Jellyfish gel and its hybrid hydrogels with high mechanical strength

Abstract

The fabrication of hydrogels with well-defined structure and high mechanical strength has become a challenging and fascinating topic. The aim of this study is to develop a new method for fabricating hydrogels with high mechanical strength by utilizing the well-developed structure of biological gels. We firstly studied the mechanical properties and microstructure of a biological gel—the mesogloea of edible jellyfish Rhopilema esculenta Kishinouye (JF gel). JF gel has much higher mechanical strength than normal synthetic hydrogels due to its layered porous structure with pore walls consisting of nano-structured layers and fibers. We have also synthesized hydrogels by radiation-induced polymerization and crosslinking and found that they are distinctly stronger than those produced by the classical thermal polymerization using a crosslinking agent. When a synthetic gel is incorporated into JF gel by the radiation-induced polymerization and crosslinking of a hydrophilic monomer, a novel type of hybrid hydrogel with very high mechanical strength results. The compressive and tensile strengths of the hybrid hydrogels are generally several times to more than ten times higher than those of JF gel and the corresponding component synthetic gels. The hybrid gels combine the well-developed structure of biological jellyfish gel and the unique microstructure of the synthetic gel produced by the radiation method, and strong interactions between the two networks are formed.

---

Introduction

Most tissues of animals are mainly composed of biological gels, and in some living beings such as jellyfish and sea cucumbers, hydrogel composes almost their whole body. The biological hydrogels in animals are usually elastic and tough at a high water content.¹ If synthetic hydrogels have similar mechanical properties as the biological hydrogels in animals, they may serve as ideal candidates for replacing damaged human tissues. Unfortunately, man-made hydrogels are usually mechanically weak, thereby strongly impeding their applications in biomedical and other fields. Except for chemical compositions, the pivotal reason for the huge difference in the mechanical properties between biological gels and synthetic gels is their different microstructures. Biological gels usually have well-developed microstructures, while normal synthetic hydrogels have no ordered structure at the molecular level.¹

There is currently a widespread interest in developing new types of hydrogels with unique microstructures and high mechanical strength.^2–4 Topological (TP) gels⁵ and nanocomposite (NC) gels⁶ exhibit high extension to break, up to 10–20 times the original length, and double-network (DN) gels^7,8 have a high modulus (just sub-MPa), with a failure compressive stress as high as 20 MPa. Malkoch et al. have synthesized new poly(ethylene glycol) (PEG)-based hydrogel materials by click chemistry, the hydrogels have well-defined networks and significantly improved mechanical properties.⁹ We have also developed a “macromolecular microsphere composite” (MMC) hydrogel with high mechanical strength.¹⁰ Some other strong hydrogels based on these hydrogels or hydrogels with novel microstructures have been reported.^11–17 A common feature of these hydrogels is that they have unique microstructures which are effective in spreading the applied load over a broad area close to a track tip, reducing the crack tip stresses and hence slowing crack propagation.

Most hydrogels are made directly from synthetic polymeric materials, such as poly(vinyl alcohol) (PVA), or by the polymerization of hydrophilic monomer(s). In recent years, hydrogels made from natural materials have drawn more attention due to their better biocompatibility and biodegradability, as well as bio-responsive properties. Pristine or artificially modified biological materials, such as peptides and proteins^18–22 and DNA,²³ have been used to make hydrogels for different applications. Many kinds of hybrid gels have been made from synthetic polymer and biological materials.^24–27 However, they made use of only the biomaterials separated from the animals, and hence the materials did not have their original well-developed complex structures seen in animals.

Jellyfish are marine invertebrates belonging to the Scyphozoan class. The umbrella of jellyfish is composed of both mesogloea and outer skin.²⁸ It is mainly made up of collagen which forms the extracellular matrix²⁹ (ECM) and contains about 96–97 wt.-% (with outer skin) water when swelled to equilibrium.³⁰ When the salts in it are extracted, as was done here, the water content can be as high as 99 wt.-%. Even with such a high water content, in our experience, jellyfish still have quite a high mechanical strength. However, little attention has been paid to the values and understanding of the mechanical properties of jellyfish.

We fabricate a new type of hybrid hydrogel by introducing a synthetic hydrogel into a biological hydrogel which is directly obtained from an animal body. The biological hydrogel has a well-developed structure and relatively high mechanical strength, the synthetic hydrogel formed in the biological hydrogel might have a microstructure different to normal hydrogels. We anticipate that the combination of the two may offer a substantial increase in mechanical strength of the hybrid hydrogel.

