Zwitterionic hydrogels crosslinked with novel zwitterionic crosslinkers: Synthesis and characterization

Article abstract of DOI:10.1016/j.polymer.2011.04.056

Two novel zwitterionic sulfobetaine dimethacrylate crosslinkersN,N- bis(methacryloxyethyl)-N-methyl-N-(3-sulfopropyl)ammonium (CL1) and N,N-bis(methacryloxyethyl)-N-methyl-N-(4-sulfobutyl)ammonium (CL2) betaines were synthesized and used for preparation of zwitterionic hydrogels formed from N-(methacryloxyethyl)-N,N-dimethyl-N-(3-sulfopropyl)ammonium betaine (SBDMA) via redox-initiated free-radical polymerization. The commercially available crosslinkers N,N′-methylene bisacrylamide (BIS) and ethylene glycol dimethacrylate (EDMA) were also used. Equilibrium water content, sorption degree, diffusion coefficient of water, state of water, degree of crosslinking and mechanical properties were determined for hydrogels crosslinked using different crosslinking conditions. A minor difference in the spacer length between the charged moieties in CL1 and CL2 crosslinkers, respectively, was shown to influence the hydrogel properties. The CL1 and CL2 crosslinkers with chemical structure similar to SBDMA resulted in hydrogels with higher stiffness, mechanical strength and crosslink density compared to hydrogels crosslinked by BIS and EDMA. This difference was assigned to suppression of the compositional drift during the hydrogel formation when crosslinkers with chemical structure similar to monomer were used. PolySBDMA hydrogels exhibited a low adhesion of RAT-2 fibroblasts-like cells.

Full text of DOI:10.1016/j.polymer.2011.04.056

Polymer 52 (2011) 3011e3020
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Zwitterionic hydrogels crosslinked with novel zwitterionic crosslinkers: Synthesis
and characterization
*
*
Peter Kasák , Zuzana Kroneková, Igor Krupa, Igor Lacík
Polymer Institute of the Slovak Academy of Sciences, Dúbravská cesta 9, 845 41 Bratislava, Slovakia
a r t i c l e i n f o
a b s t r a c t
Article history:
Two novel zwitterionic sulfobetaine dimethacrylate crosslinkersN,N-bis(methacryloxyethyl)-N-methyl-
N-(3-sulfopropyl)ammonium (CL1) and N,N-bis(methacryloxyethyl)-N-methyl-N-(4-sulfobutyl)ammo-
nium (CL2) betaines were synthesized and used for preparation of zwitterionic hydrogels formed from
N-(methacryloxyethyl)-N,N-dimethyl-N-(3-sulfopropyl)ammonium betaine (SBDMA) via redox-initiated
free-radical polymerization. The commercially available crosslinkers N,N0-methylene bisacrylamide (BIS)
and ethylene glycol dimethacrylate (EDMA) were also used. Equilibrium water content, sorption degree,
diffusion coefficient of water, state of water, degree of crosslinking and mechanical properties were
determined for hydrogels crosslinked using different crosslinking conditions. A minor difference in the
spacer length between the charged moieties in CL1 and CL2 crosslinkers, respectively, was shown to
influence the hydrogel properties. The CL1 and CL2 crosslinkers with chemical structure similar to
SBDMA resulted in hydrogels with higher stiffness, mechanical strength and crosslink density compared
to hydrogels crosslinked by BIS and EDMA. This difference was assigned to suppression of the compo-
sitional drift during the hydrogel formation when crosslinkers with chemical structure similar to
monomer were used. PolySBDMA hydrogels exhibited a low adhesion of RAT-2 fibroblasts-like cells.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 6 December 2010
Received in revised form
20 April 2011
Accepted 26 April 2011
Available online 5 May 2011
Keywords:
Polysulfobetaine hydrogel
Synthesis of sulfobetaine crosslinker
Compositional drift
1. Introduction
crosslinking agents for methacrylate network formation, although
their application may be limited due to low solubility in water [6].
Hydrogels assisting the life sciences in diverse areas of biomed-
icine and biotechnology represent a significant topic in current
polymer science [1e3]. Applications of polymeric hydrogels include
such areas as implants, medical devices and diagnostics, drug
release, scaffolds and tissue engineering, enzyme and cell engi-
neering and others.
Hydrogels based on synthetic polymers are made of mixtures of
monomers and crosslinking agents, with the polymerization process
typically carried out to the complete conversion. The chemical
similarity of crosslinker and monomer is not always obeyed and,
consequently, their different rates of incorporation to the polymeric
network lead to compositional drift during hydrogel formation. This
may negatively influence hydrogel properties due to the lack of
control of the crosslinking process, which increases the heteroge-
neity of hydrogel network [4,5].
N,N0-methylene bisacrylamide (BIS) is another regularly used
crosslinking reagent. However, in application with methacrylic
monomers the compositional drift may also impact the hydrogel
formation due to different reactivity ratios between acrylate and
methacrylate families [5,7]. In some cases, commercially available
crosslinkers are not suitable for hydrogel network formation and
then the synthesis of new crosslinkers is required [4,5,8]. For
example, a hydrophilic crosslinker with a linkage similar to the
phosphorylcholine structure was synthesized in order to prepare
a hydrogel with methacrylic monomer containing the phosphor-
ylcholine moiety [4]. The resulting hydrogel showed improved
mechanical properties compared to the hydrogel prepared in the
presence of BIS [6]. In the case of poly(N-vinyl pyrrolidone) [8], the
pyrrolidone-based crosslinker was synthesized with the aim to
eliminate the inconsistencies in the hydrogel properties due to
differences in the reactivity ratios between N-vinyl pyrrolidone and
dimethacrylate crosslinker. These examples demonstrate that the
selection of crosslinker is crucial in the strategy used for preparation
of hydrogels and in controlling their properties and performance.
Zwitterionic polymers represent a strongly developing class of
polymers. They are based on the electrically neutral monomer units
Ethylene glycol dimethacrylate (EDMA) and triethylene glycol
dimethacrylate belong to frequently used commercially available
* Corresponding authors. Tel.: þ421 2 5477 2467; fax: þ421 2 5477 5923.
E-mail addresses: peter.kasak@savba.sk (P. Kasák), igor.lacik@savba.sk (I. Lacík).
0032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2011.04.056

3012
P. Kasák et al. / Polymer 52 (2011) 3011e3020
that formally contain both positive and negative charges on different
atoms in a monomer unit. Zwitterionic polymers are of biomimetic
character, which makes them suitable for designing non-biofouling
materials and surfaces [9e13] with unique physical and chemical
properties [14e16]. Zwitterionic monomers are based on various
chemical structures and include, most typically, sulfobetaines, car-
boxybetaines and phosphorylbetaines. Among them, the sulfobetaine
monomer (N-(methacryloxyethyl)-N,N-dimethyl-N-(3-sulfopropyl)
ammonium betaine), SBDMA, is frequently used due to its low price,
commercial availabilityand simple synthetic accessibility [17,18]. This
monomer has been used in studies devoted to solution properties
[15], kinetics of polymerization [19], reduced protein adsorption
[20,21], blood biocompatibility [22,23] and in vivo tests [24].
Owing to the significance of zwitterionic hydrogels, we decided
to synthesize novel water soluble methacrylate crosslinkers based
on the sulfobetaine methacrylate monomer unit with the chemical
structure matching the structure of SBDMA monomer. This
approach should suppress the eventual compositional drift during
the hydrogel formation. The aim was to evaluate the principal
physico-chemical properties of the resulting hydrogels and to
estimate whether this strategy may lead to enhancement of these
properties. It should be mentioned that in parallel to our work,
a similar strategy was published on polycarboxybetaine hydrogels
by Carr et al. [5]. This paper demonstrated that using a structurally
similar crosslinker to the monomer unit was beneficial for both
non-biofouling and mechanical properties, which was discussed in
terms of the hydrophilicity and homogeneity of the hydrogel
network.
