Guanidine-based polycarbonate hydrogels: From metal-free ring-opening polymerization to reversible self-assembling properties

Article abstract of DOI:10.1039/c1sm05581j

Bio-inspired stabilization of aliphatic polycarbonate-based hydrogels has been carried out by the metal-free ring-opening co-polymerization (ROP) of 6-membered cyclic carbonates containing, respectively, protected guanidine and carboxylic acid functions. Polyethylene glycol (PEG) bound to methylcarboxy trimethylene carbonate at each extremity was used as the cross-linker, and the copolymerizations were performed in CH₂Cl₂ for 24 h in the presence of 1,8-diazabicyclo[5.4.0]-undec-7-ene (DBU) and N-(3,5- trifluoromethyl)phenyl-N′-cyclohexylthiourea (TU), the catalyst and co-catalyst, respectively. Well-defined hydrogels of various compositions and presenting a high gel fraction were obtained. HR-MAS NMR has been successfully employed to validate our purification technique as well as to assess the selective de-protection of the guanidine and carboxylic acid functions. Evidence of self-assembling properties has been attested by differential scanning calorimetry (DSC), HR-MAS NMR analysis and swelling test experiments in aqueous buffered solutions at pH 4 and 8.

Full text of DOI:10.1039/c1sm05581j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Soft Matter
Cite this: Soft Matter, 2011, 7, 9628
www.rsc.org/softmatter
PAPER
Guanidine-based polycarbonate hydrogels: from metal-free ring-opening
polymerization to reversible self-assembling properties†
a
a
b
b
a
Christophe Bartolini, Laetitia Mespouille,* Ingrid Verbruggen, Rudolph Willem and Philippe Dubois
Received 3rd April 2011, Accepted 25th May 2011
DOI: 10.1039/c1sm05581j
Bio-inspired stabilization of aliphatic polycarbonate-based hydrogels has been carried out by the
metal-free ring-opening co-polymerization (ROP) of 6-membered cyclic carbonates containing,
respectively, protected guanidine and carboxylic acid functions. Polyethylene glycol (PEG) bound to
methylcarboxy trimethylene carbonate at each extremity was used as the cross-linker, and the
copolymerizations were performed in CH
2
Cl
2
for 24 h in the presence of 1,8-diazabicyclo[5.4.0]-undec-
0
7
-ene (DBU) and N-(3,5-trifluoromethyl)phenyl-N -cyclohexylthiourea (TU), the catalyst and
co-catalyst, respectively. Well-defined hydrogels of various compositions and presenting a high gel
fraction were obtained. HR-MAS NMR has been successfully employed to validate our purification
technique as well as to assess the selective de-protection of the guanidine and carboxylic acid functions.
Evidence of self-assembling properties has been attested by differential scanning calorimetry (DSC),
HR-MAS NMR analysis and swelling test experiments in aqueous buffered solutions at pH 4 and 8.
Introduction
linear and cyclic polymer mixtures or large discrepancy between
expected and experimental molecular weights.
Aliphatic polycarbonates in their diversities constitute an inter-
esting class of materials with high diversity owing to their high
biocompatibility, reliable (bio)degradability and low toxicity,
making them very suitable for green technology and biomedical
In the recent years, the development of green catalysts for the
ROP of lactones and cyclic carbonates has been spawned by the
will to revolutionize polymer chemistry in order to fulfil the strict
19,20
requirements of microelectronic or biomedical applications.
1
,2
applications (scaffold for tissue engineering or drug carriers).
In this context, enzymatic catalysis of the ROP of cyclic
carbonates has received great attention as a new methodology to
They are obtained either by copolymerization of carbon dioxide
with epoxide or by ring-opening polymerisation (ROP) of cyclic
produce biodegradable polymers with environmental bene-
21,22
1,3,4
carbonate monomers.
While the former approach is CO
2
fits.
However, their low catalytic activities, lack of control
consuming and can alleviate environmental concerns, the
formation of a 5-membered cyclic by-product, the presence of
ether linkage and the use of air-sensitive catalysts hamper the
applicability of the latter approach from an up-scaled synthetic
over molecular weight distributions, the need of a highly stable
support, their non-recyclability and their relatively high cost are
23,24
not favouring their industrial development.
More recently,
metal-free organic catalysts have been developed and applied
successfully to the controlled ROP of lactones and cyclic
25,26
5,6
standpoint.
Beside the facts that only few cyclic carbonates are commer-
cially available, ring-opening polymerization is gaining
increasing attention thanks to its versatility and mild reaction
carbonates. Successful organocatalysts include phosphazene,
27–33
34
iodine trichloride, bifunctional
37–39
N-heterocyclic carbene,
35,36
thiourea-amine,
amidine and guanidine-based catalysts
7–9
10,11
conditions. As a result, cationic, anionic
and coordination–
have emerged as promising polymeri-
zation processes. However, several limitations have been
observed as loss of CO , formation of ether linkages, broad
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).
Amidine and guanidine-based catalysts were found particularly
efficient for the controlled ROP of trimethylene carbonate
12–18
insertion approaches
2
molecular weight distribution, back-biting reactions leading to
40
41
(
TMC) as well as for other functionalized cyclic carbonates,
42,43
allowing to reach complex topologies such as block-
and
44
graft-copolymers presenting good control over molecular
weights and their distribution. These catalytic systems appear as
suitable alternatives to metal-based catalysts for producing
biomaterials, and they have been investigated recently for the
a
Center of Innovation and Research in Materials and Polymers
(CIRMAP), Laboratory of Polymeric and Composite Materials,
University of Mons (UMONS), 20, Place du Parc, B-7000 Mons, Belgium
b
High Resolution NMR Centre (HNMR), Department of Materials and
Chemistry (MACH), Vrije Universiteit Brussel, Pleinlaan 2, B-1050
Brussels, Belgium
45
44
46,47
preparation of nanovectors, nanoparticles or hydrogels.