In this work, we studied the microstructure and the mechanical properties of a biological gel—the mesogloea of an edible jellyfish Rhopilema esculenta Kishinouye (hereafter referred to as JF gel)—and fabricated the hybrid hydrogels of the JF gel and synthetic gel using a radiation-induced polymerization and crosslinking method.

Experimental

Materials

The jellyfish used was Rhopilema esculenta Kishinouye, and the slabs of jellyfish umbrellas purchased from the fish market were usually treated with a mixture of table salt and alum. They were washed thoroughly with deionized water for 72 h to remove the salts in them. The outer skins of the umbrellas of the swollen jellyfishs were removed carefully, leaving only the mesogloea as the raw materials (JF gel). The JF gel was cut into cubic or cylindrical pieces. Acrylic acid (AA, AR grade) purchased form Bodi Chemicals Co. Ltd. (Tianjin, China) was redistilled before use. Acrylamide (AAm, AR grade) was from Komiou Chemicals Co. Ltd. (Tianjin, China). N,N′-Methylene-bis-acrylamide (MBAA, ultra pure grade) was from Amresco Inc. (OH, USA).

Preparation of gels

Polyacrylic acid (PAA) and polyacrylamide (PAAm) gels were prepared by direct irradiation the monomer solutions with ⁶⁰Co-γ rays (dose rate: 10 kGy/h) for 2 h in air at ambient temperature. The hybrid hydrogels were prepared as follows: the JF gel was put into an aqueous monomer solution with or without a crosslinker MBAA for 18 h to ensure the full exchange of the water in JF gel with the monomer solution, then the sample was irradiated with ⁶⁰Co-γ rays (dose rate: 10 kGy/h) for 2 h, resulting in a hybrid gel.

SEM investigations

The JF gel and the synthesized hydrogel samples for SEM analyses were cut from the inner part of the hydrogels. To avoid the formation of large ice crystals, the hydrogels were rapidly plunged into liquid nitrogen for about 5 min. The samples were subsequently freeze-dried until all water was removed. The dried samples were cracked to little pieces and the appropriate ones with fresh surfaces were analyzed with a Hitachi S-4800 scanning electron microscope (Tokyo, Japan).

Mechanical tests

The compressive stress–strain measurements were performed on as-prepared hydrogels using a CSS-2202 electronic universal testing machine (Changchun Institute of Experimental Machines, Changchun, China) at a crosshead speed of 5 mm/min. The cylindrical gel samples were 15–20 mm in diameter and 8–12 mm in thickness. The stress σ_c is calculated as follows: σ_c = Load/(πr²) (r is the initial unloaded radius). The strain ε_c under compression is defined as the change in the thickness relative to the thickness of the freestanding specimen. Tensile mechanical measurements were performed on as-prepared gels using an Instron 3365 electronic universal testing machine (Instron Co., USA) at a cross-head speed of 50 mm/min. The cubic JF and hybrid gels were 4–6 mm in thickness and 8–12 mm in width, and the gauge lengths were 15–25 mm. The cylindrical PAA and PAAm gels were 8.7 mm in diameter and the gauge lengths were 15–25 mm. The tensile stress σ_t is calculated as follows: σ_t = Load/S₀ (S₀ is the initial cross section area). The elongation ε_t is defined as the change in the length relative to the gauge length of the freestanding specimen. Stress and strain between elongations of 10% and 50% were used to calculate initial modulus of elasticity.

Results and discussion

Jellyfish gel

We tested the mechanical properties of JF gel. As shown in Fig. 1, a JF gel fractured in a compression test at a compressive stress (σ_c) of 1.16 MPa and a strain (ε_c) of 80%, and during a tensile test it broke at a tensile stress (σ_t) of 21.3 kPa and an elongation (ε_t) of 160%. The σ_c and σ_t of the JF gel were not very high, but they are much higher than those of normal synthetic hydrogels,^6,8,10 especially when considering its very high water content of 99 wt.-%. It should also be pointed out that the mechanical properties of JF gel varied from sample to sample, with one apparent reason being the individual difference between the jellyfishes. We tested more than ten JF gels from different jellyfishes and found that their σ_c and σ_t had a relative variance about 20%. To avoid the random errors arising from sample difference, the JF gels used in this study had a variance in failure stress of less than 5%.

Fig. 1 Compressive (a) and tensile (b) stress–strain curves of a JF gel.

The mechanical properties of JF gels with lower water contents were not measured because of the difficulty in preparing suitable samples for tests. But it is reasonable to predict that the mechanical strength of jellyfish gel should increase with decreasing water content.