-
SO3
O
+
N
O
SBDMA
-
SO3
O
+
N
O
O
O
CL1
-
SO3
O
+
N
O
O
O
CL2
Newly designed crosslinkers N,N-bis(methacryloxyethyl)-N-
methyl-N-(3-sulfopropyl) ammonium betaine (CL1) and
N,N-bis(methacryloxyethyl)-N-methyl-N-(4-sulfobutyl) ammo-
nium betaine (CL2) are shown in Fig. 1. They are structurally
similar to SBDMA monomer also shown in Fig. 1. These cross-
linkers differ only in the length of spacer, i.e. propyl vs butyl,
separating the sulfate and ammonium moieties. This minor
difference is expected to influence the degree of hydrophilicity of
the network and, hence, tune the properties of the resulting
hydrogel. The feasible synthesis of CL1 and CL2 is described. Pol-
ySBDMA hydrogels were characterized in terms of mechanical
properties, diffusion coefficient of water, swelling in water as well
as in aqueous solution of salts, different water state, degree of
crosslinking, mesh size and fibroblast adhesion. Some of these
properties were compared to polySBDMA hydrogels prepared by
crosslinking with commercially available crosslinkers BIS and
EDMA. The majority of work is devoted to hydrogels prepared at
the typical crossslinker level ꢀ 3 mol% to monomer. Sorption and
mechanical properties were determined for hydrogel samples
crosslinked by CL1 up to 20 mol% to monomer. This work
demonstrates that crosslinkers with a sulfobetaine moiety of the
same chemical character as monomer SBDMA can be recom-
mended for formation of polySBDMA hydrogels. In this case, the
compositional drift with conversion is expected to be minimized,
which cannot be warranted with commercial crosslinkers of
a different chemical character.
Fig. 1. Chemical structures of monomer N-(methacryloxyethyl)-N,N-dimethyl-N-
(3-sulfopropyl)ammonium betaine (SBDMA), and crosslinkers N,N-bis(methacry-
loxyethyl)-N-methyl-N-(3-sulfopropyl)ammonium betaine (CL1) and N,N-bis
(methacryloxyethyl)-N-methyl-N-(4-sulfobutyl)ammonium betaine (CL2).
(Aldrich, >99%) were used as received. N,N-bis(methacryloxyethyl)
methylamine was prepared from methyl methacrylate and
N,N-bis(2-hydroxyethyl)methylamine catalyzed with MeONa [25].
Acetone was dried over CaCl2 and distilled before use. Ultrapure
water was obtained from Ultrapure Water System NW Series (Heal
Force Bio-Meditech Holdings, Ltd, China). All solvents and salts
were of analytical grade.
RAT-2 fibroblasts-like cell line (ECACC, UK, Cat. No. 94050409)
was used for testing the cell adhesion properties of prepared
hydrogels. Cells were grown on tissue culture polystyrene flasks in
Dulbecco’s modified Eagle medium (DMEM) supplemented with 5%
fetal calf serum, 2 mmol Lꢁ1 -glutamine and 4.5 g Lꢁ1 glucose at
L
37 ꢂC in humidified atmosphere containing 5% CO2. All cell culture
reagents were purchased from Gibco BRL (Invitrogen Corporation,
Germany).
2.2. Synthesis of CL1
The crosslinker CL1 was prepared following Scheme 1. To an ice-
cooled solution of 1.020 g (4 mmol) N,N-bis(methacryloxyethyl)
methylamine
1 in 2 mL dry acetone, a solution of 520 mg
2. Experimental part
(4.2 mmol, 0.390 mL) 1,3-propanesultone in 1 mL acetone was
added drop-wise. The reaction mixture was stirred at 40 ꢂC for 5 h.
The white precipitate formed was filtered and washed with a small
amount of dry acetone and diethylether. The filtrate was evapo-
rated and 2 mL of dry acetone were added to the residue. The
precipitate was formed at room temperature within one week. The
white powder was obtained by centrifugation at a yield of 81%.
1H NMR (300 MHz, CD3OD): 6.15 d, 2 H, J ¼ 2.0 Hz (]CeH); 5.57
d, 2 H, J ¼ 2.0 Hz (]CeH); 4.64 t, 4 H, J ¼ 8.4 Hz (OeCH2); 3.88 t, 4
H, J ¼ 8.4 Hz (NeCH2); 3.70 m, 2 H (NeCH2); 3.25 s, 3 H (NeCH3);
2.1. Materials and reagents
N-(methacryloxyethyl)-N,N-dimethyl-N-(3-sulfopropyl) ammo-
nium betaine (SBDMA, Aldrich, 97%), N,N0-methylene bisacryla-
mide (BIS, SigmaeAldrich, 99%), ethylene glycol dimethacrylate
(EDMA, Aldrich, 98%), N,N,N0,N0-tetramethylethylenediamine
(TMEDA, Fluka, >99%), ammonium peroxodisulfate (APS, Aldrich,
>98%), 1,3-propanesultone (Aldrich, 98%) and 1,4-butanesultone

P. Kasák et al. / Polymer 52 (2011) 3011e3020
3013
2.82 t, 2 H, J ¼ 8.2 Hz (SeCH2); 2.24 m, 2 H (CH2); 1.96 s, 6 H (]
CeCH3).
shaped polySBDMA hydrogels were immersed for 4 days in
distilled water to remove any low molecular weight compounds.
The water was changed several times during this process. Poly-
SBDMA hydrogel slabs were cut with an appropriate cork bore to
obtain cylinders used for further characterization. PolySBDMA
hydrogels were designated based on the molar concentration of
crosslinker to SBDMA in the feed as listed in Table 1.
13C NMR (75 MHz, CD3OD): 167.7, 137.1, 127.5, 65.9, 63.3, 62.3,
58.9, 19.8, 18.4.
FTIR (cmꢁ1, KBr): 1728 s, 1638 m, 1458 w, 1456 w, 1338 m, 1329
m, 1204 s, 1171 s, 1040 s, 952 w, 938 w, 805 w, 630 w, 539 w, 529 w.
2.3. Synthesis of CL2
2.5. NMR spectroscopy
Scheme 1 depicts also the synthesis of crosslinker CL2. To an ice-
cooled solution of 1.020 g (4 mmol) N,N-bis(methacryloxyethyl)
1H NMR and 13C NMR were recorded on a Varian Gemini 300
instrument at 298 K. Chemical shifts are reported in ppm downfield
to internal standard TMS (0.00 ppm). The solvent CD3OD was used as
a reference. Working frequency was 300 MHz for 1H and 75.5 MHz
for 13C NMR. Coupling patterns were designated as s e singlet,
d e doublet, te triplet, m e multiplet. Couplingconstants are givenin
Hz. Concentration of samples was around 10 mg of crosslinker dis-
solved in 0.7 mL of solvent.
methylamine
1 in 2 mL dry acetone, a solution of 571 mg
(4.2 mmol, 0.430 mL) 1,4-butanesultone in 1 mL acetone was added
drop-wise. The reaction mixture was stirred at room temperature
for 5 h. The white precipitate formed was filtered and washed with
a small amount of dry acetone and diethylether. The filtrate was
evaporated and 1 mL of dry acetone was added to the residue. The
precipitate was formed in a freezer at ꢁ20 ꢂC for about one week.
