Exploring the versatility of hydrogels derived from living organocatalytic ring-opening polymerization

Article abstract of DOI:10.1039/b920216a

In this work we have bridged the use of mild and living organocatalytic ring-opening polymerization to facilitate the synthesis of cross-linked networks with an emphasis on hydrogels. Amidine-catalyzed ring-opening polymerization of bis-carbonate macromon

Full text of DOI:10.1039/b920216a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/softmatter | Soft Matter
Exploring the versatility of hydrogels derived from living organocatalytic
ring-opening polymerization†
Fredrik Nederberg,ab Vivian Trang,c Russell C. Pratt,a Sung-Ho Kim,a John Colson,d Alshakim Nelson,a
e
Curtis W. Frank, James L. Hedrick, Philippe Dubois and Laetitia Mespouille
a
f
aef
*
Received 28th September 2009, Accepted 9th February 2010
First published as an Advance Article on the web 9th March 2010
DOI: 10.1039/b920216a
In this work we have bridged the use of mild and living organocatalytic ring-opening polymerization to
facilitate the synthesis of cross-linked networks with an emphasis on hydrogels. Amidine-catalyzed
ring-opening polymerization of bis-carbonate macromonomers in the presence of an alcohol provides
the onset for the reaction and various building blocks issued from the initiator, macromonomer and
comonomer can be used in different proportions to tailor the swelling behavior and mechanical
integrity of final networks. Easy modifications of the building blocks additionally allow for finely
tuning the hydrogel functionality and/or promoting responsiveness in the final structure.
membranes, soft contact lens, cornea replacement, bioadhesives,
Introduction
tissue engineering scaffolds, and drug delivery applications.17,18
Several synthetic routes to hydrogels are available today and
ultimately hydrogel characteristics such as swelling, tensile
strength, and biocompatibility in combination with the applica-
tion determine which system to use.18–21 Recently we described an
organocatalytic ROP approach to network and subsequent
hydrogel formation using bis-carbonate functional poly(ethylene
glycol) (PEG) oligomers3 initiated from alcoholic initiators in the
presence of an amidine-based organic catalyst. From our initial
report a key model reaction identified the critical monomer
concentration dependent reaction regime and enhanced kinetic
control was demonstrated by the introduction of trimethylene
carbonate (TMC) as a comonomer. The addition of the como-
nomer allowed for near quantitative conversion of monomer to
polymer, once initiated from an alcoholic initiator, providing an
efficient route to hydrogel formation. The addition of the
comonomer is not only crucial for enhanced kinetic control, but
also suggests a possible means to introduce functionality
together with the judicious choice of the initiator. Here we report
the impact of the initial feed composition as the initial monomer
to initiator ratio (i.e. the degree of polymerization DP),
concentration of comonomer relative to macromonomer as well
as the introduction of functional macroinitiator or comonomers
over final hydrogel properties (such as swelling, and mechanical
behavior). This was otherwise made possible by our recent ability
to produce functional carbonate monomers from a synthetic
strategy derived from 2,2-bis(methylol)propionic acid (bis-MPA)
(MTC-COOH), a common building block for the construction of
biocompatible dendrimers.22 Our pursuit of versatility stems
from the potential to tag an arbitrary alcohol or amine onto the
free acid of MTC-COOH to generate new ROP monomers
having a wider range of functional groups and to introduce new
properties to the hydrogel.23 In the frame of this work, a partic-
ular interest was devoted to an urea-based carbonate monomer
susceptible to strengthen the resulting network by formation of
H-bonds with the PEO cross-linker. A second option to intro-
duce functionalities and therefore interesting properties is the use
Recent advances in organic catalysis have allowed living ring-
opening polymerization (ROP) to expand into new areas. In the
last few years our group has developed catalysts for a variety of
monomeric building blocks to provide novel high performance
materials with tunable properties.1 This development was
initially fueled from the need for a metal free catalyst alternative
to meet the demanding needs for microelectronic materials and
more recently for biomaterials in biomedical applications.2–4
Successful organocatalysts for living ROP of cyclic esters and
carbonates include N-heterocyclic carbenes,5–10 bifunctional
thiourea-amines,11,12 amidine and guanidine13,14 based-catalysts
such as 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD), N-methyl-
TBD (MTBD), and 1,8-diazabicyclo[5.4.0]-undec-7-ene (DBU).
Living ROP has traditionally found application in the produc-
tion of (co)polymers of various topologies (block, graft, star-
shape, comb-like, etc.) presenting targeted molecular weights and
narrow molecular weight distributions.1,15,16 These features have
now been bridged towards hydrogels, an important category of
materials finding a lot of applications in the field. Hydrogels are
a unique class of aqueous swollen networks that have found
a multitude of use both in vitro and in vivo including separation
aIBM Almaden Research Center, 650 Harry Road, San Jose, 95120 CA,
USA. E-mail: hedrick@almaden.ibm.com
bDepartment of Chemistry, Stanford University, Stanford, 94305 CA, USA
cDepartment of Chemistry, University of California, Berkeley, CA 94720,
USA
dDepartment of Chemistry and Biochemistry, University of Oklahoma,
Norman, 73069 OK, USA
eDepartment of Chemical Engineering, Stanford University, Stanford, CA
94305, USA
fCenter of Innovation and Research in Materials and Polymers, Laboratory
of Polymeric and Composite Materials, 20, Place du Parc, 7000 Mons,
Belgium.
