Structural and Dynamic Properties of Chemically Crosslinked Mammalian and Fish Gelatin Hydrogels

Abstract

Gelatin is a collagen-derived biopolymer widely used in food, pharmaceutical and biomedical applications due to its biocompatibility and gelling ability. However, gelatin hydrogels suffer from unstable mechanical strength, limited thermal resistance and susceptibility to microbial contamination. The main aim of the present study is to investigate the influence of gelatin cryostructuring followed by photo-induced menadione sodium bisulfite (MSB) chemical crosslinking on the structural and functional characteristics of mammalian and fish gelatin hydrogels. The integration of scanning electron microscopy, dielectric spectroscopy and rheological experiments provides a comprehensive view of the of molecular, morphological and mechanical properties of gelatin hydrogels under photo-induced chemical crosslinking. The SEM results revealed that crosslinked hydrogels are characterized by enlarged pores compared to non-crosslinked systems. For mammalian gelatin, multiple pores with thin partitions are formed, giving a dense and stable polymer network. For fish gelatin, large oval pores with thickened partitions are formed, preserving a less stable ordered architecture. Rheological data show strong reinforcement of the elastic and thermal stability of mammalian gelatin. The crosslinked mammalian system maintains the gel state at higher temperatures. Fish gelatin exhibits reduced elasticity retention even after crosslinking because of a different amino acid composition. Dielectric results show that crosslinking increases the portion of bound water in hydrogels considerably, but for fish gelatin, bound water is more mobile, which may explain weaker mechanical properties.

1. Introduction

Gelatin is a biopolymer derived by partial hydrolysis of protein collagen, primarily found in animal connective tissues [1]. Due to its biocompatibility, biodegradability and ability to form thermo-reversible hydrogels, gelatin is widely used in food technology, pharmaceuticals, cosmetics and other biomedical applications [2]. Upon cooling, the gelatin polypeptide chains transfer from random coils to the triple-helix collagen-like structures, forming three-dimensional physically cross-linked hydrogels, stabilized by non-covalent forces such as electrostatic and Van der Waals interactions, hydrogen bonding and hydrophobic forces. Gelatin hydrogels have diverse applications as jelly-like food components, edible coatings, functional packaging, and encapsulation systems for nutritional products [3,4,5,6], biocompatible scaffolds [7,8,9,10,11] and drug delivery carriers in biomedicine [12,13,14,15].

The versatile properties of gelatin and the growing market have led to increased production [16]. However, the functional characteristics of gelatin are highly dependent on its natural source. The main sources of mammalian gelatin are the bones and skin of cattle and pigs. Porcine gelatin contains a high content of proline and hydroxyproline, conferring excellent thermal stability and strong rheological properties to its hydrogels [17]. Recently, fish-derived gelatin has gained serious attention [18,19]. This interest stems from waste utilization goals and religious dietary requirements. Fish gelatin, composed of smaller peptide fragments, exhibits increased bioavailability and reduced immunogenicity [20]. Additionally, as a by-product of fish processing, it contributes to sustainable production by minimizing waste and resource use [21,22]. However, due to differences in amino acid profile, fish gelatin is characterized by a lower melting point and weakened viscoelastic parameters [23,24], which reduces its direct use as an alternative to mammalian gelatin.

Both gelatin types have several disadvantages, including unstable mechanical strength of gel structures, limited thermal resistance under certain conditions and susceptibility to microbial contamination, which can reduce their effectiveness in modern technological applications [25]. The rheological properties of gelatin hydrogels, such as the storage modulus, loss modulus and gelation dynamics, play a key role in their functionality, determining texture characteristics, stability and processing capabilities in food and biomedical systems [26,27,28,29]. These properties are determined by the molecular organization of gelatin and its network. However, thermal or mechanical stress can disrupt its hydrogel supramolecular architecture, leading to gel melting or structural breakdown. This instability limits gelatin’s applicability as a reusable cooling system [30,31] and scaffold materials for tissue engineering [32]. Microbial contamination is another critical issue, particularly in high-moisture hydrogels for food packaging and biomedical uses like wound dressings.

To address these limitations, particularly regarding the rheological and structural properties of gelatin-based systems, various modification approaches are employed, including physical, chemical and enzymatic treatments. Physical combination of gelatin with polysaccharides, including chitosan, sodium alginate, and κ-carrageenan, helps to improve the physicochemical parameters of gelatin hydrogels due to associative and segregative interactions [33,34,35,36,37]. An alternative approach involves blending gelatins from different sources (fish and mammalian origin), which promotes collagen-like structure formation and enhances triple helix ordering, creating new opportunities for gelling improvement [38]. Enzymatic crosslinking with transglutaminase forms iso-peptide bonds, enhancing network stability, but is limited by high cost and sensitivity to processing conditions [39]. Commonly used chemical crosslinking agents like glutaraldehyde, genipin and methacrylic anhydride strengthen gelatin through covalent bonds [40,41,42,43,44]. However, these agents face the following limitations. Glutaraldehyde is toxic [45], genipin is expensive [46], and methacrylic anhydride produces unwanted by-products [47]. Glyceraldehyde offers a biocompatible alternative but requires high acetone concentrations for effective crosslinking, which may complicate gelatin processing [45]. These disadvantages limit the use of the above-listed modification methods in food applications and drive the search for safer alternatives.

