Molecular weight distribution and endgroup functionality of poly(2-ethyl-2-oxazoline) prepolymers

Article abstract of DOI:10.1016/j.polymer.2014.11.005

Control over molecular weight distribution and endgroup functionality is important for prepolymers that are to be further incorporated into block and graft copolymers. Monofunctional, telechelic and heterobifunctional poly(2-ethyl-2-oxazoline) (PEtOx) oli

Full text of DOI:10.1016/j.polymer.2014.11.005

Polymer 56 (2015) 147e156
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Molecular weight distribution and endgroup functionality of
poly(2-ethyl-2-oxazoline) prepolymers
*
O. Celebi, S.R. Barnes, G.S. Narang, D. Kellogg, S.J. Mecham, J.S. Riffle
Macromolecules and Interfaces Institute and the Department of Chemistry, Virginia Tech, Blacksburg, VA 24061, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 July 2014
Received in revised form
Control over molecular weight distribution and endgroup functionality is important for prepolymers that
are to be further incorporated into block and graft copolymers. Monofunctional, telechelic and hetero-
bifunctional poly(2-ethyl-2-oxazoline) (PEtOx) oligomers were prepared by cationic ring-opening
polymerization with different classes of initiators including methyl triflate as the control, activated
benzyl and xylyl halides, and non-activated alkyl iodides. Endgroup functionalities and molecular weight
distributions were compared by SEC, 1H NMR and titrations. PEtOx oligomers initiated with methyl
triflate had polydispersities (PDI's) near 1.1 while those initiated with activated benzyl or xylyl halides
had PDI's of 1.30e1.45. The difference was attributed to slower initiation and to covalent species in the
29 October 2014
Accepted 1 November 2014
Available online 7 November 2014
Keywords:
Polyoxazoline
Controlled molecular weight
Telechelic
1
activated halide-initiated reactions that were present in addition to the more active cations. H NMR
showed an approximately 1:1 endgroup functionality in these cases and titrations of the amine
endgroups were in good agreement with the NMR endgroup data. Due to the control of functionality and
relatively good control over molecular weights and dispersities, it was concluded that these polymers
should be suitable as prepolymers for membrane, hydrogel and drug delivery applications. Non-activated
alkyl iodides initiated these reactions quite slowly and there was a wider discrepancy between targeted
and obtained molecular weights after isolation of the oligomers.
©
2014 Published by Elsevier Ltd.
1. Introduction
similarities to poly(ethylene oxide) and polypeptides since they are
2 2
comprised of a eNCH CH - backbone and an amide in each repeat
Monofunctional and telechelic poly(2-oxazoline) prepolymers
are candidates for components of block and graft copolymers for
many applications provided that they can be synthesized with
reasonable control over molecular weights and their distributions
and that their endgroups can be functionalized. They are prepared
from five-membered cyclic imino ethers that are polymerized via
cationic ring-opening polymerization [1,2]. Ideally, the endgroup
functionality can be obtained by employing a functional initiator
and/or a termination agent. Monomer compositions can be varied
to obtain hydrophilic (e.g. having methyl or ethyl side chains), hy-
drophobic (butyl, nonyl, phenyl), thermoresponsive (primarily
ethyl and n- and iso-propyl) and fluorinated phenyl-containing
polymers. Moreover, different side chain functionalities can be
introduced post-polymerization to attach a variety of bioactive
compounds or nanoparticles. Poly(2-oxazoline)s have structural
unit [3e6].
Structural features of poly(2-oxazoline)s offer good potential for
biomedical applications [3,4,7e10]. Poly(2-methyl-2-oxazoline)
and PEtOx are of interest due to their hydrophilicity, biocompati-
bility [11e13], stealth behavior [14e16], and prolonged blood cir-
culation times [15,17]. Promising chemical and biological properties
of poly(2-oxazoline)s have revealed a new class of polymer candi-
dates for high-capacity delivery systems for hydrophobic drugs
[18], chemotherapeutic agents [19], enhanced cellular delivery of
proteins [20], and brain delivery [21].
Mono- and difunctional PEtOx and poly(2-methyl-2-oxazoline)
prepolymers can potentially be incorporated into hydrophilic or
amphiphilic linear or graft copolymers or cross-linked networks.
These are of interest for polymeric membranes, as components of
bioimaging agents and as hydrogels for tissue engineering
matrices. A key to those copolymers, gels or nanostructures is the
preparation of polyoxazoline oligomers with controllable molecu-
lar weights and good endgroup functionality. Mono- [22,23] and
difunctional [24] allyl or benzyl halides have been studied as
*
Corresponding author.
E-mail address: judyriffle@aol.com (J.S. Riffle).
http://dx.doi.org/10.1016/j.polymer.2014.11.005
032-3861/© 2014 Published by Elsevier Ltd.
0

148
O. Celebi et al. / Polymer 56 (2015) 147e156
initiators for poly(2-oxazoline)s. These halides are good electro-
philes and many of them are commercially available or easily
synthesized. Kobayashi et al. [25] reported controlled molecular
weight hydroxyfunctional telechelic poly(alkyl oxazoline)s utilizing
difunctional allylic or benzylic initiators followed by termination
with potassium carbonate. Our group previously reported poly(-
alkyl vinyl ether-b-2-ethyl-2-oxazoline) [26] and poly(-
dimethylsiloxane-b-2-ethyl-2-oxazoline) [27] block copolymers
using non-activated iodoethyl and activated benzyl iodide-
containing macroinitiators, respectively.
Numerous studies have been conducted on the synthesis and
applications of heterobifunctional poly(ethylene oxide) oligomers
as well as some post-functionalization reactions [28e34]. However,
there are few reports of controlled molecular weight poly(2-
oxazoline)s with precise endgroup functionality. Reif and Jordan
2. Experimental
2.1. Materials
0
a,a -
Methyl trifluoromethanesulfonate (methyl triflate, ꢀ98%),
dibromo-p-xylene (97%), benzyl bromide (98%), t-butyl piperazine-
1-carboxylate (boc-piperazine, 97%), piperidine (>99.5%), triisobu-
tylsilane (TIBS, 99%), ammonium chloride (>99.5%), dichloro-
methane (>99.8%), dioxane (>99%), calcium hydride (CaH2, 95%),
isopropanol (>99.5%) and standard hydrochloric acid solution
(0.1 M) were purchased from SigmaeAldrich and used as received.