+003265373484; Tel: +003265373482
E-mail:
laetitia.mespouille@umons.ac.be;
Fax:
† Electronic Supplementary Information (ESI) available: Provided in the
supporting information are stress-strain curves of hydrogels, hydrogel
pictures, and differential scanning calorimetry thermograms. See DOI:
10.1039/b920216a/
2006 | Soft Matter, 2010, 6, 2006–2012
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Table 1 Hydrogel properties and the effect from initiator and como-
nomer (TMC) concentrationa
of functional initiator, which can range from a simple alcohol to
a biologically important nucleophile, or an environmentally
responsive hydroxyl functional macroinitiator to create
compartmental hydrogels or gels responsive to pH, pressure,
temperature or other factors.24 Not only does this approach
allow for tunable properties, but also for post modifications or
selective incorporation of functionality and/or to conjugate
drugs, proteins, etc. Moreover it provides a simple alternative to
existing condensation or radical approaches.18–21
TMCc (eq. to
Cross-linkerb c cross-linker) DPd (kPa)e break (%)e swelling (%)f
E
Strain at
Degree of
,
PEG-8k
PEG-8k
PEG-8k
PEG-8k
PEG-8k
PEG-8k
5
5
5
5
10
50
10
50
50
366
156
65
61
60
1 160
430
420
450
420
170
100 450
100 450
100 580
81
100 1 170 137
a
All hydrogels were obtained using 5 eq. of DBU catalyst relative to
cross-linker. Cross-linking performed on a 1.0 g cross-linker scale.
Result and discussion
b
c
For the cross-linking reaction the concentration of cross-linker +
TMC was held at 25 wt% in methylene chloride. Degree of
The synthetic strategy envisioned as a new route for producing
amphiphilic polycarbonate networks consists in the ring-opening
copolymerization of a bis-carbonate functional poly(ethylene
glycol) (PEG) (1) (macromonomer) with trimethylene carbonate
(TMC) initiated from the 2-benzyl-2-methylpropane-1,3-diol
and catalyzed by 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU,
pKa ¼ 24.3, Scheme 1) in CH2Cl2 at r.t. The choice of DBU as
catalyst relies on its ability to activate both the initiating and
propagating alcohol; facilitating ROP under mild reaction
conditions while the TMC comonomer has proved to enhance
the kinetic control to obtain near quantitative conversion of
monomer to polymer, providing an efficient route to network
formation.3
d
e
polymerization DP ¼ (ncross-linker + nTMC)/ninitiator
.
From tensile
testing in water at 37 ꢂC, each sample measured three times and
f
average is given. S ¼ (mass swollen gelꢁmass polymer precursor)/
(mass polymer precursor)*100, average from three samples given.
(DP 10), resulting hydrogels swelled significantly in water
(ꢀ1160%) and with an elastic modulus of about 50 kPa.
Conversely, for the highest targeted DP’s (50 and 100), hydrogels
with considerably higher elastic modulus (366 and 450 kPa
respectively) and significantly lower swelling degree were formed
(450–400%). We rationalize these data through the number of
cross-links generated per chain. That is, for the low DP, chains
form with fewer cross-links producing a low cross-linking
density, resulting in an increased swelling capability and a lower
modulus. Alternatively by increasing the DP, the cross-linking
density increases with the expected increase in modulus and
decrease in swelling.12
In the frame of this work, a series of hydrogels P(TMC-g-PEO)
were prepared in which the individual constituents as the initial
monomer-to-initiator and cross-linker-to-monomer ratios were
varied. The effect of the composition of the feed was evaluated in
terms of tensile strength properties and swelling behavior.
