University of Wollongong

Research Online
Faculty of Science, Medicine and Health - Papers Faculty of Science, Medicine and Health

2013

Ionic-covalent entanglement hydrogels from gellan

gum, carrageenan and an epoxy-amine
Leo Stevens
University of Wollongong, [email protected]

Paul Calvert
University of Massachusetts - Dartmouth, [email protected]

Gordon G. Wallace
University of Wollongong, [email protected]

Marc in het Panhuis

University of Wollongong, [email protected]

Publication Details
Stevens, L., Calvert, P., Wallace, G. G. & in het Panhuis, M. (2013). Ionic-covalent entanglement hydrogels from gellan gum,
carrageenan and an epoxy-amine. Soft Matter, 9 (11), 3009-3012.

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library:
[email protected]
Ionic-covalent entanglement hydrogels from gellan gum, carrageenan and
an epoxy-amine
Abstract
A 'one pot' preparation of interpenetrating polymer network hydrogels with double network characteristics is
presented. A small addition of biopolymer dramatically increases the stiffness and strength of the epoxy-amine
gels without affecting the large strain at failure value.

Keywords
ionic, gellan, gum, amine, hydrogels, epoxy, carrageenan, entanglement, covalent

Disciplines
Medicine and Health Sciences | Social and Behavioral Sciences

Publication Details
Stevens, L., Calvert, P., Wallace, G. G. & in het Panhuis, M. (2013). Ionic-covalent entanglement hydrogels
from gellan gum, carrageenan and an epoxy-amine. Soft Matter, 9 (11), 3009-3012.

This journal article is available at Research Online: http://ro.uow.edu.au/smhpapers/256

 Journal Name Dynamic Article Links ►

Cite this: DOI: 10.1039/c0xx00000x

www.rsc.org/xxxxxx ARTICLE TYPE
Ionic-covalent entanglement hydrogels from gellan gum, carrageenan
and an epoxy-amine
Leo Stevens,a,b Paul Calvert,c Gordon G. Wallace,b and Marc in het Panhuisa,b*
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
5 DOI: 10.1039/b000000x