SEM investigations revealed that JF gel has an ordered structure (Fig. 2), at least after drying. Although it is not possible to know the relationship between the structure of the hydrated material and that of the dried material seen in the SEM, it seems likely that they are similar as the hydrated material was quite optically cloudy in spite of containing 99 wt.-% water. It seems reasonable to assume that this structure is retained on drying. At a large scale, it shows a layered porous structure with pore size up to 10 μm (Fig. 2a). Water could be squeezed out during compression and tensile tests. We think that this is evidence for the intrinsic existence of pores as large as 10 μm in jellyfish. Higher magnification investigations indicated that the pore walls consist of layers and many fibers connected to the layers (Fig. 2b–d). The layers are mainly dense and flat slices with some pores on them, and some parts of the layers even consist of many fibers. The thicknesses of the layers and the diameters of the fibers are about 20–50 nm. The nano-structured layers and fibers are formed by the aggregation of many biological macromolecular chains (mainly collagen²⁹) into condensed structures.

Fig. 2 SEM micrographs of the freeze-dried jellyfish mesogloea.

Therefore they are much stronger than the single chains found in normal synthetic hydrogels. In addition, the layered structure provides the mechanism to broadly disperse stress near a crack tip, ensuring that very large external loads are required to bring the force on crack tip chains up to that required for chain scission. These are very possibly the main reasons for the high mechanical strength of JF gel. However, the strong interactions among the layers and fibers make them hard to deform (giving a high elastic modulus considering the high water content) and thereby cause the low fracture strain and elongation of JF gel.

Hydrogels synthesized with radiation method

High-energy radiation has been widely used to prepare hydrogels for a long time.^31–33 However, little attention has been paid to the mechanical properties of the hydrogels obtained. We synthesized hydrogels by radiation-induced polymerization and crosslinking and found with some surprise that the hydrogels had very high mechanical strengths compared with classically polymerized gels, as will be shown later.

2 h ⁶⁰Co γ-rays irradiation (absorbed dose rate: 10 kGy/h) was found to be enough for the acrylic acid (AA) and acrylamide (AAm) aqueous solutions to be transformed into hydrogels which could not be dissolved in a large excess amount of water but instead showed equilibrium swelling behavior. All the gels were synthesized with 2 h irradiation in this study.

We prepared the hydrogels with hydrophilic monomers AA and AAm, and two series of gels were obtained, the first synthesized with no added crosslinker and a range of monomer concentrations (C_M = 2, 3 and 4 M, the obtained gels are numbered as A1, A2 and A3, respectively) and the second with C_M = 2 M and a range of crosslinker concentrations (C_C = 0.05, 0.1, 0.2 and 0.3 mol/mol% to monomer concentration, and the gels are B1, B2, B3 and B4). When a monomer solution was introduced into a JF gel and then irradiated with γ-rays, a hybrid hydrogel could be obtained due to the radiation-induced polymerization and crosslinking of the monomer. Therefore, simultaneously, hybrid hydrogels were also prepared by the irradiation of JF gels with the monomer solutions inside. The hybrid hydrogels are called JF-AA or JF-AAm gels according to the monomer used.

The compressive and tensile properties were measured in the as-synthesized, rather than the equilibrium swollen conditions. For each experiment, at least five samples were tested, and the test results of three samples with relative error less than 10% were used to calculate the averages. Table S1 (ESI†) summaries the synthesis conditions, water contents of the gels and some test results.

Fig. 3 shows typical compressive and tensile stress–strain curves of the synthesized hydrogels. For convenient comparison, the stress–strain curves of JF gel were also included. Fig. 4 shows the compressive stresses (σ_c), tensile stresses (σ_t) and elastic moduli (E) of the hydrogels.

Fig. 3 The typical compressive (a, b) and tensile (c, d) stress–strain curves of JF gel, PAA and PAAm gels, as well as JF-AA and JF-AAm gels.

Fig. 4 The compressive stresses (σ_c), tensile stresses (σ_t) and elastic moduli (E) of the hydrogels versus monomer concentration (C_M) and crosslinker concentration (C_C). The samples that did not break in the tests are marked with “x” in the symbols.

We shall consider first the low strain tensile properties and elastic moduli (E) of the materials. It is evident from Fig. 3c that the JF gels had as high or higher moduli than the A series (no added crosslinker) PAA gels in-spite of the former's much higher water content. However the PAAm gels had as high or higher moduli than the JF gel. Clearly the synthesis technique caused much more crosslinking in AAm than in AA. The hybrid JF-AA gels showed just a little higher moduli than the JF gel suggesting that the JF gel provided some additional crosslinking to the PAA (or visa versa). However there was probably some reaction between the two components. The JF-AAm gels had very much higher moduli than either of the components showing that there must have been considerable reaction between the two components. As the moduli in the latter system actually decreased a little with increasing either monomer and crosslinker content it seems likely that most of the load at small strains was held on chains that bridged between the two components of the gel.