Table 1 summarizes the results including mechanical proper-
ties and swelling measurements obtained by varying either the
initiator concentration or the comonomer composition for the
reaction of bis-carbonate PEG (Mnꢀ8000 g molꢁ1, PDI 1.03) in
the presence of DBU. In a general procedure, a 1.0 g macro-
monomer scale was used and the reaction was performed in
anhydrous methylene chloride at a 25 wt% concentration of all
constituents. Initially the comonomer concentration was held
constant at 5 equivalents to macro-monomer and the monomer-
to-initiator concentration was varied to give targeted degrees of
polymerization (DP)’s of 10, 50, and 100. For the lower DP
In a second set of experiments, the amount of comonomer
(TMC) relative to the macromonomer (from 5 to 50 equivalents)
was varied, while the initiator concentration was held constant to
give a targeted DP of 100. The swelling is largely reduced with
increasing TMC concentration and the modulus increases to
exceed 1 MPa for the highest concentration. In addition,
noticeable hydrogel turbidity is observed as the concentration of
TMC increases. Poly (trimethylene carbonate) (PTMC) homo-
polymer is hydrophobic, and with a contact angle (water–poly-
mer) exceeding 70ꢂ. Thus, when increasing the concentration of
TMC it likely produces PTMC domains or regions that decrease
the overall water–hydrogel interaction. In line with previous
results25 the modulus of the hydrogels increases with TMC
concentration (from 450 kPa [5eq] to 1170 kPa [50eq]), addi-
tionally strain at break values increases (from ꢀ60% [5eq] to
>130% [50eq]). Improvement of mechanical properties by
introduction of hydrophobic segments with low Tg was already
observed with other systems. For example, Dubois et al.26
demonstrated that incorporation of poly(3-caprolactone) (PCL)
cross-linkers in poly(N,N-dimethylamino-2-ethyl methacrylate)
PDMAEMA network led to an increase of the Young modulus
from 0.69 ꢃ 0.09 MPa to 2.72 ꢃ 0.12 MPa and an increase of the
strain at break from 12 to 215%. Thermal micro-structure
analysis of dried samples using differential scanning calorimetry
(DSC) indicated a phase separated morphology in which the
melting transition of PEG and the glass transition temperature
(Tg) of PTMC are observed (Fig. S1, ESI†). This observation
supports both the mechanical behavior and visual observation as
Scheme 1 General cross-linking procedure employing bis-carbonate
PEG macromonomer (1) as cross-linker in the presence of an alcohol
initiator DBU catalyst.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2006–2012 | 2007

View Article Online
Scheme 2 Polymerization of CL and end-group derivatization with
carbonate groups.
phase separated hydrophobic PTMC domains are expected to
form when increasing the concentration of TMC, leading to
mechanical strength enhancement.
Fig. 1 Degree of swelling (A) and modulus (,) as function of PCL/
PEG macromonomer ratio.
To further explore the use of hydrophobicity to toughen
hydrogels, carbonate functional PCL, (Mnꢀ8000 g molꢁ1, PDI
1.40), polymerized from 2-benzyl-2-methylpropane-1,3-diol
(DP of 70) in the presence of triazabicyclo-[4.4.0]dec-5-ene [TBD]
(pKa ¼ 26.0), was surveyed as a co-cross-linker (Scheme 2).25
PCL, as PTMC, is non-soluble in water and with a contact angle
(water–polymer) exceeding 90ꢂ.27
both controlled radical polymerization (CRP) and ROP was
employed (Scheme 3). More precisely, N-isopropylacrylamide
(NiPAAm) was polymerized by nitroxide mediated polymeriza-
tion (NMP) (Mnꢀ5800 g molꢁ1, PDI 1.12) using an alkoxyamine
initiator bearing a hydroxyl group. The post NMP preservation
of the PNiPAAm hydroxyl group was confirmed in a chain
extension experiment. Indeed, PNiPAAm was chain-extended by
polymerization of L-lactide using a thiourea/sparteine catalyst/
co-catalyst system. As attested by 1H NMR analysis by the
presence of PLA signals, and gel permeation chromatography by
a shift of the SEC trace toward lower retention volume, (GPC,
Mnꢀ8900 g molꢁ1, PDI ¼ 1.11), PNiPAAm was cleanly chain
extended by PLA.
The bis-carbonate PCL cross-linker was used in different
concentrations to bis-carbonate functional PEG, with a target
DP of 100 and with 5eq of TMC relative to the total concen-
tration of PCL and PEG. The addition of PCL had a significant
effect on the mechanical properties of the hydrogels explained
through the formation of phase separated hydrophobic domains
similar to the behavior previously observed with PTMC (Fig. 1).
This led to a significant decrease of the degree of swelling. Visual
inspection revealed an increased turbidity of the hydrogel with
increasing PCL content, in addition the modulus increases and
swelling decreases with increasing PCL concentration. For
example, a 1 : 1 (PEG:PCL) weight ratio generates a hydrogel
with a modulus of ꢀ4.5MPa and with a degree of swelling of
about 100% (Table 2). This material was particularly tough as it
exhibited the highest tensile strength of all materials made and
was shown to possess high strain values, in fact strain exceeded
200% when the material slipped through the grips without
breaking (Fig. S2, ESI†). DSC analysis supports phase separa-
tion of PEG and PCL as both melting transitions of PEG and
PCL were observed after drying the samples (Fig. S3, ESI†).
These data demonstrates that this general synthetic route can be
bridged to other families of polymers and that such combinations
may be used to tailor hydrogel properties.
Hydrogel formation was accomplished as described before,
using the hydroxy functional PNiPAAm as the initiator for
a targeted DP of 100 and 5eq of TMC to PEO cross-linker.