Acetonitrile (EMD chemicals, 99.8%) and 2-ethyl-2-oxazoline (Sig-
maeAldrich, ꢀ99%) were dried over CaH
2
and distilled into a flame-
dried flask under nitrogen. Sodium iodide (SigmaeAldrich, ꢀ99.5%)
ꢁ
was dried under vacuum at 100 C. Diethylether (99.9%), methanol
[
35] synthesized
a
-hydroxyalkyl-
u
-amino functional amphiphilic
(99.9%), chloroform (99.98%) and acetone (99.5%) (Fisher Scientific)
were used as received. 3-Chloropropylmethyldichlorosilane and 3-
chloropropyldimethylchlorosilane (both from Gelest, Inc.), vinyl-
magnesium bromide (Tokyo Chemical Industry, Inc., 14% in THF, ca.
0.1 M), magnesium sulfate (E.M. Science), trifluoroacetic acid (TFA,
Alfa Aesar, 99%) and 0.1 M (standard) KOH (Alfa Aesar) were used as
poly(2-methyl-2-oxazoline)s from t-butyldiphenylsiloxyalkyltri-
flate hydrophobic initiators followed by termination with boc-
piperazine and subsequent deprotection reactions. Kataoka et al.
[36] described an alternative method to prepare a-amino-u-hy-
droxy-poly(2-oxazoline)s by initiating the polymerization with a
phthalimide-functional tosylate initiator and terminating the re-
actions with sodium hydroxide. This was followed by conversion of
the phthalimide group to an amine by treatment with hydrazine.
Hoogenboom and Schubert et al. [37] demonstrated the synthesis
of “clickable” poly(2-oxazoline)s with alkyne and hydroxyl groups
utilizing propargyl- and 3-butynyl tosylate initiators. Alternatively,
Volet and coworkers synthesized mono- and difunctional poly(2-
methyl-2-oxazoline) oligomers utilizing iodomethane and 1,3-
diiodopropane initiators, respectively, and terminated with so-
dium azide. Azido-functional polyoxazoline oligomers were then
transformed to acrylate, epoxide, carboxylic acid and poly(ethylene
oxide) functionalities utilizing heterobifunctional alkynes and click
chemistry via Huisgen 1,3-dipolar cycloaddition reactions [38].
Zalipsky et al. [15] prepared hydrophilic poly(2-oxazoline) oligo-
mers with carboxylic acid and hydroxyl groups using ethyl 3-
bromopropionate/potassium iodide as the initiator system and
utilized potassium hydroxide to hydrolyze the ethyl ester groups
and simultaneously terminate the polymerization. Allyl-functional
poly(2-oxazoline)s were also prepared by initiating the polymeri-
zation with allyl tosylate, then the allyl endgroup was reacted with
trimethoxysilane by hydrosilylation [39]. Hoogenboom and Schu-
bert et al. [40] also described the synthesis of a multifunctional
3 2
purchased. Deuterated solvents (CDCl and D O) were acquired
from Cambridge Isotope Laboratories, Inc. Cellulose acetate dialysis
membranes (1000 MWCO, wet in 0.05% aqueous sodium azide)
were purchased from Spectrum Laboratories, Inc. The SEC mobile
phase, N-methylpyrrolidone (NMP), was purchased from Fisher
Scientific, stirred over phosphorus pentoxide (P
2 5
O ), distilled under
vacuum, and filtered through a 0.2 m PTFE filter before use. After
m
distillation but before filtration, 4.34 g of lithium bromide (LiBr)
was added per liter of NMP to provide a 0.05 M solution. LiBr was
purchased from SigmaeAldrich and dried under vacuum before use
ꢁ
at 100 C overnight.
2.2. Synthesis
2.2.1. Synthesis of a diiodo-p-xylene initiator
Diiodo-p-xylene was prepared from dibromo-p-xylene and so-
dium iodide in acetone. A mixture of dibromo-p-xylene (2.00 g,
7.58 mmol) and sodium iodide (6.81 g, 45.5 mmol) in acetone
(18 mL) was placed in a 100-mL flask and the reaction was con-
ꢁ
ducted at 60 C for 48 h. The solvent was removed under vacuum
and the residue was dissolved in chloroform (250 mL) and extrac-
ted with DI water (500 mL ꢂ 3). The organic layer was dried over
copoly(2-oxazoline) scaffold containing
a
-anthracene and
u
-azide
4
anhydrous MgSO and filtered, and the solvent was removed under
termini as well as pendent alkenes in the side chain. In this study an
anthracene moiety with an iodomethylene substituent was intro-
duced at the initiation step, and 2-(dec-9-enyl)-2-oxazoline was
copolymerized with 2-ethyl-2-oxazoline to introduce alkene
pendent groups, then the polymerization was terminated with
sodium azide. Their approach shows promise for selective attach-
ment to bioactive agents.
vacuum. The residue was recrystallized from 1,4-dioxane, and
washed with cold dioxane and diethylether. The product was vac-
uum dried overnight at 50 C. Yield was ~50%.
ꢁ
2.2.2. Synthesis of 3-iodopropylmethyldivinylsilane and 3-
iodopropyldimethylvinylsilane
3-Iodopropylmethyldivinylsilane was prepared similarly to a
previously reported procedure [29]. 3-Chloropropylmethyl-
dichlorosilane (6.70 g, 34.5 mmol) was added via a syringe to a
flame-dried, 250-mL, round-bottom flask with a magnetic stir bar
While many homo- and copolymers containing poly(2-alkyl-2-
oxazoline) compositions have been reported, there has not been
sufficient attention given to molecular weight and functionality.
This paper describes a detailed investigation of the level of control
over molecular weight and endgroup functionality of oligomeric
PEtOx's with the aim of achieving prepolymers for incorporation
into block or graft copolymers, including stepegrowth reactions.
Monofunctional, telechelic and heterobifunctional PEtOx oligomers
were prepared using different initiators and termination agents,
and size exclusion chromatography (SEC) with multiple detectors,
2
under N . Vinylmagnesium bromide (98 mmol, 98 mL of a 1 M
vinylmagnesium bromide solution in THF) was transferred into the
flask that was in an ice bath, then the reaction was conducted at
ꢁ
60 C for 24 h. The reaction mixture was cooled to room temper-
ature and the salt by-products were allowed to precipitate over 1 h.
The liquid portion was decanted and concentrated using a rotary
evaporator. The crude product was diluted with dichloromethane
(200 mL) and washed 4x with saturated aqueous ammonium
chloride solution (500 mL each) in a 1-L separatory funnel. Anhy-
drous magnesium sulfate was added to the organic layer to remove
residual water, followed by vacuum filtration. Dichloromethane
1
H NMR spectroscopy, and titrations were combined to assess the
influence of reaction variables on molecular weights and endgroup
functionality. Prepolymers with secondary amine and vinylsilane
endgroups have been investigated.

