Allyl-Functionalized Polysaccharides for 3D Printable Hydrogels Through Thiol–Ene Click Chemistry

Abstract

This study presents the synthesis of allyl-functionalized polysaccharide carbamates (AFCs) with tailored water solubility designed for use in responsive hydrogels and 3D printing applications. A modular one-pot approach was employed to produce cellulose- and xylan-based AFCs, utilizing polysaccharide phenyl carbonates as activated compounds. By fine-tuning the degree of substitution (DS) of functional groups, the water solubility and shear-thinning properties of AFCs were controlled to enhance the gelation and printability. AFC-based hydrogels could be obtained by rapid gelation induced without harmful catalysts through UV irradiation at 365 nm. The materials displayed highly porous and interconnected microstructures, as well as mechanical resilience and high swelling ratios. The hydrogel formation was characterized, and its crosslinking degree was calculated using HR-MAS NMR. The study demonstrated that gelation behavior was sensitive to the pH value, with optimal results under neutral or acidic conditions. Initial 3D printing trials confirmed the material’s rapid shaping capabilities, which is beneficial for biomedical applications and advanced manufacturing of stimuli-responsive materials.

1. Introduction

Hydrogels are a class of materials defined as three-dimensional polymeric networks that can retain large amounts of water without deformation in a swollen state, demonstrating excellent properties due to their porous morphology and elasticity [1,2,3]. These 3D networks are mainly formed through physical or chemical crosslinking, both of which have advantages and disadvantages. Physical crosslinking is the well-known route to prepare hydrogels for several applications, and it includes the formation of non-covalent bonds through various types of interactions, such as hydrogen bonding, electrostatic interactions, or physical entanglement. These types of interactions provide the hydrogel with a balanced flexibility to rigidity structure while retaining reversibility and self-healing properties. However, due to the weak nature of these non-covalent interactions in physical crosslinking, the hydrogels obtained are amenable to instability and structure deformation due to sensitivity to external stimuli. This often results in weak mechanical properties, which limits their use in specific applications. Chemical crosslinking introduces covalent bonds, which allow the preparation of more stable hydrogels with enhanced physical and mechanical strength and toughness [4,5,6,7].

Synthetic polymers, such as polyvinyl alcohol [8], polyethylene glycol [9], polyvinyl pyrrolidone [10], and polyacrylic acid [11], are still widely used for the preparation of hydrogels due to their facile and controllable synthesis through a selection of various monomers containing different functional groups achieving defined chemical structures and molecular weights [12,13]. However, these conventional hydrogels pose significant challenges and limitations for their use in biomedical applications due to the lack of sufficient biocompatibility and biodegradability [14,15]. On the contrary, polysaccharide-based materials offer solutions to these challenges due to their excellent biocompatibility, biodegradability, and non-toxic nature. As a result, their use in biomedical applications, such as tissue engineering, has been increasing and gaining widespread acceptance [13]. Additionally, polysaccharides possess abundant reactive groups that can be chemically modified to achieve the desired properties [13,16,17].

Contrary to other polysaccharides, which have native hydrogel formation properties such as agarose and alginate, cellulose and xylan need chemical modifications to introduce reactive groups for crosslinking reactions to obtain hydrogels [18,19]. Click chemistry approaches are of particular interest in this context because they enable selective crosslinking under mild and aqueous conditions. Cellulose derivatives with azido and alkyne moieties have been prepared by conventional synthesis and were used to form hydrogels in aqueous media [20]. Similarly, xylan-based organogels and their resulting hydrogels have been prepared using the same crosslinking approach [21]. However, despite the widespread use of copper-catalyzed reactions in chemical synthesis and crosslinking, they still pose challenges due to copper trace contamination, which can be difficult to remove [21,22]. Thus, it is of great interest to explore other selective crosslinking reactions. Of great advantage in this context are modular synthesis approaches based on activated polysaccharide aryl carbonates that enable the predictable preparation of polysaccharide derivatives with different types of reactive groups, as well as tailored solubility behavior [21].