Interestingly, when the dry gel is immersed in water at ambient
temperature a clear gel formed with a degree of swelling of about
700%, however increasing the temperature to 37 ꢂC (physiological
conditions) the hydrogel became opaque (white) due to the lower
critical solubility temperature (LCST) transition of PNiPAAm
around 32–35 ꢂC (Fig. S4, ESI†). This phase separation confirms
the successful incorporation of the macro-initiator into the
hydrogel and demonstrates how a thermally induced response
may be provided to the final hydrogel. The tensile strength
increased above the LCST transition due to the collapse of the
PNiPAAm block and the modulus was measured to 0.47 MPa at
37 ꢂC and with a strain at break value of ꢀ150%. The same
material at room temperature (21 ꢂC) possessed a modulus of
0.31 MPa and a strain at break value of ꢀ50% (Table 2).
To further demonstrate the versatility of the synthetic
approach, the initiating alcohol was replaced with a macro-
initiator. For this purpose, a bifunctional initiator able to promote
The use of functional carbonates as comonomer provides
another route to influence the hydrogel properties or to introduce
Table 2 Mechanical properties of functional polycarbonate-based networksa
Initiatorb
Cross-linker
Comonomer
E (kPa)d
Strain at break (%)d
Degree of swelling (%)f
2
PEO + PCL
PEO
PEO
TMC
TMC
TMC
TMC/sIPN
4500Pa
310e
470
>200
50e
120
150
100
700
—
PNiPAAm-OH
PNiPAAM-OH
2c
PEO
100
376
a
All hydrogels were obtained using 5eq. of DBU catalyst relative to macromonomer. b Cross-linking performed on a 1.0 g PEO scale. For the cross-
c
linking reaction the concentration of cross-linker + TMC was held at 25 wt% in methylene chloride. P(TMC-g-PEO) primary network semi-
d
e
interpenetrated by PUC. From tensile testing in water at 37 ꢂC, each sample measured three times and average is given. From tensile testing in
water at 25 ꢂC, each sample measured three times and average is given. f S ¼ (mass swollen gel-mass dry gel)/(mass dry gel)*100, average from three
samples given at r.t.
2008 | Soft Matter, 2010, 6, 2006–2012
This journal is ª The Royal Society of Chemistry 2010

View Article Online
the reactive medium was poured in to a Petri dish already
charged with TMC and PEG cross-linker dissolved in DCM. The
targeted DP was fixed at 100 and the composition in TMC at 5
equivalents to PEO cross-linker. Surprisingly, after 48 h of
reaction, the solution remained viscous and no gel was formed.
This unsuccessful observation cannot be ascribed to the less
active (ꢁ)-sparteine catalyst since preliminary tests aiming at
synthesizing P(TMC-co-PEG) hydrogels or P(MTC-pentylphe-
nylurea) homopolymers (conv. ¼ 90%, Mw/Mn ¼ 1.14) were
conducted successfully (data not shown). Therefore, it was
assumed that H-bonding interactions were taking place in the
polymerization medium, inhibiting the propagating site due to
the resulting steric hindrance. This assumption was verified by
DSC of the P(MTC-pentylphenylurea) (Mn ¼ 10 500 g molꢁ1 and
Mw/Mn ¼ 1.14) and PEG (Mn ¼ 3.4 kg molꢁ1) precursors but
also of P(MTC-pentylphenylurea)/PEG blends of various
compositions. As illustrated by Fig. 2 only one glass transition
temperature (Tg) comprised between the individual Tg’s of the
two respective polymers was observed and corresponded in each
composition to the value predicted by the Fox equation.
Scheme 3 Synthesis of hydroxyl terminated poly(NiPAAm) macro-
initiator.
selective functional groups. We have recently demonstrated how
1,3-diols may be used as universal building blocks for functional
carbonates, and their controlled ROP using our organocatalytic
platform. More precisely, bis-MPA is the starting building block
that after simple COOH derivatization and ring-closure provides
a functional cyclic carbonate tag for further hydrogel function-
alization.23 In this work we targeted a different MTC monomer
carrying an urea group and its synthesis is shown in Scheme 4.
Initially, 5-amino-1-pentanol was reacted with phenyl iso-
thiocyanate to provide a phenylurea containing pentanol. This
intermediate was subsequently used to acylate MTC-COCl
yielding the MTC-pentylphenylurea monomer, which was
obtained in high purity and yield. The choice of MTC-pentyl-
phenylurea as comonomer was motivated by its ability to form
H-bonds with itself or with PEO. Bulk characterization of the
The miscibility between the two polymer chains was further
evidenced by the absence of the melting point related to the PEG
cross-linker. The miscibility observed between P(MTC-pentyl-
phenylurea) and PEG gives evidence for the secondary interac-
tions present between the two polymer counterparts and may
explain the absence of gelation. To lower the impact of
H-bonding over the gel formation, the MTC-pentylphenylurea
was directly copolymerized with TMC and the PEG cross-linker.
The homogeneous dispersion of MTC-pentylphenylurea into the
matrix is expected to suppress the cooperative effect observed in
the polymer and reduce the steric hindrance of the self-assembled
domains. Various compositions of 14.5, 21.5 and 26.6 mol% were
targeted by varying the [MTC-pentylphenylurea]0/[TMC]0 initial
ratio while the experimental conditions were kept unchanged.
Interestingly, only the hydrogel containing the lowest amount of
MTC-pentylphenylurea was obtained successfully with a gel
fraction of 80%. Increasing the composition in MTC-urea to 21.5
and 26.6 mol% led only to viscous solutions corresponding to
polymer chains with high molecular weight as observed by SEC
(Mnapp ¼ 28 600 and 26 700 g molꢁ1, respectively, multimodal
signals). Again, it seems that urea-urea and urea-PEG
H-bonding have a strong inhibiting effect over the cross-linking
process.