O. Celebi et al. / Polymer 56 (2015) 147e156
149
was removed under vacuum, and the product, 3-
chloropropylmethyldivinylsilane, was vacuum distilled at 75 C,
enclosed with a rubber septum, and the solution was stirred for
~5 min to dissolve the initiator. The temperature was raised to
ꢁ
ꢁ
~0.8 Torr, and collected as a colorless liquid. A 2-fold excess of NaI
60 C. The polymerization was maintained for 4 h and 40 min (~85%
1
(
0.034 mol, 5.1 g) was charged to a round-bottom flask equipped
conversion as measured by H NMR). The reaction mixture was
with a magnetic stir bar. The reaction vessel was sealed with a
septum and flame-dried. Acetone (25 mL) was added via syringe to
the flask. 3-Chloropropylmethyldivinylsilane (0.017 mol, 3.0 g) was
charged to the reaction vessel via syringe. The reaction was stirred
cooled to RT and the polymer was terminated with boc-piperazine
(1.10 g, 5.91 mmol) in 10 mL of acetonitrile and stirred overnight.
The telechelic polymer was recovered by precipitation in dieth-
ꢁ
ylether, collected by filtration and dried under vacuum at 50
C
ꢁ
at 60 C for ~48 h. The acetone was removed by rotary evaporation,
overnight. The polymer solution was dialyzed against 4 L of DI
water for 48 h using a 1000 g/mol MWCO cellulose acetate dialysis
membrane, then freeze-dried. The product yield was ~80%.
Deprotection of the boc-piperazine endgroups followed the
same procedure described previously using a bocPip-PEtOx59-boc-
pip (700 mg, 0.111 mmol, 0.222 mmol boc-piperazine endgroups).
Telechelic PEtOx oligomers prepared from dibromo-p-xylene
were synthesized using the same procedures for the polymeriza-
tion and deprotection reactions described above with adjusted
stoichiometric amounts of reagents.
and the reaction mixture was dissolved in chloroform (150 mL). The
excess NaI and the NaCl salt byproducts were removed by vacuum
filtration. The clear, colorless 3-iodopropylmethyldivinylsilane
ꢁ
1
product was vacuum distilled at 70 C and ~0.8 Torr. H NMR
confirmed the expected structures in each case. Yield was ~40%.
Synthesis of 3-iodopropyldimethylvinylsilane was performed in
a similar procedure utilizing 3-chloropropyldimethylchlorosilane
(
7
10.0 g, 58.4 mmol) and vinylmagnesium bromide (70.2 mmol,
0.2 mL) and a 2-fold excess of NaI was used in the second step.
Yield was ~40%.
2.2.5. Synthesis of heterobifunctional poly(2-ethyl-2-oxazoline)s
2.2.3. Synthesis of monofunctional poly(2-ethyl-2-oxazoline)
with one vinylsilane endgroup and one piperazine endgroup
oligomers with piperidine or piperazine endgroups
An exemplary procedure for preparing a heterobifunctional PEtOx
with one terminalvinyl group and a piperazine group at the otherend
is provided. EtOx monomer (8.06 mL, 7.91 g, 79.8 mmol) was charged
via a syringe to a 100-mL, flame-dried, round-bottom flask enclosed
with a rubber septum. 3-Iodopropyldimethylvinylsilane (0.405 g,
1.59 mmol) and acetonitrile (8 mL) were syringed into the flask, and
the solutionwas stirred for ~5 min at RT. The reaction flask was placed
A procedure for the synthesis of a PEtOx oligomer initiated with
methyl triflate and terminated with an aliphatic amine is provided.
Methyl triflate (0.280 mL, 0.404 g, 2.46 mmol), EtOx monomer
(
12.5 mL, 12.3 g, 124 mmol), and acetonitrile (12.5 mL) were sy-
ringed into a flame-dried, 100-mL, round-bottom flask containing a
magnetic stir bar and enclosed with a rubber septum bound with
steel wire at room temperature. The solution was stirred for ~
ꢁ
in an oil bath and the temperature was raised to 60 C. The cationic
1
5
min. Four different oligomers were prepared with the same
polymerizationwasmonitored using H NMR, and at~74%conversion
initiator and monomer concentrations, and the polymerizations
of monomer (7 h, 20 min), the growing chains were terminated with
anexcess ofboc-piperazine (1.48 g, 7.95 mmol)solution inacetonitrile
(15 mL). The solution was stirred overnight. The polymer was recov-
ered by precipitation in diethylether, then dialyzed and freeze-dried
as described above.
ꢁ
were conducted at 25, 50, 65 and 80 C. The monomer conversion
1
was monitored using H NMR spectroscopy by analyzing samples
prepared from 0.1-mL aliquots from the reaction mixture in
0
.4e0.5 mL of CDCl
excess of piperidine (2.43 mL, 2.10 g, 24.6 mmol) or boc-piperazine
1.83 g, 9.84 mmol) solutions in 20 mL of acetonitrile. The solutions
3
. The polymerization was terminated with an
(
2.2.6. Synthesis of divinyl and piperidine-functional poly(2-ethyl-2-
oxazoline) oligomers
were stirred overnight and the polymers were isolated by adding
each solution dropwise into stirring diethylether (300 mL). Each
Synthesis of a 3-iodopropylmethyldivinylsilane-initiated and
piperidine-terminated PEtOx oligomer followed a similar proce-
dure to that described above, with EtOx (8.38 mL, 8.23 g,
83.0 mmol), 3-iodopropylmethyldivinylsilane (0.365 g, 1.37 mmol),
acetonitrile (8.20 mL), and piperidine (0.677 mL, 0.584 g,
6.85 mmol). The polymerization was terminated at ~67% conver-
sion of monomer. These heterobifunctional initiators yielded ~ 60%
of polymer.
ꢁ
PEtOx oligomer was dried under vacuum at 50 C overnight and
diluted to 200 mL with DI water. The polymer solution was placed
in a 1000 g/mol MWCO cellulose acetate dialysis membrane and
dialyzed against 4 L of DI water for 48 h. The contents of the dialysis
1
membrane were transferred to a 250-mL flask and freeze-dried. H
NMR confirmed the expected chemical structures. The product
yields were ~70%.
Deprotection of the boc-piperazine endgroups was performed
utilizing a similar procedure to that reported previously [11]. A
PEtOx62-bocpip (700 mg, 0.113 mmol) was dissolved in a 4-mL
mixture of 95:2.5:2.5 TFA:water:TIBS (v:v:v) and stirred for
2.3. Characterization
1H NMR spectra were obtained using a JEOL Eclipse Plus 500
NMR operating at 500.16 MHz or a Varian Unity Plus spectrometer
operating at 399.87 MHz at room temperature with 64 or 128 scans.
The spectra of the polymers and initiators were obtained in D
(0.10 g/mL) and CDCl (0.03 g/mL), respectively.
