2.1. UV-induced sol to gel transition of PNiPAM solution
Before constructing the IPN network, the conditions for PNiPAM gel formation were optimized. Unlike other systems where temperatures varied from around 50°C to as high as 75 °C [
17,
31], the IPN preparation was operated explicitly at room temperature. Therefore, the gelation time needed to be identified to ensure the formation of a three-dimensional network. Several techniques, such as light scattering (LS) [
12,
32], beam diffraction [
13], and rheology [
33,
34,
35], have been widely used to elucidate the mechanism of material gelation and its molecular structure. In this work, the rheological technique was used to investigate the crosslinking behavior due to its direct correlation with the evolving physical and mechanical properties of the system during the crosslinking process.
Figure 1(a) illustrates the relationship between zero-shear viscosity (
η0) and UV irradiation time (
t) for the mixture of NiPAM (1 M) and BIS (0.01 M) containing I2965 (0.02 M).
η0 gradually increased with increasing UV irradiation time within 0 - 60 minutes. Subsequently, it exhibited a rapid increase after exposure to UV radiation for more than 60 minutes. The relationship between
and
t has intermittently changed within a UV irradiation time of 120 – 240 minutes, exhibiting linearity. The gel point (
tg) is tentatively identified as the intersection at 120 minutes. In another experiment, the gelation point has also been associated with the parallel lines of shear and loss modulus (
G′ and
G″), as shown in
Figure 1(b). As gelation proceeds beyond the sol-gel transition, log–log plots of
G′ and
G″ versus angular frequency produce a parallel line at
t = 120 minutes, with the corresponding relaxation exponent,
n, for
G′ ~
G″~
ωn [
36]. The slope,
n, was found to be 0.29, which corresponds to the chemical gel in previous work (n < 0.5) [
37]. Therefore, a UV irradiation time of 120 minutes is considered to be an appropriate condition for initiating the formation of the gel network of PNiPAM in our experimental condition. However, to achieve a fully developed three-dimensional network of PNiPAM, subsequent material construction was carried out using a UV irradiation time of 180 minutes.
2.2. Effect of crosslinker concentration on viscoelastic and mechanical behaviors of single and interpenetrating polymer networks
The composition of the starting materials and the abbreviations for the resulting hydrogels are listed in
Table 1. The viscoelastic properties of the hydrogels were analyzed by studying the frequency-dependent behavior of the shear modulus,
G′, loss modulus,
G″, and tan
δ at room temperature. As shown in
Figure 2, the
G′ was significantly higher than
G″, and both
G′ and
G″ exhibited slight variations with increasing frequency for all IPN hydrogels. This indicated that the hydrogels have a stable network with elastic characteristics owing to the high molecular intra- and/or inter-actions level. The results demonstrate that the viscoelastic properties of chitosan are significantly enhanced by incorporating the PNiPAM chain structure.
At a single frequency of 1 Hz (6.28 rad s
-1), the
G′ of all the hydrogels was compared as a function of the crosslinker content, as shown in
Figure 3(a). The
G′ of CS hydrogel increased significantly with an increase in the crosslinking agent. This can be attributed to the inherent structure of pure PNiPAM, as illustrated in the inset of
Figure 3(b), where the
G′ of pure PNiPAM prepared in an ethanol solution also showed a significant increase with an increase in the crosslinking agent. These hydrogels prepared in ethanol (IPN
E systems) provided a higher value of
G′ compared to those prepared in deionized water (IPN
w systems). Interestingly, the
G′ of the IPN
E system showed a substantial concentration dependency for the crosslinker content, whereas that of the IPN
w system remained almost constant value even when the crosslinker increased. This might be due to the excess crosslinker content above 0.01 M (1.45wt%) in the aqueous solution, where the conversion of NiPAM to PNiPAM remained almost constant regardless of the crosslinking agent used in this polymerization condition. The
G′ of the IPN hydrogels (0.6–3.37 MPa) was approximately 70 to 400 times higher than that of the CS hydrogel. Previous studies have shown that the
G′ value for the hydrogel containing polyvinyl pyrrolidone/carboxymethyl cellulose at a 1:4 ratio was approximately 0.43 MPa [
38]. On the other hand, the
G′ of Salecan and poly(2-acrylamido-2-methylpropanesulfonic acid-co-N-hydroxymethyl acrylamide) semi-IPN hydrogels ranged from 0.27 to1.14 kPa [
39]. This indicates a relatively high modulus for hydrogels prepared in this study.