ꢂ
homopolymer has revealed a Tg of ꢀ15 C for poly(MTC-pen-
tylphenꢂylurea) which is significantly higher than that of PTMC
(ꢀꢁ20 C).
In the present work, various strategies to incorporate the urea
containing polymer were attempted demonstrating both the
strength and synthetic limitations dictated by the functional
group.
Our initial strategy envisioned to introduce the MTC-pentyl-
phenylurea monomer into the polymer network upon the
macroinitiator method, following the same strategy as with the
PNiPAAm-OH. Practically, MTC-urea was initiated in CH2Cl2
at r.t. using benzyl alcohol as initiator and (ꢁ)-sparteine with
N-(3,5-trifluoromethyl)benzyl-N0-cyclohexylthiourea (TU) as
catalyst and cocatalyst, respectively. The initial [MTC-urea]0/[I]0/
[(ꢁ)-sparteine]0/[TU]0 ratio was fixed at 32/1/5/5 and the initial
monomer concentration was fixed to 1.5 mol Lꢁ1 (Mn NMR ¼
10 500 g molꢁ1 and Mw/Mn ¼ 1.14). After 2 h of polymerization,
In order to introduce H-bonding within the network and overcome
the synthetic limitations observed above, semi-interpenetrating
Fig. 2 Experimental Tg’s obtained by DSC (5 ꢂC minꢁ1, 2nd scan) for
PEG (PEO) (3.4 k), P(MTC-pentylphenylurea) (Mn ¼ 10 500 g mol-1,
Mw/Mn ¼ 1.14), and PEG/P(MTC-pentylphenylurea).
Scheme 4 Synthesis and structure of urea-functional MTC monomer
(MTC-pentylphenylurea).
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2006–2012 | 2009

View Article Online
(all Aldrich) were used as received. Poly(ethylene glycol) (PEG):
8000g molꢁ1 (PDI: 1.03) [Fluka], trimethylene carbonate (TMC)
and L-lactide (Bohringer-Ingelheim) were all azeotropically
dried from toluene prior of use. 3-Caprolactone (Aldrich) was
distilled from calcium hydride prior of use. 1,8-Dia-
zabicyclo[5,4,0]undec-7-ene (DBU) (Aldrich) was distilled twice
prior of use. 1H NMR was performed on a Bruker Avance
400 MHz instrument. Gel permeation chromatography (GPC)
was performed in THF using a waters column chromatograph
with refractive index detection and compared with known
polystyrene standards. Differential scanning calorimetry (DSC)
was measured on a TA Instrument Q1000. Tensile testing was
measured on an Instron 5844 using a 10N load cell and a stan-
dard video extensometer setup at 37 ꢂC using a Biopuls
controlled water bath and Watlow thermostat, error range are
approximately of about 10%. Swelling studies were performed in
distilled water and an average from three parallel measurements
was given. Degree of swelling (%) ¼ (mass swollen gelꢁmass
polymer precursor)/(mass polymer precursor)*100.
Fig. 3 Swelling profiles for semi-interpenetrating networks (sIPNS)
containing various amounts of P(MTC-pentylphenylurea) (PUC).
network (SIPN) were envisioned as the more feasible route to
incorporate the urea groups. In the present case, the P(TMC-g-
PEO) primary network was made by standard methods and
subsequently swollen in a solution containing P(MTC-pentyl-
phenylurea) allowing the latter to penetrate the initial network
during the course of 24 h. In detail, P(TMC-g-PEO) hydrogels
slabs (m ꢀ 16 mg, diameter ¼ 0.25 inch) were pre-swollen in
50 mg of THF in order to dissolve the PEG crystalline micro
domains and to open the pores. P(MTC-pentylphenylurea)
(Mn NMR ¼ 10 500 g molꢁ1 and Mw/Mn ¼ 1.14), previously
dissolved in 0.1 mL of THF was added to the pre-swollen slabs to
reach a composition of 25, 50 or 75 wt% of P(MTC-pentylphe-
nylurea) in the network. Excess polymer on the surface was
washed out with water and the slabs dried overnight under
vacuum. After this treatment, SIPNs containing 23.8, 39.1 and
53.4 wt% of P(MTC-pentylphenylurea), respectively, were
successfully obtained. Swell tests of the SIPNs in deionized water
revealed that increasing amount of urea in the polymer network
decreased the swelling behavior of the primary network (Fig. 3).
Tensile testing on a 39.1 wt% P(MTC-pentylphenylurea) SIPN
revealed very high strain at break values (ꢀ150%) and an elastic
modulus of about 0.1 MPa (Fig. S5, ESI†) (Table 2).