The photo-crosslinking of gelatin chains offers precision and biocompatibility, particularly with visible light [48]. This approach enables the rapid formation of hydrogels directly at wound sites under gentle polymerization conditions, offering notable benefits for biomedical applications [48,49,50]. Visible light provides gentle conditions with minimal toxicity, whereas UV radiation, though efficient, risks causing oxidative damage to tissues upon direct exposure. However, when direct contact with living tissue is not required, UV crosslinking offers high efficiency, particularly using a recently discovered photosensitizer—menadione sodium bisulfite (MSB), a water-soluble derivative of vitamin K [51]. A critical step to obtain the required product involves controlled freezing and thawing cycles, which create ice crystals that template a macroporous architecture, promoting the formation of additional intra- and intermolecular junctions and the gelatin association into a dense chain-ordered structure. These conditions promote a structural framework for gelatin photo-crosslinking when exposed to UV light, creating strong covalent bonds (Figure 1) that considerably improve the hydrogel’s thermal stability and mechanical properties [30,31,51,52]. Thus, for gelatins of different origins, the resulting structural difference is determined by their amino acid content and disposition to form extra hydrogen bonding and MSB-induced covalent bonds. A key advantage of MSB is its ability to impart antibacterial properties to gel structures via the generation of reactive oxygen species, providing protection against microbial contamination. This makes MSB especially important for the creation of gelatin-based ice substitutes that can be used in the food industry as cooling agents with minimal risks of cross-contamination [50]. The combined application of physical linkage and chemical photo-crosslinking can enhance gel properties noticeably, increasing mechanical properties several-fold [9].

Nevertheless, the impact of MSB-mediated crosslinking on the rheological properties of gelatin remains poorly understood, limiting the rational design of these materials for specific applications. Moreover, there are no the published results of MSB-based cross-linking of fish gelatins. Thus, the aim of this study is to study the structure-property relationships in MSB-crosslinked hydrogels based on the mammalian and fish gelatins. In the present study, the rheological characterization was complemented by scanning electron microscopy and dielectric spectroscopy, providing a comprehensive view of the crosslinking influence on viscoelastic behavior, microstructural organization and water dynamics in gelatin hydrogels.

The basis for this study stems from the limitations of pure gelatin hydrogels, such as unstable mechanical strength, low thermal resistance, and microbial susceptibility, which restrict their use in food (e.g., reusable cooling agents) and biomedical applications. Fish gelatin, derived from processing waste, offers sustainability and reduced immunogenicity but suffers from weaker gelling and mechanical properties compared to mammalian (porcine) gelatin. We addressed this problem by combining cryostructuring (freeze-thaw cycles) with photo-induced MSB crosslinking, a novel biocompatible method that enhances the number of covalent bonds and imparts potential antimicrobial effects without toxic by-products.

2. Materials and Methods

2.1. Materials

In this study, we used porcine gelatin (Type A, 300 bloom food grade, Sigma, G2500, Sigma-Aldrich, St. Louis, MO, USA). The laboratory-extracted fish gelatin was obtained from Atlantic cod skin by acid treatment, as defined in detail previously [53]. Previous studies showed that this fish gelatin exhibited thermal stability and rheological properties more similar to mammalian gelatin than its commercial alternatives [53]. Menadione sodium bisulfite (MSB) (Sigma, St. Louis and Burlington, MA, USA, M5750) was used for chemical crosslinking.

2.2. Preparation of Gelatin Hydrogels

Porcine gelatin (PG) or fish gelatin (FG) was dissolved in MilliQ water prepared with the Arium mini ultrapure water system (Sartorius, Gottingen, Germany). The 12.5% w/v gelatin was kept at room temperature overnight. The gelatin solution was heated at 50 °C for one hour with stirring at 1000 rpm on a JOAN Lab thermal shaker (JOANLAB, Huzhou City, China). The 5% MSB solution was added to the heated protein solution to obtain a mixture of 10% gelatin and 1% MSB. The resulting solution was stirred at 40 °C for 1 h. The sample was then refrigerated to 4 °C overnight to complete the sol-gel transition, followed by freezing at −40 °C for 18 h. After freezing, samples were thawed at 4 °C for 6 h. After the freeze-thaw cycle, the samples were heated to 35–40 °C before the rheological experiment to melt the gel and allow the sample to be placed into the rheometer cell. The sample in the sol state was placed in the rheometer ring cell, which was then cooled to 1 °C with maximum possible speed. After fast cooling, the sample was irradiated for 20 min with a UV lamp (6 UV lamps, 9 W each). During irradiation, the rheometer cell was maintained at 1 °C. After irradiation, the lamp was removed and the sample was equilibrated to the temperature of measurement. The 10% porcine gelatin (PG) and fish gelatin (FG) gels were used as control samples. They were prepared using a similar procedure to that for the crosslinked samples, the difference being the absence of menadione and UV irradiation (samples were kept in the rheometer cell for 20 min at 1 °C before measurements).