The theory of rubber elasticity states that the elastic modulus of a soft polymer network is directly proportional to the density of the crosslink points. In the case of a phantom network, the
G′ of gels is determined by the crosslinking density (
ve) and the extent of solvent swelling, as depicted in Equation (1) and Equation (2) [
39]:
where
v2 is the volume fraction of cross-linked polymer in the hydrogel;
R and
T are the gas constant and absolute temperature, respectively;
ϕ is the functionality of the crosslinks. The term
qF represents the mass of swollen gel in equilibrium divided by the constant weight of the hydrogel after solvent evaporation. This value increases as the crosslinker content increases and reaches its maximum at 2.65 wt% crosslinker. The details of this phenomenon are discussed in
Section 2.5, specifically referred to as the equilibrium swelling ratio (
SRe). The term
ρ represents the polymer density and
d represents the density of the solvent [
39]. The crosslink densities
ve were calculated from Equation (1) and Equation (2) and they are depicted in
Figure 3(b) for the IPN
E system. At the concentration of 21.4wt% of BIS content (IPN
E4),
ve increased further, but the
G′ decreased, suggesting a high degree of brittleness in the IPN
E4 structures. It has been reported adding an excess amount of crosslinking agent enhances the density of the CS network and decreases the flexibility of the chains due to the reinforced chain entanglements [
40].
To gain insights into the influence of BIS concentration on the viscoelastic properties of liquid and solid gels, we have scrutinized important parameters such as the relaxation exponent and gel strength of the materials. Following the Winter-Chambon criterion [
41], the power law relation is also evident in dynamic mechanical experiments, as expressed below:
where
Γ(1 -
n) represents the gamma function,
n denotes the relaxation exponent,
δ is the phase angle that remains independent of frequency but is proportional to the relaxation exponent, given by
δ = nπ/2. Additionally,
S represents the gel strength parameter, which relies on the crosslinking density and molecular chain flexibility.
The power law of mechanical behavior represents the self-similar (fractal) structure of clusters at the gel point (
GP) [
36,
42]. The statistical self-similarity of a polymer is quantitatively represented by a fractal dimension,
df, which describes the relation between the mass of a molecular cluster (
M) in the network to its radius (
R) through the expression of
Rdf ~
M. Muthukumar established an expression for the relaxation exponent,
n in terms of the
df for polydisperse material, where the excluded volume effect of the polymer chain is completely screened. The equation is given as follows [
42]:
where
d (= 3) is the spatial dimension. As mentioned earlier, the values of relaxation exponent,
n are determined by analyzing a log-log plot and the scaling relation of
G′ and
G″. The
df values are obtained in the range of 2.22-2.44 for IPN
E, which is close to 2.50 of complete screening of excluded volume interactions. It indicates a tight and compact network structure develops [
43].
The effect of different amounts of BIS on gel strength
S in Equation (3) is shown in
Figure 4. When ethanol was used as a solvent, a clear dependence of the
S value on the crosslinker concentration was observed. The gel strength
S of CS hydrogel was outstandingly enhanced with the crosslinker content. Interestingly, the IPN
E3 showed a synergistic character resulting from the crosslinking density. At low concentrations of BIS, the gel network becomes fragile and susceptible to additional cross-links, resulting in an increase in the strength parameter
S. However, at the highest concentration of BIS, the strength of the network is primarily influenced by polymer entanglements, and the impact of additional cross-links on the gel strength parameter is not as significant as it is at lower levels of polymer cross-linking. These investigations have shown that the parameter
S is sensitive to changes in the strand length between crosslinks. When the strand length shrinks (e.g., due to increased BIS concentration), the cross-linking density increases, resulting in a “harder” gel with a higher value of
S.
2.3. Temperature dependence on mechanical stability of swollen hydrogel networks
Hydrogels typically contain a significant amount of moisture, which results in poor mechanical properties. This is due to the high degree of swelling and low density of the polymer chains. Consequently, this poses a significant challenge for their practical application. We manipulated the water content (%) within the hydrogels to assess their mechanical stability in their swollen state.
Figure 5 (and
Figure S7 in details) illustrates the temperature influence on
G′ and
G″ for the chitosan hydrogel, comparing it with IPN
E3 hydrogel in the 30 – 80
oC temperature range. In the case of the chitosan hydrogel, the slope of log
G′ vs. log
ω increased with the temperature rise. For example,
G′ decreases more rapidly as the frequency decreases at 80 °C, compared to the other temperatures. The difference between
G′ and
G″ at low frequency became smaller as temperature increased from 30 to 80
oC. At 80
oC, a crossover between
G′ and
G″ occurred at approximately 0.06 Hz. On the other hand, the crossover of the
G′ and
G″ was not observed for IPN
E3 hydrogel. The network structure of chitosan relies on several key interactions: hydrophobic forces, molecular entanglements, and secondary forces such as ionic and hydrogen bonding. These interactions become more time-sensitive under stress. When we decrease the testing frequency, which corresponds to longer experimental durations, the elastic property of the gel, represented as
G′, decreases, and its viscous nature,
G", increases. At high temperatures and low frequencies,
G′ and
G″ become more similar, eventually crossing over. At this point,
G" takes over, indicating a shift from a solid-like to a liquid-like behavior due to a thermal disruption in the physical network.