Polymerization of N-isopropylacrylamide
N-Isopropylacrylamide, 2,2,5-trimethyl-3-(4¢-p-hydroxymethyl-
phenylethoxy)-4-phenyl-3-azahexan (concentration determined
by the desired degree of polymerization), and DMF (1 mL gꢁ1
monomer) were charged in a Schlenk tube and degassed by three
pump/freeze/thaw cycles. The mixture was heated under stirring
at 125 ꢂC and typically stopped when the conversion was
ꢀ80–90%. The reaction mixture was diluted with dichloro-
methane and precipitated in diethyl ether. The formed white
powder was filtrated, dried until constant weight under vacuo,
1
and further analyzed with H NMR (CDCl3), and GPC (THF).
General procedure for synthesis of bis-carbonate PCL
PCL and MTC-COOH (3eq to PEG) was charged in a 50 mL
round bottom flask equipped with a stir bar and dry THF was
added to dissolve the compounds. DCC (4eq to PEG) dissolved
in THF was added the flask and the formed DCC-urea derivative
started to precipitate after about 5 min. The solution was left
under stirring for an additional 2 h for the reaction to complete.
Following the reaction the DCC-urea was filtrated off and the
polymer precipitated in cold methanol. The formed precipitate
was collected and dried until constant weight, yield typical >85%.
1H-NMR (400 MHz, CDCl3): d ¼ 4.73 (d, 4H, 2 ꢄ CH2OCOO),
4.38 (m, 4H, 2 ꢄ PCL-(CH2)4CH2-OCO), 4.22 (d, 4H, 2 ꢄ
CH2OCOO), 4.06 (m, PCL-backbone), 2.30 (m, PCL-backbone),
1.65ꢁ1.33 (PCL-backbone).
Conclusions
In summary, our report has explored various routes to hydrogels
from an organocatalytic ring-opening polymerization approach.
The various constituents used for the cross-linking, i.e. initiator,
comonomer, and (co) cross-linker, may be used in different
proportions in order to affect swelling behavior and tensile
strength of the final network. In addition, selective functional
groups and responsiveness may be added to the material by the
use of functional macro initiators or comonomers. Not only is
our approach simple and versatile, it also provides a simple
alternative to the existing condensation or radical methods to
provide hydrogels.
Polymerization of L-lactide from a PNiPAAm macroinitiator
Experimental part
Hydroxy terminated PNiPAAm macro initiator was azeotropi-
cally distilled from toluene, dried under vacuum, and brought
inside a glove-box. The PNiPAAm, thiourea catalyst (1eq to
PNiPAAm), and sparteine (1eq to PNiPAAm) were dissolved in
dry methylene chloride (3 g methylene chloride/0.5 g polymer)
and left under stirring for 10 min. L-Lactide monomer was added
(ratio dependent on the targeted molecular weight) and the
polymerization left under stirring for 14 h. Benzoic acid (2eq to
PNiPAAm) was added to quench the catalyst and the formed
Materials and instrumentation
For the synthesis of bis-carbonate PEG, and monomeric
carbonate building block see previous work.13 For the synthesis
of PCL see previous work.28 2,2,5-Trimethyl-3-(40-p-hydrox-
ymethyl-phenylethoxy)-4-phenyl-3-azahexan was prepared
according to literature procedures.1 Solvents were dried using
activated alumina columns from Innovative Systems. N-Iso-
propylacrylamide, phenyl isothiocyanate, 5-amino-1-pentanol
2010 | Soft Matter, 2010, 6, 2006–2012
This journal is ª The Royal Society of Chemistry 2010

View Article Online
diblock copolymer was precipitated in cold methanol. The white
precipitate was collected by filtration, rinsed with additional
methanol and dried under vacuum until a constant weight was
reached. 1H-NMR (CDCl3): 6.47 (bs, 1H, PNiPAAm-NH), 5.20
(q, 1H, PLA-CH), 4.0 (bs, 1H, PNIPAAm-CH), 2.20ꢁ1.20
(m, 3H, PNIPAAm-CH2CH), 1.60 (d, 3H, PLA-CH3) 1.13
(s, 6H, PNiPAAm-(CH3)2). GPC: typical PDIꢀ1.10–1.15.
monomers combined). DBU catalyst (2eq. to cross-linker) and
benzyl-2,2-bis(methylol)propionate initiator (for a degree of
polymerization (DP) relative to both monomers) were added and
the dish was sealed and left for a total of 14 h. Following the
reaction benzoic acid (1.2eq relative to DBU) was added to
deactivate the catalyst after which the gel was washed extensively
with further methylene chloride and dried until a constant weight
was reached, typical gel fraction ꢀ85–90%.
Preparation of semi-interpenetrated network. The primary
PEG-network was produced following the general procedure
described here above. TMC (5 equivalents to PEG cross-linker)
and benzyl-2,2-bis(methylol)propionate were used as como-
nomer and initiator, respectively and a degree of polymerization
of 100 was targeted. DBU (2 equivalents to PEG cross-linker)
and TU (10 equivalents to initiator) were used as catalyst and
co-catalyst respectively. Gel fraction of 91.7% was calculated
after extraction of unreactive monomer by swelling in CH2Cl2. In
parallel, P(MTC-pentylphenylurea) was produced in CH2Cl2 for
a targeted DP of 32. The polymerization was initiated by benzyl
alcohol and catalyzed by (ꢁ)-sparteine (5 equivalents to initiator)
in presence of TU cocatalyst (5 equivalents to initiator).