Hydrogels likewise constitute an emergent class of materials with
high diversity thanks to their biocompatibility, their soft and
†
1
Electronic supplementary information (ESI) available. See DOI:
0.1039/c1sm05581j
9
628 | Soft Matter, 2011, 7, 9628–9637
This journal is ª The Royal Society of Chemistry 2011

View Article Online
1
MHz for H nuclei, with a triple channel dedicated HR-MAS 500
1
SB BL4 probe for measuring H, Sn and C nuclei with
rubbery nature as well as their mechanical properties mimicking
well natural tissues. Nevertheless, their development for tissue
117
13
48
engineering or orthopaedic applications has been limited
a z-gradient coil. The sample was prepared in a 4 mm ZrO rotor
2
18
because of the structural instability of aliphatic polycarbonates.
3
(120 mL) in CDCl . The measurement was done at room
1
Such limitations can be addressed jointly by introducing func-
tionalities in the network able to self-assemble through
secondary interactions and by using living polymerization tech-
niques. In this work, we take advantage of the natural self-
assembling properties of guanidinium functions and carboxylic
acids giving rise to cross-linked aliphatic polycarbonate-based
hydrogels stabilized by secondary interactions. With these
objectives in mind, we present the synthesis of functional cyclic
carbonates bearing BOC-protected guanidines and tert-butyl-
protected carboxylic acids, respectively, using the protocol
temperature at a magic angle spinning rate of 5 KHz. The H HR-
MAS NMR spectra were obtained either in the standard
one-pulse-acquire mode or in the diffusion-filtered mode. In
1
particular, the diffusion-filtered HR-MAS H spectra were
acquired using the ‘‘Longitudinal Eddy Current Relaxation Delay
49
with Bipolar Gradient Pulses’’ (LEDBPP), pulse sequence
available in the TOPSPIN 2.0 pulse sequence library of Bruker as
‘‘ledgp2s1d’’, with 70% of the maximum gradient intensity, using
SIN.100 gradient pulse, according to a protocol previously
50
described. Size exclusion chromatography (SEC) was performed
ꢁ
41
revisited by Pratt et al. and copolymerized by amidine-medi-
ated ROP in the presence of poly(ethylene glycol)-a,u-methyl-
carboxy trimethylene carbonate cross-linker (MTC-PEO-MTC).
Selective de-protection of the above mentioned functions was
carried out in trifluoroacetic acid (TFA) in solution in methylene
chloride in order to investigate possible interactions occurring
between guanidinium and carboxylic acid functions. It is
in CHCl
3
at 30 C, using an Agilent 1200 chromatograph
equipped with a degasser, an isocratic HPLC pump (flow rate ¼
ꢀ1
1 mL min ), an auto-sampler (loop volume ¼ 200 mL, solution
ꢀ1
concentration ¼ 1 mg mL ), refractive-index and UV-Vis
detectors, and three columns: a PL gel 10 mm guard column and
two PL gel Mixed-B 10 mm columns (linear columns for separa-
6
tion of MWPS ranging from 500 to 10 Dalton). Poly(styrene)
1
demonstrated that H HR-MAS NMR is a suitable and flexible
standards were used for calibration. DSC measurements were
performed under nitrogen flow using a DSC apparatus Q2000
from TA Instruments, in Tzero aluminium pan (heat–cool–heat
tool not only to monitor the de-protection of the chemical groups
considered but also to validate the purification technique inves-
tigated. Finally, evidence for self-assembling properties is
ꢁ
ꢀ1
ꢁ
ꢀ1
mode, heating rate 10 C min , cooling rate 5 C min ). FTIR
analysis was carried out with a Tensor 27 FTIR spectrometer
equipped with an ATR from Bruker. Spectra were recorded (from
assessed by DSC, HR-MAS NMR analysis in D
testing at pH 4 and 8.
2
O and swelling
ꢀ1
4
000 to 500 cm ) owing to a single reflection crystal system and
a DTGS detector. MALDI mass spectra were recorded using a
Waters QToF Premier mass spectrometer equipped with
a nitrogen laser, operating at 337 nm with a maximum output of
Experimental part
Materials and instrumentation
ꢀ
2
5
00 J m delivered to the sample in 4 ns pulses at 20 Hz repeating
ꢀ
1
a,u-Hydroxyl polyethylene glycol (M ¼ 2000 g mol , Aldrich)
rate. Time-of-flight mass analyses were performed in the reflection
mode at a resolution of about 10 000. The matrix, trans-2-[3-(4-t-
butyl-phenyl)-2-methyl-2-propenylidene]malononitrile (DCTB),
n
was dried by three successive azeotropic distillations with
0
anhydrous toluene. N-(3,5-Trifluoromethyl)phenyl-N -cyclo-
35
ꢀ1
hexylthiourea (TU) was prepared as previously reported. 2,2-
Bis(methylol)propionic acid (bis-MPA, 98% Aldrich), benzyl
bromide (98% Aldrich), tert-butyl bromoacetate (98% Aldrich),
was prepared in chloroform with a concentration of 20 mg mL .
The matrix solutions (1 mL) were applied to a stainless steel target
and air dried. Polymer samples were dissolved in acetonitrile to
ꢀ1
1
,3-di-boc-2-(2-hydroxyethyl)guanidine (>96%, Fluka), tri-
0
obtain 1 mg mL solutions. 1 mL aliquots of these solutions were
applied onto the target area already charged with the matrix
crystals, and then dried. Finally, 1 mL of a NaCl solution (2 mg
phosgene (98% Aldrich), oxalyl chloride (99% Aldrich), N,N -
dicyclohexylcarbodiimide (DCC, 99% Aldrich), cyclohexylamine
ꢀ1
(
(
99% Aldrich), 3,5-bis(trifluoromethyl)phenyl isothiocyanate
98% Aldrich), Pd/C (Aldrich), trifluoroacetic acid (TFA, 99%
mL in acetonitrile/water, 1/1 v/v) was applied on the target plate.
For recording the single-stage MALDI-MS spectrum, the quad-
rupole (rf-only mode) was set to pass ions from 3500 to 6500 Th,
and all ions were transmitted to the pusher of the time-of-flight
analyser where they were mass analyzed with 1 s integration time.
Data were acquired in the continuum mode until acceptable
averaged data were obtained.