Centrifugation resulted in a white powder of 64% yield.
1H NMR (300 MHz, CD3OD): 6.15 d, 2 H, J ¼ 1.8 Hz (]CeH); 5.52
d, 2 H, J ¼ 1.8 Hz (]CeH); 4.63 t, 4 H, J ¼ 8.4 Hz (OeCH2); 3.87 t , 4
H, J ¼ 8.4 Hz (NeCH2); 3.55 m, 2 H (NeCH2); 3.24 s, 3 H (NeCH3);
2.86 t, 2 H, 3J ¼ 8.2 Hz (SeCH2); 1.96 s, 6 H (]CeCH3), 1.96 m, 2H
(CH2); 1.85 m, 2H (CH2).
2.6. FTIR spectroscopy
FTIR data were recorded on a Specord M 80 spectrophotometer
in KBr pellets with 32 scans. The spectra were characterized in
wavenumbers with abbreviations for determination of relative
intensities of signals as s e strong, m e medium, w e weak.
13C NMR (75 MHz, CD3OD): 167.7, 137.1, 127.5, 64.4, 62.2, 59.0,
51.3, 23.0, 22.1, 18.4.
FTIR (cmꢁ1, KBr): 1717 s, 1635 m, 1474 m, 1458 w, 1319 m, 1293
m, 1237 m, 1202 s, 1162 s, 1044 m, 956 m, 818 w, 793 w, 650 w, 608
w, 571 w, 533 w.
2.7. Swelling and sorption measurements
Hydrogel cylinders with diameter of 10 mm were lyophilized for
one day and stored in a desiccator over CaCl2 at room temperature
before measurement. Lyophilized hydrogels were immersed in
water at 24 ꢂC and the water sorption was followed by gravimetry.
Before measuring the weight of hydrogel samples during swelling
experiments, any excess water on the hydrogel surface was gently
wiped off using Kimwipe paper. The equilibrium water content
(EWC) in weight percent, which expresses the maximum amount of
water swelling the hydrogel at given conditions, was determined
from Eq. (1) [6]:
2.4. Formation of polySBDMA hydrogel
SBDMA aqueous solution (2.0 mol Lꢁ1, 1.0 mL), various amounts
of crosslinkers and 40 m
L of 0.22 mol Lꢁ1 APS aqueous solution (final
APS concentration equals to 8.8 mmol Lꢁ1) as initiator were placed
in a vial. The solution was stirred for 30 min to ensure that all
components are dissolved and the solution is homogenous. Argon
gas was bubbled for 30 min through the solution to remove oxygen.
A 14 m
L (0.093 mol Lꢁ1) aliquot of TMEDA as the accelerator was
added and stirring was continued for an additional 30 s. The solution
was then injected between two glass plates of dimensions (in mm)
55 ꢃ15 ꢃ1, with the thickness adjusted with a glass spacer, and
purged byan argon gas for about 1 min before injecting the solution.
The side walls were secured with adhesive tape. After the gelling
process was completed within 24 h at 23 ꢄ1 ꢂC, the obtained slab-
WN ꢁ W
EWC ¼
d$100
(1)
WN
where WN and Wd are the weights of the swollen hydrogel in
equilibrium and of the lyophilized hydrogel, respectively.
O
O
OH
N
-
-
SO3
85 %
2
SO3
O
O
O
S
[MeONa]
- MeOH
O
O
S
O
O
O
O
O
O
+
+
N
2
N
2
N
2
O
acetone
64 %
acetone
81 %
CL2
1
CL1
Scheme 1. Synthesis of sulfobetaine crosslinkers CL1 and CL2.

3014
P. Kasák et al. / Polymer 52 (2011) 3011e3020
ꢀ
À
ꢁ
Table 1
2=3
42
40
PolySBDMA hydrogel samples synthesized in the presence of various crosslinkers of
s
various concentrations. Molar concentration of SBDMA was equal to 2.0 mol Lꢁ1
.
ne
Á
¼
(3)
aꢁ2
V
RT
a
ꢁ
Sample name
Crosslinker
CL1
Concentration in feed
mol% to monomer
where
s
is the strain stress or compression stress, 42 is the volume
CL1-0.5
CL1-1
CL1-3
CL1-10
CL1-20
0.5
1
3
10
20
fraction of polymer at the equilibrium (in the fully swollen hydro-
gel; it was calculated from the change of the volumes of dried and
fully swollen cylinders), and 40 is the volume fraction of polymer in
the relaxed state (non-swollen, but not dehydrated hydrogel), R is
the gas constant, T is the absolute temperature and
a is the defor-
CL2-0.5
CL2-1
CL2-3
CL2
0.5
1
3
mation ratio related to the stress determined as the ratio of elas-
tically deformed length L to initial length L0 of the hydrogel.
Crosslink densities were arbitrarily calculated from 40% strain and
BIS-1
EDMA-1
BIS
EDMA
1
1
averaged crosslink distance (mesh size),
crosslink density using Eq. (4) [30]:
x
, was estimated from the
ꢀ
ꢁꢁ1=3
neNA
V
The rate of water sorption by hydrogel is expressed by the
sorption degree (SD), which (in weight percent) is defined by Eq. (2)
[26]:
x
¼
(4)
where NA is Avogadro’s number. The deformation of 40% was
chosen arbitrarily to estimate the effectiveness of crosslinkers.
Wt ꢁ W
SD ¼
d$100
(2)
Wd
2.10. Cell adhesion
where Wt is the weight of wet hydrogel at different time intervals
and Wd is the weight of lyophilized hydrogel. The sorption degree
at equilibrium was determined after equilibration of hydrogel
samples in water, 100 mmol Lꢁ1 phosphate buffered saline (PBS)
and 155 mmol Lꢁ1 (0.9%) NaCl solutions.
Cell adhesion tests were carried out using polySBDMA thin
hydrogel films (CL1-1 in Table 1) deposited on cover glass slips by
a spin-coating technique. Cover glass slips were cleaned with
Piranha solution for 10 min and then washed extensively with
deionized water and dried in oven at 120 ꢂC prior to use. The spin-
coated films were prepared as follows: A stock solution was
prepared consisting of 2 mol Lꢁ1 of SBDMA, 1 mol% to monomer
2.8. Differential scanning calorimetry
content of CL1 and 0.013 mol Lꢁ1 of APS in distilled water. 300
stock solution was mixed with 7 L TMEDA for 20 s and the solution
mL of
The differential scanning calorimetry (DSC) measurements were
carried out by employing a DSC Mettler-Toledo 821 differential
scanning calorimeter. Nitrogen gas was passed through the
instrument at a flow rate of 50 mL minꢁ1. Before measurement, the
DSC was calibrated from ꢁ100 to 250 ꢂC at a heating rate of
5 ꢂC minꢁ1 with an indium standard. Hydrogel samples swollen to
the EWC were gently wiped with Kimwipe paper to remove surface
water, then cut, weighed accurately and immediately sealed in
aluminum pans. The sample weight ranged from 4 to 11 mg. DSC
curves were obtained by heating from ꢁ45 to 20 ꢂC and by cooling
from 20 to ꢁ45 ꢂC with a heating rate of 10 ꢂC minꢁ1. All experi-
ments were done in duplicate.
m
was dropped onto the cover glass slip mounted to a spin coater. The
spin-coating was performed for 30 s at 5000 rpm. Then the cover
glass slip was transferred to a Petri dish, where the hydrogel was
formed for 70 min under the argon atmosphere. Consequently, the
slip was immersed in distilled water for 2 days and stored at 4 ꢂC.