3
0 min. The reaction mixture was diluted with 5 mL of methanol
and 2 mL of water. The solution was dialyzed against 4 L of DI water
for 48 h, then freeze-dried.
Synthesis of boc-piperazine-terminated PEtOx oligomers with
benzyl bromide as the initiator followed a similar procedure to that
described above and the yields obtained were ~60%.
2
O
3
Size exclusion chromatography (SEC) was conducted on the
materials to measure molecular weight distributions. The column
set consisted of 3 Agilent PLgel 10-mm Mixed B-LS columns
2
.2.4. Synthesis of diiodo-p-xylene or dibromo-p-xylene-initiated
300 ꢂ 7.5 mm (polystyrene/divinylbenzene) connected in series
with a guard column having the same stationary phase. An isocratic
pump (Agilent 1260 infinity, Agilent Technologies) with an online
degasser (Agilent 1260), autosampler and column oven was used
for mobile phase delivery and sample injection. A system of mul-
tiple detectors connected in series was used for the analysis. A
multi-angle laser light scattering (MALLS) detector (DAWN-HELEOS
II, Wyatt Technology Corporation), operating at a wavelength of
and piperazine-functional telechelic poly(2-ethyl-2-oxazoline)
oligomers
An exemplary procedure for synthesizing a diiodo-p-xylene-
initiated and piperazine-functional telechelic PEtOx is provided.
Diiodo-p-xylene (0.212 g, 0.593 mmol) was placed in a dry, 100-mL,
round-bottom flask. EtOx monomer (3.61 mL, 3.54 g, 35.7 mmol)
and acetonitrile (3.5 mL) were added via a syringe to the flask

150
O. Celebi et al. / Polymer 56 (2015) 147e156
6
58 nm, a viscometer detector (Viscostar, Wyatt Technology Corp.),
and a refractive index detector operating at a wavelength of 658 nm
Optilab T-rEX, Wyatt Technology Corp.) provided online results.
same monomer and initiator concentrations and their molecular
weights and endgroup structures were analyzed (Fig. 1, Table 1). It
was reasoned that this set would serve as control materials since
fast initiation relative to propagation and narrow molecular weight
distributions have been reported earlier by several investigators [2].
(
The system was corrected for interdetector delay, band broadening,
and the MALLS signals were normalized using a 21,000 g/mol
polystyrene standard obtained from Agilent Technologies or Var-
ian. Data acquisition and analysis were conducted using Astra 6
software from Wyatt Technology Corp. A universal calibration
curve was constructed using polystyrene standards with narrow
PDI's. Seventeen standards were used to establish the calibration
curve with molecular weights ranging from 2000 to 1.4 million g/
mol, but the low molecular weight range was of particular interest
for this study. Five standards in the range of 2000 to 10,000 g/mol
were used to cover this range. The uncertainty of the universal
calibration molecular weight measurements of the PEtOx homo-
polymers was determined by preparing and running replicate
samples of 3 different methyl triflate-initiated PEtOx oligomers in
the range of 7000 g/mol. Each sample was prepared 3 times and
each specimen was run and analyzed 3 separate times producing 9
ꢁ
The propagating cationic chains were terminated at 25 C with
either piperidine or boc-piperazine in a similar manner to that
described by Luxenhofer et al. [11]. Monomer conversion was
1
monitored using H NMR spectroscopy from the integral ratios of
resonances corresponding to the monomer and polymer backbone
(eNCH CH e) [23,41], and the polymers were terminated at
2 2
80e100% conversion. Polymers with boc-piperazine endgroups
were deprotected by cleaving the t-butyloxycarbonyl groups under
acidic conditions to form secondary amine-functional chains.
The structures and endgroups of these monofunctional PEtOx
1
oligomers were characterized using H NMR (Fig. 2aec), titrations
and SEC (Table 1). Polymerization resulted in broad NMR peaks
around 3.45 ppm due to methylene backbone protons
(eNCH
2
CH
2
e) and pendent group protons at 2.25 (eCH
2
e) and
analyses of each of the 3 samples. The variation in M
calibration was defined as 2 standard deviations (7e8%). The vari-
ation in M by light scattering was also defined as 2 standard de-
viations (11e13%). All of these measurements were carried out
based on the same calibration curve and therefore this evaluation
takes into account the sample preparation and run-to-run varia-
tions of the SEC samples. Validation of the system was performed
by monitoring the molar mass of a known molecular weight
polystyrene sample by light scattering with every sample set. The
accepted variance of the 21,000 g/mole sample was defined as 2
n
by universal
0.90 ppm (-CH
3
). The methyl endgroup introduced in the initiation
step resonated at 2.8 and 3.0 ppm, thus reflecting cis and trans
amide endgroups. The signals around 1.40 (Fig. 2a) and 1.30 ppm
(Fig. 2b) were assigned to piperidine (6H) and t-butyl (9H) in boc-
piperazine endgroups, respectively. The integral ratios of methyl
protons on the initiator end to piperidine or t-butyl groups on the
terminated end were compared to assess the efficiency of termi-
nation (1:1 endgroup ratio). It should be noted that the t-butyl
protons on the boc-piperazine terminated oligomers appear as a
sharp singlet, while the piperidine proton resonances are broad,
n
1
standard deviations (11.5% M
-month time period.
Titrations were performed to determine the number average
n
, 9% M
w
) from a set of 34 runs over a
and thus it is expected that the H NMR analyses may be more
accurate for the boc-piperazine-functional polymers. However, a ±
6
~5% error was applied for all of the NMR analyses. In all cases, the
1
molecular weights of poly(2-oxazoline) oligomers by endgroup
analysis. An Orion 3 Star pH meter with gel-filled pH electrodes was
utilized for pH measurements. A piperidine- or piperazine-
functional polymer solution (~200e300 mg in 20 mL) in water or
isopropanol, respectively, was charged to a 50-mL, 3-necked,
round-bottom flask containing a stir bar. The solution, while being
stirred slowly, was titrated with a 0.05 M KOH(aq) or HCl(aq) stan-
dard solution by addition of 0.05-mL increments using a micro-
pipette. Equivalence (inflection) points were determined by plot-
H NMR integrals were consistent with the expected 1:1 ratio of
methyl groups on the initiator to secondary amines on the termi-
nated end. Number average molecular weights were calculated by
comparing the ratio of the integrals of PEtOx proton resonances to
3
the endgroup protons (both CH and either piperidine or t-butyl).
Disappearances of the t-butyl resonances upon deprotection of boc-
piperazine were also indicative of successful deprotection (Fig. 2c).