A 'one pot' preparation of interpenetrating polymer network application in tissue engineering 14,15,16,17.
hydrogels with double network characteristics is presented. A Interpenetrating gel networks based on cross-linked acrylates
small addition of biopolymer dramatically increases the 50 and polysaccharides have recently been reported to have high
stiffness and strength of the epoxy-amine gels without strength and toughness 18,19. These gels are notable for a “Mullins
10 affecting the large strain at failure value. effect”, a slow recovery of stiffness after extension to high
strains18,19.
Hydrogels have been employed commercially in food 1, Here, we report on IPN hydrogels based on polysaccharides
pharmaceuticals 2 and wearable prosthetics 3, applications that 55 and epoxy-amine chemistry that form in ‘one pot’. These ionic-
make use of the inherent non-toxic properties of many hydrogel- covalent entanglement (ICE) gels show mechanical effects that
forming polymers. However, fully swollen hydrogels have low differ from polysaccharide/acrylate and double network gels. This
15 strength and toughness 4. As a result, few current applications new combination of ionic and covalent binding (Fig. 1) allows
require the hydrogels to bear significant mechanical load. the two polymer backbones to be cross-linked simultaneously,
Overcoming this natural weakness will open the field to an array 60 requiring fewer steps than sequential network formation. In this
of applications including flexible electrodes, soft tissue mimics process the cross-linking of the biopolymer network is likely
for tissue engineering and soft robotics 5,6. completed first, as they are known to gel on a timescale of
20 In 2003, Gong et al. demonstrated that hydrogel materials can minutes. The epoxy-amine curing proceeds on a slower timescale,
be toughened by interweaving two separate polymer networks to increasing the strength and rigidity of the ICE gel over several
create a hybrid material 7. The strength and strain to failure of 65 days.
these 'double network' (DN) hydrogels is much greater than those It is suggested that the stiffness, strength and curing time of the
of the parent networks 5,7,8,9. The formation of these gels was ICE systems can be tailored by tuning some or all of the
25 achieved by sequential network formation, in which a primary following parameters: biopolymer concentration, ionic cross-
network was formed photo-catalytically before a second linker concentration, different ionic cross-linkers (e.g. Na+,
monomer solution was swollen into it and subsequently cross- 70 Mg2+), molecular weight of PEDGE, molecular weight of
linked. The process required UV irradiation in nitrogen Jeffamine and weight ratio between the biopolymer and epoxy
atmospheres7, 10. By selecting polymers whose gelling chemistries amine networks.
30 do not interact, interpenetrating polymer networks (IPN) can be
formed simultaneously, rather than sequentially. However, there
is currently no clear understanding of how the mechanical
properties of a hybrid gel formed in this fashion will compare to
those of the component gels.
35 Gellan gum (GG) and the carrageenans (CG) are anionic
polysaccharides which are USA FDA and European Union
approved as food additives, and have found wide application as
gelling, stabilising and suspending agents 11,12. Gellan is a
biopolymer with carboxylic acid groups, whereas the
40 carrageenans are sulfonated. These biopolymers form hydrogels
in the presence of divalent cations such as calcium. Epoxy-amine
(EA) chemistry is a simple covalently binding reaction that yields
linkages similar to the natural peptide bond. Unlike most forms of
Fig. 1. Schematic representation of ionic-covalent entanglement gels: Top
acrylate polymerisation, epoxy-amine chemistry provides a
left, the epoxy-amine linking reaction between Jeffamine (y=39, x+z = 6)
45 simple route for forming covalent networks that are not readily 75 and PEGDGE (n=9). Bottom left, the ionic cross-linking sites on the
disrupted by physical factors 13. Both the epoxy-amines and the kappa-carrageenan biopolymer (bottom left) and right, the resulting
biopolymers are actively investigated as materials for future interpenetrating network. The photo shows a typical ICE hydrogel.
This journal is © The Royal Society of Chemistry [year] [journal], [year], [vol], 00–00 | 1
 The hydrogels used in this study were prepared as follows. A 10% from the plateau value) corresponds with the maximum
calcium chloride (CaCl2, 0.15 mol/L) solution was generated 60 shear strain that can be applied to the material before it undergoes
from anhydrous calcium chloride (≥95%, Fluka Analytical, batch irreversible deformation. As shown in Table 2, the addition of a
0001251790) and Milli-Q water (resistivity 18.2 MΩ cm). small amount of the carrageenans (to the EA) greatly increases
5 Biopolymer solutions (2% w/v) were made by dissolving low- the modulus compared to that of the epoxy-amine gel without
acyl gellan gum (CP Kelco, batch 9K6968A), iota carrageenan (i- reducing the yield strain. As with the compression tests, k-CG
CG, CP Kelco, Batch SK93842) or kappa-carrageenan (k-CG, CP 65 was more effective than i-CG in stiffening the EA gel. However,
Kelco, batch SK92650) in Milli-Q H2O on a laboratory hotplate this change was relatively modest when compared to the large
(Stuart, CB162) with heating (70 °C) and magnetic stirring before difference in the moduli of i-CG and k-CG gels. The modulus of
10 capping and storing overnight at 70 °C in a laboratory oven single network gels of i-CG is 1/50 the modulus of k-CG gels, but
(Binder Inc., 1.2 kW). A solution of epoxy-amine (50% w/v) was when combined with the epoxy-amine system the relative
then prepared by mixing 50 mL Milli-Q, 34 g of a molten (40- 70 improvement is much greater (Table 1).
50˚C) polyetheramine (Jeffamine ED-2003, Huntsman, Batch
Table 1. Summary of values for water content (WC), stress at failure
1F518, molecular weight 2000 g/mol), and 14 g of poly(ethylene (σf), strain at failure (εf) and tangent modulus (Etan, 20-30% strain) of the
15 glycol) diglycidyl ether (PEGDGE, Aldrich, Batch MKBC9721, hydrogels formed from gellan gum (GG), kappa-carrageenan (k-CG),
Mn = 500 g/mol). These precursor solutions were then combined iota-carrageenan (i-CG), and epoxy-amine (EA). Where failure was not
hot (70 °C), at ratios dictated by the gel type: Epoxy-amine gel 75 observed, stress at 80% (σ80) is reported in lieu of failure stress. The data
was Q-tested (confidence interval ≥ 95 %) and the reported values and
solutions (25% w/v EA) were formed from 50 mL each of EA numerical errors are averages of the values obtained and ± 1 standard
solution and Milli-Q. Biopolymer gel solutions (0.9 % w/v deviation, respectively.
20 biopolymer, 13.6 mM CaCl2) were formed from 50 mL each of
biopolymer solution and Milli-Q water with 10 mL of 0.15M Hydrogel WC Etan σf σ80 εf
(%) (kPa) (kPa) (kPa) (%)
CaCl2. ICE gel solutions (0.9% w/v biopolymer, 22.7% w/v EA, EA 75 79 ± 39 - 175 ± 43 ≥ 80
13.6mM CaCl2) were formed from 50 mL each of EA and i-CG 99 1.3 ± 0.4 1.8 ± 0.2 - 79 ± 2
k-CG 99 77 ± 9 57 ± 2 - 54 ± 1
biopolymer solutions with 10 mL of 0.15M CaCl2. Once mixed, GG 99 86 ± 35 180 ± 130 - 40.0 ± 0.3
25 gel solutions were decanted into PVC molding wells, sealed and i-CG/EA 76 250 ± 100 - 592 ± 7 ≥ 80
stored under controlled ambient conditions (45% relative k-CG/EA 76 360 ± 150 - 930 ± 130 ≥ 80
GG/EA 76 59 ± 32 - 300 ± 170 ≥ 80
humidity, 21˚C) for 1 week prior to testing. All gels were tested
as-made.
Compression testing was conducted on a universal mechanical
30 tester (Shimadzu, EZ-S). Quadruplicate samples were placed into
individual 60 mm Petri dishes lined with moist, absorbent paper
and linearly compressed with a 15 mm diameter head at a rate of
25 mm/min to a maximum compression of 8 mm (80%). Ionic-
covalent entanglement gels containing i-CG and k-CG (Fig. 2a,
35 Fig. S1, Supporting Information and Table 1) showed
substantially higher moduli and strengths than the epoxy-amine
networks, with the reinforcement effect being most prominent at
high strain. For example, the stress at 80% strain value of the k-
CG/EA gels (930 ± 130 kPa) is 5.3 fold higher than the
40 corresponding EA hydrogels. This observation combined with
reinforcement compared to k-CG gels is characteristic of double
network behaviour7. Although forming the strongest network as a
single network (Fig. 2b, Table 1), gellan gum does not
significantly change the modulus and failure properties of the
45 epoxy-amine gels.
Hydrogels produced in this study were further examined by
rheological testing using a controlled strain rheometer (Anton
Paar Physica MCR 301, 15 mm parallel plates). Quadruplicate Fig. 2. Typical compressive stress-strain curves of (a) kappa-carrageenan
samples were tested under a dynamic amplitude sweep mode 80 (k-CG), epoxy-amine (EA) and k-CG/EA ICE hydrogels and (b) gellan
gum (GG), EA and GG/EA ICE hydrogels.
50 between 0.01% and 100% strain at a constant angular frequency
of 5 Hz (Fig. 3 and Fig. S2, Supporting Information). Using this Although gellan gum is the stiffest polysaccharide gel, both
technique, complex shear modulus may be separated into its compression and rheological analysis revealed it has almost no
elastic 'storage modulus' (G') and viscous 'loss modulus' (G"), effect when incorporated into the epoxy-amine gels. At present
which reflect the solidity and liquidity of the material, 85 we do not have a definitive explanation for this behaviour,
55 respectively. Soft, non-Newtonian materials are known to exhibit however, it is suggested that this may be caused by GG losing its
a finite 'linear viscoelastic' (LVE) region in which the storage and ability to become cross-linked by the Ca2+ ions. The absence of
loss moduli are invariant with strain 20. The end of the LVE region this occurrence for the carrageenans may point towards a
(determined as the strain where the G' decreases by more than deactivation or complexation of the GG carboxyl functionality