Aldrich), and potassium hydroxide (KOH, Acros) were used as
received. Pyridine (>99% Fiers), triethylamine (>99% Aldrich),
1
,8-diazabicyclo[5.4.0]undec-7-ene (DBU, >99% Fluka) and
(
ꢀ)-sparteine (>99% Aldrich) were all dried over BaO for 48 h,
and the latter two compounds distilled under reduced pressure.
Toluene, tetrahydrofuran and chloroform (p.a. Chemlab) were
dried using activated aluminoxide column from a MBraun
Solvent Purification System (MB-SPS-800), methylene chloride
Synthesis of trimethylenecarbonate-5-methyl-5-
carboxyethylguanidine-1,3-tert-butyloxycarbonyl (MTC-
GuaBOC) 1
(
2 2 2
CH Cl , HPLC grade, Fisher) was dried over CaH for 48 h and
distilled under reduced pressure, heptane (p.a. Chemlab) and
ethyl acetate (Merck), and dimethylformamide (DMF, >99.8%
Aldrich) were used as received.
In a flamed dried round bottom-flask under nitrogen, 5-methyl-
2-oxo-[1,3]dioxane-5-carboxylic acid (MTC-COOH) (1.6 g,
10 mmol) was initially converted to MTC-Cl using standard
procedures with oxalyl chloride (1.4 g, 11 mmol) in 22 mL of
1
H NMR spectra were recorded by using a Bruker AMX-300 or
AMX-500 apparatus at room temperature in CDCl
3
or in DMSO-
51
d
6
(30 mg per 0.6 mL). HR-MAS NMR spectra were acquired on
anhydrous THF. After 1 hour reaction, the reactor is plunged
in an ice bath and a solution of 1,3-di-boc-2-(2-hydroxyethyl)
a Bruker AVANCE II NMR spectrometer resonating at 500.08
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 9628–9637 | 9629

View Article Online
guanidine (3 g, 10 mmol) and triethylamine (2 g, 20 mmol) in
0 mL of anhydrous THF was added dropwise over 30 min.
(1.5 mmol) were introduced. The polymer was dried by
3 successive azeotropic distillations with anhydrous toluene prior
2
ꢁ
The solution obtained was stirred for another 30 min prior to
removing the ice bath, and subsequently the solution stirred for
another 4 hours under nitrogen. The triethylamine salt formed
was filtered off and the solvent evaporated under vacuum. The
crude product was then solubilised in ethyl acetate and purified
by filtration through silica column. Fractions of monomer
identified by TLC were gathered and the solvent eliminated
to being dissolved in 50 mL of anhydrous THF at 40 C. The
polymer solution was then transferred to the anhydride solution
under nitrogen by means of a stainless steel capillary. After 15 h
of reaction at room temperature, the urea salt was removed by
filtration and the PEO cross-linker is recovered by precipitation
1
in 8 volumes of cold diethyl ether. Yield: 94%. H NMR (CDCl
3
)
d: 4.69 and 4.19 (d, 4H, 2 CH
(s, 3H, CH ).
2 2
), 3.64 (t, 4nH, 2nCH ), 1.35
1
under vacuum to yield 1 as white crystals. Yield: 95%. H NMR
3
(
CDCl ) d: 11.5 (s, 1H, NH), 8.65 (t, 1H, NH), 4.70 (d, 2H, CH ),
3 2
4
.31 (t, 2H, CH ), 4.19 (d, 2H, CH ), 3.73 (q, 2H, CH ), 1.47
2 2 2
Synthesis of polycarbonate-based hydrogels by organocatalytic
ROP
(
s, 18H, CH ), 1.36 (s, 3H, CH ).
3 3
In a typical experiment, a glass vial in a glove box, equipped with
Synthesis of trimethylenecarbonate-5-methyl-5-carboxy-tert-
butylacetate (MTC-tBAc) 2
a magnetic stirrer, was charged with 59 mg of 1 (0.13 mmol),
2
88 mg of 2 (0.1) (see Scheme 1), 300 mg of MTC-PEO-MTC
0.13 mmol) and 34 mg of TU (0.09 mmol). Reactants were
dissolved in 0.5 mL of anhydrous CH Cl . In a second glass
(
In a round bottom flask, 2,2-bis(methylol)propionic acid (11.5 g,
8
6 mmol) and potassium hydroxide (4.8 g, 85 mmol) were dis-
ꢁ
2
2
vial,
3
mg of 2,2-bis(methylol)-3-methyl-3-carboxy-benzyl
solved in 70 mL of DMF. The medium was stirred at 100 C for
(
0.013 mmol) and 40 mg of DBU (0.26 mmol) were dissolved in
1
h. Once the medium was homogeneous, 15 mL of tert-butyl
9
2 2
0 mL of CH Cl to reach a final 1 M monomer concentration.
bromoacetate (102 mmol) were added. After 15 h of reaction at
ꢁ
Once the monomer mixture was homogeneous, the initiator
solution was added quickly. The mixture was stirred for a few
seconds and the magnetic stirrer was then removed from the
medium. The polymerization was allowed to occur for 24 h,
subsequently stopped by addition of few drops of benzoic acid
solution in CH Cl . The hydrogel obtained was finally washed by
1
00 C, the KBr salt was filtered off and DMF was evaporated
under vacuum. The 2,2-bis(methylol)-3-methyl-3-carboxy-tert-
butylacetate was then recovered by extraction with CHCl . The
3
solvent was eliminated under vacuum to yield a light yellowish
1
oil. Yield: 99%. H NMR (CDCl ) d: 4.51 (s, 2H, CH ), 3.71
3
2
(
d, 2H, CH ), 3.66 (d, 2H, CH ), 2.44 (OH), 1.38 (s, 9H, CH ),
2
2
2
2
3
swelling in a H O/THF (1/1 v/v) mixture in order to extract
2
1
.10 (s, 3H, CH3).
catalysts and soluble fractions and dried under vacuum until
In a second step, 20 g of 2,2-bis(methylol)-3-methyl-3-car-
boxy-tert-butylacetate (81 mmol) and 250 mL of anhydrous
CH Cl and 80 mL of pyridine (486 mmol) were introduced in
constant weight, F
G
¼ 92%.