During this period distilled water was exchanged periodically. The
film on a cover glass slip was sterilized by autoclaving prior to cell
adhesion tests. The thickness of films of around 3 mm was deter-
mined by a confocal laser scanning microscopy.
Growing RAT-2 fibroblasts-like cells at concentration of 1 ꢃ105
per plate in DMEM were seeded on a 60 mm tissue culture poly-
styrene Petri dish to which a glass cover slip with the spin-coated
polySBDMA hydrogel was placed. Cells were incubated at 37 ꢂC in
a humidified atmosphere containing 5% CO2 for 5 days. Prolifera-
tion of RAT-2 fibroblasts-like cells on the sample surface and their
morphology were analyzed using optical microscopy (XDS-IM1,
Optika microscopes) by a digital camera (Canon, DS 126071) using
10ꢃ and 2.5ꢃ objectives. Cell density on the hydrogel surface and
on the non-coated cover glass slip, used as a positive control, was
determined using a hemocytometer (Glaswarenfabrik Karl Hecht
KG, Germany).
2.9. Mechanical properties
Hydrogels were tested on a Texture Analyzer TA-XT2i (Stable
Micro Systems, Godalming, UK) equipped with a force transducer of
1 mN resolution and Texture Expert software version 1.16 used for
data acquisition and evaluation. The mechanical stability was
measured on hydrogel cylinders of diameter 3.83 mm in
compression mode using a vertically moving mobile probe of
diameter 4 mm at a constant speed of 0.6 mm sꢁ1. The cylinders
were deformed up to about 90% of their initial height. The initial
height values were determined from the compression stressestrain
curves based on the probe position at which the force values started
to increase. Average values and standard deviations were obtained
from the analysis of at least six replicates. Mechanical properties
were evaluated from the stressestrain curves using the approach
described recently [27].
3. Results and discussion
3.1. Synthesis of CL1 and CL2
The synthesis of new crosslinkers was carried out from dime-
The crosslink density and distance between crosslinks were
determined for hydrogel samples using their swelling characteris-
tics and mechanical properties. The elastic modulus is related to the
effective network chain concentration of the swollen hydrogel. The
crosslink density ne/V can be calculated from Eq. (3) [28,29]:
thacrylate 1 according to Scheme 1. 1 was prepared from
commercially available N,N-bis(2-hydroxyethyl)methylamine and
methyl methacrylate with azeotropic distillation of methanol
catalyzed with MeONa following the previously reported method
[25]. A chromatographic separation provided product of 92% yield,

P. Kasák et al. / Polymer 52 (2011) 3011e3020
3015
which is slightly higher than reported (85%). Synthesis of CL1 and
CL2 was done in acetone with a sufficient yield of 81 and 64%,
respectively. Crosslinkers were characterized by means of 1H NMR,
13C NMR and FTIR (both contained in the Supplementary Data file).
1H NMR spectra for CL1 and CL2 demonstrate the purity of
synthesized crosslinkers. In the FTIR spectra of CL1 and CL2, strong
peaks are observed at about 1320 and 1640 cmꢁ1, which indicate
the presence of methacrylic C]C double bonds. Esteric carbonyl is
DSC measurements were accompanied by swelling experiments
of lyophilized samples to determine EWC values from Eq. (1) that
combine total freezing and non-freezing water in the polymeric
network. The EWC values and DSC results were then used to
determine the weight percentage of non-freezing water (Wnf):
Wnf ¼ EWC ꢁ Wtf
(6)
Fig. 2 shows DSC endothermic curves of the first heating run for
hydrogel samples between ꢁ45 and 20 ꢂC. The peaks correspond-
ing to freezable-bound water (lower temperature shoulder) and
freezing free water (higher temperature shoulder), respectively, are
overlapped and shifted to temperatures above 0 ꢂC due to the fast
heating rate used in this study. These data are, nevertheless, suffi-
cient to determine the melting enthalpy calculated from the area
under the integrated endothermic curves from ꢁ15 to 15 ꢂC. The
content of total freezing water was determined from Eq. (5), which,
together with the equilibrium water content, provides information
on the content of non-freezing water from Eq. (6). Table 2
summarizes obtained EWC, Wtf and Wnf data in addition to
Wtf/Wnf ratios in order to compare the characteristics of hydrogels
prepared in the presence of various crosslinkers. It should be
mentioned that integration of exothermic curves from cooling runs
between 20 and ꢁ45 ꢂC resulted in lower values of enthalpy (data
not shown). These values were not used for further analysis as they
were assigned to artifacts (water evaporation and condensation)
similar to those discussed in the work of Ahmad and Huglin [32].
EWC values for all hydrogels with the lower crosslinker content
up to 3 mol% to monomer are about 60 wt% with some slight
systematic differences depending on both type and concentration
of crosslinker. For hydrogels formed in the presence of CL1 and CL2,
EWC decreases by about 4 wt% for both crosslinkers upon
increasing their concentration. This follows the expected increase
in the degree of crosslinking and, therefore, the slightly reduced
capability of the hydrogels to be swollen by water. An increase in
CL1 level to 10 and 20 mol% to monomer further reduces the EWC
values to around 47 and 42 wt%, respectively.
detected at about 1720 cmꢁ1
. Peaks referring to quaternary
ammonium and organic sulfate are seen at about 1030 and
1170 cmꢁ1, respectively.
3.2. Formation of polySBDMA hydrogels
PolySBDMA hydrogels were prepared using SBDMA concentra-
tion of 2 mol Lꢁ1 (56 wt%) and various crosslinkers of concentra-
tions (in mol% to monomer), given in Table 1, by redox-initiated
free-radical polymerization using as a redox couple initiator APS
and accelerator TMEDA. Polymerizations were carried out to reach
complete conversion that was confirmed by Soxhlet extraction of
prepared discs showing no solid residues in the extracted fraction.
Moreover, FTIR analysis of the freeze-dried hydrogels did not show
the presence of peaks from C]C double bonds at 1640, 1320 and
1300 cmꢁ1 (data not shown), which also demonstrates that the
xerogels did not contain any residual monomer. Thus the solid
content of prepared hydrogels is approximately 56e62% given by
the amounts of monomer and crosslinkers. After polymerization,
hydrogels were colorless with transmittance over 95% for around
1 mm thick slabs within the wavelength range between 400 and
800 nm determined by UVeVIS spectrometry (data not shown).
The hydrogels exhibited uniform and smooth surfaces without the
occurrence of bubbles or other imperfections. Their shape was
stable in aqueous solutions demonstrating that an effective
network was formed and that the hydrogels exhibit a high capacity
for water swelling.
A higher water content for hydrogels prepared using CL1
compared to CL2 (at the same crosslinker concentration) reflects
the more hydrophobic character of CL2 (4-sulfobutyl pendant) than
CL1 (3-sulfopropyl pendant). This trend of somewhat different
hydrophilicity of hydrogels prepared in the presence of CL1 and
CL2, respectively, is further seen with respect to Wtf and Wnf rep-
resenting the state of water related to the EWC values via Eq. (6).
3.3. Determination of the state of water by DSC
The state of water in hydrogels is a material property that can be
correlated with applications especially in biorelated fields, where
the aqueous environment determines the performance of proteins
and cells [31]. In swollen hydrogels, states of water are classified to
freezing free water, freezable-bound water, and non-freezing water
[32]. Freezing free water does not participate in hydrogen bonding
and/or complexation with the polymer chains. Freezable-bound
water is the intermediate water that interacts only weakly with
polymer chains and affects transport properties [6,33]. It exhibits
the phase transition temperature lower than 273 K. The non-
freezing water is the fraction of water bonded to the hydrophilic
sites of the polymer chain through hydrogen bonds. This water does
not exhibit a detectable phase transition over the temperature
range from 200 to 273 K [32] and may influence the biocompati-
bility of a hydrogel [34].