For scanning electron microscopy (SEM) and dielectric measurements, the prepared mixture of 10% gelatin and 1% menadione was poured into a prepared platform. The sample was then refrigerated at 4 °C overnight to complete the sol-gel transition, followed by freezing at −40 °C for 18 h. After freezing, samples were thawed at 4 °C for 6 h. Then, the samples were irradiated for 20 min with a UV lamp. During irradiation, ice cubes were placed under the platform to prevent the sample from melting.

For all experimental methods, the thickness of the sample during irradiation was approximately 1 mm for better passage of UV irradiation.

2.3. Scanning Electron Microscopy

SEM was used to study the morphology of lyophilized gelatin hydrogel samples using the Merlin field emission scanning electron microscope (Carl Zeiss, Oberkochen, Germany). The surface morphology was examined at an accelerating voltage of 15 kV. For this purpose, hydrogel samples were frozen in liquid nitrogen, fractured with a scalpel, and lyophilized for 48 h using a Labfreez FD-10-MR^® freeze dryer (Labfreez Co., Ltd., Beijing, China) at a condenser temperature of <−55 °C, vacuum pressure of 5 Pa, and pump capacity of 2 L/s. The fractured surfaces were sputter-coated with gold for SEM observations. The experiments were performed at the Interdisciplinary Center for Analytical Microscopy (Kazan Federal University, Kazan, Russia). For quantitative analysis of the hydrogel microstructure, the images were processed using MountainsLab software (Version 2) (Digital Surf, Besançon, France).

2.4. Dielectric Spectroscopy

Dielectric experiments across 10 decades of frequencies from 1 Hz to 65 GHz were carried out in three stages.

At the first stage, the dielectric spectra were recorded in the frequency range 1 Hz–10 MHz using the Alpha Frequency Response Analyzer as part of the Novocontrol BDS-80 measuring complex. The results were recorded using the built-in licensed WinDeta software (version 2.0). A plane-parallel capacitor with electrodes 12 mm in diameter was used as a measuring cell; a 0.5 mm thick Teflon ring sets the distance between electrodes. The cell was calibrated with air and benzene at 20 °C.

At the second stage, the measurements were carried out in the frequency range 1 MHz–1 GHz using the E4991A radio frequency analyzer (Keysight, Santa Rosa, CA, USA), also being a block of the Novocontrol BDS-80 measuring complex. The samples were placed in a similar measuring cell as described previously.

At the third stage, the dielectric spectra were measured in the frequency range 100 MHz–65 GHz with the help of a PNA-X Agilent N5247A network analyzer (Agilent, Santa Clara, CA, USA). The results were recorded using the built-in licensed Agilent 85070 software (Version 1). A coaxial probe with a diameter of 10 mm, calibrated using deionized water Milli-Q at 20 °C, was used as the measuring cell.

The temperature range of dielectric experiments of 5–50 °C was overlapped with a step of 5 °C. Temperature control was carried out using the Quatro system at the first two measurement stages and the LOIP LT 900 thermal stabilizer (LOIP Ltd., Saint-Petersburg, Russia) at the third stage. The spectra obtained during the three measurement stages were combined to obtain broadband dielectric spectra in the frequency range from 1 Hz to 65 GHz. The dielectric relaxation parameters were calculated using the Datama software package (version 2.0) [54].

Unfortunately, the high level of ions in the studied hydrogels did not permit obtaining real relaxation data across the whole frequency range due to high values of near-electrode polarization, critical for the real part of complex dielectric permittivity, and high values of through conductivity, interfering with obtaining correct values of its imaginary part. That is why we employed for analysis the dielectric characteristics of initial non-linked hydrogels using the results in the frequency range 10⁵–10¹¹ Hz, and only 10⁷–10¹¹ Hz for chemically cross-linked systems.

2.5. Rheology

The experiments were performed on an MCR-302 automatic rheometer (Anton Paar GmbH, Graz, Austria) with a PP50-SN37449 cone-and-plate measuring system (diameter 50 mm, angle 1°, gap between cone apex and plate 0.1 mm) in two repeats and their mean values were used for analysis. The following experimental protocols were used:

• Frequency sweep: 0.1–300 rad/s at a constant strain amplitude of 1%;
• Strain amplitude sweep: 0.05–300% at a constant frequency of 6.28 rad/s;
• Temperature ramp: 0–50 °C at 1 °C/min with a constant amplitude of 1% and frequency of 1 Hz [23].

3. Results

3.1. Effect of MSB Crosslinking on the Morphology of Mammalian and Fish Gelatin

The sample preparation for SEM experiments involves dehydration, which may distort the native hydrogel structure, which is transferred to xerogel form [55,56,57]. Nevertheless, SEM provides valuable information on the morphological features of material and trends of changes during crosslinking. Figure 2 shows SEM images of fish (FG) and porcine (PG) gelatin hydrogels.

The 2% FG hydrogel shows an ordered structure with predominantly elongated, rectangular pores (Figure 2a). Pore walls are thin with transverse filaments observed in some regions. The average pore size is 2.1 ± 1.7 μm. In contrast, the PG hydrogel demonstrates a less ordered structure with elongated, irregular pores averaging 1.6 ± 0.8 μm (Figure 2d). The pore walls form continuous sinuous structures, creating a dense network with lower porosity compared to fish gelatin. Such differences in the xerogel microstructure are due to the physicochemical properties of gelatins of different origins; higher porosity is characteristic of fish gelatin obtained as a result of using alkaline or acidic treatment [24].