2
2
a dried round bottom flask maintained under nitrogen. The
ꢁ
Selective de-protection of guanidinium and carboxylic acid
functions of the hydrogels with trifluoroacetic acid
mixture was cooled down to ꢀ75 C with an acetone/dry ice bath
and a solution of triphosgene (12 g, 45 mmol) in 150 mL of
In a typical experiment, a hydrogel slab was immersed in
2 2
CH Cl was added dropwise in 30 minutes. The medium was
a CH
stirrer. The hydrogel was then thoroughly washed by swelling in
O/THF (1/1 v/v) in order to extract away tert-butanol and
2 2
Cl /TFA mixture (85/15 v/v) and left for 24 h on an orbital
allowed to return to room temperature and a saturated aqueous
solution of NH Cl (200 mL) was added in order to quench tri-
4
H
2
phosgene and stop the reaction. The organic layer was washed 2
times with 200 mL of a 1 M HCl aqueous solution and finally
TFA and was finally dried under vacuum until constant weight.
with 200 mL of saturated solution of NaHCO
was dried over MgSO and the monomer recovered by evapo-
ration of CH Cl under vacuum. The resulting monomer 2 was
recrystallized from diethyl ether. Yield: 28%. H NMR (CDCl
d: 4.71 (d, 2H, CH ), 4.57 (s, 2H, CH ), 4.21 (d, 2H, CH ), 1.47
s, 9H, CH ), 1.41 (s, 3H, CH ).
3
. The organic layer
Swelling experiments
4
2
2
Dry hydrogel slabs were immersed in buffered aqueous solutions
1
3
)
(
pH 4 and pH 8) and withdrawn at determined time intervals,
2
2
2
blotted with tissue paper to absorb excess surface water, and
weighed. The swelling degree was determined at room tempera-
ture as a function of time using the following equation:
(
3
3
Synthesis of poly(ethylene oxide)-a,u-methylcarboxy
w d d
S (%) ¼ (m /m )/m
trimethylene carbonate (MTC-PEO-MTC)
In a flame-dried round bottom flask kept under nitrogen and
ꢁ
in which m and m represent the masses of the wet and dry gels,
w
d
thermostatized at 0 C in an ice bath, MTC-COOH (1.45 g,
respectively.
9
mmol) and 5 mL of anhydrous THF were mixed. Once the
medium was homogeneous, DCC (0.93 g, 4.5 mmol) prelimi-
narily dissolved in a minimal amount of anhydrous THF was
Results and discussion
0
added to the medium. The immediate precipitation of N,N -
Six-membered cyclic carbonates bearing BOC-protected guani-
dicyclohexylurea in the medium indicates the formation of the
anhydride. In a second flame-dried round bottom flask under
dine group 1 and tert-butoxy-protected carboxylic acid 2,
1
respectively, have been synthesized and characterized by
H
nitrogen, 3 g of a,u-hydroxyl polyethylene oxide PEO(OH)
2
NMR (Scheme 1 and Schemes S1 and S2 in the ESI†), using the
9
630 | Soft Matter, 2011, 7, 9628–9637
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Scheme 1 Synthesis of trimethylenecarbonate-5-methyl-5-carboxyethylguanidine-1,3-tert-butyloxycarbonyl (MTC-GuaBOC) 1 and trimethylenecar-
bonate-5-methyl-5-carboxy-tert-butylacetate (MTC-tBAc) 2.
41,52
protocol described by Pratt et al.
Poly(ethylene oxide)-a,u-
methylcarboxy trimethylene carbonate (MTC-PEO-MTC)
modulated from 0 to 90 mol%. All polymerizations were carried
out at room temperature in a sealed glass vial in the glove box
ꢀ1
cross-linker was obtained by esterification of the commercially
with an initial monomer concentration of 1 mol L . After 24 h of
reaction, the catalyst is deactivated by addition of a few drops of
benzoic acid solution in toluene and the swollen network disks
were removed from their container. All polymer networks are
finally purified by immersion in a solvent mixture of distilled
water/THF (1/1 v/v) in order to release catalyst and soluble
fractions (non-reacted monomers, cross-linker and poly-
carbonate oligomers containing or not some PEO grafts).
ꢀ1
available poly(ethylene oxide)-a,u-hydroxyl (M
by 5-methyl-2-oxo-[1,3]dioxane-5-carboxylic acid (MTC-
n
¼ 2 kg mol )
0
COOH) preliminarily converted into anhydride using the N,N -
dicyclohexylcarbodiimide (DCC) coupling agent in anhydrous
THF. After 15 h of reaction, the urea salt was removed by
filtration and the PEO cross-linker recovered by selective
1
precipitation in cold diethylether. H NMR data in CDCl
3
reveal
the presence of methylene oxocarbonate protons at 4.2 and
.7 ppm while quantitative esterification of PEO was further
Finally, gel fractions (F
at room temperature for 24 h were determined by weighing the
insoluble parts (w ) after solvent evaporation and knowing the
initial weight (w ) of co-monomers introduced: F /w
¼ w
G
) of the hydrogels dried under vacuum
4
evidenced by MALDI-TOF where only a population corre-
+
sponding to [MTC-PEO-MTCNa] was observed. No trace of
g
p
G
g
p
.
PEO(OH)
2
or mono-esterified HO-PEO-MTC were detected in
Table 1 gathers the initial molar composition and the gel frac-
tions obtained for each sample. It turns out that, whatever the
targeted composition, monolithic and transparent hydrogels
with FG as high as 92% or more were obtained. In order to
validate our purification technique and make sure that no non-
cross-linked monomers remained entrapped in the polymer
the mass spectrum (Scheme S3, ESI†).