Considering now the fracture properties, in the compression tests (Fig. 3 a,b, Fig. 4 a,b and Table S1, ESI†), PAA hydrogels fractured in a σ_c range 0.31–4.82 MPa and a ε_c range 77.5–92.3%. The σ_c increased with increasing C_C, but it initially increased with increasing C_M, from 0.35 MPa (2 M) to 4.82 MPa (3 M), and then decreased. In contrast, the PAAm gels (A series) didn't fracture even at ε_c >95%, the σ_c (in these cases the stress at ε_c = 95%) increased with C_M, from 1.90 MPa (2 M) to 7.98 MPa (4 M). The comparison between the two materials is particularly interesting as the PAAm materials (A series) had by far the higher moduli. However, all the PAAm gels (B series) synthesized with the addition of a crosslinker broke in the compression tests. When C_M was 2 M, the σ_c firstly increased with C_C to a maximum of 5.16 MPa (0.10%) and then decreased dramatically. The fracture ε_c monotonously decreased with C_C.

In the tensile tests (Fig. 3c,d, Fig. 4 c–f and Table S1, ESI†), PAA and especially PAAm gels synthesized without a crosslinker exhibited very high elongations, about 800–1200% for PAA gels and 1500–1700% for PAAm gels, indicating they were highly elastic. The PAAm gels (A series) didn't break even at such high elongations. These elongation values are very close to those for the hydrogels exhibiting high extensions.^5,6,9,34 However, with the addition of a crosslinker, the elongation of the gels (B series) decreased dramatically, with a decrease to about 200–300% for PAA gels and less than 100% for PAAm gels. The σ_t of the PAA gels synthesized with higher values of C_M (3 M and 4 M) and C_C (0.30%) were higher than that of JF gel (21.3 kPa). The σ_t and the moduli of PAAm gels (A series) were usually higher than those of JF gel and the corresponding PAA gels. Also the σ_t of the PAAm gels with crosslinker (B series) were lower than that of PAAm gel (A1) synthesized with the same C_M but without a crosslinker, but the elastic moduli of the former were higher than those of the latter. This latter observation is consistent with the tendency of increasing the crosslinking to increase the modulus of a network (by the Gaussian theory) but decrease its toughness (by the Lake–Thomas model).

Hybrid gels

Since we intended to make use of the original structure of the JF gel, it was necessary to make sure that JF gel did not decompose during the preparation process, and so two experiments were carried out: one was the irradiation of JF gel with γ-rays for 2 h, the other was the immersion JF gel in monomer solutions for 18 h. In both cases there was no significant change in the appearance and mechanical strength of the JF gel. We have also observed the JF gels with SEM and found that there was no significant change in their microstructures (Fig. S1, ESI†), suggesting that JF gel kept its original microstructure.

The compressive and tensile test results for the hybrid gels are included in Fig. 3 and 4. They exhibited strikingly high mechanical strengths, and they were usually much stronger than JF gel and the corresponding PAA and PAAm gels.

For the JF-AA hybrid hydrogels, in compression tests only two samples (B3 and B4) broke and their ε_c were more than 90%. It has to be mentioned that the other samples did not fracture even at ε_c > 95%, so their fracture σ_c could not be obtained and the quoted σ_c values were chosen at ε_c=95%. The σ_c increased with increasing C_M but, with the increase of C_C, it increased until C_C was 0.20%, and then decreased. The JF-AA gel (A3) synthesized with 4 M AA had the highest σ_c of 28.8 MPa at 95% strain. In tensile tests, the σ_t of the JF-AA gels were higher than those of PAA gels and usually 3 times that of JF gel. The JF-AA gels exhibited higher ε_t than JF gel but lower than the corresponding PAA gels. The addition of a crosslinker induced a significant decrease in ε_t, but σ_t increased with C_C till 0.20% and then decreased. The moduli of the JF-AA gels were normally higher than that of JF gel, and especially than those of PAA gels.

The mechanical strengths of the hybrid JF-AAm gels were usually higher than those of the JF-AA gels. The σ_c of the JF-AAm gels (A series) was about 2–4 times higher than those of the corresponding PAAm gels and 7–30 times higher than that of JF gel. The highest σ_c was 34.9 MPa (A3). The JF-AAm gel (A2) with an 84.5 wt.-% water content still had a high σ_c of 19.5 MPa. The addition of a proper amount of crosslinker could make the JF-AAm gel stronger in compression but not tension, the σ_c of JF-AAm gel (B2) was 21.3 MPa which is about twice of the σ_c (11.5 MPa) of the corresponding JF-AAm gel (A1) without a crosslinker. However, an excess amount of crosslinker induced a dramatic decrease in σ_c. The σ_t of the JF-AAm gels (A series) were usually 9–11 times that of JF gel and 2–6 times higher than those of PAAm gels (A series). The JF-AAm gels exhibited similar elongations as that of the JF gel but much lower than those of corresponding PAAm gels. The addition of a crosslinker led to the small decrease in σ_t and modulus.