Recently, this approach was employed to synthesize cellulose and xylan derivatives with furan or maleimide moieties. Cellulose- and xylan-based hydrogels were prepared from these compounds through the Diels–Alder crosslinking reaction, which proved successful and could be tuned further to obtain hydrogels that are temperature sensitive and have reversibility properties, which could be exploited for self-healing [23]. Based on these findings, it was also of great interest to explore polysaccharide derivatives that can be crosslinked through the thiol–ene reaction. Radical thiol–ene reaction displays wide advantages, including rapid kinetics, no by-products, high yields and conversions, and no requirement for harmful catalysts [24,25]. In the past decade, the thiol–ene reaction became an interesting tool and was employed for the preparation of functional polysaccharide materials or for crosslinking reactions to form hydrogels for a wide range of applications [26,27,28,29].

Hydrogels with a well-defined 3D-printed morphology and tailored shape are of great interest for biomedical applications, such as tissue engineering and regenerative medicine [30,31]. Contrary to frequently applied thermoplastic or ablative 3D printing procedures, bioprinting of gels requires specific bioinks [32,33,34]. Polymeric systems that exploit the thiol–ene reaction are especially attractive in this context [35,36]. The crosslinking reaction that initiates the transition from a processible liquid to a stable gel proceeds very fast (within seconds) and enables 3D printing of well-defined objects. Moreover, it is triggered by irradiation with UV light, which allows for very precise localization of when and where the sol–gel transition occurs.

Due to the inherent advantages of polysaccharides over synthetic polymers, there is a high demand for water-soluble polysaccharide derivatives that feature reactive groups for the thiol–ene reaction in the context of hydrogels and biobased 3D printing [37]. Although conventional modification methods for introducing alkene or thiol groups into the polymer backbone of certain polysaccharides have been reported in the literature, these approaches often suffer from inefficiencies and difficulties in controlling the degree of substitution (DS) of the resulting derivatives [38]. Additionally, they are limited to specific types of polysaccharides. Consequently, there is a need for modern synthetic approaches that address these challenges by enabling the preparation of customized polysaccharide derivatives, which are accessible for selective thiol–ene reactions. It has been demonstrated that polysaccharide phenyl carbonates are well suited for this purpose. They can be synthesized with well-defined DS values up to the theoretical maximum (e.g., 3.0 for cellulose and 2.0 for xylan). The phenyl carbonate moiety can be replaced quantitively by conversion with amines to introduce various functionalities, including allyl groups that are reactive towards thiols [21,39,40,41]. Moreover, it is possible to introduce hydrophilic groups that can tune the solubility in aqueous media, which is important because the desired groups for click chemistry reactions are hydrophobic.

Based on the previous studies, it was hypothesized that (i) water-soluble cellulose carbamates and xylan carbamates with alkene moieties can be synthesized by a modular approach, (ii) hydrogels can be prepared from these compounds by covalent crosslinking through thiol–ene reaction, (iii) the properties of the hydrogels can be controlled by the molecular structure of the derivatives, as well as the crosslinking conditions, and (iv) the materials are suitable for 3D printing applications.