CL1-0.5
CL1-1
^
CL1-3
endo
CL1-10
CL1-20
-
CL2 0.5
CL2-1
CL2-3
Wg-1
5
BIS-1
In our study, DSC was employed as the most common technique
[6] to determine the content of total freezing (Wtf) water, in weight
percentage, in hydrogels from Eq. (5):
EDMA-1
D
H
Wtf ¼ Wff þ Wfb
¼
$100
(5)
-40
-30
-20
10
0
-10
D
HW
(°C)
Temperature
where Wff and Wfb are the weight percentages of free freezing and
freezable-bound water, respectively, H is the melting enthalpy of
water in the hydrogel and HW is the melting enthalpy of bulk
water equal to 333 J gꢁ1 [32].
D
Fig. 2. DSC endothermic curves of first heating run for polySBDMA hydrogels prepared
from 2 mol Lꢁ1 SBDMA concentration and various concentrations of crosslinkers in mol
% to monomer.
D

3016
P. Kasák et al. / Polymer 52 (2011) 3011e3020
Table 2
350
300
250
200
150
100
50
Equilibrium water content, EWC, total freezable water, Wtf, and non-freezing water,
Wnf, for zwitterionic polySBDMA hydrogels prepared from 2 mol Lꢁ1 SBDMA and
various concentrations of CL1 and CL2 (mol% to monomer) and 1 mol% to monomer
of BIS and EDMA crosslinkers.
Sample
EWC (wt%)
Wtf (wt%)
Wnf (wt%)
Wtf/Wnf
CL1-0.5
CL1-1
CL1-3
CL1-10
CL1-20
63.0
61.5
59.8
47.1
42.2
47.3
43.4
41.6
28.8
18.6
15.7
18.1
18.2
18.3
23.6
3.0
2.4
2.3
1.6
0.8
CL1-0.5
CL1-1
CL1-3
BIS-1
CL2-0.5
CL2-1
CL2-0.5
CL2-1
CL2-3
60.8
58.9
57.3
49.1
48.2
47.9
10.8
10.7
9.4
4.5
4.5
5.1
CL2-3
EDMA-1
0
BIS-1
EDMA-1
60.8
60.5
41.4
48.1
19.4
12.4
2.1
3.8
0
50
100
150
200
250
Time (min)
Fig. 3. The sorption degree (SD) dependence on time for polySBDMA hydrogels
crosslinked with all crosslinkers used in this study (in mol% to monomer). Lines
connect the experimental data points to guide the eye.
For CL1 the Wtf values decrease from 47.3 to 18.6 wt% with
increasing CL1 concentration from 0.5 to 20 mol%. The Wtf values at
the crosslinker concentration up to 3 mol% to monomer are
significantly lower than Wtf values for hydrogels prepared in the
presence of CL2 crosslinker. This indicates that hydrogel crosslinked
by CL2 is less hydrophilic than that crosslinked by CL1. It is inter-
esting to note that only a minor difference in the crosslinker
chemistry between CL1 and CL2 represents a significant factor
influencing the state of water, and interactions of water with the
hydrogel network. In this sense, Table 2 shows that Wnf values are
higher and ratios Wtf/Wnf are lower for hydrogels prepared using
CL1 compared to CL2. The duplicate experiments exhibited less
than 2% variation, hence, the discussed differences among the data
in Table 2 are outside the error range. In summary, these data
demonstrate that a higher degree of hydrogen bonding interactions
with water molecules occurs for hydrogels crosslinked with CL1
than with CL2.
EWC, Wtf and Wnf values in Table 2 are also shown for hydrogels
prepared in the presence of 1 mol% to monomer of BIS and EDMA
crosslinkers. The hydrogel prepared in the presence of BIS is more
hydrophilic than that prepared in the presence of EDMA. Then the
state of water in the former hydrogel is similar to the hydrogels in
the presence of CL1 and the state of water for the latter hydrogel is
closer to the situation observed for hydrogels prepared in the
presence of CL2. This observation can be related to the chemical
structures where the methylene bisacrylamide moiety of BIS is
more hydrophilic than is the ethylene glycol dimethacrylate moiety
of EDMA.
a hydrogel in time t and at the equilibrium, respectively. According
to the RitgerePeppas theory [37,38], these experimental data can
be divided into two regions. The region of “short times” corre-
sponds to the early stage of diffusion, for which the short-time
approximation is valid for the first 60% of the equilibrium water
content. The region of “long times” corresponds to the late stage of
diffusion. In order to study the water transport mechanism in the
hydrogel, the RitgerePeppas model described by Eq. (7) was
considered to fit the experimental data:
Mt
MN
¼ ktn
(7)
here Mt/MN is the fractional water content, k is the kinetic constant,
t is the diffusion time and n is the time exponent that can be related
to the solute transport. For a thin hydrogel film, and when n ¼ 0.5,
the water diffusion mechanism is described by Fickian diffusion.
The comparison of the model with the experimental data for
sample CL1-0.5 is shown as an inset in Fig. 4.
3.4. Diffusion coefficient of water in hydrogels
Fig. 3 reveals the experimental data for hydrogels crosslinked by
all crosslinkers obtained from the water sorption experiments. The
weight of hydrogel after placing the lyophilized hydrogel samples
in water was determined at different time intervals until the
saturation level was reached. Overall, the crosslinker concentration
seems to be a more significant factor for sorption of water than the
selection of crosslinker. Results shown in Fig. 3 can be used for
calculation of the diffusion coefficient of water in hydrogels for the
case when diffusion controls the transport process [35]. Generally,
the sorption behavior in polymers is a complex process. It can range
from pure Fickian to pure relaxation behavior [35e37] and is
controlled either by Fickian diffusion or by segmental relaxation
depending on which of these processes is slower. This phenomenon
can be analyzed using various models [38e41].
Fig. 4. Water sorption experiment for sample CL1-0.5 showing the fractional water
content Mt/MN in hydrogel as a function of time: (I) short-time approximation, and (II)
long-time approximation ranges. Line connects the experimental data points to guide
the eye. Inset: Comparison of experimental data with model given by equation (6) for
the fractional water content Mt/MN in hydrogel as a function of time for sample
CL1-0.5 in the short-time range.
A typical dependence for the water sorption experiment is
shown in Fig. 4. Mt and MN are the characteristic water contents in

P. Kasák et al. / Polymer 52 (2011) 3011e3020
3017
The n values for the investigated hydrogels are summarized in
500
400
300
200
100
0
water
Table 3. They are all close to the value of 0.5, which suggests that
the water diffusion into hydrogels can be considered as Fickian.
Then, the water diffusion into the hydrogel in the region of short
times can be described by Eq. (8) [37]:
100 mmol.L-1 PBS
155 mmol.L-1 NaCl
ꢀ
ꢁ
0:5
Mt
MN
Dt
L2
¼ 2
(8)
where L is the sample thickness and D is the diffusion coefficient.
Eq. (8) is restricted to thin layers. In our studies, the ratio of sample
thickness to diameter is w1/15, which can be considered as the
diffusion of water into a planar sheet. This validates using of Eq. (8)
to estimate diffusion coefficients of water in investigated hydrogels,
which are contained in Table 3.