Polymerization was carried out at r.t. for 2 h before to be stopped
Synthesis of phenylureapentanol
In a dry 100 mL round bottom flask equipped with a stir bar was
charged amino pentanol (5.0 g, 48.5 mmol, 1eq). Dry THF
(30 mL) was added and the resulting solution cooled to 0 ꢂC
using an ice bath. A dropping funnel was attached in which
phenylisocyanate (5.19 g, 4.74 mL, 43.6 mmol, 0.9eq) and 30 mL
of dry THF was charged. The resulting solution was added drop
wise during a period of 30 min. The resulting solution was
allowed to warm to ambient temperature and then left under
stirring for an additional 16 h. THF was removed through
rotational evaporation the following morning. The crude
product was recrystallized from ethyl acetate and then stirred
rigorously for an additional 4 h. The solids thus formed were
removed by filtration, washed with further ethyl acetate and
dried until a constant weight was reached, yield 7.0 g (ꢀ80%).
1H-NMR (DMSO-d6) d: 8.19 (s, 1H, NH), 7.39 (d, 2H, ArH),
7.21 (t, 2H, ArH), 6.88 (s, 1H, ArH), 6.10 (t, 1H, NH), 4.40
(t, 1H, OH), 3.40 (q, 2H, CH2), 3.05 (q, 2H, CH2), 2.43 (m, 4H,
CH2), 2.32 (m, 2H, CH2).
by addition of benzoic acid. (Mn NMR ¼ 10 500 g molꢁ1
,
Mw/Mn ¼ 1.19). Semi-interpenetration was performed in THF
by swelling the primary network in a minimal amount of THF to
open the pores. The solution of P(MTC-pentylphenylurea),
previously dissolved in THF was then added over the pre-swollen
sample and the network was allowed to absorb the polymer
ꢂ
solution overnight at 40 C. Then the network was dried under
vacuum and the surface of the gel washed by deionized water.
Synthesis of MTC-pentylphenylurea
MTC-COOH (4.3 g, 26.8 mmol) was initially converted to MTC-
Cl using standard procedures with oxalylchloride. The formed
intermediate was dissolved in 50 mL of dry methylene chloride
and charged in an addition funnel. In a dry 500 mL round
bottom flask equipped with a stir bar was charged phenyl-
ureapentanol (5.55 g, 25 mmol), pyridine (1.97 g, 2.02 mL,
25 mmol) and dry methylene chloride (150 mL). The addition
funnel was attached under nitrogen and the flask cooled to 0 ꢂC
using an ice bath. The MTC-Cl solution was added drop wise
during a period of 30 min and the solution allowed an additional
30 min under stirring. The ice bath was removed and the solution
allowed to gently heat to ambient temperature and left under
stirring for an additional 16 h. The crude product was purified by
column chromatography the following morning using silica gel.
Methylene chloride was initially used as eluent before gently
increasing the polarity finishing with a final concentration of 5
vol% methanol. The product fractions were collected and the
solvent removed through rotational evaporation. The isolated
product was dried under vacuum until a constant weight was
used yielding 8.0 g (ꢀ80%) of an off-white oil which crystallized
Acknowledgements
FN thanks the Swedish Research Council (VR) for financial
support. LM was supported by IBM, the Center for Polymeric
Interfaces and Molecular Assemblies (CPIMA) under NSF
funding, and the Center of Innovation and Research in Materials
ꢀ
and Polymers (CIRMAP). CIRMAP is very grateful to ‘‘Region
Wallonne’’ and European Union (FEDER, FSE) for general
financial support in the frame of Objectif 1-Hainaut: Materia
Nova, as well as to the Belgian Federal Government Office of
Science Policy (SSTC-PAI 6/27). Further support comes from
NSF-GOALI grant (NSF-CHE-0645891). Authors thank Ted-
die Magbitang, Dolores Miller, and Dave Myung for analytical
support, and Debbi Price and Bohringer-Ingelheim for kindly
donating the TMC monomer.
References
1 N. E. Kamber, W. Jeong, R. M. Waymouth, R. C. Pratt,
B. G. G. Lohmeijer and J. L. Hedrick, Chem. Rev., 2007, 107,
5813–5840.
1
upon standing. H NMR (DMSO-d6) d: 8.39 (s, 1H, NH), 7.40
2 K. Fukushima, R. C. Pratt, F. Nederberg, J. P. K. Tan, Y. Y. Yang,
R. M. Waymouth and J. L. Hedrick, Biomacromolecules, 2008, 9,
3051–3056.
3 F. Nederberg, V. Trang, R. C. Pratt, A. F. Mason, C. W. Frank,
R. M. Waymouth and J. L. Hedrick, Biomacromolecules, 2007, 8,
3294–3297.
4 S. H. Kim, J. P. K. Tan, F. Nederberg, K. Fukushima, Y. Y. Yang,
R. M. Waymouth and J. L. Hedrick, Macromolecules, 2009, 42, 25–
29.