The increase of FG content to 10% in combination with freeze-thaw cycles leads to morphological transformations of the structure: irregularly shaped cavities with an average diameter of 3.1 ± 1.1 μm are formed, the walls between them show moderate thickening, and transverse fibers are absent (Figure 2b). In the case of PG hydrogels with a concentration of 10%, subjected to freeze-thaw treatment (Figure 2e), a heterogeneous architecture is formed. The structure contains zones with large cavities exceeding 20 μm in diameter and with distorted geometry, along with areas of fine-porous structure (about 1.8 μm), morphologically similar to 2% PG the freeze-thaw treatment. The average size of cavities increases to 4.5 ± 4.3 μm.

MSB crosslinking reorganizes the FG architecture into ordered oval pores; the interpore walls become thicker and the surface topography becomes smoother. The average cavity diameter increases to 6.6 ± 3.1 μm, representing a 3-fold increase compared to initial values (Figure 2c).

Similar structural changes also characterize porcine gelatin: crosslinking reduces structural tortuosity. The cavities lose their elongated shape but do not develop a clearly defined oval morphology. The average cavity size increases to 5.6 ± 2.4 μm, representing a 3.5-fold increase compared to non-crosslinked material (Figure 2f).

Figure 3 shows the pore size distribution histogram for primary and modified hydrogel systems of fish and mammalian gelatin, demonstrating a shift in distribution toward an increase in pore dimensions after modification. Upon rapid cooling, the aqueous phase crystallizes to form finely dispersed ice inclusions, promoting local gelatin concentration and formation of organized, compact protein structures [31].

The subsequent photoactivated modifications with MSB stabilize the locally formed compact three-dimensional architecture. Under UV radiation, MSB becomes photoactive and interacts with gelatin macromolecules. Free radicals form protein chain segments. These radicals then react with each other, creating covalent crosslinks between gelatin molecules [30,31]. A reinforced gelatin hydrogel matrix forms local dense pseudocrystalline regions homogeneously distributed throughout the hydrogel volume [51]. The local increases in gelatin network density, coupled with increased spacing between dense regions, explain the observed increase in average pore size after modification [57].

Thus, the results of SEM experiments show that despite the general trend toward pore enlargement and structural ordering after photo-crosslinking, the morphology of fish and mammalian gelatin hydrogels remains distinct. Crosslinked FG forms more regular, oval-shaped pores with thickened walls, whereas PG is characterized by a less ordered porous architecture with thinner partitions. These differences reflect variations in the organization of the polymer network. Mammalian gelatin forms a denser and more uniform matrix due to the high content of imino acids, which promote stable intermolecular crosslinks, whereas fish gelatin exhibits a less compact but more morphologically ordered structure.

3.2. Kinetics of Water in Gelatin Hydrogels with Cross-Linked Modification

In dielectric experiments, we used the set of our broad-band equipment overlapping about 10 frequency decades from 1 Hz to 65 GHz [58]. However, the high conductivity of hydrogels did not permit us to obtain dielectric data in the entire possible frequency range, and we were limited. Figure 4 shows, as an example, the obtained dielectric relaxation data for PG hydrogels.

One can see that the initial physical hydrogel (Figure 4a,b) and chemically cross-linked hydrogel (Figure 4c,d) show some different relaxation pictures. The first system, which has some lower conductivity, shows three relaxation processes, but for the crosslinked system, we observe only two high-frequency processes.

The high-frequency process (red lines in Figure 4), centered about dozens of GHz, is typical for relaxation of bulk water [59] with activation energy obtained from the corresponding Arrhenius plot equal to 18 kJ·mol⁻¹ (Figure 4b), which is very close to the known value of activation energy for dielectric relaxation of pure water 15.9 kJ·mol⁻¹ [60,61]. The next process, centered around hundreds of MHz, is determined by relaxation of bound water, as we’ll ground below. For the non-crosslinked gelatin hydrogels, one can see the dielectric relaxation of gelatin molecules an order of magnitude lower in frequency. This process is out of our interest in the present study, since it is seen only in non-crosslinked systems being screened in chemically crosslinked ones.

Figure 4. Real (a,c) and imaginary (b,d) parts of dielectric spectrum of porcine gelatin (non-crosslinked and cross-linked) at 20 °C. Open squares represent experimental data; the dark blue curve shows the approximating function; the relaxation process of free water is shown in red; bound water in green; gelatin in blue; the magenta curve shows the Jonscher term [62,63,64,65]; and the orange curve represents the conductivity contribution.

Figure 4. Real (a,c) and imaginary (b,d) parts of dielectric spectrum of porcine gelatin (non-crosslinked and cross-linked) at 20 °C. Open squares represent experimental data; the dark blue curve shows the approximating function; the relaxation process of free water is shown in red; bound water in green; gelatin in blue; the magenta curve shows the Jonscher term [62,63,64,65]; and the orange curve represents the conductivity contribution.

