried under vacuum (1023 torr)
at 60 8C to remove any left unreacted furfuryl mercaptan. Finally, colu Mn chromatography of the product (ethylacetate: n-heptane, 1: 2) was performed to eliminate traces, if any, of DMPA
species (Rf 5 0.6–0.7) and furfuryl mercaptan (Rf 5 0.82) (Fig.
S12, Supporting Information). The purified polymer product
was remained in the silica colu Mn (Rf 5 0 in ethylacetate:
n-heptane, 1: 2) and finally eluted by ethylacetate (Rf 5 0.94,
Fig. S13, Supporting Information).


JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

ARTICLE

SCHEME 1 The reaction pathway to allyl-functionalized PCLs and subsequent functionalization via thiol–ene chemistry.

Crosslinking of Furan-Functionalized PCLs
(Demonstration Test)
A mixture of a furan-functionalized PCL and 1,10 -(methylenedi-4,1-phenylene)bismaleimide in a 1:1 furan to maleimide
equivalent ratio in tetrahydrofuran was injected in molds
and cured at 30 8C for 48 h. After opening the molds, the
samples were washed by a Soxhlet extraction in acetone at
60 8C to eliminate un-crosslinked materials and solvent.

RESULTS AND DISCUSSION

The straightforward and up-scalable synthetic pathway to
allyl-functionalized PCLs and subsequent functionalization
via thiol–ene reaction is described in Scheme 1. The incorporation of allyl groups to PCL chain ends was carried out by
the addition of an allyl glycidyl ether (AGE) unit (proceeded
by earlier activation of AGE by the protic acid HBF4ÁEt2O) in
the presence of the polymeric diol. Two commercial PCLdiols with average Mn values given by the suppliers of 2000
and 4000 g mol21 were used as macroinitiators for cationic
polymerization of AGE. The accurate Mn values of the PCLdiol macroinitiators were determined by 1H NMR (Figs. S1
and S2, Supporting Information). In all cases, AGE was slowly introduced to the system containing macroinitiator and
HBF4ÁEt2O as catalyst. An addition of the protonated AGE to
the terminal hydroxyl groups gives rise to the modified PCLdiol containing multiple repeating AGE units, located at both
polymer chain ends. The characteristics of the PCL-diol macroinitiators and the obtained modified PCLs are presented in
Table 1.

WWW.MATERIALSVIEWS.COM

The allyl-functionalized PCLs were characterized by 1H NMR
analysis. Taking into consideration on one hand the Mn values of the PCL-diol macroinitiators previously determined,
and on the other hand the molar ratio between AGE and CL
units determined from 1H NMR spectra, the average number
of allyl groups and Mn values of the obtained products were
calculated.
Figure 1(a) shows a representative 1H NMR spectrum of the
obtained AGE-functionalized PCL (entry 1, Table 1). The
spectra of all other samples can be found in the Supporting
Information (Figs. S4–S9). In the spectra, all expected signals
corresponding to CL and AGE monomer units are present.
By comparing the intensity of signals corresponding to the
alkene group (signal f at 5.27–5.04 ppm) with that of the
separate signal assigned to repeating CL units (signal m at
2.55–2.05 ppm) and taking into account the degree of polymerization of the starting PCL-diols, the total number of
attached AGE units per polymer chain could be calculated as
shown in Table 1. An additional 1H NMR analysis of the reaction mixture before purification via precipitation showed
that the number ratio of AGE and CL units was in good
agreement with the feed amount ratio of AGE and PCL-diol
(Fig. S3, Supporting Information). It is worth noting that side
reactions such as the polymerization of AGE proceeded via
the activated chain end mechanism or a reaction between an
activated AGE and moisture, resulting in homopoly(AGE)
byproduct, might not be completely excluded.62 Unreacted
AGE and poly(AGE) byproduct are soluble in methanol,90–93
and thereby could be removed by multiple precipitation
in methanol. Accordingly, for a relatively low feed AGE to

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

931


ARTICLE

WWW.POLYMERCHEMISTRY.ORG

JOURNAL OF
POLYMER SCIENCE

FIGURE 1 1H NMR spectra in CDCl3 of an allyl-functionalized PCL (Entry 1, Table 1) before (a) and after (b) thiol–ene reaction with
furfuryl mercaptan (Entry 4, Table 2). The signal denoted as a0 corresponds to the methine protons of HOACH(R)A groups of terminal HO–AGE units. The signal denoted as i0 corresponds to the methylene protons of A(AGE)yACH2(R)A groups of CL units next to
AGE units. [Color figure can be viewed at wileyonlinelibrary.com]

PCL-diol molar ratio (3.1–20.0), 1H NMR analysis of the purified product indicated that the number of allyl groups incorporated into a PCL chain was slightly lower than the
theoretical value (Entries 1–6, Table 1). In these cases, the
average composition of the products corresponds to 3–16.1
AGE units per PCL chain. Nevertheless, the oligomerization
of AGE initiated by PCL-diol was less efficient with further
increasing the feed amount ratio of AGE and PCL-diol. As