2 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]
 that prevents the cross-linking of the GG network. Supporting Information). The average ratios of the tangent
moduli of the unloading to the loading part of EA, k-CG/EA, i-
Table 2. Summary of rheological properties. Linear viscoelastic storage
modulus (G'), linear viscoelastic loss modulus (G") and linear viscoelastic 35 CG/EA and GG/EA ICE gels over 10 cycles are 0.93 ± 0.09,
region endpoint (γmax) of hydrogels formed from gellan gum (GG), kappa- 0.71 ± 0.06, 0.84 ± 0.06, and 0.60 ± 0.07, respectively. This
5 carrageenan (k-CG), iota-carrageenan (i-CG), and epoxy-amine (EA). could suggest that the ICE gels dissipate more energy compared
to the EA gels when cycled between 20 and 40% strain.
Hydrogel G' (kPa) G" (kPa) γmax (%)
EA 21.2 ± 5.0 0.074 ± 0.021 41 ± 37 Previous research on the behaviour of interpenetrating gels has
i-CG 0.131 ± 0.013 0.0059 ± 0.0043 ≥100 40 revealed two distinct modes of reinforcement: (i) double network
k-CG 72.4 ± 6.3 6.73 ± 0.74 0.61 ± 0.22 hydrogels, in which the initial moduli of double network gels are
GG 176 ± 25 14.7 ± 1.6 0.288 ± 0.053
similar to the component gels, but they exhibit a dramatic
EA/i-CG 92 ± 26 0.188 ± 0.054 44 ± 13
EA/k-CG 120 ± 57 2.3 ± 1.0 46 ± 23 increase in stiffness in the high strain region 7; and (ii) gellan
EA/GG 25 ± 17 0.219 ± 0.021 58 ± 10 gum/poly(acrylamide) hydrogels, combining gellan gum with
45 poly(acrylamide) into a hybrid gel resulted in a modest increase
Previous research on the polysaccharide/acrylate hybrid gels in initial modulus, but a large increase in the strain to failure
showed a slow recovery effect from large strains18,19, whereas value compared to gellan gum gels18. Hence, combining a low
double network gels display irreversible softening under cyclic modulus/high strain poly(acrylamide) network with high
10 loading 21,22. It has been suggested that hybrid gels have the modulus/low strain gellan gum network results in a high
potential ability to withstand strain without any significant 50 modulus/high strain hybrid gel. The results presented in this
reduction in mechanical characteristics, which is important for paper suggest a third variant, the carrageenan/epoxy-amine
the future application of these materials as load tolerant hybrid gels are much stiffer than either gel alone, and retain the
hydrogels18,19. This was investigated by subjecting the hydrogels high strain to failure of the epoxy-amine gels.
15 to a cyclic testing regime between 20% and 40% compressive
strain (rate 50 mm/min). The gels were held for 2 s between
loading and unloading cycles. The evolution of typical EA, k-CG
and k-CG/EA gels is shown in Fig. 4. The tangent moduli of the
loading and unloading cycles are approximately constant for the
20 EA and k-CG/EA gels, but decrease significantly for k-CG gels.
The relative mechanical characteristics are maintained, i.e. ICE
gels of k-CG/EA have higher tangent modulus compared to EA
gels. These results indicate that the mechanical characteristics of
the EA and ICE gels are not adversely affected by the imposed
25 testing regime over 10 cycles. In contrast, the k-CG gels are
unable to withstand repeated mechanical compression.