The piperazine and piperidine endgroups were titrated to
determine the concentrations of amine and to correlate this data
1
ting the first derivative of the pH curve (
D
pH/
DV) against titrant
with the H NMR endgroup analyses. Titrations showed the ex-
volume.
pected two inflection points for piperazine endgroups and one for
piperidine-terminated polymers. Piperazinium-functional poly-
mers were titrated with base and the titrant volumes between the
two inflection points corresponded to the number of amine
endgroups. To determine the equivalence points that resulted from
abstraction of protons from piperazinium endgroups, the first de-
3
. Results and discussion
Prepolymers synthesized with three classes of initiators based
on their rates of initiation and the relative nucleophilicity of the
counterions have been compared. These include methyl triflate
rivative of the pH curve (Fig. S1a) (DpH/DV) was plotted against
(
fast initiation, weakly nucleophilic counterion), activated alkyl
volume (Fig. S1b). When pH/ V is at its maximum value, it rep-
D
D
halides including initiators with benzyl and xylyl groups (slower
initiation, more nucleophilic counterions), and non-activated alkyl
iodides (very slow initiation, also more nucleophilic counterions).
While triflates are ideal with regard to oligomerization conditions,
their sensitivity to moisture makes triflate macromonomers more
difficult to handle and purify. Alkyl halide initiators are much less
sensitive to moisture and would be preferable if oligomers with
good molecular weight distributions and terminal functionality can
be achieved.
resents the inflection point of the titration. Titrant volumes be-
tween the two inflection points were utilized to determine the
concentration of amine endgroups. Piperidine-functional polymers
were titrated with acid. The NMR and titration data shown in
Table 1 are within the error range for all of the triflate-initiated
polymers. It is reasoned that the titration data is likely more ac-
curate due to inherent line broadening in the NMR spectra. The
endgroup data showed that the molecular weights were somewhat
higher than the targeted values and this was attributed to loss of
some low molecular weight species in the dialysis process used to
purify the oligomers.
3.1. Monofunctional poly(2-ethyl-2-oxazoline) oligomers initiated
by methyl triflate and terminated with secondary amines
Molecular weights of the PEtOx oligomers were measured by
SEC in N-methylpyrrolidone containing 0.05 M LiBr and compared
A series of methyl triflate-initiated PEtOx oligomers were pre-
pared at different temperatures (25, 50, 65 and 80 C) with the
to number average molecular weights (M
endgroup analyses (Table 1). All of the polymerizations initiated
n
's) calculated from the
ꢁ

O. Celebi et al. / Polymer 56 (2015) 147e156
151
Fig. 1. Synthesis of controlled molecular weight poly(2-ethyl-2-oxazoline) oligomers with piperidine or piperazine endgroups.
with methyl triflate that were conducted over a range of temper-
atures (25e80 C) and that were terminated at 70 to near 100%
monomer conversion produced unimodal, symmetrical chromato-
grams (Fig. 3). While reactions proceeded faster as expected as
temperature was increased, no effect on molecular weight distri-
bution was observed with temperature in this range as long as the
reactions were not continued after the monomers were depleted.
ꢁ
n
M 's were calculated from the SEC curves via both a universal
Table 1
Methyl triflate-initiated PEtOx oligomers prepared at different temperatures.
a
b 1H NMR
c,d
c,e
Reaction
temperature ( C)
Reaction time (h) Termination agent
M
n
target
M
n
M
n
titration
M
n
SEC universal calibration
M
n
SEC light scattering PDI
ꢁ
8
6
5
2
0
5
0
5
4
7
24
boc-Piperazine
Piperidine
Piperidine
5200
4500
4850
4400
6250 ± 300
5450 ± 250
5650 ± 300
4550 ± 250
5750 ± 250 6800 ± 250
5650 ± 150 6600 ± 250
6500 ± 350
6200 ± 350
6800 ± 350
5900 ± 350
1.13
1.15
1.12
1.11
6000 ± 50
7000 ± 250
6100 ± 250
215
Piperidine
e
a
b
c
Corrected for monomer conversion at termination and includes the terminal boc-piperazine or piperidine endgroups.
Error in the NMR spectra were estimated to be ± ~5% in the calculated M
The dn/dc was 0.0515 ± 3%.
n
values.
d
e
The standard deviation varied from ±120 to 240.
The standard deviation falls in the range of ±260 to 370.

152
O. Celebi et al. / Polymer 56 (2015) 147e156
calibration that was prepared with a series of monodisperse poly-
styrene standards and viscometric and differential refractive index
detectors, and also by MALLS static light scattering and differential
calibration method is also more dependent on the higher molecular
weight polymers because it depends on solution viscosity. All of the
SEC analyses of the triflate-initiated polymers showed somewhat
refractive index detectors. The intensity of the Raleigh ratio (R(
q
))
n
higher M 's than the endgroup analyses (NMR and titration)
signal from the light scattering detector is a function of the
refractive index increment (dn/dc), the sample concentration, and
the molar mass of the molecules in solution. Due to the dependence
on molar mass, light scattering is more sensitive to higher molec-
ular weight polymers as opposed to the lower oligomeric range
investigated herein, whereas the refractive index signal only de-
pends on concentration (not on molar mass). The universal
(Table 1). This can likely be attributed to the SEC weighted
dependence on the higher molecular weight side of the distribu-
tions with both light scattering and viscometric measurements.
The refractive index increment (dn/dc) is critical to calculating
molar mass by light scattering. It is also important for calculating
molar mass using a universal calibration because an accurate
measure of the concentration of the polymer molecules at each
increment is necessary to determine the molar mass averages for a
polymer distribution. The dn/dc's of PEtOx in NMP with 0.05 M LiBr
were calculated using two methods. The dn/dc of the samples used
for the uncertainty study was measured for each replicate based on
the assumption of 100% recovery. The average dn/dc of the 27 data
points was 0.0514. A high molecular weight standard was used to
determine a dn/dc of 0.0516 offline. Therefore a dn/dc value of
0
0
.0515 was used to calculate M
n
s of the methyl triflate-initiated
polymers shown in Table 1.
We also compared molecular weight distributions of two PEtOx
ꢁ
oligomers that were polymerized at 80 C, where one was termi-
nated immediately after monomer conversion (4 h), and the other
ꢁ
was maintained at 80 C well after all of the monomers had been
consumed (24 h). The polymer that had been heated for ~24 h had a
prominent high molecular weight shoulder in the SEC chromato-
gram, whereas the polymer heated for only 4 h had a symmetrical
unimodal peak and a PDI of 1.13 (Fig. S2). Moreover, our studies
indicated that as these side reactions occurred, there was no cor-
relation between the integral ratios of protons from initiator frag-
ments and terminal groups. This may be attributed to enamine
formation and chain coupling as monomer was depleted as previ-
ously suggested by Litt et al. [42]. All of this data on molecular
weights and endgroup structures of PEtOx oligomers suggests that
materials with controllable molecular weights and narrow PDI's
with the expected endgroup functionality can be prepared with
methyl triflate as an initiator (Table 1). It is important, however, to
terminate the reactions at times close to complete monomer
conversion.