The hybrid gels could also deform under high torsion stresses to large strains without any sign of damage. The JF-AA gel (B4) was twisted with hand for several turns and it totally recovered its original shape immediately after the release of torsional forces (ESI,† Video 1). On the contrary, the PAA gel (B4) and JF gel were readily fractured after only about 180° torsion (ESI,† Video 2 and 3).

These mechanical test results indicate that the incorporation of a synthetic gel into jellyfish gel by the radiation-induced polymerization and crosslinking method is a convenient and effective way to fabricate hybrid hydrogels with very high mechanical strength.

The photos of JF gel, PAA gel, JF-AA gel and JF-AAm gels are shown in Fig. S2 (ESI†). The JF gel was semi-transparent, the PAA gel (and PAAm gel) was highly transparent, and the hybrid gels were semi-transparent (JF-AAm gels) or opaque (JF-AA gel). Fig. 5 shows the typical SEM micrographs of freeze-dried PAA gel, JF-AA gel and JF-AAm gel. The PAA gel and PAAm gel have very similar porous structures that are normally seen in synthetic hydrogels. Presumably these structures form during the drying process, as the swollen gels are highly transparent. However, JF-AA gel and JF-AAm gel show porous structures different to that of the PAA gel or PAAm gel. They have a layered distribution of small and large pores and the distance between two layers of small pores is about several to ten micrometres, which is similar to the size of the pores in JF gel (Fig. 2a). Since the layered porous structure of JF gel (Fig. 2a) is not damaged during the preparation processes and is visible in the SEM micrographs, and the hybrid gels are semi-transparent or opaque, we can conclude that the hybrid gels inherit the layered structure from the JF gel. However, the nano-structured layers and fibers in the JF gel could not be observed in the hybrid gels. The JF-AA and JF-AAm hybrid gels did not dissolve in a large excess of water, but showed equilibrium swelling behavior and retention of the AA or AAm after a long time (more than 1 month). These results and the significant increases in elastic moduli show that PAA or PAAm gel has been successfully incorporated into JF gel to obtain a hybrid gel.

Fig. 5 SEM micrographs of the PAA gel (B4) with 84.4 wt.-% water (a), JF-AA gel (B4) with 85.5 wt.-% water (b) and JF-AAm gel (B2) with 84.5 wt.-% water (c).

The PAA and PAAm gels synthesized by radiation-induced polymerization and crosslinking are much stronger than the gels synthesized by the traditional chemical method,¹⁰ suggesting their microstructures are different. Normal synthetic hydrogel has an irregular distribution of crosslinking points and uneven distribution of chain length between crosslinking points, therefore the stress can not be evenly distributed among the polymer chains. The hydrogels synthesized by radiation-induced polymerization and crosslinking can not have an ordered microstructure at the molecular level either. However, the radiation method has some distinct features and may lead to differences in the microstructure of the gel. Firstly, the crosslinking of polymer chains in the radiation method in the absence of any common crosslinking agent is produced by the combination of two radicals formed on vicinal polymer chains. High energy γ-ray irradiation produces radicals on polymer chains homogeneously and may lead to a more even distribution of crosslinking points than in a gel produced with a crosslinker, where reactivity ratios can affect crosslink distribution. In addition, the polymer chains in the hydrogels synthesized without a crosslinker must be long and with a low crosslink density, particularly for the PAA with its low E, and hence the gels can sustain high deformation, as evidenced by the high ε_c and ε_t of the gels. When a small amount of crosslinker was added, the elongation of the gels decreased dramatically, especially for the PAAm gels. The increase in crosslink density leads to a decrease in chain length between crosslinking points, and hence the decrease in elongation. Conversely the addition of a proper amount of crosslinker can lead to the increase in σ_c, σ_t and E of the gels, greater number of polymer chains between crosslinks per unit volume. Secondly, high energy irradiation is a very effective way to initiate grafting copolymerization, radiation can cause grafting on pre-existing polymer chains. Grafted chains are covalently attached to polymer chains, this leads to the increase in crosslinking density, and, more importantly, the polymer chain with grafted chains on it can function as a flexible long chain “crosslinking agent” which can sustain much higher deformation than normal low molecular weight (LMW) crosslinking agent. In summary, the radiation method perhaps causes a chain length distribution that is more favorable for even load distribution between the chains.