(d, 2H, ArH), 7.20 (t, 2H, ArH), 6.88 (t, 1H, ArH), 6.10 (t, 1H,
NH), 4.57 (d, 2H, CH2), 4.39 (d, 2H, CH2), 4.16, t, 2H, CH2),
3.10 (q, 2H, CH2), 1.62 (m, 2H, CH2), 1.45 (m, 2H, CH2), 1.35
(m, 2H, CH2), 1.18 (s, 3H, CH3).
General procedure for gel formation. Bis-carbonate PEG
(1.0g) and TMC were charged in a Petri-dish and dissolved in
methylene chloride to give a final concentration of 25 wt% (both
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2006–2012 | 2011

View Article Online
5 E. F. Connor, G. W. Nyce, M. Myers, A. Mock and J. L. Hedrick,
J. Am. Chem. Soc., 2002, 124, 914–915.
16 K. Matyjaszewski and J. H. Xia, Chem. Rev., 2001, 101, 2921–
2990.
6 G. W. Nyce, T. Glauser, E. F. Connor, A. Mock, R. M. Waymouth
and J. L. Hedrick, J. Am. Chem. Soc., 2003, 125, 3046–3056.
7 O. Coulembier, A. P. Dove, R. C. Pratt, A. C. Sentman, D. A. Culkin,
L. Mespouille, P. Dubois, R. M. Waymouth and J. L. Hedrick,
Angew. Chem., Int. Ed., 2005, 44, 4964–4968.
8 O. Coulembier, B. G. G. Lohmeijer, A. P. Dove, R. C. Pratt,
L. Mespouille, D. A. Culkin, S. J. Benight, P. Dubois,
R. M. Waymouth and J. L. Hedrick, Macromolecules, 2006, 39,
5617–5628.
17 A. S. H. B. D. Ratner, F. J. Schoen, J. E. Lemons, Biomaterials
Science, Academic press, 1st edn, 978-0125824613, 1996.
18 S. Lin-Gibson, R. L. Jones, N. R. Washburn and F. Horkay,
Macromolecules, 2005, 38, 2897–2902.
19 M. P. Lutolf, G. P. Raeber, A. H. Zisch, N. Tirelli and J. A. Hubbell,
Adv. Mater., 2003, 15, 888.
20 M. Malkoch, R. Vestberg, N. Gupta, L. Mespouille, P. Dubois,
A. F. Mason, J. L. Hedrick, Q. Liao, C. W. Frank, K. Kingsbury
and C. J. Hawker, Chem. Commun., 2006, 2774–2776.
21 D. A. Ossipov, S. Piskounova and J. Hilborn, Macromolecules, 2008,
41, 3971–3982.
9 O. Coulembier, L. Mespouille, J. L. Hedrick, R. M. Waymouth and
P. Dubois, Macromolecules, 2006, 39, 4001–4008.
10 S. Csihony, D. A. Culkin, A. C. Sentman, A. P. Dove,
R. M. Waymouth and J. L. Hedrick, J. Am. Chem. Soc., 2005, 127,
9079–9084.
11 A. P. Dove, R. C. Pratt, B. G. G. Lohmeijer, R. M. Waymouth and
J. L. Hedrick, J. Am. Chem. Soc., 2005, 127, 13798–13799.
12 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, P. N. P. Lundberg,
A. P. Dove, H. B. Li, C. G. Wade, R. M. Waymouth and
J. L. Hedrick, Macromolecules, 2006, 39, 7863–7871.
22 M. Malkoch, E. Malmstrom and A. Hult, Macromolecules, 2002, 35,
8307–8314.
23 R. C. Pratt, F. Nederberg, R. M. Waymouth and J. L. Hedrick, Chem.
Commun., 2008, 114–116.
24 X. Yin, A. S. Hoffman and P. S. Stayton, Biomacromolecules, 2006, 7,
1381–1385.
25 D. S. Jones, D. W. J. McLaughlin, C. P. McCoy and S. P. Gorman,
Biomaterials, 2005, 26, 1761–1770.
13 B. G. G. Lohmeijer, R. C. Pratt, F. Leibfarth, J. W. Logan,
D. A. Long, A. P. Dove, F. Nederberg, J. Choi, C. Wade,
R. M. Waymouth and J. L. Hedrick, Macromolecules, 2006, 39,
8574–8583.
14 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, R. M. Waymouth and
J. L. Hedrick, J. Am. Chem. Soc., 2006, 128, 4556–4557.
15 C. J. Hawker and K. L. Wooley, Science, 2005, 309, 1200–1205.
26 L. Mespouille, J.-M. Raquez, C. W. Frank and P. Dubois, Controlled/
Living Radical Polymerization: Progress in ATRP, American
Chemical Society, Washington DC, 2009.
27 Z. G. Tang, R. A. Black, J. M. Curran, J. A. Hunt, N. P. Rhodes and
D. F. Williams, Biomaterials, 2004, 25, 4741–4748.
28 A. W. Bosman, R. Vestberg, A. Heumann, J. M. J. Frechet and
C. J. Hawker, J. Am. Chem. Soc., 2003, 125, 715–728.
2012 | Soft Matter, 2010, 6, 2006–2012
This journal is ª The Royal Society of Chemistry 2010