Thus, to extract relaxation parameters, for non-crosslinked systems (in the frequency range 10⁵–10¹¹ Hz), we have fitted the obtained spectral data by superposition of all contributions, including three Cole-Cole relaxation functions (Equation (1)), and for crosslinked hydrogels (in the frequency range 10⁷–10¹¹ Hz), by two Cole-Cole relaxation functions (Equation (2)) with additional contributions from electrical conductivity:

$${\mathsf{\epsilon }}^{*}\left(\mathsf{\omega }\right)={\mathsf{\epsilon }}_{\infty }+{\displaystyle \frac{\Delta {\mathsf{\epsilon }}_{\mathrm{FW}}}{1+{\left(i{\mathsf{\omega }\mathsf{\tau }}_{\mathrm{FW}}\right)}^{{\mathsf{\alpha }}_{\mathrm{FW}}}}}+{\displaystyle \frac{\Delta {\mathsf{\epsilon }}_{\mathrm{BW}}}{1+{\left(i{\mathsf{\omega }\mathsf{\tau }}_{\mathrm{BW}}\right)}^{{\mathsf{\alpha }}_{\mathrm{BW}}}}}+{\displaystyle \frac{\Delta {\mathsf{\epsilon }}_{\mathrm{gelatin}}}{1+{\left(i{\mathsf{\omega }\mathsf{\tau }}_{\mathrm{gelatin}}\right)}^{{\mathsf{\alpha }}_{\mathrm{gelatin}}}}}+A{\left(i\mathsf{\omega }\right)}^{n-1}+{\displaystyle \frac{\mathsf{\sigma }}{i{\mathsf{\omega }\mathsf{\epsilon }}_0}},$$

(1)

$${\mathsf{\epsilon }}^{*}\left(\mathsf{\omega }\right)={\mathsf{\epsilon }}_{\infty }+{\displaystyle \frac{\Delta {\mathsf{\epsilon }}_{\mathrm{FW}}}{1+{\left(i{\mathsf{\omega }\mathsf{\tau }}_{\mathrm{FW}}\right)}^{{\mathsf{\alpha }}_{\mathrm{FW}}}}}+{\displaystyle \frac{\Delta {\mathsf{\epsilon }}_{\mathrm{BW}}}{1+{\left(i{\mathsf{\omega }\mathsf{\tau }}_{\mathrm{BW}}\right)}^{{\mathsf{\alpha }}_{\mathrm{BW}}}}}+{\displaystyle \frac{\mathsf{\sigma }}{i{\mathsf{\omega }\mathsf{\epsilon }}_0}},$$

(2)

where ${\mathsf{\epsilon }}^{*}$ is the complex permittivity, $\mathsf{\omega }$ is the angular frequency, ${\mathsf{\epsilon }}_0$ is the permittivity of free space, ${\mathsf{\epsilon }}_{\infty }$ is the high-frequency limit of dielectric permittivity, $\Delta \mathsf{\epsilon }$, τ and α are the dielectric strength (amplitude of dispersion), relaxation time and relaxation time distribution (0 < α < 1) for corresponding relaxation processes, A is the exponent of the Jonscher function [62,63,64,65], and σ is the through conductivity. The subscripts FW and BW are related to relaxation parameters of free and bound water. The temperature dependence of the dielectric strength $\Delta \mathsf{\epsilon }$ and characteristic relaxation times τ are shown in Figure 5.

The temperature dependence of relaxation strength $\Delta \mathsf{\epsilon }$ for free water in all systems studied (Figure 5a) replicates that for pure water, shown in blue [66,67]. The dielectric strength of any dielectric process is proportional to the polarization of dipoles and reflects the concentration of definite dipoles in the system [68]. Thus, one can see the decrease in the $\Delta \mathsf{\epsilon }$ value for pure water in all hydrogels in comparison with pure water because of hydration of gelatin and transformation of part of the free water into the bound state. The main result following from the information depicted in Figure 5a is that there is a great increase in the quantity of bound water in hydrogels as a result of gelatin crosslinking. At the same time, there is no principal difference in the volume of bound water in the PG and FG systems. Also, it may be interesting for further studies that the dielectric strength of the gelatin relaxation process shows some turning point nearby 25 °C, possibly as a result of some structural transition in the hydrogel, which is more remarkable for the PG-based system.

The analysis of the temperature dependence of relaxation times in the Arrhenius representation (Figure 5b) gives a view on the dynamics of water motion in hydrogels. Free water in all four samples—PG and FG hydrogels, both crosslinked and non-crosslinked—shows so identical relaxation times that it is rather difficult to distinguish them. All systems are characterized by activation energy of free water relaxation of 18 kJ·mol⁻¹, which is rather close to the value for pure bulk water [60,61].