The most interesting and important result of this work is that the introduction of a gel, synthesized by radiation-induced polymerization and crosslinking, into a JF gel will lead to a hybrid gel which is much stronger than the constituent gels. The mechanical strengths of the hybrid hydrogels are generally several to several tens times higher than those of the JF gel and the corresponding PAA gels and PAAm gels, indicating there is a strong synergism rather than a simple addition effect between the JF gel and the synthetic gel. The mass fractions of JF gel in the solid content of hybrid hydrogels were less than one tenth, leading to the question, how can such a small amount of JF gel cause the dramatic increase in the mechanical strengths of the hybrid hydrogels? There is an interesting parallel here with the DN gels introduced by Gong and co-workers.⁷ In the latter system the first network is typically only 5–10% of the mass of the polymer in the gel but its existence not only considerably increases the strength of the combined system over that of just the second network but also it produces the main contribution to the elastic modulus.

In our system, firstly, there probably are strong interactions between the JF gel and the PAA or PAAm gels formed inside it. One very possible mechanism for such interaction is that, in addition to homopolymerization, the monomers can also be grafted onto the layers and the fibers of the JF gel to form covalent interaction points between the two networks. The occurrence of the grafting reaction on to the JF gel is possible as the high-energy irradiation causes breakage of C–H or other bonds on the biomacromolecular chains and hence the formation of macromolecular radicals which can initiate grafting. In addition to the covalent coupling, the polymer network formed inside JF gel is bound to be entangled with the nanofibers in JF gel and it can also form hydrogen bonding with the biomacromolecules.

To confirm the contribution of grafting to the increase of mechanical strength of hybrid gel, two experiments were carried out. We firstly polymerized a monomer inside a JF gel with a common chemical initiator ammonium persulfate (APS) which would not initiate grafting and in some cases a crosslinker N,N′-methylene-bis-acrylamide (MBAA) and found that the obtained hybrid gels had very similar mechanical properties as those of JF gel, i.e., there was no significant enhancement in the mechanical properties of this kind of hybrid gels. An interesting phenomenon found in these hybrid gels is that the synthetic gel inside the JF gel could be squeezed out during a compressive test and even in a swelling process, leading to the separation of brittle and transparent synthetic gel from opaque JF gel. These results suggest that there is no strong interaction between the gel synthesized by the common chemical method and the JF gel.

The second experiment is a series of cyclic tensile tests of a JF-AA hydrogel (B3) similar to the experiments done by Webber et al. on DN gels.³⁵ The loading portion of the strain–stress curves is shown in Fig. 6. The first test was stopped at a strain of 170% which is a little higher than the fracture strain of JF gel (160%) but lower than that of PAA gel (B3) (365%). The same gel was tested for the second time in which extension was stopped at a strain of 175% and the third test which was continued until the gel broke at a strain of 200%. A significant change in the stress–strain curves between the first and the second tests can be seen, suggesting there was permanent damage to the hydrogel. Very possibly, the JF gel structure in the hybrid gel began to break, particularly when the strain was higher than the fracture strain of JF gel, leaving the unbroken PAA gel structure. The stress–strain curves of the second and third tests almost overlap in the same strain range, indicating there was no further structural damage on the second loading and that the damage did not recover with time. The fracture σ_t of the third test is 92.2 kPa which is a little lower than that of JF-AA gel (B3) (104 kPa) but is much higher than that of PAA gel (B3) (28.5 kPa), indicating that the partially broken JF gel still contributes a major component to the elastic modulus and also to the mechanical strength and so the structure of the hybrid gel after the first tensile test must still be very different from that of PAA gel. These results suggest that strong physical (entanglement) and chemical (grafted chains) interactions must have been formed between the JF gel and the PAA gel in the hybrid gel during the radiation process, the first test damaged only part of the gel structure of JF gel, and perhaps broke it into pieces, the remaining JF gel still had strong physical and/or chemical (grafting) interactions with the PAA gel, thus the hybrid gel still had very high mechanical strength. Similar effects have been observed in classic DN gels and this breakup of the first network is thought to be the major energy dissipation process that gives the DN gels a high toughness and strength.³⁶ The microstructure of JF gel is thought to be important for the mechanical properties of the hybrid gels. The hybrid gel has a structure similar to that of DN gels proposed by Gong⁷ in that the two networks are entangled and sometimes grafted together. In contrast to the DN gel, the first network of the hybrid gel is a biological gel which has a well-developed structure, and the second network is also different from a normal synthetic gel which normally has uneven distribution of crosslinking points and chain lengths. In our hybrid hydrogels, the JF gel acts as the first network and may be a template for the formation of the second network. Given the unusual microstructure of the JF gel, consisting of fine fibers and layers, it seems very likely that the energy required to break it into small blocks is the main contribution to the fracture energy that gives the hybrid gels high strength. Due to the presence of the first network, the second network is confined in the interspaces between the porous layers. The hybrid gel inherits the structural heterogeneity from the JF gel, as evidenced by the SEM investigations. As pointed out by Gong et al.,² the structural inhomogeneity correlates with the drastic improvement of the strength of the DN gels. Very possibly, the structural inhomogeneity in the hybrid JF-AA gel can also improve its mechanical strength.