2. Materials and Methods

2.1. Materials

Solvents used for the synthesis of polysaccharide derivatives, including 1-butyl-3-methylimidazolium chloride (BMIMCl; abcr GmbH, Karlsruhe, Germany), N,N-dimethylformamide (DMF), and pyridine, were purchased in anhydrous grade and stored as received (Thermo Fisher Scientific, Henningsdorf, Germany). Arabinoxylan from oat spelt (monosaccharide composition: xylose: 53.7%, arabinose: 33.3%, glucose: 11.5%; molecular weight was determined by SEC using pullulan as calibration, Mw = 28,900 g/mol; see Figures S1–S6 in Supporting Information for spectroscopic characterization) was provided by Jäckering Mühlen- und Nährmittelwerke GmbH (Hamm, Germany). Microcrystalline cellulose (Avicel PH-101; molecular weight determined by viscometry, Mw = 25,300 g/mol) and sodium alginate (from brown algae, content of sodium carboxylate groups: 2.63 ± 0.12 mmol g⁻¹) were purchased from Sigma Aldrich (Taufkirchen, Germany). Allylamine (AA), 3-(dimethylamino)propyl amine (DMAPA), dithiothreitol (DTT; racemic mixture), 2,2′-(ethylenedioxy)diethanethiol (EDT), 2-hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone (Irgacure 2959), and potassium persulfate were purchased from Sigma Aldrich (Taufkirchen, Germany) and were used as received. Trifluoroacetic acid was purchased from TCI chemicals (Eschborn, Germany). Cellulose phenyl carbonate (CPC; DS_PC = 3.00) and xylan phenyl carbonates (XPC; DS_PC = 2.00) were synthesized according to the literature [42]. Nanofibrillated cellulose (NFC; 3 wt%) was kindly provided by Sappi Europe, (Maastricht, The Netherlands).

2.2. Measurements

Nuclear Magnetic Resonance (NMR) spectra of the polysaccharide derivatives were obtained using either a Bruker Avance 400 MHz or 500 MHz spectrometer (Billerica, MA, USA), with 16 scans with a relaxation delay of 2 s for ¹H NMR and up to 12,288 scans with a relaxation delay of 0.7 to 5 s for ¹³C NMR) and sample concentrations set at 50 mg/mL. To quantify phenylcarbonate groups, approximately 15 drops of trifluoroacetic acid were added to the solutions. Fourier-transform infrared (FTIR) spectra were recorded on a Nicolet iS5 spectrometer from Thermo Fischer Scientific (Kandel, Germany) using attenuated total reflectance (ATR); for each sample, 64 scans were taken over a range of 500 to 4000 cm⁻¹. Elemental composition was analyzed with a VARIO EL III CHNS analyzer (Elementaranalysensysteme GmbH, Langenselbold Germany). The degrees of substitution (DS) were determined by ¹H NMR spectroscopy for carbonate derivatives, while a combination of quantitative ¹³C NMR spectroscopy and elemental analysis was used for carbamate derivatives, as detailed in the Supporting Information. The molecular weight of the carbamate derivatives was determined by size exclusion chromatography using dimethylsulfoxide (DMSO) with 0.5 wt% LiBr as eluent (65 °C, flow rate: 0.5 mL/min) and pullulan as calibration standard (see Table S1 in Supporting Information). A JASCO system (isocratic pump PU-980, RI-930 refractive index detector) with a Novema 3000 and a Novema 300 column in series (Jasco Deutschland GmbH, Pfungstadt, Germany) was employed for the experiments. Sheer thinning behavior was measured using a Haake RotoVisco1 device provided by Thermo Fischer Scientific (Karlsruhe, Germany), and the viscosity curves were measured from a shear rate between 0.1 and 100 s⁻¹ at 25 °C using a cone-plate geometry (diameter: 60 mm, angle: 1°, material: titanium). Scanning electron microscopy (SEM) images were acquired using a Sigma VP Field Emission Scanning Electron Microscope (Carl-Zeiss AG, Oberkochen, Germany) equipped with an SE detector and operated at an accelerating voltage of 10 kV. Prior to imaging, the samples were sputter-coated with a thin layer of platinum using a CCU-010 HV system (Safematic GmbH, Zizers, Switzerland).

2.3. Synthesis of Allyl-Functionalized Polysaccharide Carbamate (Typical Example, AFC 02)

CPC (50 g, 95.78 mmol) was dissolved in DMF (500 mL) at 60 °C, and AA (13.12 g, 229.80 mmol, 0.80 equivalents per phenyl carbonate moiety) was dissolved in DMF (20 mL) and added to CPC solution. The mixture was allowed to react at 60 °C for 1 h under stirring, then a solution of DMAPA (123.32 g, 4.20 equivalents per phenyl carbonate moiety) in DMF (50 mL) was added, and the reaction was continued for 20 h at 60 °C. Afterward, the reaction mixture was cooled to room temperature and dialyzed extensively against deionized water for up to 5 days, during which water was exchanged regularly. The desired polysaccharide carbamate was isolated by lyophilization and dried at 50 °C under vacuum.