932

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

demonstrated by Entries 6 and 7 in Table 1, the average
number of AGE end units only slightly increased from 16.1
to 17.7 despite a large increase in the feed AGE to PCL-diol

molar ratio from 20 to 30. The decreased efficiency of the
AGE oligomerization via the AM mechanism is attributed to
the increase in the ratio of the dropwise-added AGE and the
hydroxyl polymer end groups over reaction time. Consequently, more competing side reactions could occur. This


JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

explanation can also be applied to the observation that the
AGE oligomerization was generally more efficient when initiated with the PCL-diol of lower Mn. We have been interested
to introduce multiple allyl groups to PCL chain termini. However, although not within the scope of this study, we anticipate that it is still possible to incorporate one or two AGE
units by the present approach, since the concentration ratio
between the AGE and the hydroxyl end groups in the reaction medium could be more easily maintained low.
In addition, analysis of 1H NMR spectra could reveal whether
AGE units were attached at both ends of PCL chains. In the
spectra of AGE-functionalized PCLs, the methine resonance
of HOACH(R)A groups of terminal HO–AGE units (signal
denoted as a0 ) may be identified. As an example shown in
Figure 1(a), although signal a0 partially overlaps with others
(i.e., signals i and d), the total integral value of peaks i, d
and a0 in the range of 4.15–3.88 ppm was obtained separately. Thus, by taking into consideration the integrals of peak i
(calculated from the intensity of signal m and the intensity
ratio of peak i and m previously obtained from the spectrum
of the starting PCL-diol in Figure S1, Supporting Information)
and peak d (equal to the integral of peak f), an estimation of
the integral value of signal a0 was possible. Assuming that
each PCL chain has two AGE end groups, from the degree of

polymerization of the starting PCL-diol and the intensity of
the 1H NMR signal corresponding to repeating CL units (signal m), the theoretical integral value of the terminal
HOACH(R) A group signal (a0 ) can be estimated and is compared with that obtained from the spectra (Table 1). The calculation of intensity ratio of observed-to-theoretical terminal
a0 signal can be expressed as follows:


Int: a0 obs:
Int: ½4:15–3:88 ppmŠ Int: i
Int: f
2
2
52DP
diol
PCL2
Int: m
Int: m Int: m
Int: a0 theor:

(1)
where Int. is the abbreviation for integral, Int: i=Int: m
equals to 0.877, as estimated from the 1H NMR spectrum of
the starting PCL-diol in Supporting Information Figure S1,
and -DPPCL2 diol is the degree of polymerization of the starting
PCL-diol.
In principle, the observed-to-theoretical intensity ratio of
signal a0 is 1 for PCL chains bearing two AGE end groups
and is 0.5 for those with one AGE terminal group. For all
samples, an observed-to-theoretical intensity ratio of signal
a0 equal to approximately 1 confirms the structure of PCL
chains with both AGE end groups.

Transmission FT–IR spectra of the allyl-functionalized PCLs
additionally confirmed the incorporation of allyl groups by
the appearance of the alkene @CAH stretching vibration at
3070 cm21 (Fig. S10, Supporting Information). The addition
of AGE units to PCL chains was also indicated by GPC
analysis (Fig. S11, Supporting Information) of the allylfunctionalized PCL revealing a shift to a higher molecular
weight (Mn 5 5970 g mol21; Ð 5 1.26), as compared with

WWW.MATERIALSVIEWS.COM

ARTICLE

that of the corresponding starting PCL-diol (Mn 5 4750 g
mol21; Ð 5 1.21).
PCLs with pendant allyl groups were used for the UVinitiated thiol–ene reaction with furfuryl mercaptan as a
model thiol (Scheme 1). Therefore, 1.1–5 equivalents of furfuryl mercaptan with respect to one allyl group were used in
the coupling reactions with allyl-functionalized PCLs. It
should be noted that the unreacted thiol after coupling reactions and initiator species were removed mainly by precipitation and evaporation. Moreover, any trace of these
remaining molecules was eliminated completely by colu Mn
chromatography purification, as evidenced by TLC analysis
(Fig. S12, Supporting Information). Representative 1H NMR
spectra of an allyl-functionalized PCL before and after thiol–
ene reaction with furfuryl mercaptan (Entry 4, Table 2) are
shown in Figure 1; spectra for other thiol–ene reactions are
shown in Supporting Information.
The occurrence of thiol–ene reactions on the allylfunctionalized PCLs was evidenced by a decrease in the 1H
NMR signal of the alkene protons (signal e at 5.92–5.71 ppm
and f at 5.27–5.04 ppm, Fig. 1) and the concurrent appearance of new signals corresponding to the coupled thiol molecule, that is signals of furan group at 7.29, 6.23 and 6.11
ppm [signals s, r, q in Fig. 1(b), respectively], methylene
group of the formed thio-ether linkage at 2.50 ppm [signals