Fig. 4. Typical tangent moduli (20-30% strain) calculated using the (a)
55 loading and (b) unloading parts of the cyclic testing regime for kappa-
carrageenan (k-CG, triangles), epoxy-amine (EA, squares) and k-CG/EA
ICE (diamonds) hydrogels.

Conclusions
In summary, we have introduced ionic covalent entanglement
60 (ICE) hydrogels based on interpenetrating polymer networks of
covalently bound epoxy-amines and ionically cross-linked
polysaccharides. Networks of biopolymers of the carrageenan and
gellan gum families were entangled with covalently linked
epoxy-amine in a ‘one pot’ synthesis with minimal processing. It
Fig. 3. Typical amplitude sweeps of (a) kappa-carrageenan, and (b)
kappa-carrageenan/epoxy-amine ICE hydrogels. Storage and loss moduli 65 was found that carrageenan biopolymers provided substantial
are indicated by squares and triangles, respectively. mechanical reinforcement when entangled with epoxy-amine
systems, whereas gellan gum did not. Specifically, the addition of
30 Similar results were obtained for the other types of gels, i.e. kappa-carrageenan to the epoxy-amine yielded a 5-6 fold increase
biopolymer gels were unable to withstand the cyclic compression, in tangent modulus and compressive stress at failure value when
but i-CG/EA and GG/EA ICE gels were not affected (Fig. S3,

This journal is © The Royal Society of Chemistry [year] Journal Name, [year], [vol], 00–00 | 3
 compared to epoxy-amine gels of comparable water contents. The
ICE gels were formed under mild conditions and are seemingly
tough, flexible and resilient. Consequently, ICE gels provide a
promising avenue for research towards load tolerant hydrogels.