The mobility of bound water is an order of magnitude less. This is rather clear because of the strong interaction of water with the gelatin matrix. Furthermore, in the crosslinked systems, a sufficient amount of water can be bound by sodium ions, also decreasing its average mobility and the total relaxation rate. The coupling agent MSB is used in the form of sodium salt. Free Na⁺, which is known as an active kosmotrope towards water [69], forms ions in the hydrogel bulk with a tight hydration shell from three to five water molecules, held electrostatically around the alkali metal ion [70,71]. Different binding sites do not necessarily result in diverse relaxation rates. Usually, all bound water gives a common relaxation process due to the fast exchange of water molecules between binding centers. However, there is a strong difference in mobility of bound water in the FG-based crosslinked hydrogel comparing both with non-crosslinked systems and even with the PG-based crosslinked hydrogel. The PG and FG crosslinked hydrogels differ from one another not only in relaxation rate of bound water but also in activation energy of its relaxation process. Bound water in the crosslinked FG system is not only more mobile but also has twice as less activation energy of dipole reorientation. It seems that it has less strong binding in the hydrogel with crosslinked fish gelatin. Obviously, it is a result of less strong supramolecular structure of fish gelatin hydrogel in comparison with the mammalian one.

The parameter α reflects the obviousness of relaxation time distribution or, obviously, the divergence of the relaxation process from the standard statistical process or the nonequivalence of the microenvironment of relaxing dipoles. The non-crosslinked hydrogels demonstrate α values close to 1 for both free and bound water (Figure 5c). Similar behavior is determined for free water in the crosslinked systems, which show high homogeneity of relaxation conditions. A different picture is determined for relaxation of bound water in the hydrogels based on crosslinked PG and FG hydrogels. First of all, one can see a strong decrease in the α value, which shows a broad distribution of relaxation rates of bound water in these systems, being, apparently, the result of the more complicated structure of the gelatin biopolymer network in the crosslinked hydrogels with a variety of water binding sites. Another observation is the presence of some crossover in the obtained temperature dependences. It is interesting, that this transition coincides rather well with the turning point in gelatin relaxation strength. Probably, our dielectric results fix some temperature transition of the gelatin network in the crosslinked hydrogels, which, in turn, reflects the changes in water relaxation characteristics.

Thus, dielectric spectroscopy reveals some consistent differences between fish and mammalian crosslinked gelatin hydrogels. Mammalian gelatin exhibits higher relaxation times and activation energies for bound water in both crosslinked and non-crosslinked systems. After crosslinking hydrogels show considerable increase of the volume of bound water in comparison with non-crosslinked analogs. At the same time the crosslinked fish gelatin shows faster water relaxation which aligns with larger pore size in this system. The behavior of parameter α suggests a more complicated hydration environment of water and tighter molecular confinement, reflecting a denser and more interconnected polymer network.

3.3. Mechanical Characteristics of Gelatin Hydrogels

The viscoelastic behavior of gelatin hydrogels was studied using dynamic mechanical analysis. Oscillatory measurements are widely used to characterize the rheological properties of hydrogel materials [26].

Figure 6 shows the elastic-viscous properties of all studied hydrogels in the gelatinous state at 4 °C. One can see that the hydrogel based on the crosslinked mammalian gelatin at 4 °C exhibits a storage modulus G′ of 10,700 Pa, which is 2.5 times higher compared to its non-crosslinked analog (4210 Pa). The crosslinking effect for fish gelatin is less pronounced: the G′ value increases from 1290 Pa for the initial hydrogel to 1670 Pa as a result of gelatin crosslinking. Notably, even the non-crosslinked mammalian gelatin demonstrates elastic properties twice as high as those of the crosslinked fish gelatin. The storage modulus G′ determines the elastic component of the material and its resistance to deformation, while the loss modulus G″ reflects the viscous component and flow tendency of the hydrogel. The dominance of G′ over G″ indicates a gel state with predominant elastic behavior [72].

The strain amplitude sweeps (Figure 6b) show that photo-crosslinked mammalian gelatin at 4 °C exhibits reduced deformability. The sample fails at approximately 11% strain with a maximum stress of 1250 Pa. In contrast, the non-crosslinked mammalian gelatin withstands strain up to 51% at 3750 Pa, indicating greater elasticity. Fish gelatin hydrogels demonstrate higher stress tolerance, with the photo-crosslinked analog sustaining 6400 Pa and the non-crosslinked sample sustaining 5400 Pa without structural failure within the tested strain range.

The thermal stability of gels exhibits significant differences at elevated temperatures (Figure 7). At 25 °C, the storage modulus G′ of crosslinked PG decreases to 1020 Pa compared to the value of 10,700 Pa at 4 °C, while for its non-crosslinked analog, it reaches 835 Pa. It is necessary to focus one’s attention on the fact that that we could not obtain any results for non-crosslinked FG because it had lost the gel-like state at 25 °C. The crosslinked fish gelatin demonstrates a storage modulus of 4.87 Pa at this temperature. The rheological behavior shows that both mammalian gelatin hydrogels and the crosslinked fish gelatin maintain their gel-like state at 25 °C, with storage modulus G′ exceeding loss modulus G″ across the frequency range. However, the mechanical properties of all three samples are significantly reduced compared to 4 °C due to the thermal effects on gel network structure. Strain amplitude sweeps demonstrate that at 25 °C, the critical stress is 36.9 Pa for crosslinked mammalian gelatin, while for the non-crosslinked gelatin, it reaches 426.7 Pa. The crosslinked fish gelatin maintains elastic behavior up to 25 °C without critical stress failure despite its lower storage modulus values.