f0 , Fig. 1(b)] and the methylene group next to the thio-ether
group at 1.75 ppm [signal e0 in Fig. 1(b)].
The efficiency of thiol–ene reactions was confirmed by integration of the signals in the 1H NMR spectra of the products
of allyl-functionalized PCLs coupled with furfuryl mercaptan.
By comparing the signal intensities before and after coupling
reactions, using a separate signal corresponding to the polymer backbone as the reference, both the conversion of allyl
groups and the number of attached thiol molecules per allyl
group could be determined. For all samples, good agreement
between conversion of allyl groups and the amount of
attached furfuryl mercaptan was observed [Fig. 1(b) and
Figs. S14–S30, Supporting Information]. The conditions and
corresponding degree of functionalization by thiol–ene reactions are summarized in Table 2.
The results presented in Table 2 show that the applied conditions of the thiol–ene reaction considerably impact the
conversion of allyl groups, particularly the content of photoinitiator (DMPA). For instance, a higher functionalization
degree was obtained with increasing photoinitiator content
between 1 and 25 mol% with respect to allyl groups (comparing Entries 1–4 and Entries 7, 9–12, Table 2). Thus, a sufficient amount of photoinitiator is necessary to achieve a
high functionalization degree, as also previously noted.71,94,95
Apparently, the amount of photoinitiator used for the optimal conditions compensates for any inhibition of the photoreaction by oxygen quenching. We observed that, with 2
equivalents of thiol, the thiol–ene reaction conversion was
optimal at a photoinitiator content of 25 mol% (Entries 4, 7,
Table 2). Further increasing photoinitiator content led to no

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

933


ARTICLE

JOURNAL OF

POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

TABLE 2 Conditions and Functionalization Degree of Thiol–Ene Reactions on Allyl-Functionalized PCLs

934

Thiola

[–SH]/[allyl]
in Feed

Mol% of
DMPAb

Functionalization
Degree (%)b

Entry

Structure, y 1 z

1

HOA(AGE)yA(CL)18A(AGE)zAOH, 5

FM

1.1

1

0

54

2

HOA(AGE)y–(CL)18–(AGE)z–OH, 5

FM

1.1

15

0

62

3

HOA(AGE)y–(CL)18–(AGE)z–OH, 5

FM

2

15

0

71

4

HOA(AGE)y–(CL)18–(AGE)z–OH, 5

FM

2

25

0

76

5

HOA(AGE)y–(CL)18–(AGE)z–OH, 5

FM

5

15

0

62

6

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

1.1

25

0

65

7

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

25

0

74

8

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

4

25

0

74

9

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

1

0

57

10

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

5

0

61

11

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

10

0

70

12

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

15

0

70

13

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

35

0

74

14

HOA(AGE)y–(CL)35–(AGE)z–OH, 8.8

FM

2

25

0

70

15

HOA(AGE)y–(CL)18–(AGE)z–OH, 5

FM

2

25

10

$100

16

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

FM

2

25

10

$100

17

HOA(AGE)y–(CL)35–(AGE)z–OH, 8.8

FM

2

25

10

98

18

HOA(AGE)y–(CL)35–(AGE)z–OH, 4

TG

2

25

10

$100

[TPP]/[-SH]

a

FM is furfuryl mercaptan and TG is thioglycerol.
Mol% with respect to allyl groups.

b
Defined as the percentage of allyl groups coupled with thiol, which
was estimated based on 1H NMR spectra in Figure 1(b) and Figures
S14–S30, Supporting Information.

further enhancement of functionalization extent (comparing
Entry 7 with Entry 13, Table 2). Indeed, a too high initiator
concentration could give rise to considerable termination
reactions as a result of radical re-combinations71 On the other hand, as side reactions such as disulfide bond formation
or thiyl radical combination are inevitable in thiol addition
reactions,71,72,94,95 the use of an excess amount of thiol was
necessary to increase functionalization efficiency. With closeto-equimolar ratios of thiol to allyl, coupling efficiencies of
only 54–65% were obtained (Entries 1, 2 and 6, Table 2).
Two equivalents of the thiol with respect to allyl groups, in
combination with a high initiator concentration, were sufficient to obtain maximum conversions of approximately 75%.
The use of more than 2 equivalents of the thiol did not further improve the reaction conversion (see Entries 5 and 8,
Table 2). It is also worth noted that when the photoinitiator
content was not sufficiently high, a highly increased amount

of thiol could result in a decrease in coupling efficiency as it
decreases the overall alkene and photoinitiator concentration
in the system (comparing Entries 3 and 5, Table 2). The fact
that a full functionalization was not achieved due to termination reactions and mainly disulfide formation has also been
reported earlier for thiol addition reactions involving polymeric species.71,72,94 In an attempt to suppress disulfide formation, triphenylphosphine (TPP) as a tertiary phosphine
was added in the reaction mixture as a disulfide reducing
agent. Such approach has been previously reported to successfully enhance thiol-ene functionalization efficiencies.71

With the use of a ten-fold excess of the reducing agent with
respect to thiol, complete reduction of all disulfide bonds
formed in the reaction was achieved, resulting in nearly
100% conversion of thiol–ene coupling reactions (Entries
15–17, Table 2). In addition, applying the same optimal reaction conditions, a full conversion of the thiol–ene coupling of
the allyl-functionalized PCL with thioglycerol as an alternative functional thiol was also obtained (Entry 18, Table 2
and Fig. S30, Supporting Information). Generally, with a coupling efficiency of 100% with the addition of a tertiary phosphine and around 75% without the use of a disulfide
reducing agent, this thiol–ene coupling approach appears
very promising to readily attach on-demand functional
groups.