In a second set of experiments, cross-linked hydrogels were
produced by ring-opening copolymerization of 1, 2 and MTC-
PEO-MTC initiated by 2,2-bis(methylol)-3-methyl-3-carboxy-
2 2
benzyl in CH Cl using DBU and TU as a catalyst and
1
co-catalyst, respectively. The choice for this catalytic system was
motivated by its high selectivity for ring-opening polymerization
at the expense of non-desired transesterification reactions. This
control stems from the high selectivity of this catalytic system
toward the strained cyclic carbonate of the monomer relative to
the acyclic carbonate and the ester moieties of the polymer
chains. Therefore, very good control over macromolecular
parameters can be achieved with molar molecular weight
depending on the initial monomer to –OH initiator ratio. The
initial [1 + 2] /[MTC-PEO-MTC] /[I] /[DBU] /[TU] molar ratio
network, diffusion-filtered HR-MAS H NMR analysis spectra
3
were recorded on each network swollen in CDCl . This technique
proved very efficient to check that release of all non-grafted
molecules was complete. Indeed, as demonstrated previously,
application of an appropriate diffusion-filtered pulse sequence
(Longitudinal Eddy Current Relaxation Delay with Bipolar
49,50
Gradient Pulses’’ (LEDBPP), see Experimental section)
reveals for each purified hydrogel sample the absence of reso-
nances arising from the monomer or catalyst species, the reso-
1
nances of which are spectroscopically suppressed from the H
0
0
0
0
0
was fixed to 90/10/1/20/7, which leads to a total degree of poly-
merization of 100. A molar fraction in cross-linker was fixed to
NMR spectrum on the basis of their translational mobility upon
application of the above LEDBPP pulse sequence. As isotropic
rotational mobility of the chemical moieties is a necessary
1
0 mol% for each sample while the ratio of 1 and 2 was
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 9628–9637 | 9631

View Article Online
Table 1 Initial monomer composition and gel fraction (F
hydrogels obtained by ROP catalyzed by DBU/TU and initiated by 2,2-
bis(methylol)-3-methyl-3-carboxy-benzyl in CH Cl at r.t. for 24 h
G
) of the
In order to promote self-assembling within the network and
strengthen the mechanical properties of the resulting hydrogels,
de-protecting the guanidine and carboxylic acid functions is
required. For this purpose, model de-protection reactions were
2
2
Hydrogels 1 (mol%) 2 (mol%) MTC-PEO-MTC(mol%)
G
F (%)
2 2
first performed on the respective homopolymers in a CH Cl /
H
H
H
H
H
90–0
0
10
30
50
90
90
80
60
40
0
10
10
10
10
10
98
92
96
98
92
TFA mixture solution (85/15, v/v) for 24 h at room temperature.
80–10
60–30
40–50
0–90
The effectiveness of the de-protection of guanidine and
1
carboxylic acid was evidenced by H NMR analyses in DMSO-d
6
revealing that at most 1 mol% of the residual tert-butoxy methyl
proton resonance at 1.48 ppm was left for the homopolycar-
bonate guanidine, while the homopolycarbonate carboxylic acid
was found to be fully de-protected. It is worth mentioning that
this de-protection is selective for tert-butoxy functions and does
not affect the polycarbonate backbone, as further confirmed by
4
9,50,53
condition for HR-MAS NMR to be applicable,
rotation-
ally mobility-refrained functions (e.g. methylene protons from
the polycarbonate backbone) also merge into the baseline.
Taking into account the almost quantitative gel fraction and the
1
H NMR spectra of the homopolymers from which an excellent
agreement between expected and experimental degrees of poly-
merizations was obtained. This procedure was applied to
hydrogel slabs of known weight by immersion in CH Cl /TFA
1
effective purification step as assessed by HR-MAS H NMR, it
can be stated that the final composition of the hydrogels is similar
to the composition in the feed.
2
2
mixtures (85/15, v/v) (Scheme 2) for 24 h at room temperature
and gentle shaking on an orbital shaker. The hydrogels were then
For the sake of clarity, all samples are depicted by the symbol
H
x–y (with H for ‘‘Hydrogel’’) with in subscript the respective
thoroughly washed in a mixture of THF/H O (50/50 v/v) for 48 h
2
composition in monomer 1 (y) and 2 (x).
at room temperature and dried under vacuum. De-protected
The transparency of the hydrogels in the dried state is note-
worthy, even though the weight fraction in PEO is ranging from
hydrogels are denoted H dep where x and y stand for the
xꢀy
molar composition in carboxylic acid and guanidine carbonate,
3
8 to 44%, indicating that PEO is not able to crystallize within
respectively, and dep for de-protected. Diffusion filtered HR-
1
MAS H NMR spectra in CDCl revealed the total absence of
the network. This assumption was confirmed by DSC analysis in
which all thermograms obtained after the second scan under
3
resonance arising from the rotationally mobile BOC and tert-
butyl groups usually observed around 1.5 ppm, therefore con-
firming the complete de-protection of the gels. The weight of each
de-protected sample is in satisfactory agreement with the
expected one, evidencing not only the full de-protection of the
inert atmosphere (N
2
) revealed the absence of PEO melting point
ꢁ
usually observed at 50 C (Fig. 1). This absence of crystallinity
can be ascribed to a homogeneous distribution of the PEO cross-
linker within the polymer network.
nd
2
Fig. 1 DSC thermograms of PEO (2 K) and hydrogel H40–50 recorded after the 2 scan under N .
9
632 | Soft Matter, 2011, 7, 9628–9637
This journal is ª The Royal Society of Chemistry 2011

View Article Online
2 2
Scheme 2 De-protection of guanidine and carboxylic acid functions within the polymer networks by trifluoroacetic acid (TFA) in CH Cl solution at
room temperature.
aforementioned functionalities but also the absence of hydrogel
damages in such an acidic environment. Differential scanning
calorimetry (DSC) analyses as performed under nitrogen on the
de-protected hydrogels revealed an increase of the glass transi-
Interestingly as revealed by Fig. 3, HR-MAS NMR analysis
performed on de-protected hydrogels in D O reveals that upon
2
increasing the guanidine composition, the oxocarbonyl carbonyl
methylene proton signal shifts down from 4.77 to 4.68 ppm near
the stoichiometric composition, which can reasonably be
ascribed to a specific interaction between the aforementioned
chemical groups.
g
tion temperature, T , indicating a modification of the nature of
the hydrogel samples (Table 2).