At 35 °C, the crosslinked PG shows a storage modulus G′ of 197 Pa, while the crosslinked FG exhibits extremely low values of 0.23 Pa (Figure 8). When heated to 35 °C, the non-crosslinked PG undergoes the gel-sol transition, with the loss modulus G″ beginning to dominate over the storage modulus G′ at frequencies above 40 rad/s, indicating the loss of elastic behavior. In contrast, the crosslinked PG retains elastic properties across the entire frequency range even at 35 °C, as confirmed by the storage modulus exceeding the loss modulus. For crosslinked FG at 35 °C, the characteristic crossover points between storage modulus G′ and loss modulus G″ appear in the frequency spectra at 2.1 rad/s with a value of 0.42 Pa, marking the onset of gel-sol transformation when viscous behavior begins to dominate over the elastic one. At 35 °C, the crosslinked porcine sample sustains 29.9 Pa in strain amplitude sweeps, while the non-crosslinked sample loses structural integrity.

Temperature scanning allows the estimation of the effect of chemical crosslinking on the thermal stability of hydrogels. For mammalian gelatin, a significant difference is observed between crosslinked and non-crosslinked samples (Figure 9). In the non-crosslinked gel, temperature exposure results in a stepwise decrease in the storage modulus G′ and loss modulus G″. At approximately 35.5 °C, a crossover of moduli occurs, which corresponds to gel-sol transition typical for gelatin systems. The crossover temperature is often interpreted as the melting temperature of the hydrogel [53,73]. However, this temperature usually exceeds the melting temperature determined by DSC, since gel-like consistency can be maintained until the modulus crossover point [74]. After this transition, the storage modulus decreases, indicating loss of elasticity and destruction of gel architecture. For the crosslinked mammalian gelatin hydrogel, the transition shows a different character. At approximately 30 °C, a sharp drop in both storage and loss moduli is observed, which is associated with the destruction of gelatin triple helices and physical interactions that stabilize the hydrogel structure [75]. Despite these changes, the storage modulus G′ exceeds the loss modulus G″ throughout the entire studied temperature range, confirming preservation of the gel state. Chemical crosslinks formed during hydrogel synthesis provide structural stability and elastic properties even at increased temperatures [43]. Thus, the observed transition at 30.1 °C does not represent the classical gel-sol transition but rather a conformational change in gelatin polypeptide chains. This change affects the mechanical characteristics without leading to complete loss of the gel state due to stable chemical crosslinks.

Fish gelatin exhibits reduced thermal stability. In the non-crosslinked hydrogel, the modulus crossover occurs at 13.3 °C, after which the values decrease to nearly zero, indicating gel destruction and transition to a viscous-flowing state. Exceptionally low G′ values at temperatures above 15 °C indicate the absence of a distinct structure. Similar behavior is observed in the crosslinked hydrogel. Despite the presence of chemical crosslinks, the moduli G′ and G″ crossover occurs at 16.1 °C, after which G′ becomes smaller than G″. This indicates that chemical crosslinking slightly increases the thermal stability of the fish hydrogel but does not guarantee preservation of the gel state under thermal exposure, unlike the case of crosslinked mammalian gelatin. This phenomenon may be attributed to a lower density of chemical bonds, differences in molecular composition, or weaker initial architecture of fish gelatin.

4. Discussion

The main goal of the present study was to investigate the effects of gelatin cryostructuring and subsequent chemical crosslinking of its molecules with menadione sodium bisulfite (MSB) on the structural and functional characteristics of mammalian and fish gelatin hydrogels. The integration of scanning electron microscopy (SEM), dielectric spectroscopy, and rheological data provides a comprehensive view on the modulation of molecular, morphological and rheological properties of gelatin hydrogels under the action of photo-induced chemical crosslinking.

The crosslinking of gelatin chains with MSB in combination with freeze-thaw cycles modifies the macroporous microstructure of hydrogels. SEM analysis revealed that crosslinked hydrogels are characterized by enlarged pores compared to the non-crosslinked systems. For mammalian gelatin, multiple pores with thin partitions are formed, showing a dense and stable polymer network [76]. In fish gelatin, larger oval pores with thickened partitions are formed, preserving an ordered but less stable morphological architecture [77]. This phenomenon is associated with the displacement and reorientation of gelatin molecules by ice crystals during freezing with subsequent reinforcement of hydrogel structure through the gelatin chains crosslinking with covalent bonding and extra hydrogen bonds [51,57]. Despite the formation of enlarged pores, the crosslinked hydrogels exhibit increased mechanical strength; this becomes particularly apparent for mammalian gelatin, where the storage modulus increases by 2.5 times compared to the non-crosslinked analog. This enhancement is due to the formation of dense microcrystalline domains in the pore walls, reinforced by covalent bonds and extra hydrogen binding between initially oriented gelatin triple helices. This phenomenon is associated with the high content of proline and hydroxyproline in mammalian gelatin, whereas fish gelatin forms a less durable network due to the low imino acid content in its primary structure [23].