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

In addition, the coupling of the allyl group with furfuryl mercaptan was also confirmed by transmission FT–IR analysis. Figure 2
demonstrates a comparison of the IR spectra of an allylfunctionalized PCL (Entry 1, Table 1) and corresponding products after coupling it with furfuryl mercaptan. With increasing
the functionalization degree, the intensity of the absorption signal corresponding to double bonds at 3079 cm21 decreases.96
Concurrently, the intensities of the signals corresponding to
furan groups at 3146, 3117 va 1011 cm21 increase.97,98
Changes in polymer structure as a result of the introduction
of pendant functional units at the PCL chain termini were
first demonstrated by DSC analysis. As shown from both DSC

JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

ARTICLE

FIGURE 2 Comparison of the transmission FT–IR spectra (normalized to the band area of the C@O stretching vibrations at 1814–
1664 cm21) in the range of 3220–3000 and 1300–650 cm21 of an allyl-functionalized PCL (a, Entry 1, Table 1) and furanfunctionalized PCL products after thiol–ene reactions with furfuryl mercaptan corresponding to functionalization efficiencies of 54
and 71% (b–Entry 1 and c–Entry 3, Table 2, respectively). [Color figure can be viewed at wileyonlinelibrary.com]

heating curves in Figure 3, allyl- and furan-functionalized PCLs
show no glass transition in the range of 230 to 170 8C, but
exhibit multiple melting peaks. The multiple melting behavior
is attributed to a “melting2recrystallization2melting” process,99 and thus, the total melting enthalpy did not reflect the
degree of crystallinity of the polymer samples. This suggests

that for these samples, perfect crystals were not formed upon
quenching, and they recrystallize in the heating process. The
low endotherm is attributed to a superposition of early melting
of secondary crystals with almost simultaneous exothermic
recrystallization, and the final endotherm contains contributions from the melting of primary crystals and the melting of

FIGURE 3 DSC thermograms of PCL (a, Mn 5 4090 g mol21) and corresponding allyl-functionalized PCL (b, Entry 4, Table 1) and
furan-functionalized PCLs (c and d, corresponding to Entries 9 and 7, Table 2 with thiol–ene coupling efficiencies of 57 and 74%,
respectively). The curves are vertically shifted for clarity. [Color figure can be viewed at wileyonlinelibrary.com]

WWW.MATERIALSVIEWS.COM

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

935


ARTICLE

JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

TABLE 3 Thermal Data Obtained Form DSC Analysis of PCL, Allyl- and Furan-Functionalized PCLsa
1st Heating

Cooling

2nd Heating

Low Endotherm
Tm (8C)b

High Endotherm
Tm (8C)b

Tc (8C)c

DH (J/g)d

Low Endotherm

Tm (8C)b

PCL (Mn 5 4090 g mol21)

–

57.3

13.8

66.9

–

50.8

Allyl-functionalized PCL
(Entry 4, Table 1)

36.0

55.0

10.9

60.6

42.9

48.5

Furan-functionalized PCL
(Entry 9, Table 2)

34.1

52.2

4.2

55.4

38.7

47.6

Furan-functionalized PCL
(Entry 7, Table 2)

34.1

48.5

2.8

55.8

37.0

44.1

a
b
c

Data obtained from thermograms in Figure 3.
Melting temperature.
Crystallization temperature.

recrystallized regions formed during heating. The melting temperatures and crystallization enthalpy during cooling of PCL
and corresponding allyl- and furan-functionalized PCLs are
summarized in Table 3. The results show that the incorporation of AGE units to PCL resulted in decreases in melting temperature and crystallization (during cooling) enthalpy. The
same trend was observed when the allyl-functionalized PCL

High Endotherm
Tm (8C)b

d
Crystallization enthalpy, determined via integration of the crystallization signal in the cooling scan.

was further coupled with furfuryl mercaptan. Hence, increasing the number of furan methylene groups (for example comparing the samples with furan functionalization degrees of 57
and 75%) led to a slight decrease in melting temperature. The
insertion of pendant furan methylene moieties to the polymer
chain ends resulted in molecular defects, giving rise to destabilization of the ordered packing.
Comparative TGA thermograms of a PCL sample before and
after functionalization with allyl and furan groups additionally
indicated changes in polymer structure (Fig. S31, Supporting
Information). The functionalization of PCL with furan groups
resulted in a slight increase in the char yield. The char-yielding
behavior is typical of furanic polymers.100 This is ascribed to

the formation of char on the upper part of the material, which
prevents the formation of volatile compounds from the inner
part.