• DS_ene = 1.19; DS_sol = 0.79.
• Yield: 47 g (129.82 mmol; 75% molar yield).
• ¹³C NMR (125 MHz, DMSO-d6): (δ, ppm) 156.1, 155.3, 135.8, 115.4, 102,7, 100.9, 73.1, 62.6, 56.9, 45.4, 43.1, 39.1, 27.6.
• Elemental analysis: theoretical: C% 51.41, H% 7.01, and N% 10.72; experimental: C% 50.21, H% 6.82, and N% 10.69.

2.4. Preparation of Polysaccharide Hydrogels Through Thiol–Ene Click Reaction (General Procedure)

AFC 02 (3 g, 9.86 mmol) and the low molecular weight crosslinker DTT (1.0 eq, 9.86 mmol) were dissolved in 100 mL aqueous acetate buffer solution with pH = 5 (acetate buffer, 0.1 M) separately. The two solutions were mixed in such a way that a final polymer mass concentration of 3 wt% was reached. Irgacure 2959 (0.05 wt%) was used as a photoinitiator, and the mixture was homogenized for 1 min using a lab vortex shaker. The final mixture containing AFC, DTT, and irgacure 2959 was filled in cubic molds (1 cm³) and irradiated under UV light (365 nm) for 30 min. A different reaction medium was used to prepare hydrogels in a similar manner as above, in which a mixture of EtOH/H₂O and two different crosslinkers (DTT and EDT) were employed.

2.5. Hydrogel Characterization

2.5.1. High-Resolution Magic-Angle Spinning NMR Spectroscopy (HR-MAS NMR)

HR-MAS NMR spectra were taken using a Bruker Avance 500 MHz for ¹H and 125 MHz for a ¹³C spectrometer with a 4 mm MAS rotor (Billerica, MA, USA). In total, 128 scans were taken with a relaxation of 2 s for ¹H HR-MAS NMR, and 40,960 scans with a relaxation delay of 2 s were taken for ¹³C HR-MAS NMR, while D₂O was used to allow the hydrogel to swell. For the calculation of the degree of crosslinking, the following equation adopted from the literature was used [43]:

$$\mathrm{D}\mathrm{C}\%={\displaystyle \frac{\frac{I_9^i}{I_{9\prime }^i}-\frac{I_9^c}{I_{9\prime }^c}}{\frac{I_9^i}{I_{9\prime }^i}}}\times 100\%=\left(1-{\displaystyle \frac{I_9^c\times I_{9^{\prime }}^i}{I_9^i\times I_{9^{\prime }}^c}}\right)\times 100\%,$$

(1)

where the superstitions i and c refer to the intensities measured initially and after crosslinking, respectively. The superstition numbers 9 and 9′ refer to the carbons of the allyl moiety C9 and the DMAPA moiety C9′, respectively.

2.5.2. Swelling Behavior

The swelling capacity of the hydrogels was measured gravimetrically after immersion of lyophilized samples in appropriate solvents. Hydrogels were prepared in cubic forms (1 cm³), and their dimensions and weight were measured prior to lyophilization. The samples were then immersed in sucrose solution (2 wt%) and lyophilized. The lyophilized samples (dimensions and weight were measured prior to immersion in buffer solutions) were immersed in aqueous solutions of different pH values until equilibrium (when the weight of the swollen samples attained a constant state). The swollen samples were taken out of solution, carefully removing any excess water with filter paper before weighing and measuring the dimensions. The relative volume (RV%) and relative weight (RW%) percentages were calculated using the following equations:

$$\mathrm{R}\mathrm{V}\%={\displaystyle \frac{\mathrm{Vs}}{\mathrm{Vd}}}\times 100\left(\%\right),$$

(2)

$$\mathrm{R}\mathrm{W}\%={\displaystyle \frac{\mathrm{W}\mathrm{s}}{\mathrm{W}\mathrm{d}}}\times 100\left(\%\right),$$

(3)

where Vd and Vs are the volumes of dried and swollen samples, and Wd and Ws are the masses of dried and swollen samples, respectively.