CL1-0.5 CL1-1 CL1-3 CL2-1 BIS-1 EDMA-1
The data listed in Table 3 can be summarized as follows: the
increase in crosslinker concentration is associated with lowering
the sorption degree and slightly decreased diffusion coefficients of
water in hydrogels. In correlation with the DSC and EWC data, the
diffusion of water is slower (i) into less hydrophilic hydrogels
crosslinked by CL2 compared to CL1, and (ii) into the hydrogels with
a higher degree of crosslinking. The results for hydrogels cross-
linked by commercial crosslinkers also follow this line of explana-
tion, i.e., a higher value of diffusion coefficient for BIS compared to
EDMA reflects the differences between hydrophilic character of
these crosslinkers.
Typical diffusion coefficients for diffusion of water in polymers
are in the order of 10ꢁ11 to 10ꢁ10 m2 sꢁ1 [42]. Hence, the diffusion
coefficients of about 2 ꢃ 10ꢁ10 m2 sꢁ1 determined in this work
demonstrate that the water diffusion in prepared sulfobetaine
hydrogels is relatively fast. For comparison, the same approach as
used in this work for characterization of the buffer sorption process
into the hydrophobically modified N,N-dimethylacrylamide hydro-
gels [43] and resulted in the early- and late-time diffusion coeffi-
cients ranging from 7.3 ꢃ10ꢁ11 to 1.35 ꢃ10ꢁ10 m2 sꢁ1 and
6.5 ꢃ10ꢁ11 to 1.21 ꢃ10ꢁ10 m2 sꢁ1, respectively. The values of diffu-
sion coefficients of water determined in our work also agree with
those determined for (N-isopropylacrylamide-co-acrylamide-co-2-
hydroxyethyl methacrylate) hydrogels ranging from about
2.5 ꢃ10ꢁ10 to 5.0 ꢃ 10ꢁ10 m2 sꢁ1 [44], glycol chitosan superporous
hydrogels ranging from 8.3 ꢃ 10ꢁ10 to 1.7 ꢃ 10ꢁ9 m2 sꢁ1 [45], and the
network of poly(vinyl alcohol) and poly(N-vinyl pyrrolidone)
crosslinked by radiation ranging from 1.8 ꢃ 10ꢁ11 to
4.4 ꢃ10ꢁ11 m2 sꢁ1 [46].
Fig. 5. Sorption degree at equilibrium for polySBDMA hydrogels crosslinked with
various crosslinkers (for CL1 at various crosslinker concentrations between 0.5 and
3 mol%, and for CL2, BIS and EDMA at crosslinker concentration of 1 mol%) in water,
100 mmol Lꢁ1 PBS and 155 mmol Lꢁ1 NaCl.
intermolecular associated structures typical for zwitterionic poly-
mers [14e16]. Thus the sorption degree at equilibrium depends on
the level of crosslinker as well as is proportional to the concentration
ofsimple electrolytes. Thisalso points out that identification ofsubtle
differences among the hydrogels may not be possible in case when
the zwitterionic hydrogels are studied in the presence of simple
electrolytes.
3.5. Mechanical properties and degree of crosslinking
Mechanical properties of the investigated hydrogels prepared in
the presence of different concentrations of crosslinkers synthesized
in this work (CL1 and CL2) as well as commercially available
crosslinkers (BIS and EDMA) are contained in Table 4.
Increased crosslinker concentration leads both to higher stiff-
ness and hardness (represented by Young’s modulus), pronounced
strength (expressed by stress parameters sb and s40), whereas
deformability of hydrogels (3b) decreases. This is a common
behavior not only for hydrogels [47] but also for crosslinked rubbers
as well as thermoplastics [48,49]. The comparison between
hydrogels crosslinked by CL1 and CL2 at the same crosslinker
concentration does not lead to an unambiguous conclusion about
the effect of each of these crosslinkers. The stress at arbitrary 40%
deformation, s40, was selected as another quantity since it enables
the comparison among all the hydrogels in this study (Table 4).
Fig. 6 and the data in Table 4 represent the comparison of
mechanical properties for hydrogels crosslinked by various cross-
linkers at the equal crosslinker concentration of 1 mol% based on
Studies on sorption degree of water in hydrogels are extended by
the determination of the sorption degree at equilibrium in the
presence of simple electrolytes in order to demonstrate the anti-
polyelectrolyte behavior of zwitterionic polymers [14,15]. Fig. 5
shows the swelling ratio for hydrogels immersed and equilibrated
in water, 100 mmol Lꢁ1 PBS and 155 mmol Lꢁ1 saline solution. Non-
surprisingly, the sorption degree is higher in the presence of
simple electrolytes in aqueous solutions, which disrupt intra- and
Table 4
Mechanical properties of hydrogels prepared by using different crosslinking
conditions: Young’s modulus, E, elongation at break, 3b, stress at break, sb, stress at
Table 3
deformation of 40%, s40, crosslink density, ne/V, and crosslink distance, x.
Sorption degrees at equilibrium, time exponents, n, and diffusion coefficients, D, of
water for polySBDMA hydrogels prepared using different crosslinking conditions.
Sample E (kPa)
3b (%) sb (kPa)
s40 (kPa)
ne/V (mol mꢁ3
) x (nm)
CL1-0.5
CL1-1
CL1-3
63 ꢄ 12 57 ꢄ 1 251 ꢄ 22
127 ꢄ 15 46 ꢄ 1 241 ꢄ 58
147 ꢄ 64 46 ꢄ 3 550 ꢄ 72
110 ꢄ 17
162 ꢄ 26
46 ꢄ 7
3.3 ꢄ 0.5
3.0 ꢄ 0.5
2.4 ꢄ 0.5
1.8 ꢄ 0.2
1.7 ꢄ 0.6
Sample
Sorption degree (wt%)
n
D.1010 (m2 sꢁ1
)
59 ꢄ 10
CL1-0.5
CL1-1
CL1-3
334
220
178
0.56 ꢄ 0.04
0.58 ꢄ 0.06
0.56 ꢄ 0.07
2.5
2.3
2.0
366 ꢄ 68 122 ꢄ 23
CL1-10
437 ꢄ 224 72 ꢄ 2 2185 ꢄ 170 695 ꢄ 95 280 ꢄ 44
CL1-20 1450 ꢄ 375 85 ꢄ 4 4400 ꢄ 260 1235 ꢄ 115 348 ꢄ 53
CL2-0.5
CL2-1
CL2-3
310
226
164
0.55 ꢄ 0.07
0.62 ꢄ 0.04
0.58 ꢄ 0.08
2.1
1.6
1.5
CL2-0.5
CL2-1
CL2-3
64 ꢄ 9
57 ꢄ 1 297 ꢄ 72
95 ꢄ 2
41 ꢄ 1
3.5 ꢄ 0.1
2.5 ꢄ 0.3
2.3 ꢄ 0.3
90 ꢄ 35 55 ꢄ 3 574 ꢄ 61
226 ꢄ 72 41 ꢄ 3 404 ꢄ 31
281 ꢄ 34 105 ꢄ 13
383 ꢄ 51 130 ꢄ 17
BIS-1
EDMA-1
230
218
0.60 ꢄ 0.05
0.59 ꢄ 0.06
2.3
1.8
BIS-1
EDMA-1
13 ꢄ 1
81 ꢄ 2 532 ꢄ 66
43 ꢄ 6
79 ꢄ 8
16 ꢄ 2
29 ꢄ 3
4.7 ꢄ 0.6
3.8 ꢄ 0.4
52 ꢄ 12 70 ꢄ 2 436 ꢄ 99

3018
P. Kasák et al. / Polymer 52 (2011) 3011e3020
Apart from obtaining quantities characterizing the mechanical
BIS
600
500
400
300
200
100
0
properties and crosslink density of hydrogels, these data are
significant also from the point of view of characterization of the
mechanism of crosslinking. The swelling and sorption as well as
state of water characteristics of polySBDMA hydrogels have already
indicated some differences among hydrogels prepared using
synthesized and commercial crosslinkers, respectively. Impor-
tantly, mechanical properties seem to demonstrate more clearly the
benefit of using crosslinkers CL1 and CL2. The stronger CL1- and
CL2-crosslinked hydrogels show that crosslinkers chemically
similar to the monomer unit are advantageous because they
minimize the compositional drift during crosslinking reaction to
complete conversion. Consequently, the hydrogel network should
be more homogeneous than in the case when chemically different
crosslinkers, such as BIS and EDMA, are used for preparation of
a hydrogel. The compositional drift may be either more signifi-
cantly (assumed for BIS; an acrylamide type of monomer with
a chemically very different character to SBDMA) or less significantly
(assumed for EDMA; a methacrylate type of monomer as is SBDMA)
taking part during the hydrogel formation. The polySBDMA
hydrogel samples crosslinked with either BIS or EDMA should be
viewed as hydrogel networks with a more significant fluctuation in
the network density that is reflected in the mechanical properties
(Table 4). The reactivity ratios for acrylamide with methacrylates
determined in the past [50] as well as for acrylate and methacrylate
monomers determined recently more precisely by pulsed-laser-
initiated copolymerizations [7] suggest that the compositional
drift occurs during formation of the hydrogels and, hence, influ-
ences the hydrogel properties. This should be considered in the
preparation strategies for polySBDMA hydrogels and, naturally,
other hydrogels.