FIGURE 4 Crosslinking reaction between furan-functionalized
PCLs and 1,10 -(methylenedi-4,1-phenylene)bismaleimide, the
image of the thermoset T1 obtained from PCL4000–8.8furan (Entry
17, Table 2) without shape-memory (a), and the images of the
thermoset T2 obtained from PCL4000–4furan (Entry 16, Table 2)
with shape-memory behavior (b: permanent shape; c: programmed shape; shape recovery is shown in Fig. S34, Supporting
Information). [Color figure can be viewed at wileyonlinelibrary.
com]

936

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

The effect of varying the number of pendant functional units at
the PCL chain termini was further demonstrated by comparing
the properties of two thermosets crosslinked from two PCL
products bearing in average 8.8 and 4 furan groups per chain
(samples T1 and T2 corresponding to the furan-functionalized
PCLs in Entry 17 and 16, Table 2, respectively). Crosslinking
was carried out via Diels–Alder reaction between furan groups
and 1,10 -(methylenedi-4,1-phenylene)bismaleimide, with maleimide to furan equimolar ratio (Fig. 4), giving samples with high
crosslinking contents (97–98%) and insoluble in common solvents at room temperature. The occurrence of Diels–Alder reaction was confirmed by the ATR FT-IR result, showing the
disappearance of the furan signal at 1011 cm21 and the maleimide signals at 830 and 687 cm21, as well as the increase in
intensity of the signal at 863 cm21 attributed to the Diels–Alder
linkage (Fig. S32, Supporting Information).101 It has been
reported that in both physically and thermally crosslinked PCL

materials, the crystallized PCL phase with melting and crystallization transitions can act as thermally triggered switching
domains, resulting in shape-memory properties. For sample T1,
the higher crosslinking density as a result of the higher number
of pendant furan groups hindered the crystallization of PCL


JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

chains. Thus, no PCL melting endotherm was found in the DSC
heating curve (Fig. S33a, Supporting Information), and the
materials showed no shape-memory behavior. On the other
hand, sample T2 showed a PCL melting transition at 44 8C (Fig.
S33b, Supporting Information), as well as shape-memory behavior. Temporary shapes (for example, spiral and stretched strip)
of T2 could be programmed by twisting or stretching the samples at temperatures above 50 8C, followed by cooling down to
room temperature to fix the temporary shapes (Fig. 4 and Fig.
S34, Supporting Information). The original permanent shapes
were recovered by heating the samples at the above deformation temperature (Fig. S34, Supporting Information).
Besides, the DSC heating curves of the thermosets showed a
broad endotherm in the range of 80–170 8C, assigned to the
breakage of the Diels–Alder bonds.97 Because the Diels–Alder
bonds reform at low temperatures (20–60 8C),97 these thermosets were recyclable (demonstrated in Fig. S35, Supporting Information).
CONCLUSIONS

Activated monomer (AM) oligomerization of AGE using commercially available PCL-diols as macroinitiators in combination with thiol–ene coupling reactions provides a convenient
synthetic route for the synthesis of PCLs functionalized with
multiple functional side groups at the chain termini. A disadvantage of previously reported approaches to end-cap PCLs
generally with functionalities and more specifically with furyl

groups via coupling reactions of PCL hydroxyl end groups is
the impossibility of varying the number of the terminal functional groups. Apparently, the strategy used in this study
providing PCLs with adjustable, multiple furyl end-block
groups as reactive polymer precursors allows for a broader
ability to tune the final material structure and properties. In
general, this synthetic platform is envisaged to be promising
for introduction of several various functional groups such as
biomolecules or natural compounds containing free thiols to
PCLs for on-demand biomaterial applications for example.

ARTICLE

2 E. A. Rainbolt, K. E. Washington, M. C. Biewer, M. C. Stefan,
Polym. Chem. 2015, 6, 2369–2381.
3 A. Seema, R. Liqun, Macromolecules 2009, 42, 1574–1579.
4 M. Labet, W. Thielemans, Chem. Soc. Rev. 2009, 38, 3484–3504.
5 R. T. MacDonald, S. K. Pulapura, Y. Y. Svirkin, R. A. Gross, D.
L. Kaplan, J. Akkara, G. Swift, S. Wolk, Macromolecules 1995,
28, 73–78.
6 E. Stavila, G. O. R. Alberda van Ekenstein, A. J. J. Woortman,
K. Loos, Biomacromolecules 2014, 15, 234–241.
7 L. A. Henderson, Y. Y. Svirkin, R. A. Gross, D. L. Kaplan, G.
Swift, Macromolecules 1996, 29, 7759–7766.
8 G. O. Jones, Y. A. Chang, H. W. Horn, A. K. Acharya, J. E.
Rice, J. L. Hedrick, R. M. Waymouth, J. Phys. Chem. B 2015,
119, 5728–5737.
9 H. Kim, J. V. Olsson, J. L. Hedrick, R. M. Waymouth, ACS
Macro Lett. 2012, 1, 845–847.
10 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, R. M. Waymouth,
J. L. Hedrick, J. Am. Chem. Soc. 2006, 128, 4556–4557.