The presence of selective interactions occurring inside the de-
protected polymer networks has been first evidenced by differ-
ential scanning calorimetry (DSC). Indeed, it was already
reported that macromolecular interactions in miscible polymer
blends influence the composition dependence of the glass tran-
Selective de-protection was also evidenced by swelling exper-
iments in demineralised water and comparison of the swelling
profile of the de-protected and protected hydrogels. Practically,
dry hydrogel slabs were immersed in a large volume of demin-
eralised water at room temperature and withdrawn at deter-
mined time intervals, blotted with tissue paper to absorb the
sition temperature, T , and the increment of heat capacity,
g
5
4,55
DC
display a linear dependence on the composition. In strongly
interacting blends, T and DC present a positive or a negative
deviation from linearity, with the T lying no longer between
p
.
g p
In a non-interacting miscible system, the T and DC
g
p
g
those of the pure components. In this work, each hydrogel
sample was characterized by DSC under nitrogen using a
ꢁ
ꢀ1
heat–cool–heat mode (heating rate: 10 C min ; cooling rate:
ꢀ1
ꢁ
5
C min ). Each thermogram exhibits not only a single T ,
g
evidencing the good miscibility between the PEO cross-linker
and the functional polycarbonate backbone, but also a positive
deviation from linearity in the plot of T against composition,
g
proving unambiguously the presence of segmental interactions
54,55
occurring into the network (Fig. 2).
g
The maximum T value
being reached around the stoichiometric composition of guani-
dine and carboxylic acid functions, it can be assumed that the
positive T
g
deviation is the result of secondary interactions
g
Fig. 2 Glass transition temperature (T ) of the de-protected hydrogels
against their molar composition.
bridging guanidine and carboxylic acid moieties.
Table 2 Glass transition temperature of the protected and de-protected hydrogels determined by DSC under nitrogen in a heat–cool–heat mode
ꢀ1
ꢁ
ꢀ1
ꢁ
(
heating rate: 10 C min , cooling rate 5 C min ) and the conversion of de-protection of the polymer networks
ꢁ
ꢁ
a
Conversion of de-protection (%)
Hydrogels
T
g
/ C
De-protected hydrogels
T
g
/ C
H
H
H
H
H
90–0
ꢀ22.4
ꢀ19.3
ꢀ2.3
H
H
H
H
H
90–0dep
80–10dep
60–30dep
40–50dep
0–90dep
ꢀ19.7
ꢀ6.4
0
100
100
100
100
100
80–10
60–30
40–50
0–90
ꢀ9.7
3
ꢀ23.8
ꢀ4.25
a
1
3
As determined by diffusion filtered HR-MAS H NMR in CDCl , see text.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 9628–9637 | 9633

View Article Online
Fig. 3 Chemical shift of the oxocarbonyl carbonyl methylene proton (H
a
) resonance with the molar composition of the gel in D
2
O measured by HR-
MAS NMR in a diffusion filter pulse sequence.
excess of water at surface and weighed. The swelling degree was
in Fig. 5, the gel presents exactly the same swelling profile and
reaches the same degree of swelling at equilibrium, demon-
strating the stability of the network towards hydrolysis in
deionised water.
determined as
relationship:
a function of time, using the following
S (%) ¼ (m
w
ꢀ m
d
)/m
d
Self-assembling properties between carboxylic acid and gua-
nidinium functions have also been evidenced by swelling test,
both in acetate buffer (pH 4, ionic strength: 0.20) and in phos-
phate buffer (pH 8, ionic strength: 0.27). In Fig. 6 the swelling
degree at equilibrium reached pH 4 and pH 8 and is plotted as
a function of the hydrogel composition.
w d
with m and m the mass of the wet and dry gel, respectively.
All tests were performed in triplicate. Fig. 4 presents the
swelling profile of gels H40–50 and H40–50dep in deionised water at
r.t. As clearly evidenced, the de-protection of carboxylic acid and
guanidinium functions led to a significant increase of water
As expected, the hydrogel containing only carboxylic acid
functions (H90–0dep) presents a higher degree of swelling at
equilibrium in the basic condition. Indeed, at pH 4 and therefore
uptake, i.e., S : 175.7%, when compared to the protected
eq
hydrogel (S : 62.4%), owing to the de-protection of highly
eq
hydrophilic groups.
below the pKa of the carboxylic acid (pK : 4.75), all acidic
a
In order to rule out possible hydrolysis of carbonate links of
hydrogels, that could likewise be responsible for water uptake
increase, the reproducibility of the swelling behaviour of
hydrogels H40-50 in deionised water was assessed. As displayed
functions are protonated and can form H-bonds in between
themselves. These secondary interactions constitute additional
cross-linking points, triggering a reduced swelling degree (Seq
H90–0dep ¼ 71%, Scheme 3A). In contrast, at pH 8, all functions
9
634 | Soft Matter, 2011, 7, 9628–9637
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 4 Swelling profile of H40–50 (red curve) and H40–50dep (blue curve) in deionised water at r.t. Mean relative error over S
t
¼ ꢂ4%.