3.2. Comparison of prepolymers initiated with activated benzyl and
xylyl halides
Benzyl and xylyl halide initiators were investigated due to their
ease of handling and because mono- or telechelic prepolymers,
respectively, would be accessible by this method [25,43].
Fig. 2. 1H NMR spectra of a) CH
the initiator at 65 C and terminated with piperidine, b) CH
synthesized using methyl triflate as the initiator at 80 C and terminated with pro-
-PEtOx-piperidine synthesized using methyl triflate as
3
ꢁ
3
-PEtOx-boc-piperazine
ꢁ
tected piperazine, and c) CH
zine) prepared at 80 C obtained in D
3
-PEtOx-piperazine (deprotected CH
O.
3
-PEtOx-boc-pipera-
Fig. 3. Refractive index chromatograms of PEtOx oligomers initiated with methyl tri-
flate and polymerized at 25, 50, 65 and 80 C.
ꢁ
ꢁ
2

O. Celebi et al. / Polymer 56 (2015) 147e156
153
Monofunctional PEtOx oligomers were initiated with benzyl bro-
ꢁ
mide at 25, 40, 50 and 65 C under identical reaction conditions to
those conducted with methyl triflate so that the degree of control
over molecular weight and endgroup functionality could be directly
compared. Telechelic PEtOx oligomers were prepared using
1
dibromo- and diiodo-p-xylene initiators. H NMR indicated the
successful conversion of dibromo-p-xylene to diiodo-p-xylene as
confirmed by the upfield shift of aromatic and methylene protons
in diiodo-p-xylene in reference to dibromo-p-xylene. The melting
ꢁ
point of the diiodo-functional initiator was 178e179 C and
ꢁ
dibromo-p-xylene had a melting point of 145e147 C, consistent
with literature values [44].
1H NMR again showed an approximate 1:1 correlation between
boc-piperazine (t-butyl) resonances (9H) and benzyl endgroup ar-
omatic protons (5H) at 7.10e7.35 ppm (Fig. 4). Chemical structures
of the difunctional PEtOx oligomers with boc-piperazine endgroups
1
and their deprotected forms were confirmed by H NMR and ti-
trations (Fig. 5aec). Aromatic and methylene protons from the
initiator resonated at 7.15 and 4.50 ppm, respectively. The integral
values of t-butyl protons (18H) on the endgroups and resonances
from the initiator (4H) suggested that difunctional polymers were
formed (Fig. 5b). Disappearance of the t-butyl proton resonances
upon deprotection of the chain ends indicated conversion to sec-
1
ondary amine endgroups (Fig. 5c). H NMR also showed the ex-
pected downfield shifts of the protonated piperazinium resonances
as they were formed under acidic conditions.
n
M values were calculated using the relative integral values of
PEtOx protons in the repeat units and the endgroups (Table 2). As
with the triflate-initiated oligomers, the amine endgroups of these
oligomers were also titrated. With the exception of entry 4, Table 2
shows good agreement between M 's derived from the titration
n
and NMR endgroup data for the polymers that were initiated with
benzyl and xylyl halides. We and others have observed that 2-ethyl-
2
-oxazoline polymerizations conducted using benzyl (and also
initiated by non-activated alkyl halides) halides proceed via a
mixture of covalent and cationic oxazolinium species due to some
attack on the terminal cationic ring by the relatively nucleophilic
halide counterions [23,41,45]. The reactivity of cationic propagating
chains is much higher compared to the covalently-bound species
[46]. However, since the NMR data show an ~1:1 ratio of benzyl to
piperazine functionality and since both covalent and ionic propa-
gating species were present in the benzyl halide-initiated poly-
merizations, this suggests that both the covalent and ionic species
react with piperazine in the termination step. This is important
Fig. 5. 1H NMR spectra of a) dibromo- and diiodo-p-xylene initiators obtained in
ꢁ
CDCl
terminated with boc-piperazine obtained in D
chelic PEtOx (deprotected form) obtained in D
3
, b) a telechelic PEtOx oligomer polymerized using diiodo-p-xylene at 60 C and
2
O, and c) piperazine-functional tele-
O.
2
because even though two types of propagating species were pre-
sent, the desired endgroup functionality was still obtained.
SEC results, however, showed that when employing the same
dn/dc value for the benzyl and xylyl halide-initiated polymers that
was measured for the methyl triflate-initiated oligomers, the SEC
sample recovery values yielded by the Astra software were higher
than 100% (signifying that the dn/dc value was too low for those
polymers). The refractive index increment is independent of molar
mass as long as the chemical structure is independent of molar
mass. For the polymers in Table 2 that were initiated with benzyl or
Fig. 4. 1H NMR spectrum of a PEtOx oligomer that was initiated with benzyl bromide
ꢁ
at 25 C and terminated with boc-piperazine obtained in D
2
O.

154
O. Celebi et al. / Polymer 56 (2015) 147e156
Table 2
Alkyl halide-initiated monofunctional and telechelic PEtOx oligomers terminated with boc-piperazine.
ꢁ
Targetb
c 1H NMR
d
Initiator
Temperature ( C)
Reaction time (h)
M
n
M
n
M
n
titration
M
n
SEC universal calibration
PDI
Benzyl bromide
Benzyl bromide
Benzyl bromide
Benzyl bromide
Dibromo-p-xylene
Diiodo-p-xylene
Monovinyl alkyl iodide
Divinyl alkyla iodide
25
40
50
65
60
60
60
60
304
33
25
4650
4550
4850
5000
7200
5550
4000
4300
6050 ± 300
5500 ± 300
6250 ± 300
6150 ± 300
8300 ± 400
6750 ± 350
7300 ± 350
7250 ± 350
5900 ± 100
5600 ± 300
6100 ± 250
5650 ± 150
8600 ± 150
6975 ± 125
e
4700
4900
4900
5600
6300
5000
6500 ± 250
5600 ± 250
1.36
1.34
1.33
1.30
1.45
1.37
1.26
1.33
19
8.5
4.5
13.5
10.5
6150 ± 125
a
Terminated with piperidine.
b
c
Corrected for the extent of monomer conversion at termination and includes the boc-piperazine or piperidine endgroups.
Error in the NMR spectra were estimated to be ± ~5% for boc-piperazine functional oligomers and ±7% for piperidine functional polymers of the M
n
values.