This phenomenon anticipated from our work has recently been
demonstrated for polycarboxybetaine hydrogels [5], where the
concentration of crosslinker was in the range from 2 to 100 mol%.
We increased the level of CL1 crosslinker up to 20 mol% to mono-
mer for which, similarly as in Ref. [5], crosslink density increases
and crosslink distance decreases upon increasing the CL1 concen-
tration. This is accompanied by significantly increased stiffness and
strength of formed hydrogels, which emphasizes the possibility to
tune the physical properties of sulfobetaine hydrogels by using
newly prepared CL1 and CL2 crosslinkers. This is not possible by
using commercial crosslinkers such as BIS and EDMA due to their
limited solubility. The future work will be carried out to recognize
the behavior of polySBDMA hydrogels prepared and tested in the
solutions of simple electrolytes to further correlate the information
on polysulfobetaines to polycaboxybetaine hydrogels [5].
CL2
EDMA
CL1
0
20
40
60
80
deformation (%)
Fig. 6. The compression stressedeformation curves for hydrogels crosslinked with
various crosslinkers at concentration of 1 mol% on monomer content. Dashed line
indicates stress values s40 at the arbitrary deformation of 40%.
monomer content. Commercial crosslinkers form a more elastic
network and softer hydrogels compared to hydrogels crosslinked by
CL1 and CL2. This is implied from the about 1.3-fold higher elon-
gation at break as well as from the Young’s moduli, which are lower
by about one order of magnitude and about two-fold in case of the
hydrogels crosslinked by BIS and EDMA, respectively. Since the
stresssteeply increases with increased deformation, sb values do not
show any systematic variation with the type of crosslinkers, i.e., the
most deformable hydrogels show the highest values of the stress at
break. On the other hand, the comparison of stress values at the
lower deformation, arbitrarily set to 40%, clearly identifies that the
stress s40 required to achieve this deformation is several times
higher for hydrogels crosslinked by CL1 and CL2 than for those
crosslinked by EDMA and BIS. Although mechanical properties for
hydrogel samples crosslinked by CL1 and CL2 exhibit some differ-
ences, overall it may be stated that, at the same crosslinker
concentration between 0.5 and 3 mol%, the hydrogels do not differ
significantly. Data contained in Fig. 6 and Table 4 also indicate that
hydrogels crosslinked by EDMA are stronger than hydrogels cross-
linked by BIS. These data are confirmed by quantities characterizing
the level of crosslinking: the crosslink density and the distance
between crosslinks. At the same concentration of crosslinker of
1 mol%, both indicate a more effective crosslinking reaction, i.e.
higher crosslink densities and lower crosslink distance, in the
presence of CL1 and CL2 than in the presence of commercial cross-
linkers with different chemical structure than SBDMA monomer.
Fig. 7. Microscopic picture of a cover glass slip spin-coated by polySBDMA hydrogel (a) and non-coated cover glass slip used as a control (b) after 5 days of cultivation with the
RAT-2 fibroblasts-like cells.

P. Kasák et al. / Polymer 52 (2011) 3011e3020
3019
3.6. Cell adhesion to the hydrogel surface
VEGA Grant Agency under the contract No. 2/0152/10. Dr. Dusan
Chorvat, Jr., from International Laser Centre in Bratislava, is thanked
for analysis of spin-coated films by confocal laser scanning
microscopy. Dr. I. Janigova from the Polymer Institute SAS is
acknowledged for the DSC measurements. This publication is also
the result of the project implementation: Centre for materials,
layers and systems for applications and chemical processes under
extreme conditions supported by the Research & Development
Operational Programme funded by the ERDF.
Numerous factors influence the affinity of cells to polymeric
surfaces including general chemistry of monomers and crosslinkers
[51], hydrophobic and hydrophilic properties [52], presence of
functional groups [53], and others [54]. Zwitterionic hydrogels are
known as polymers with non-biofouling properties [5,9,16,24,55],
which was also tested in this work.
Fig. 7 demonstrates the adherence of RAT-2 fibroblast-like cells
to the hydrogel layer prepared by spin-coating on a cover glass slip
using the CL1-1 recipe (Table 1). The cell adhesion to the non-coated
slip served as a positive control. After cultivation for 5 days, the
density of cells on a hydrogel surface was around 2.2 ꢃ104 cells/cm2
with 5% confluence of cells exhibiting a round-like shape (Fig. 7a),
while the control sample reveals the cell density of around 4 ꢃ105
cells/cm2 and 90% confluence of firmly attached cells spread on the
surface (Fig. 7b). This behavior was consistently seen for hydrogels
formed by spin-coating on a glass cover slip as well as for hydrogel
slabs, which demonstrates low biofouling of polysulfobetaines
crosslinked by newly synthesized crosslinker. It should be noted
that at this low level of crosslinker, the crosslinker chemistry should
not play a significant role since the hydrogels consist almost
exclusively of the zwitterionic polymers. The beneficial effect of the
zwitterionic crosslinker compared to the non-zwitterionic cross-
linker can be expected at the higher levels, where the non-
zwitterionic crosslinker is expected to interrupt the restructuring
of water within the zwitterionic polymer network thus reducing the
non-biofouling character as demonstrated recently [5].
Appendix. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.polymer.2011.04.056.
References
[1] Peppas NA. Hydrogels in medicine and pharmacy. Boca Raton, FL: CRC; 1987.
[2] Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Adv Mater 2006;18:
1345e60.
[3] Hoffman AS. Adv Drug Deliv Rev 2002;54:3e12.
[4] Kiritoshi Y, Ishihara K. Polymer 2004;45:7499e504.
[5] Carr LR, Xue H, Jiang S. Biomaterials 2011;32:961e8.
[6] Goda T, Watanabe J, Takai M, Ishihara K. Polymer 2006;47:1390e6.
[7] Buback M, Feldermann A, Kowollik C, Lacík I. Macromolecules 2001;34:
5439e48.
[8] White LA, Jönson S, Hoyle CE, Mathias LJ. Polymer 1999;40:6597e605.
[9] Banerjee I, Pangule RC, Kane RS. Adv Mater 2011;23:690e718.