11 A. P. Dove, ACS Macro Lett. 2012, 1, 1409–1412.
12 M. Bouyahyi, M. P. F. Pepels, A. Heise, R. Duchateau, Macromolecules 2012, 45, 3356–3366.
13 X. Shen, M. Xue, R. Jiao, Y. Ma, Y. Zhang, Q. Shen, Organometallics 2012, 31, 6222–6230.
14 F. Wang, C. Zhang, Y. Hu, X. Jia, C. Bai, X. Zhang, Polymer
2012, 53, 6027–6032.
15 W. A. Ma, Z. X. Wang, Organometallics 2011, 30, 4364–4373.
16 I. Barakat, P. Dubois, R. Jerome, P. Teyssie, Macromolecules
1991, 24, 6542–6545.
17 S. M. Guillaume, Eur. Polym. J. 2013, 49, 768–779.
18 D. Takeuchi, T. Nakamura, T. Aida, Macromolecules 2000,
33, 725–729.
 , J. P. Dijkstra, C. Birg, M.
19 Z. Zhong, K. M. J. Ankone
Westerhausen, J. Feijen, Polym. Bull. 2001, 46, 51–57.
20 J. Libiszowski, A. Kowalski, A. Duda, S. Penczek, Macromol.
Chem. Phys. 2002, 203, 1694–1701.
, P. J. Dijkstra, J. Feijen, Mac21 W. M. Stevels, M. J. K. Ankone
romolecules 1996, 29, 8296–8303.
22 D. Appavoo, B. Omondi, I. A. Guzei, J. L. van Wyk, O.
Zinyemba, J. Darkwa, Polyhedron 2014, 69, 55–60.
 ro
^ me, R. Je
ro
^ me, P. Lecomte,
23 R. Riva, S. Schmeits, C. Je
Macromolecules 2007, 40, 796–803.
24 S. E. Habnouni, V. Darcos, J. Coudane, Macromol. Rapid
Commun. 2009, 30, 165–169.

ACKNOWLEDGMENTS

This research is funded by the Department of Science and Technology (DOST) – Ho Chi Minh City under grant number VLM (11-KH 2014). We thank Tri M. Phan, Nhi K. D. Nguyen, Phuong T.
Hoang, and Viet Q. Nguyen for their assistance to the experiments and analysis. Coenraad Schaap (Perstorp AB) and Alvin
Kim (Perstorp Chemicals Asia Pte Ltd) are acknowledged for
advice and for supplying polycaprolactone products. Elvira
Schlatter (Bruno Bock) and Matthias Rehfeld (Bruno Bock) are
acknowledged for advice and for kindly providing the thioglycerol product.

REFERENCES AND NOTES
1 M. A. Woodruff, D. W. Hutmacher, Prog. Polym. Sci. 2010,
35, 1217–1256.

WWW.MATERIALSVIEWS.COM

25 S. Gimenez, S. Ponsart, J. Coudane, M. Vert, J. Bioact. Compat. Polym. 2001, 16, 32–46.
ro
^ me, Macromol. Rapid Com26 X. Lou, C. Detrembleur, R. Je
mun. 2003, 24, 161–172.
27 J. Hao, E. A. Rainbolt, K. Washington, M. C. Biewer, M. C.
Stefan, Curr. Org. Chem. 2013, 17, 930–942.
28 F. Ercole, A. E. Rodda, L. Meagher, J. S. Forsythe, A. P.
Dove, Polym. Chem. 2014, 5, 2809–2815.
29 J. Hao, J. Servello, P. Sista, M. C. Biewer, M. C. Stefan, J.
Mater. Chem. 2011, 21, 10623–10628.
30 J. Yan, Y. Zhang, Y. Xiao, Y. Zhang, M. Lang, React. Funct.
Polym. 2010, 70, 400–407.
31 C. Vaida, M. Takwa, M. Martinelle, K. Hult, H. Keul, M.
€ ller, Macromol. Symp. 2008, 272, 28–38.
Mo
32 T. C. Boire, M. K. Gupta, A. L. Zachman, S. H. Lee, D. A.

Balikov, K. Kim, L. M. Bellan, H. J. Sung, Acta Biomater. 2015,
24, 53–63.

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

937


ARTICLE

WWW.POLYMERCHEMISTRY.ORG

33 X. Yang, C. Cui, Z. Tong, C. R. Sabanayagam, X. Jia, Acta
Biomater. 2013, 9, 8232–8244.

64 S. Ponsart, J. Coudane, M. Vert, Biomacromolecules 2000,
1, 275–281.

34 J. J. Wurth, V. P. Shastri, J. Polym. Sci. A: Polym. Chem.
2013, 51, 3375–3382.

65 C. Mangold, F. Wurm, H. Frey, Polym. Chem. 2012, 3, 1714–
1721.

35 B. A. Van Horn, R. K. Iha, K. L. Wooley, Macromolecules
2008, 41, 1618–1626.
36 H. Shen, J. Chen, M. Taha, Polym. J. 2014, 46, 598–608.