Errors were not assigned to the SEC values for the cases of aromatic initiators due to the unknown error caused by deviations in the dn/dc across the molecular weight
d
distribution.
xylyl halides (Fig. S3), this was not the case. Just one aromatic ring
on the endgroup (or in the middle of the chain for xylyl halide
initiation) impacted the dn/dc sufficiently to cause deviances in
calculations of molecular weights, and the impact was greater for
lower molecular weights. This implied that dn/dc varied as a
function of molecular weight and therefore it varied significantly
across the distribution. Thus, it was necessary to estimate the dn/dc
values of the benzyl and xylyl-initiated PEtOx oligomers shown in
Table 2 by adjusting them for each sample until the SEC software
suggested 100% recovery of the sample. It is noted that the mo-
lecular weights derived from SEC for the polymers in Table 2 with
the aromatic initiators were lower than their titrated and NMR
values. This may indeed be attributable to a variance in dn/dc across
the molecular weight distribution for those in Table 2 prepared
with aromatic initiators due to the high refractive index of the
initiator.
confirmed efficient termination by either piperidine or boc-piper-
azine. Methyl and vinyl groups attached to the silicon atom
remained intact under acidic conditions during cleavage of the
protecting groups (Fig. 7c). However, the initiation rate with these
initiators relative to propagation was too slow and there were
substantial amounts of low molecular weight species in the mo-
lecular weight distributions (Fig. S5). Yields of the polymerizations
initiated by the two alkyl iodide initiators were ~60%. The data in
Table 2 show a wider discrepancy between the targeted and ac-
quired molecular weights than the polymers initiated with the
activated benzyl halides, and this is attributed to more loss of the
low molecular weight fractions during isolation and purification in
cases where non-activated halides were the initiators. Moreover,
the SEC chromatograms of polymers initiated with the non-
activated halides had significant low molecular weight tails
(Fig. S5) that were not observed with the activated halide initiators
(Fig. S4). Thus, it was reasoned that the combination of very slow
initiation with the non-activated alkyl iodides combined with the
nucleophilicity of the iodide counterions that caused some slowly-
propagating species resulted in products with significant low mo-
lecular weight fractions.
n
While absolute values for M 's from SEC for the benzyl and
xylyl-initiated polymers were not reliable, the molecular weight
distributions for those materials were uniformly higher than for
polymers initiated with methyl triflate (~1.30e1.45 versus ~1.1,
Tables 1 and 2). RI chromatograms of the benzyl bromide initiated
oligomers showed a slight tail on the low molecular weight side
relative to the very symmetrical curves from the methyl triflate-
initiated polymers (Fig. S4). The rather broad PDI values were
attributed to slower initiation rates and slow propagation of co-
valent species that were present in addition to the cationic oxa-
zolinium ions.
3.3. Heterobifunctional PEtOx oligomers from vinylsilane-
functional non-activated alkyl halide initiators
Heterobifunctional PEtOx's with vinylsilane functionalities on
one chain end and an amine group on the other were attempted
using mono- and divinylsilylpropyl iodide initiators (Fig. 6). These
initiators containing one or two vinyl groups were synthesized
from
3-chloropropylchlorodimethylsilane
and
3-
chloropropyldichloromethylsilane, respectively, by reaction with
vinylmagnesium bromide. The alkyl bromides were then converted
to the corresponding iodides by a Finkelstein reaction as described
previously [28e30]. The molecular structures of these mono- and
1
divinylsilyl alkyl iodides were confirmed by H NMR (Figs. 7a and
8
a). The methylene protons adjacent to the silicon atom reso-
nated at 0.6 ppm, and the middle (-CH ) protons on the propyl
2
group were observed at 1.7 ppm. The signals at 3.1 ppm were
assigned to the methylene group directly attached to the iodine.
1
Vinyl protons appeared at 5.6e6.2 ppm. H NMR analyses of the
heterobifunctional PEtOx oligomers also showed the expected
chemical structures (Figs. 7b and 8b). Endgroup analyses again
Fig. 6. Synthesis of heterobifunctional PEtOx oligomers initiated by mono- or divinyl
alkyl iodides and terminated with aliphatic amines.

O. Celebi et al. / Polymer 56 (2015) 147e156
155
Fig. 8. 1H NMR spectra of a) 3-iodopropylmethyldivinylsilane obtained in CDCl
a methyldivinylsilylpropoxy-PEtOx-Piperidine oligomer obtained in D O.
and b)
3
2
such as the formation of chain-coupled products were
minimized.
Activated and non-activated halide initiators were investigated
and the polymer molecular weights and endgroup functionalities
were compared with the methyl triflate-initiated controls. For the
polymers initiated with methyl triflate and the activated halides,
NMR data suggested that the endgroup structures were preserved.
Thus, it is concluded that amines used in the termination step
reacted with both the cationic and covalent endgroups on the
polymers initiated with activated halides. Endgroup structures and
1
Fig. 7. 1H NMR spectra of a) 3-iodopropyldimethylvinylsilane obtained in CDCl
dimethylvinylsilylpropoxy-PEtOx-boc-Piperazine obtained in O, and
dimethylvinylsilylpropoxy-PEtOx-Piperazine obtained in D O.
, b)
c)
their relative concentrations obtained from H NMR and titration
3
D
2
were in good agreement. In all cases, the polymers prepared with
the halide-functional initiators had broader molecular weight dis-
tributions than the alkyl triflate-initiated controls and this was
attributed to a combination of slower initiation and also to the
relative nucleophilicity of the counterions that led to some covalent
propagating species in addition to the cations in those cases. The
purification steps that included dialysis also removed some low
molecular weight polymers and this was more pronounced with
the non-activated halide initiators, thus leading to wider discrep-
ancies between targeted and obtained molecular weights.
SEC chromatograms showed some differences between oligo-
mers prepared from fast and slow initiating species. The PEtOx
oligomers had small dn/dc values in the SEC solvent and even a
slight change caused significant deviations in molecular weights as
in the case of benzyl and xylyl halide initiated oligomers due to the
presence of an aromatic ring with a high refractive index. The
variation in refractive index increment with concentration across
the molecular weight distribution caused deviation in SEC molec-
ular weights.
2
4
. Conclusions
Poly(2-alkyl-2-oxazoline) prepolymers were prepared with
triflate, and activated and non-activated alkyl halide initiators,
and their molecular weights and endgroup functionalities were
examined with the objective of delineating synthetic parameters
for oligomers that could be used as components of block and
graft copolymers. Results showed that mono-functional PEtOx
oligomers with the expected functionality utilizing triflate initi-
ators as the control materials were prepared. The polymeriza-
ꢁ
tions were conducted between 25 and 80 C and the reactions
were terminated at 70e100% monomer conversion. No effects on
molecular weight distributions were observed with temperature
in this range. By terminating each reaction without allowing
prolonged heating in the absence of monomer, side reactions

1
56
O. Celebi et al. / Polymer 56 (2015) 147e156
[14] Konradi R, Pidhatika B, Muhlebach A, Textor M. Langmuir 2008;24(3):613.