The IPNE3 exhibits a stable network when subjected to stress at varying temperatures. This characteristic of our system can be utilized in the development of shape memory hydrogels. Applying temperature or stress can cause the permanent shape of IPNE3 to disengage, transforming it into a temporary shape. This temporary shape can be restored by cooling the deformed sample while under load or reducing stress. Shape recovery occurs when the gel in its temporary shape is reheated and/or when a sufficiently large stress is applied to break the physical network. From a molecular perspective, the shape memory cycle regulates the relaxation times of the temporary shape through the amplitude of stress and/or temperature.
2.4. Chemical structure of single and IPN hydrogels
We assessed the chemical structures of the materials using Fourier-transform infrared spectroscopy (FT-IR).
Figure 6 shows the FT-IR spectra of the NiPAM monomer, PNiPAM
E hydrogel, CS hydrogels, and IPN
E3. Several changes became apparent in the FT-IR spectrum of the PNiPAM
E hydrogel compared to the NiPAM monomer, as illustrated in
Figure 6(a) and 6(b). The sharp peak at 3274 cm
-1 in the NiPAM monomer, attributed to the N-H stretching vibrations, became significantly broader in the PNiPAM
E hydrogel. This broadening was due to overlap with O-H vibrations from alcohol residues in the solvent. The distinctive spectral peaks of the NiPAM monomer at 1618 and 960 cm
-1, associated with C=C bending and the vinyl group, vanished in the PNiPAM
E hydrogel, indicating polymerization proceeds. Moreover, specific peaks related to the isopropyl group vibrations appeared at 1386 and 1367 cm
−1, shifting to a lower wavelength than the monomer (1400 cm
-1), suggesting that these isopropyl groups act as crosslinking points. This formation of crosslinking points was supported by interactions between the
tert-C atom of the side isopropyl group and the main chain isopropyl group, as explained in a previous study [
44].
For the CS hydrogel presented in
Figure 6(c), we found overlapping peaks at 3356 and 3288 cm
-1, indicating -OH stretching, symmetric N-H vibration, and intermolecular H-bonds between polysaccharide chains. The carbonyl stretching vibration (amide-I), N-H stretching vibration (amide-II), and C-N stretching vibration (amide-III) of CS were observed at 1643, 1556, and 1311 cm
-1, respectively. Additionally, the symmetrical deformation of the methyl (CH
3) groups of CS was identified at 1373 cm
-1, consistent with prior research [
45].
The characteristic peaks of the IPN hydrogel indicate the presence of functional groups from both neat hydrogels. Some shifts in peak positions were observed as shown in
Figure 6(d). These shifts suggest that the initial reactants could penetrate the CS hydrogel, and upon exposure to UV light, the PNiPAM structure was formed. Further shifts in wavenumber and band broadening indicate the formation of intermolecular associations between these pure hydrogels, which display compatible characteristics of both CS and PNiPAM hydrogels.
2.5. Equilibrium swelling ratio (SRe) of hydrogel networks
Figure 7 shows the swelling ratio
SRe of the CS and IPN hydrogels with varying crosslinker contents, calculated using Equation (5) [
46].
where
wd and
we represent the weight of dried hydrogels before being immersed in deionized water and the swollen sample reaching its equilibrium state at room temperature, respectively.
The interpenetration of the PNiPAM network resulted in an increased SRe for the CS hydrogel. The CS hydrogel is formed through physical crosslinking, primarily held together by hydrophobic interactions of the N-acetyl group and inter- and/or intra-molecular hydrogen bondings. This results in a polymer network that can easily deform, leading to disintegration and dissolution in water. Moreover, the CS network lacks stable crosslinking points, making it capable of absorbing less water than the IPNs, which have a stable network with chemically crosslinked structures.
The hydrophilicity of CS improved when the PNiPAM network was introduced, leading to an increased
SRe with a rise in crosslinker content. At lower crosslinker contents, a less stable network was formed, allowing polymer chains to partially dissolve in water and easily diffuse from the polymer structure. The network structure exhibited high hydrophilicity, interacting with water molecules most effectively at a crosslinker content of 2.65 wt%, resulting in the highest water uptake. However,
SRe decreased at crosslinking concentrations exceeding 2.65 wt% due to the increased density of crosslinks and the entanglement of polymer chains within the gel network. This led to a dense structure that hindered further swelling, consistent with our crosslink density findings. As a result, water molecules diffused more slowly into the network, limiting the relaxation of network chains in the hydrogels [
47].