Fig. 6 The stress–strain curves of the loading portion of cyclic tensile tests of the same JF-AA gel (B3). The tensile stress–strain curves of JF gel and PAA gel (B3) are also shown for comparison.

Based on the experimental results and discussion above, the fabrication mechanism of the hybrid gels and the microstructures of JF gel and hybrid gel are proposed as Fig. 7.

Fig. 7 The proposed microstructures of JF gel and hybrid gel.

Conclusions

Jellyfish gel has quite high compressive and tensile strengths even with a very high water content of 99 wt.-%. When freeze-dried the jellyfish gel has a layered porous structure and its pore walls consist of nano-structured layers with many fibers connected to the layers. Presumably a similar structure remains when the material is water swollen as it is optically cloudy. The well-developed microstructure provides the mechanism for the jellyfish gel to disperse stress on it and hence suppress stress concentrations at flaws and crack tips. The hydrogels synthesized using the radiation-induced polymerization and crosslinking technique usually are much stronger than those synthesized with traditional chemical methods, possibly because the radiation techniques form a chain length distribution that is more favorable for the even load distribution amongst the chains. The most interesting and important result in this work is the preparation of hybrid gels with very high mechanical strengths. The compressive and tensile strengths of the hybrid hydrogels are generally several to several tens times higher than those of the JF gel and the corresponding PAA gels and PAAm gels.

Our hybrid hydrogel system possesses several distinct features. Firstly, the hybrid hydrogels are the first hydrogels made by the direct incorporation of a synthetic gel into a biological gel obtained directly from an animal body. This hybrid gel makes use of not only the biomaterials, but also, more importantly, their well-developed microstructures. Secondly, this method is very simple and may become a general strategy for hydrogel fabrication due to its broad choices of monomers and the living beings. Many hydrophilic monomers can be used due to the non-selective initiation effect of high-energy irradiation. The use of jellyfish as a building material should be applicable to other biological gels, including the tissues in animal bodies. The radiation method is especially applicable to the fabrication of hydrogels for biomedical uses, because of its mild reaction conditions, lack of initiator residue and the in situ sterilization of products in the radiation process. Thirdly, the hybrid gels have very high compressive, tensile and torsional strengths, with strengths comparable to the strongest hydrogels even synthesized.^6,7,10 Some of the mechanical properties are close to those of the tissues in human body. It is also possible to improve the mechanical properties and biocompatibility to meet the requirements for tissue engineering. We believe that this work will trigger the fundamental and application studies on the new hybrid hydrogels.

It should be noted that this study focused only on the preparation and mechanical properties of the new hybrid gels. Some interesting and important problems still remain. For example, what is the microstructure of jellyfish gel at the molecular level? Understanding this problem may open a new way to make hydrogels with similar microstructure and mechanical properties as those of jellyfish and improve them further. Although we have discussed the reasons for the high mechanical strengths of the hydrogels synthesized using radiation-induced polymerization and crosslinking and the hybrid gels, more detailed studies should be carried out to confirm our speculations.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant No.50673013), China-Australia Fund for S&T Cooperation (Grant No. 50811120112) and Beijing Municipal Commission of Education.