The materials that were initiated with methyl triflate and benzyl
[
[
15] Zalipsky S, Hansen CB, Oaks JM, Allen TM. J Pharm Sci 1996;85(2):133.
16] Bauer M, Lautenschlaeger C, Kempe K, Tauhardt L, Schubert US, Fischer D.
Macromol Biosci 2012;12(7):986.
and xylyl halides are considered to be potential candidates for
hydrophilic functional macromonomers or telechelic difunctional
oligomers for incorporation into novel block and graft copolymer
structures that may find use in novel membranes and various
medical applications from hydrogels to drug delivery.
[17] Woodle MC, Engbers CM, Zalipsky S. Bioconjug Chem 1994;5(6):493.
[18] Luxenhofer R, Schulz A, Roques C, Li S, Bronich TK, Batrakova EV, et al. Bio-
materials 2010;31(18):4972.
[19] Han Y, He Z, Schulz A, Bronich TK, Jordan R, Luxenhofer R, et al. Mol Phar-
macol 2012;9(8):2302.
[
[
20] Tong J, Luxenhofer R, Yi X, Jordan R, Kabanov AV. Mol Pharm 2010;7(4):984.
21] Tong J, Yi X, Luxenhofer R, Banks WA, Jordan R, Zimmerman MC, et al. Mol
Pharm 2012;10(1):360.
Acknowledgments
The authors gratefully acknowledge the support of the National
Science Foundation under contracts DMR 1263248 (REU), DMR
[22] Saegusa T, Kobayashi S, Yamada A. Makromol Chem 1976;177(8):2271.
[23] Liu Q, Konas M, Riffle JS. Macromolecules 1993;26(21):5572.
[24] Kobayashi S, Uyama H, Narita Y. Macromolecules 1990;23(1):353.
[25] Kobayashi S, Uyama H, Narita Y, Ishiyama J. Macromolecules 1992;25(12):
3232.
1126534 (MRI) and DMR 1106182 for financial support of the work
described in this manuscript.
[26] Liu Q, Konas M, Davis RM, Riffle JS. J Polym Sci Part A Polym Chem 1993;31(7):
1
709.
Appendix A. Supplementary data
[
[
27] Liu Q, Wilson GR, Davis RM, Riffle JS. Polymer 1993;34(14):3030.
28] Thompson MS, Vadala TP, Vadala ML, Lin Y, Riffle JS. Polymer 2008;49(2):345.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.polymer.2014.11.005.
[29] Vadala ML, Thompson MS, Ashworth MA, Lin Y, Vadala TP, Ragheb R, et al.
Biomacromolecules 2008;9(3):1035.
[30] Goff JD, Huffstetler PP, Miles WC, Pothayee N, Reinholz CM, Ball S, et al. Chem
Mater 2009;21(20):4784.
References
[31] Oishi M, Nagasaki Y, Kataoka K. Functional PEG for drug delivery. In: Kwon GS,
editor. Polymeric drug delivery systems, vol. 148. Boca Raton: Taylor &
Francis; 2005. p. 93.
[
[
1] Aoi K, Okada M. Prog Polym Sci 1996;21(1):151.
2] Kobayashi S. 4.14-Polymerization of oxazolines. In: Matyjasewski K, Moller M,
editors. Polymer science: a comprehensive reference. Amsterdam: Elsevier;
[
[
32] Brocas A-L, Mantzaridis C, Tunc D, Carlotti S. Prog Polym Sci 2013;38(6):845.
33] Roberts MJ, Bentley MD, Harris JM. Adv Drug Deliv Rev 2012;64(Suppl. (0)):
116.
2012. p. 397.
[34] Cardoen G, Burke B, Sill K, Mirosevich JJ. Polym Res 2012;19(4):1.
[35] Reif M, Jordan R. Macromol Chem Phys 2011;212(16):1815.
[36] Yang Y, Kataoka K, Winnik FM. Macromolecules 2005;38(6):2043.
[37] Fijten MWM, Haensch C, van Lankvelt BM, Hoogenboom R, Schubert US.
Macromol Chem Phys 2008;209(18):1887.
[
[
[
[
3] Barz M, Luxenhofer R, Zentel R, Vicent MJ. Polym Chem 2011;2(9):1900.
4] Hoogenboom R. Angew Chem Int Ed 2009;48(43):7978.
5] Hoogenboom R. Macromol Chem Phys 2007;208(1):18.
6] Hoogenboom R. Polyethers and polyoxazolines. Handbook of ring-opening
polymerization. Wiley-VCH Verlag GmbH & Co. KGaA; 2009. p. 141.
7] Luxenhofer R, Han Y, Schulz A, Tong J, He Z, Kabanov AV, et al. Macromol
Rapid Commun 2012;33(19):1613.
[38] Volet G, Lav T, Babinot J, Amiel C. Macromol Chem Phys 2011;212(2):118.
[39] Chujo Y, Ihara E, Ihara H, Saegusa T. Macromolecules 1989;22(5):2040.
[40] Kempe K, Hoogenboom R, Jaeger M, Schubert US. Macromolecules
[
[
[
8] Knop K, Hoogenboom R, Fischer D, Schubert US. Angew Chem Int Ed
2
011;44(16):6424.
41] Fijten MWM, Hoogenboom R, Schubert US. J Polym Sci Part A Polym Chem
008;46(14):4804.
2010;49(36):6288.
[
9] Adams N, Schubert US. Adv Drug Deliv Rev 2007;59(15):1504.
2
[
[
[
[
10] Schlaad H, Diehl C, Gress A, Meyer M, Demirel AL, Nur Y, et al. Macromol
Rapid Commun 2010;31(6):511.
11] Gaertner FC, Luxenhofer R, Blechert B, Jordan R, Essler M. J Control Release
[
[
42] Litt M, Levy A, Herz J. J Macromol Sci Chem 1975;9(5):703.
43] Hoogenboom R, Fijten MWM, Schubert US. J Polym Sci Part A Polym Chem
2004;42(8):1830.
2007;119(3):291.
[
[
44] Moore JS, Stupp SI. Macromolecules 1986;19(7):1815.
45] Hoogenboom R, Paulus RM, Fijten MWM, Schubert US. J Polym Sci Part A
Polym Chem 2005;43(7):1487.
12] Goddard P, Hutchinson LE, Brown J, Brookman LJ.
J
Control Release
1989;10(1):5.
13] Kronek J, Kronekov aꢀ Z, Lusto nꢁ J, Paulovi ꢁc ov ꢀa E, Paulovi ꢁc ov aꢀ L, Mendrek BJ.
[46] Dworak A. Macromol Chem Phys 1998;199(9):1843.
Mater Sci Mater Med 2011;22(7):1725.