[10] Zhang Z, Chen S, Chang Y, Jiang S. J Phys Chem B 2006;110:10799e804.
[11] Chang Y, Chen S, Zhang Z, Jiang S. Langmuir 2006;22:2222e6.
[12] Zhang Z, Finlay JA, Wang LF, Gao Y, Callow JA, Callow ME, et al. Langmuir
2009;25:13516e21.
[13] Cheng G, Li G, Xue H, Chen S, Bryers JD, Jiang S. Biomaterials 2009;30:
5234e40.
4. Conclusion
[14] Lowe AB, McCormick CL. Chem Rev 2002;102:4177e89.
[15] Singh PK, Singh VK, Singh M. e-Polymers 2007;30:1e34.
[16] Lowe AB, McCormick CL. Polyelectrolytes and polyzwitterions: synthesis,
properties, and applications. Washington, DC: ACS Books; 2006.
[17] Bayer AG. Verfahren zur Herstellung von ungesaettigten Sulfonsaeur-
ebetainen durch Umsetzen eines tertiaeren Amins mit einem Sulton. Patent
DE1211156; 1963.
[18] Rohm, Haas. Polymers of quaternary ammonium compounds. Patent
NL6411736; 1965.
[19] Wang H, Hirano T, Seno M, Sato T. Eur Polym J 2003;39:2107e14.
[20] Chen S, Jiang S. Adv Mater 2008;20:335e8.
[21] West SL, Salvage JP, Lobb EJ, Armes SP, Billingham NC, Lewis AL, et al.
Biomaterials 2004;25:1195e204.
[22] Zhang Z, Zhang M, Chen S, Horbett TA, Ratner BD, Jiang S. Biomaterials 2008;
29:4719e25.
[23] Ladd J, Zhang Z, Chen S, Hower JC, Jiang S. Biomacromolecules 2008;9:
1357e61.
[24] Zhang Z, Chao T, Liu L, Cheng G, Ratner BD, Jiang S. J Biomater Sci Polym Ed
2009;20:1845e59.
The goal of this work was to contribute to the topic of synthesis
and characterization of zwitterionic hydrogels and to identify new
features leading to more controlled properties of this class of
hydrogels. Two novel sulfobetaine dimethacrylate crosslinkers
were synthesized and applied for preparation of polysulfobetaine
hydrogels based on the sulfobetaine monomer of a similar chemical
structure as crosslinkers. The crosslinkers differ slightly in terms of
the length of spacer between the charged groups, i.e., 3-sulfopropyl
vs. 4-sulfobutyl spacers. Such a small difference in the crosslinker
structure introduces a detectable effect on the equilibrium water
content, state of water, diffusion coefficient of water, mechanical
properties and degree of crosslinking even at
a crosslinker
concentration not exceeding 3 mol% to monomer. These differences
can be ascribed to the affinity of water to the hydrogel network
depending on the chemical structure of sulfobetaine crosslinker.
The diffusion of water in selected hydrogels was found to obey
Fickian diffusion. Diffusion coefficients of about 2 ꢃ10ꢁ10 m2 sꢁ1
reveal some correlation with degree of crosslinking and type of
crosslinker. In addition, no limitations in the solubility of CL1 and
CL2 crosslinkers can be used to control the mechanical properties,
water content and degree of crosslinking of polySBDMA hydrogels,
which cannot be done using commercial crosslinkers, such as BIS
and EDMA, due to their limited solubility in water.
This work on polysulfobetaine hydrogels demonstrates, in addi-
tion to those on polycarboxybetaine [5] and polyphosphobetaine [6]
ones, that similar chemical structure of crosslinker and monomer is
a prerequisite for enhancement of mechanical properties and
maintaining the non-biofouling character of zwitterionic hydrogels.
[25] Korschunov MA, Bodnaryuk FN. Zhur Org Khim 1968;4:1157e61.
[26] Wang C, Yu B, Knudsen B, Harmon J, Moussy F, Moussy Y. Biomacromolecules
2008;9:561e7.
ꢀ
ꢀ
[27] Krupa I, Nedelcev T, Racko D, Lacík I. J Sol Gel Sci Technol 2010;53:107e14.
[28] Park K, Shalaby WSW, Park H. Biodegradable hydrogels for drug delivery.
Lancaster, PA, USA: Technomic Publishing Company; 1993.
[29] Dusek K, Prins W. Adv Polym Sci 1969;6:1e102.
[30] Cluff EF, Gladding EK, Pariser R. J Polym Sci 1960;45:341e5.
[31] Vogler EA. Role of water in biomaterials. In: Ratner BD, Hoffman AS, Schoen FJ,
Lemons JE, editors. Biomaterials science: an introduction to materials in
medicine. 2nd ed. Elsevier Academic Press; 2004. p. 59e65 [chapter 1.5].
[32] Ahmad MB, Huglin MB. Polym Int 1994;33:273e7.
[33] Mirejovsky D, Patel AS, Rodriguez DD. Curr Eye Res 1991;10:187e96.
[34] Morisaku T, Watanabe J, Konno T, Takai M, Ishihara K. Polymer 2008;49:
4652e7.
[35] Serra L, Domenech J, Peppas NA. Biomaterials 2006;27:5440e51.
[36] Hilt JZ, Byrne ME, Peppas NA. Chem Mater 2006;18:5869e75.
[37] Ritger PL, Peppas NA. J Control Release 1987;5:23e36.
[38] Ritger PL, Peppas NA. J Control Release 1987;5:37e42.
[39] Higuchi T. J Pharm Sci 1963;52:1145e8.
[40] Peppas NA, Sahlin JJ. Int J Pharm 1989;57:169e72.
Acknowledgements
[41] Anseth KS, Brannon-Peppas L, Bowman CN. Biomaterials 1996;17:1647e57.
[42] Korsmeyer RW. In: Tarcha PJ, editor. Polymers for controlled drug delivery.
Boca Raton: CRC Press; 1991. p. 16e34.
[43] Mullarney MP, Seery TAP, Weiss RA. Polymer 2006;47:3845e55.
[44] Lee WF, Shieh CH. J Appl Polym Sci 1999;71:221e31.
[45] Park J, Kim D. J Appl Polym Sci 2010;115:3434e41.
This research was supported by the Sixth Framework Program of
the EU, IP-031867, P. Cezanne, Slovak Research and Development
Agency under the contract Nos. RPEU-0007-06 and 51-037905, and

3020
P. Kasák et al. / Polymer 52 (2011) 3011e3020
[46] Hill DJT, Whittaker AK, Zainuddin. Radiat Phys Chem 2011;80:213e8.
[47] Zhang Z, Chao T, Jiang S. J Phys Chem B 2008;112:5327e32.
[48] Tobolsky AV. Properties and structure of polymers. New York: John Willey &
Sons, Inc.; 1960.
[49] Chodak I. Prog Polym Sci 1998;23:1409e42.
[50] Plochodzka KJ. Macromol Sci-Rev Macromol Chem 1981;C20:67e148.
[51] Cima L, Lopina S, Merrill E. Ann Biomed Eng 1994;22:184.
[52] Smetana K. Biomaterials 1993;14:1046e50.
[53] Ratner BD. Hydrogel surfaces. In: Peppas NA, editor. Hydrogels in medicine
and pharmacy: fundamentals. Boca Raton, FL: CRC Press; 1986. p. 85e91.
[54] Hilborn J, Bjursten LM. J Tissue Eng Regen Med 2007;1:110e9.
[55] Zhang Z, Chen S, Jiang S. Biomacromolecules 2006;7:3311e5.