References

1. T. Miyazaki, K. Yamaoka, J. P. Gong and Y. Osada, Macromol. Rapid Commun., 2002, 23, 447–455.
2. Y. Tanaka, J. P. Gong and Y. Osada, Prog. Polym. Sci., 2005, 30, 1–9.
3. P. Calvert, Adv. Mater., 2009, 21, 743–756.
4. J. Kopecek, Biomaterials, 2007, 28, 5185–5192.
5. Y. Okumura and K. Ito, Adv. Mater., 2001, 13, 485–487.
6. K. Haraguchi and T. Takehisa, Adv. Mater., 2002, 14, 1120–1124.
7. J. P. Gong, Y. Katsuyama, T. Kurokawa and Y. Osada, Adv. Mater., 2003, 15, 1155–1158.
8. A. Nakayama, A. Kakugo, J. P. Gong, Y. Osada, M. Takai, T. Erata and S. Kawano, Adv. Funct. Mater., 2004, 14, 1124–1128.
9. M. Malkoch, R. Vestberg, N. Gupta, L. Mespouille, P. Dubois, A. F. Mason, J. L. Hedrick, Q. Liao, C. W. Frank, K. Kingsbury and C. J. Hawker, Chem. Commun., 2006, 2774–2776.
10. T. Huang, H. Xu, K. Jiao, L. Zhu, H. R. Brown and H. Wang, Adv. Mater., 2007, 19, 1622–1626.
11. M. F. Zhu, Y. Liu, B. Sun, W. Zhang, X. L. Liu, H. Yu, Y. Zhang, D. Kuckling and H. J. P. Adler, Macromol. Rapid Commun., 2006, 27, 1023–1028.
12. M. K. Shin, G. M. Spinks, S. R. Shin, S. I. Kim and S. J. Kim, Adv. Mater., 2009, 21, 1712–1715.
13. W. C. Lin, W. Fan, A. Marcellan, D. Hourdet and C. Creton, Macromolecules, 2010, 43, 2554–2563.
14. T. Y. Dai, X. T. Qing, H. Zhou, C. Shen, J. Wang and Y. Lu, Synth. Met., 2010, 160, 791–796.
15. M. E. Seitz, D. Martina, T. Baumberger, V. R. Krishnan, C. Y. Hui and K. R. Shull, Soft Matter, 2009, 5, 447–456.
16. G. Q. Jiang, C. Liu, X. L. Liu, Q. R. Chen, G. H. Zhang, M. Yang and F. Q. Liu, Polymer, 2010, 51, 1507–1515.
17. Y. Tan, K. Xu, P. X. Wang, W. B. Li, S. M. Sun and L. S. Dong, Soft Matter, 2010, 6, 1467–1471.
18. J. D. Ehrick, S. K. Deo, T. W. Browning, L. G. Bachas, M. J. Madou and S. Daunert, Nat. Mater., 2005, 4, 298–302.
19. S. Frey, R. P. Richter and D. Goerlich, Science, 2006, 314, 815–817.
20. W. A. Petka, J. L. Hardin, K. P. McGrath, D. Wirtz and D. A. Tirrell, Science, 1998, 281, 389–392.
21. Z. J. Sui, W. J. King and W. L. Murphy, Adv. Funct. Mater., 2008, 18, 1824–1831.
22. A. Saiani, A. Mohammed, H. Frielinghaus, R. Collins, N. Hodson, C. M. Kielty, M. J. Sherratt and A. F. Miller, Soft Matter, 2009, 5, 193–202.
23. S. H. Um, J. B. Lee, N. Park, S. Y. Kwon, C. C. Umbach and D. Luo, Nat. Mater., 2006, 5, 797–801.
24. C. Wang, R. J. Stewart and J. Kopecek, Nature, 1999, 397, 417–420.
25. C. Wang, J. Kopecek and R. J. Stewart, Biomacromolecules, 2001, 2, 912–920.
26. S. Halstenberg, A. Panitch, S. Rizzi, H. Hall and J. A. Hubbell, Biomacromolecules, 2002, 3, 710–723.
27. K. Nam, T. Kimura and A. Kishida, Biomaterials, 2006, 28, 1–8.
28. V. Schmid, S. I. Ono and S. Reber-Muller, Microsc. Res. Tech., 1999, 44, 254–268.
29. E. Song, S. Y. Kim, T. Chun, H.-J. Byun and Y. M. Lee, Biomaterials, 2006, 27, 2951–2961.
30. A. G. Lowndes, Nature, 1942, 150, 234–235.
31. H. M. Nizam El-Din, S. G. Abd Alla and A. W. M. El-Naggar, J. Macromol. Sci., Part A: Pure Appl. Chem., 2007, 44, 47–54.
32. A. Hiroki, Y. Maekawa, R. Katakai, T. Yamashita, Y. Muroya, Y. Katsumura and M. Yoshida, Macromolecules, 2006, 39, 4132–4137.
33. D. Solpan, S. Duran and O. Guven, J. Appl. Polym. Sci., 2002, 86, 3570–3580.
34. Y. H. Na, Y. Tanaka, Y. Kawauchi, H. Furukawa, T. Sumiyoshi, J. P. Gong and Y. Osada, Macromolecules, 2006, 39, 4641–4645.
35. R. E. Webber, C. Creton, H. R. Brown and J. P. Gong, Macromolecules, 2007, 40, 2919–2927.
36. H. R. Brown, Macromolecules, 2007, 40, 3815–3818.

---

Footnote

† Electronic Supplementary Information (ESI) available: Supporting figures and videos. See DOI: 10.1039/c0sm00632g

---