Rheological measurements confirm that crosslinking with MSB greatly improves greatly the elasticity and thermal stability of mammalian gelatin. The crosslinked system maintains a gel state at higher temperatures, while the non-crosslinked mammalian gelatin undergoes gel-sol transformation. Fish gelatin exhibits reduced elasticity retention even after crosslinking due to weaker intra- and intermolecular interactions caused by its amino acid composition. The comparison with other crosslinking agents demonstrates the effectiveness of MSB application. In studies of hydrogels crosslinked with methacrylic anhydride, the mammalian gelatin showed lower elasticity, and fish gelatin was markedly less durable compared to our results under similar conditions [43]. This indicates more efficient covalent bonding of gelatin chains with the help of MSB. Moreover, MSB provides antimicrobial properties, expanding the potential applications of hydrogels in biomedical and food systems [31].

Dielectric data have deepened our understanding of the obtained SEM and rheology results to the molecular level: it was shown that crosslinking violently increased the portion of bound water in the systems. Apparently, the dense biopolymer matrix with cryostructural fragments creates favorable conditions for water binding, changing hydrogel elasticity. The more organized water environment in crosslinked gels, resulting from gelatin cryostructuring, new covalent bonding, and additional hydrogen bonds, promotes modified bulk properties of hydrogels, which is consistent with rheological strength. At the same time, we have determined that mammalian and fish gelatin hydrogels differ strongly in the mobility of bound water, which is more weakly coupled and more mobile in comparison to the crosslinked mammalian gelatin. Bound water determines the inner and overall mobility of gelatin chains, and possibly the difference in the mobility of bound water is the key parameter in the explanation of different mechanical characteristics of crosslinked gelatin hydrogels.

Therefore, the gelatin crosslinking with MSB combined with cryostructuring effectively strengthens the polymeric network of mammalian gelatin, providing a balance between structural stability and mechanical properties. Fish gelatin remains less stable due to weak molecular interactions, emphasizing the importance of selecting appropriate starting materials for target applications. Further studies may focus on optimizing crosslinking parameters for fish gelatin and exploring the antimicrobial properties of MSB to develop functional hydrogels.

5. Conclusions

In this work, we investigated the effects of gelatin cryostructuring followed by chemical crosslinking of gelatin molecules on the structural and functional characteristics of mammalian and fish gelatin hydrogels. The integration of scanning electron microscopy, dielectric spectroscopy, and rheological data provides a comprehensive view of the modulation of molecular, morphological and rheological properties of gelatin hydrogels under the action of photo-induced chemical crosslinking.

Photo-induced chemical crosslinking of gelatin chains with MSB in combination with freeze-thaw actions modifies the microstructure of hydrogels, creating an original macroporous architecture. Our SEM results revealed that crosslinked hydrogels are characterized by enlarged pores compared to non-crosslinked systems. For mammalian gelatin, multiple pores with thin partitions are formed, indicating the formation of a dense, stable polymer network. In the fish gelatin hydrogel, larger oval pores with thickened partitions are formed, preserving an ordered but less stable supramolecular architecture. This phenomenon is associated with the displacement and reorientation of gelatin molecules by forming ice crystals during freezing, with subsequent crosslinking fixing this structure through covalent bonds. Despite the enlarged pores, the crosslinked hydrogels exhibit increased mechanical strength, particularly the mammalian gelatin, where the storage modulus increases by 2.5 times compared to the non-crosslinked version. This enhancement is due to the formation of dense microcrystalline domains in pore walls, reinforced by covalent bonds and extra hydrogen bonding between initially oriented triple helices. This phenomenon is associated with the high content of proline and hydroxyproline in mammalian gelatin, whereas fish gelatin forms a less durable network due to the low imino acid content.

Rheological measurements confirm that MSB crosslinking greatly improves the elasticity and thermal stability of mammalian gelatin. The crosslinked system maintains a gel state at higher temperatures, while the non-crosslinked mammalian gelatin undergoes gel-sol transformation. Fish gelatin exhibits reduced elasticity retention even after crosslinking due to weaker molecular interactions because of its amino acid composition. Comparison with other crosslinking agents demonstrates the effectiveness of MSB in gelatin crosslinking. In studies of hydrogels crosslinked with methacrylic anhydride, the mammalian gelatin showed lower elasticity and fish gelatin was markedly less durable compared to our results under similar conditions. This indicates more efficient covalent bonding of gelatin chains with the help of MSB. In addition, the adjacency of neighboring gelatin chains provokes the formation of extra hydrogen bonds. Moreover, MSB provides antimicrobial properties to gelatin hydrogels, expanding their potential applications in biomedical and food systems.

Dielectric data have deepened our explanation of the obtained SEM and rheology results to the molecular level: it was shown that gelatin crosslinking increased violently a portion of bound water in these systems. We have determined that mammalian and fish gelatin hydrogels differ strongly in binding strength and mobility of bound water, which is more weakly coupled and more mobile for gelatin of fish origin in comparison with crosslinked mammalian gelatin. Possibly, the difference in mobility of bound water is key to the explanation of different mechanical characteristics of crosslinked gelatin hydrogels.

Therefore, gelatin crosslinking with MSB, combined with cryostructuring, effectively strengthens the polymeric network of mammalian gelatin, providing additional balance between structural stability and mechanical properties. Fish gelatin remains less stable due to weak intra- and intermolecular interactions, emphasizing the importance of selecting appropriate starting materials for target applications. Further studies may focus on optimizing crosslinking parameters for fish gelatin and exploring the antimicrobial properties of MSB to develop functional hydrogels.