66 S. Kelch, S. Steuer, A. M. Schmidt, A. Lendlein, Biomacromolecules 2007, 8, 1018–1027.
67 A. T. Neffe, B. D. Hanh, S. Steuer, A. Lendlein, Adv. Mater.

2009, 21, 3394–3398.
68 I. Bellin, S. Kelch, R. Langer, A. Lendlein, Proc. Natl. Acad.
Sci. 2006, 103, 18043–18047.
69 S. Hvilsted, Polym. Int. 2012, 61, 485–494.

37 B. Nottelet, M. Vert, J. Coudane, Macromol. Rapid Commun.
2008, 29, 743–750.
38 B. Nottelet, G. Tambutet, Y. Bakkour, J. Coudane, Polym.
Chem. 2012, 3, 2956–2963.
39 B. Nottelet, A. El Ghzaoui, J. Coudane, M. Vert, Biomacromolecules 2007, 8, 2594–2601.

70 P. Espeel, F. E. Du Prez, Macromolecules 2015, 48, 2–14.

40 B. Nottelet, J. Coudane, M. Vert, Biomaterials 2006, 27,
4948–4954.

71 L. T. T. Nguyen, J. Devroede, K. Plasschaert, L.
Jonckheere, N. Haucourt, F. E. Du Prez, Polym. Chem. 2013,
4, 1546–1556.

41 V. Jacquier, C. Miola, M. F. Llauro, C. Monnet, T. Hamaide,
Macromol. Chem. Phys. 1996, 197, 1311–1324.

72 L. Billiet, O. Gok, A. P. Dove, A. Sanyal, L. T. T. Nguyen, F.
E. Du Prez, Macromolecules 2011, 44, 7874–7878.

42 S. Doran, E. Murtezi, F. B. Barlas, S. Timur, Y. Yagci, Macromolecules 2014, 47, 3608–3613.

73 L. T. T. Nguyen, M. T. Gokmen, F. E. Du Prez, Polym. Chem.
2013, 4, 5527–5536.

43 I. Javakhishvili, S. Hvilsted, Polym. Chem. 2010, 1, 1650–
1661.
€
€ m, K. Hult, M.
44 C. Hedfors, E. Ostmark,
E. Malmstro

74 M. J. Kade, D. J. Burke, C. J. Hawker, J. Polym. Sci. A:
Polym. Chem. 2010, 48, 743–750.

Martinelle, Macromolecules 2005, 38, 647–649.
45 G. Carrot, J. G. Hilborn, M. Trollsa˚s, J. L. Hedrick, Macromolecules 1999, 32, 5264–5269.
46 K. M. Lee, P. T. Knight, T. Chung, P. T. Mather, Macromolecules. 2008, 41, 4730–4738.
 ro
^ me, M.
47 T. Defize, R. R. Riva, P. J. M. Dubois, C. Je
Alexandre, Macromol. Rapid Commun. 2011, 32, 1264–1269.
48 F. J. Xu, Z. H. Wang, W. T. Yang, Biomaterials 2010, 31,
3139–3147.
49 Q. Wu, T. Yoshino, H. Sakabe, H. Zhang, S. Isobe, Polymer
2003, 44, 3909–3919.
50 L. Najemi, T. Jeanmaire, A. Zerroukhi, M. Raihane, Starch–
Sta€rke 2010, 62, 90–101.
51 H. Shen, G. Quintard, J. Chen, M. Taha, J. Appl. Polym. Sci.
2015, 132, 41295/1–10.

75 C. E. Hoyle, C. N. Bowman, Angew. Chem. Int. Ed. 2010, 49,
1540–1573.
76 A. B. Lowe, Polym. Chem. 2010, 1, 17–36.

77 M. Bednarek, React. Funct. Polym. 2013, 73, 1130–1136.
78 D. E. Borchmann, N. ten Brummelhuis, M. Weck, Macromolecules 2013, 46, 4426–4431.
79 A. Dondoni, A. Marra, Chem. Soc. Rev. 2012, 41, 573–586.
80 I. Javakhishvili, W. H. Binder, S. Tanner, S. Hvilsted, Polym.
Chem. 2010, 1, 506–513.
 ska, R. Szyman
 ski, P. Kubisa, S. Penczek, Die Mak81 K. Brzezin
romol. Chem. Rapid Commun. 1986, 7, 1–4.
82 M. Shi, M. S. Shoichet, J. Biomater. Sci. Polym. Ed. 2008,
19, 1143–1157.
83 M. Shi, J. H. Wosnick, K. Ho, A. Keating, M. S. Shoichet,
Angew. Chem. Int. Ed. 2007, 46, 6126–6131.
84 Y. L. Liu, T. W. Chuo, Polym. Chem. 2013, 4, 2194–2205.