are in anionic carboxylate form. The resulting electrostatic
repulsion occurring between negative charges is responsible for
the gel expansion and the significantly larger water uptake (Seq
stressed that the hydrogels were found to be very unstable at pH
8 resulting in H0–90 dep to completely degrade within 24 h. This
was ascribed to the hydrolysis of carbonate links under basic
conditions. Between these two extreme compositions, interesting
swelling behaviour of the gels has been observed. At pH 4,
carboxylic acid functions are protonated, allowing H-bonding to
occur between acidic functions and possibly also with neigh-
bouring PEO fragments. As the molar composition in guanidi-
nium cations increases, carboxylic acid functions can establish
H-bonds with guanidinium ions as well, explaining the smooth
increases of Seq. Beyond the stoichiometric composition, the
additional cationic guanidinium moieties are not stabilized by
H-bonding and the electrostatic repulsions present in the
network contribute to the large increase of Seq. At pH 8, the
carboxylic acid functions are fully de-protonated to carboxylates
H
90–0dep ¼ 680%, Scheme 3B). The hydrogel containing 100% of
guanidine functions (H0–90 dep) displays a very important degree
of swelling at pH 4, at which all guanidine moieties are proton-
ated to guanidinium cations. Similarly to the gel H90–0dep, the gel
H0–90dep swells considerably to alleviate the electrostatic repul-
sion between cationic charges (Seq ¼ 1221%). At pH 8, a clear
decrease of Seq is measured (H0–90dep : S ¼ 586%). While the
eq
56
a
reported pK value of guanidine ions is 13.6, no value for
guanidinium alkyl derivatives is, to the best of our knowledge,
available in the literature. Acetyl guanidine is meanwhile char-
5
6
acterized by a pK
a
of 8.3. Therefore, and based on the hydrogel
is close to 8–9. It is
behaviour, it can be assumed that the pK
a
Fig. 5 Reproducibility of the swelling behaviour of hydrogel H40–50dep in deionised water. Mean relative error over S
t
¼ ꢂ4%.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 9628–9637 | 9635

View Article Online
Fig. 6 Effect of the molar composition of de-protected hydrogels on the degree of swelling at equilibrium (Seq) at pH 4 (red curve) and pH 8 (blue
curve). Mean relative error over S
t
¼ ꢂ6%.
Scheme 3 Schematic representation of the swelling behaviour of carboxylic acid-containing hydrogels at (A) pH 4 and (B) pH 8.
even though guanidine functions are partially protonated. When
the hydrogel composition tends to the stoichiometric 1/1 ratio,
cationic and anionic charges are in balance and lead to neutral-
ization of the hydrogel, and therefore to a decrease of the degree
of swelling. These swelling tests in acidic and basic conditions
demonstrate not only the adaptive properties of these hydrogels
but also confirm the presence of secondary interactions occurring
inside the polymer networks.
within the network. Moreover, it appeared that the maximum of
the deviation was close to the 1/1 stoichiometric composition in
carboxylic acid and guanidine functions. This self-assembling
2
property was further supported by HR-MAS NMR in D O with
a composition dependent shift of the methylene oxocarbonyl
carbonyl proton resonance. Last but not least, swelling testing in
buffer aqueous solution of pH 4 and pH 8 allowed us to provide
some support to the stabilisation of the network either by
H-bonding or by electrostatic interactions. This work opens the
way to new polycarbonate-based materials with improved
performances allowing development of new materials for tissue
engineering or drug delivery systems.
Conclusion
The synthesis of well-defined polycarbonate hydrogels contain-
ing lateral groups able to self-assemble through secondary
interactions has been carried out with success by metal-free ring-
opening polymerization. HR-MAS NMR was found to be an
efficient tool to discriminate between rotationally mobile groups
like tert-butyl functions and constrained functions as the poly-
carbonate backbone. Even better, this technique enabled us to
assess our purification protocol of hydrogels since no signal
arising from translationally mobile non-reacted monomers or
free catalyst was detected. Selective de-protection of tert-butyl
and BOC groups was achieved successfully in TFA/CH Cl
Acknowledgements
This work was supported by the European Commission and
Region Wallonne (FEDER Program) in the frame of Materia
2
Nova and OPTI MAT program of excellence, by the Interuni-
versity Attraction Pole program of the Belgian Federal Science
Policy Office (PAI 6/27) and FNRS-FRFC. Dr Pascal Gerbaux,
FNRS Research associate, and Dr Julien De Winter, FNRS
fellow, are thanked for the mass analysis experiments. R.W. is
indebted to the Fund for Scientific Research-Flanders (FWO,
Belgium) for financial support (Grant G.0064.07) and to the
Research Council (Onderzoeksraad) of the Vrije Universiteit
Brussel for financial support (Concerted Research Action, Grant
GOA31 and other grants).
2
2
(
15/85) solution without any alteration of the polycarbonate
backbone as supported by NMR. Finally, self-assembling of
carboxylic acid and guanidine functions has been fully evidenced
first by DSC analysis where a positive deviation from linearity of
g
the plot of T versus composition was observed, proving unam-
biguously the presence of segmental interactions occurring
9
636 | Soft Matter, 2011, 7, 9628–9637
This journal is ª The Royal Society of Chemistry 2011

View Article Online
3
1 O. Coulembier, L. Mespouille, J. L. Hedrick, R. M. Waymouth and
P. Dubois, Macromolecules, 2006, 39, 4001.
References
1
2
G. Rokicki, Prog. Polym. Sci., 2000, 25, 259.
L. B. Lu and K. L. Huang, J. Polym. Sci., Part A: Polym. Chem.,
32 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.
2005, 43, 2468.
3
4
5
B. Ochiai and T. Endo, Prog. Polym. Sci., 2005, 30, 183.
D. J. Darensbourg, Chem. Rev., 2007, 107, 2388.
M. Kroger, C. Folli, O. Walter and M. Doring, Adv. Synth. Catal.,
33 N. E. Kamber, W. Jeong, S. Gonzalez, J. L. Hedrick and
R. M. Waymouth, Macromolecules, 2009, 42, 1634.
34 O. Coulembier, F. Meyer and P. Dubois, Polym. Chem., 2010, 1, 434.
35 A. P. Dove, R. C. Pratt, B. G. G. Lohmeijer, R. M. Waymouth and
J. L. Hedrick, J. Am. Chem. Soc., 2005, 127, 13798.
36 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.
37 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.
38 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, R. M. Waymouth and
J. L. Hedrick, J. Am. Chem. Soc., 2006, 128, 4556.
39 L. Zhang, R. C. Pratt, F. Nederberg, H. W. Horn, J. E. Rice,
R. M. Waymouth, C. G. Wade and J. L. Hedrick, Macromolecules,
2010, 43, 1660.
2
006, 348, 1908.
R. C. Jeske, J. M. Rowley and G. W. Coates, Angew. Chem., Int. Ed.,
008, 47, 6041.
H. R. Kricheldorf, R. Dunsing and A. S. I. Albet, Makromol. Chem.,
987, 188, 2453.
6
7
8
9
2
1
A. C. Albertsson and M. Sjoling, J. Macromol. Sci., Part A: Pure
Appl. Chem., 1992, 29, 43.