The IPNE system exhibited a higher magnitude of SRe than the IPNw system. This is because less water could penetrate the dense IPNw structure, whereas the IPNE networks had a more porous structure, illustrating the morphology in FE-SEM results. The mechanism of water absorption will be thoroughly investigated in our future research.
2.6. Microstructure of the CS and IPN hydrogels
We used FE-SEM to examine the surface and cross-sectional structures of the fresh hydrogels.
Figure 8 illustrates the morphologies of CS and IPN hydrogels with different BIS concentrations. The CS hydrogel exhibited a network of interconnected fibers with both coarse and smooth surfaces, consistent with previous findings [
48]. Such surface structures are typical of polyelectrolyte components like CS and sodium alginate, as well as their mixtures [
49]. When the PNiPAM network was added, the surface of the CS hydrogel became smoother and more uniform as the BIS crosslinker content increased. The roughness of the hydrogel surface played a crucial role in altering its morphology. A smoother surface enhances the antifouling properties of hydrogel materials, making them more resistant to unwanted adhesion.
Cross-sectional images of the samples revealed different structures based on the BIS concentration. The CS hydrogel exhibited a well-defined lamellar structure due to the system’s homogeneity, allowing the NiPAM solution to penetrate easily. In the IPNs, a smooth and dense structure with irregular pores was observed, indicating excellent compatibility between the two polymers. The lamellar layer of the CS hydrogel was replaced by the PNiPAM network, creating a porous structure ideal for absorbing wound fluid and facilitating oxygen supply, which promotes faster wound healing [
50]. Furthermore, it is a useful material for loading and releasing preservative substances in food processing applications [
51]. The IPN structure becomes denser with a higher crosslinker content, resulting from the increased crosslink density of PNiPAM within the CS network. Moreover, this material finds application in food processing for loading and releasing preservatives [
51]. With a higher crosslinker content, the IPN structure became denser due to increased crosslink density of PNiPAM within the CS network. This homogeneity confirmed the compatibility between CS and PNiPAM, contributing to the high mechanical properties observed in the
G′ results.
2.8. HeLa cell adhesion and proliferation on the hydrogels’ surface
The biocompatibility of biomaterials can be assessed by examining how cells adhere to them. In this study, HeLa cells, a commonly used immortalized human cancer cell line in research labs worldwide, were used [
55]. The morphology of HeLa cells on both the CS and IPN
E hydrogels was observed and compared to those on the PS control surface after 1 hour of cell seeding, as shown in
Figure 9(a) to 9(c). No significant differences were observed in cell morphology between the control and hydrogel surfaces. However, it was evident that the hydrogel surfaces had more cell colonies than the control surface. The positively charged sites in the CS structure likely enhanced electrostatic interactions with the negatively charged cell membranes and proteins, facilitating cell adhesion to the CS surface [
56]. Additionally, the hydrophobic nature of PNiPAM-based IPN might contribute to its cell adhesion properties, as it can switch from a hydrophilic to a hydrophobic state above its VPTT (37°C), as shown in
Figure S9. The appearance of the swollen IPN
E3 in the PBS solution changed from transparent at room temperature to opaque at the incubation temperature (37°C).
After 24 hours of cell culture initiation, HeLa cell proliferation was observed. Cells on the polystyrene (PS) control surface displayed a flattened morphology, indicating strong cell adhesion to the surface, as shown in
Figure 9(a′). In contrast, cells on the hydrogel surfaces exhibited a rounder morphology, with lower adhesion and less proliferation, as shown in
Figure 9(b′) to 9(c′). Cell populations were counted after detaching from the material surface, and revealing that more cells had grown and spread on the PS control surface (3.6 ± 0.10 × 10
6 cells/cm
3) compared to the hydrogels (
Table 1). The hydrogels likely had a significant impact on the strength of interactions with HeLa cells, leading to improved resistance against fouling [
57]. The number of adhered cells on the IPN surfaces was slightly lower than on the CS surface, possibly due to the smoother surface, as indicated by the FE-SEM results. In general, cells adhere well to stiff surfaces [
58,
59], and materials with higher roughness are more prone to fouling as contaminants tend to accumulate in the "valleys" of rough surfaces [
60,
61]. Furthermore, the surface of IPN
E3 underwent a thermal reversal, transitioning from a hydrophobic state at the incubation temperature (37°C) to its original hydrophilic state at room temperature. This demonstrated the detachment or release of HeLa cells from the surface. In other words, the excellent thermo-reversible gelation of this material offers numerous benefits for various applications. The gel’s strength is significantly enhanced when the IPN
E3 is swollen and used above the VPTT, as described in
Section 2.3 and in the
Supplement Information. Conversely, it can be reversed to its original transparent state when cooled down below the VPTT.