52 A. Lendlein, A. M. Schmidt, M. Schroeter, R. Langer, J.
Polym. Sci. A: Polym. Chem. 2005, 43, 1369–1381.

85 D. J. Martin, G. F. Meijs, G. M. Renwick, P. A. Gunatillake, S.
J. McCarthy, J. Appl. Polym. Sci. 1996, 60, 557–571.

53 A. M. Schmidt, Macromol. Rapid Commun. 2006, 27, 1168–
1172.

86 M. Okada, K. Tachikawa, K. Aoi, J. Polym. Sci. A: Polym.
Chem. 1997, 35, 2729–2737.

54 R. M. Baker, J. H. Henderson, P. T. Mather, J. Mater. Chem.
B 2013, 1, 4916–4920.

87 Y. L. Liu, C. I. Chou, J. Polym. Sci. A: Polym. Chem. 2005,

43, 5267–5282.

55 E. D. Rodriguez, X. Luo, P. T. Mather, ACS Appl. Mater.
Interfaces 2011, 3, 152–161.

88 R. Crossley, P. Schubel, A. Stevenson, J. Reinf. Plast. Compos. 2014, 33, 58–68.
89 A. Gandini, T. M. Lacerda, A. J. F. Carvalho, E. Trovatti,
Chem. Rev. 2016, 116, 1637–1669.
90 A. O’Connor, J. N. Marsat, A. Mitrugno, T. Flahive, N.
Moran, D. Brayden, M. Devocelle, Molecules 2014, 19,
17559–17577.
 Kon
a
k, K. Ulbrich, J. Appl. Polym. Sci. 2005,
, C.
91 M. Hruby
95, 201–211.
€ ller, Macromolecules 2007, 40,
92 M. Erberich, H. Keul, M. Mo
3070–3079.
93 Y. Koyama, T. Ito, H. Matsumoto, A. Tanioka, T. Okuda, N.
Yamaura, H. Aoyagi, T. Niidome, J. Biomater. Sci. Polym. Ed.
2003, 14, 515–531.

56 M. Schappacher, A. Soum, S. M. Guillaume, Biomacromolecules 2006, 7, 1373–1379.
57 M. Schappacher, N. Fur, S. M. Guillaume, Macromolecules
2007, 40, 8887–8896.
58 I. Javakhishvili, S. Hvilsted, Biomacromolecules 2009, 10,
74–81.
59 M. Bednarek, P. Kubisa, J. Polym. Sci. A: Polym. Chem.

2005, 43, 3788–3796.
60 M. Basko, M. Bednarek, L. T. T. Nguyen, P. Kubisa, F. Du
Prez, Eur. Polym. J. 2013, 49, 3573–3581.
61 T. Endo, Y. Shibasaki, F. Sanda, J. Polym. Sci. A: Polym.
Chem. 2002, 40, 2190–2198.
62 P. Kubisa, S. Penczek, Prog. Polym. Sci. 1999, 24, 1409–1437.
63 M. S. Kim, K. S. Seo, G. Khang, H. B. Lee, Macromol. Rapid
Commun. 2005, 26, 643–648.

938

JOURNAL OF
POLYMER SCIENCE

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

94 P. Derboven, D. R. D’hooge, M. M. Stamenovic, P. Espeel,
G. B. Marin, F. E. Du Prez, M. F. Reyniers, Macromolecules.
2013, 46, 1732–1742.


JOURNAL OF
POLYMER SCIENCE

WWW.POLYMERCHEMISTRY.ORG

95 S. P. S. Koo, M. M. Stamenovic, R. A. Prasath, A. J. Inglis, F.
E. D. Prez, C. Barner-Kowollik, W. V. Camp, T. Junkers, J.
Polym. Sci. A: Polym. Chem. 2010, 48, 1699–1713.
96 J. Coates, In Encyclopedia of Analytical Chemistry; R. A.

Meyers, Ed.; John Wiley & Sons, Ltd: Chichester, 2006;
pp 10815–10837.
97 G. Rivero, L. T. T. Nguyen, X. K. D. Hillewaere, F. E. Du Prez,
Macromolecules 2014, 47, 2010–2018.

WWW.MATERIALSVIEWS.COM

ARTICLE

98 A. Mellouki, M. Herman, J. Demaison, B. Lemoine, L.
Margule`s, J. Mol. Spectrosc. 1999, 198, 348–357.
99 B. B. Sauer, W. G. Kampert, E. Neal Blanchard, S. A.
Threefoot, B. S. Hsiao, Polymer 2000, 41, 1099–1108.
100 S. Boufi, M. N. Belgacem, J. Quillerou, A. Gandini, Macromolecules 1993, 26, 6706–6717.
101 C. Gaina, O. Ursache, V. Gaina, Polym. Plast. Technol. Eng.
2011, 50, 712–718.

JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2017, 55, 928–939

939