5 Acknowledgements
This work was supported by funding from the University of
Wollongong, and Australian Research Council Centre of
Excellence, Laureate Fellowships (G.G. Wallace) and Future
Fellowship (M. in het Panhuis) Programs. Dr R.C. Clark, Mr P.
10 Jackson (both C.P. Kelco), and Mr P. Balding (UoW) are thanked
for provision of gellan gum and technical assistance, respectively.

Notes and references

a
Soft Materials Group, School of Chemistry, University of Wollongong,
Wollongong, NSW 2522, Australia. Tel: +61 24221 3155; E-mail:
15 [email protected]
b
Intelligent Polymer Research Institute, ARC Centre of Excellence for
Electromaterials Science, AIIM Facility, University of Wollongong,
Wollongong, NSW 2522, Australia; E-mail: [email protected]
c
Department of Bioengineering, University of Massachusetts Dartmouth,
20 North Dartmouth, MA, 02747, USA; E-mail: [email protected]
† Electronic Supplementary Information (ESI) available: [typical
compressive strain curves for i-CG/EA hydrogels, typical strain sweeps
for i-CG/EA and GG/EA hydrogels and cyclic testing of GG/EA
hydrogels]. See DOI: 10.1039/b000000x/
25

1 D. Gómez-Díaz, J.M Navaza, J. Food Eng., 2003, 56, 387-392.

2 P. Gupta, K. Vermani, S. Garg, Drug Discov. Today, 2002. 7, 569-
579.
3 P.C. Nicolson, J.R. Vogt, Biomaterials, 2001, 22, 3273-3283.
4 N.A. Peppas, Polymer, 1977, 18, 403-407.
5 R. Kishi, K. Hiroki, T. Tominaga, K.I. Sano, H. Okuzaki, J.G.
Martinez, T.F. Otero, Y. Osada, J. Polym. Sci. Pt. B-Polym. Phys.,
2012, 50, 790-796.
6 P. Calvert, Adv. Mater. 2009, 21, 743-756.
7 J. P. Gong, Y. Katsuyama, T. Kurokawa, Y. Osada, Adv. Mater.
2003, 15, 1155-1158.
8 M.A. Haque, T. Kurokawa, J.P. Gong, Polymer, 2012, 53, 1805-
1822.
9 D. Myung, Polymer, 2007, 48, 5376-5387.
10 J.P. Gong, T. Kurokawa, T. Narita, G. Kagata, Y. Osada, G.
Nishimura, M. Kinjo, J. Am. Chem. Soc., 2001, 123, 5582-5583.
11 E. R. Morris, K. Nishinari, M. Rinaudo, Food Hydrocolloids 2012,
28, 373-411.
12 T. Thrimawithana, Carbohyd. Polym., 2010. 82(1): p. 69-77.
13 Z.A.A. Hamid, et al., Biomaterials 2010, 31, 6454-6467.
14 C.J. Ferris, M. in het Panhuis, Soft Matter 2009, 5, 3430-3437.
15 Y. Yoshioka, P. Calvert, Exp. Mech. 2002. 42, 404-408.
16 D. R. Pereira, J. Silva-Correia, S. G. Caridade, J. T. Oliveira, R. A.
Sousa, A. J. Salgado, J. M. Oliveira, J. F. Mano, N. Sousa, R. L. Reis,
Tissue Eng. Part C 2011, 17, 961-972.
17 V.E. Santo, A.M. Frias, M. Carida, R. Cancedda, M.E. Gomes, J.F.
Mano, R.L. Reis, Biomacromolecules 2009, 10, 1392-1401.
18 S.E. Bakarich, G.C. Pidcock, P. Balding, L. Stevens, P. Calvert, M.
in het Panhuis, Soft Matter 2012, 8, 9985-9988.
19 J.-Y. Sun, X. Zhao, W. R. K. Illeperuma, O. Chaudhuri, K. H. Oh, D.
J. Mooney, J. J. Vlassak, Z. Suo, Nature, 2012, 489, 133–136.
20 H.A. Barnes, J. Non-Newton. Fluid 1997, 70, 1-33.
21 R. E. Webber, C. Creton, H. R. Brown, J. P. Gong, Macromolecules
2007, 40, 2919-2927.
22 J. P. Gong, Soft Matter 2010, 6, 2583-2590.

4 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]