T. Ariga, T. Takata and T. Endo, Macromolecules, 1997, 30, 737.
1
1
1
1
1
1
1
1
0 J. Matsuo, K. Aoki, F. Sanda and T. Endo, Macromolecules, 1998, 31,
432.
1 H. Keul, R. B €a cher and H. H o€ cker, Makromol. Chem., 1986, 187,
579.
2 H. R. Kricheldorf, J. Jenssen and I. Kreisersaunders, Makromol.
Chem., 1991, 192, 2391.
4
2
40 F. Nederberg, B. G. G. Lohmeijer, F. Leibfarth, R. C. Pratt, J. Choi,
A. P. Dove, R. M. Waymouth and J. L. Hedrick, Biomacromolecules,
2007, 8, 153.
3 K. R. Carter, R. Richter, H. R. Kricheldorf and J. L. Hedrick,
Macromolecules, 1997, 30, 6074.
4 P. Dobrzynski and J. Kasperczyk, J. Polym. Sci., Part A: Polym.
Chem., 2006, 44, 3184.
5 M. Helou, O. Miserque, J. M. Brusson, J. F. Carpentier and
S. M. Guillaume, ChemCatChem, 2010, 2, 306.
6 H. R. Kricheldorf and A. Stricker, Macromol. Chem. Phys., 1999, 200,
41 R. C. Pratt, F. Nederberg, R. M. Waymouth and J. L. Hedrick, Chem.
Commun., 2008, 114.
42 L. Mespouille, F. Nederberg, J. L. Hedrick and P. Dubois,
Macromolecules, 2009, 42, 6319.
43 S. Tempelaar, L. Mespouille, Ph. Dubois and A. P. Dove,
Macromolecules, 2011, 44, 2084.
44 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.
1726.
7 H. R. Kricheldorf, G. Behnken, G. Schwarz, P. Simon and
M. Brinkmann, J. Macromol. Sci., Part A: Pure Appl. Chem., 2009,
46, 353.
1
1
8 T. F. Al-Azemi and K. S. Bisht, Polymer, 2002, 43, 2161.
45 C. B. Cooley, B. M. Trantow, F. Nederberg, M. K. Kiesewetter,
J. L. Hedrick, R. M. Waymouth and P. A. Wender, J. Am. Chem.
Soc., 2009, 131, 16401.
46 F. Nederberg, V. Trang, R. C. Pratt, S. H. Kim, J. Colson, A. Nelson,
C. W. Frank, J. L. Hedrick, P. Dubois and L. Mespouille, Soft
Matter, 2010, 6, 2006.
47 F. Nederberg, V. Trang, R. C. Pratt, A. F. Mason, C. W. Frank,
R. M. Waymouth and J. L. Hedrick, Biomacromolecules, 2007, 8,
3294.
9 N. E. Kamber, W. Jeong, R. M. Waymouth, R. C. Pratt,
B. G. G. Lohmeijer and J. L. Hedrick, Chem. Rev., 2007, 107, 5813.
0 D. Bourissou, S. Moebs-Sanchez and B. Martin-Vaca, C. R. Chim.,
2
2
2
2007, 10, 775.
1 K. S. Bisht, Y. Y. Svirkin, L. A. Henderson, R. A. Gross,
D. L. Kaplan and G. Swift, Macromolecules, 1997, 30, 7735.
2 I. K. Varma, A. C. Albertsson, R. Rajkhowa and R. K. Srivastava,
Prog. Polym. Sci., 2005, 30, 949.
2
2
3 J. Feng, F. He and R. X. Zhuo, Macromolecules, 2002, 35, 7175.
4 R. A. Gross, B. Kalra and A. Kumar, Appl. Microbiol. Biotechnol.,
48 S. Anseth Kristi, C. Svaldi Dina, T. Laurencin Cato and R. Langer, in
Photopolymerization, American Chemical Society, 1997, vol. 673,
p. 189.
2001, 55, 655.
2
2
2
2
2
5 L. Zhang, F. Nederberg, J. M. Messman, R. C. Pratt, J. L. Hedrick
and C. G. Wade, J. Am. Chem. Soc., 2007, 129, 12610.
6 L. Zhang, F. Nederberg, R. C. Pratt, R. M. Waymouth, J. L. Hedrick
and C. G. Wade, Macromolecules, 2007, 40, 4154.
7 E. F. Connor, G. W. Nyce, M. Myers, A. Mock and J. L. Hedrick,
J. Am. Chem. Soc., 2002, 124, 914.
8 G. W. Nyce, T. Glauser, E. F. Connor, A. Mock, R. M. Waymouth
and J. L. Hedrick, J. Am. Chem. Soc., 2003, 125, 3046.
49 R. Warrass, J. M. Wieruszeski and G. Lippens, J. Am. Chem. Soc.,
1999, 121, 3787.
50 S. Iqbal, F. Rodriguez-Llansola, B. Escuder, J. F. Miravet,
I. Verbruggen and R. Willem, Soft Matter, 2010, 6, 1875.
51 P. A. Wender, E. Kreider, E. T. Pelkey, J. Rothbard and
C. L. VanDeusen, Org. Lett., 2005, 7, 4815.
52 D. P. Sanders, K. Fukushima, D. J. Coady, A. Nelson, M. Fujiwara,
M. Yasumoto and J. L. Hedrick, J. Am. Chem. Soc., 2010, 132, 14724.
53 J. C. Martins, F. A. G. Mercier, A. Vandervelden, M. Biesemans,
J. M. Wieruszeski, E. Humpfer, R. Willem and G. Lippens, Chem.–
Eur. J., 2002, 8, 3431.
9 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.
3
0 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,
54 M. Song, D. J. Hourston, H. M. Pollock and A. Hammiche, Polymer,
1999, 40, 4763.
55 T. K. Kwei, J. Polym. Sci., Polym. Lett. Ed., 1984, 22, 307.
56 A. Albert, R. Goldacre and J. Phillips, J. Chem. Soc., 1948, 2240.
5617.
This journal is ª The Royal Society of Chemistry 2011
Soft Matter, 2011, 7, 9628–9637 | 9637