The swelling ratio (SR%) was calculated using the following equation:

$$\mathrm{S}\mathrm{R}\%={\displaystyle \frac{(\mathrm{W}\mathrm{s}-\mathrm{W}\mathrm{d})}{\mathrm{W}\mathrm{d}}}\times 100\left(\%\right),$$

(4)

where Ws and Wd are the weights of the swollen and dried hydrogel samples, respectively.

2.5.3. Mechanical Testing

The mechanical properties of the prepared hydrogels were measured using a universal testing machine Zwick Roell compression device (Zwick/Roell, Ulm, Germany) with a force reaching a maximum of 500 N; the measurements were carried out at 25 °C. Hydrogels were prepared in cubic forms, and their dimensions were measured prior to the compression tests. Different polymer and crosslinking concentrations, as well as different media for hydrogel formation, were investigated.

Young’s modulus (E) was calculated using the following equation:

$$E={\displaystyle \frac{\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{s}}{\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}}}={\displaystyle \frac{\mathsf{\sigma }}{\mathsf{\epsilon }}}\left(\mathrm{k}\mathrm{P}\mathrm{a}\right)={\displaystyle \frac{\mathsf{\sigma }}{\raisebox{1ex}{$dL$}\!\left/ \!\raisebox{-1ex}{$L$}\right.}}\left(\mathrm{k}\mathrm{P}\mathrm{a}\right),$$

(5)

where σ is the measured stress, dL is the difference between the length at stress and the initial length, and L is the initial length.

2.6. Preliminary Printing Trials

The allyl-functionalized derivative AFC 01 was dissolved in buffer solution (pH value of 5), and dithiothreitol with Irgacure 2959 (0.05 wt%) was added to the polymer solution and mixed thoroughly using a lab vortex. The mixture was covered with aluminum foil to prevent crosslinking in the solution. The mixed solution was loaded into a syringe that was covered in aluminum foil and equipped with a 27-gauge needle. The solution was injected into a Petri dish under UV irradiation (365 nm) without premature gelation inside the syringe.

2.7. Three-Dimensional Printing Experiments

The printing ink was prepared as follows: APC 02 (100 mg) was dissolved in water (2.5 g), and the pH values were adjusted to 4 using 0.1 M HCl. NFC dispersion (7.5 g, 4 wt%), together with alginate (667 mg), DTT (21.3 mg), and Irgacure 2959 (6.2 mg), were added, and the ink was mixed thoroughly using a mechanical stirrer.

A GeSiM Robotics BioScaffolder 3.2 printer (GeSim, Radeberg, Germany) with a printing speed of 10 to 20 mm s⁻¹ was employed. The GeSim software (version 1.17.2.4575) and the 3D Builder 18.0.1931.0 (Microsoft, Redmond, WA, USA) were used to create the models for printing (FSU letters: 30 × 20 mm², cubes: 5 × 5 × 5 mm³). The composite inks were extruded from a 10 mL polyethylene-based plastic barrel UV/Vis light block (Nordson UK Limited, Manchester, UK) equipped with tips with an inner diameter of 410 µm (covered with tape to block UV light) onto polystyrene Petri dishes. The specimens were printed with extrusion pressures between 150 and 180 kPa. A UV lamp OmniCure Series 1500 (365 nm, Excelitas Technologies Corp., Pittsburgh, PA, USA) was mounted directly on the printer head, and the extruded ink was exposed to UV light for 2 s after printing each layer. After finishing the printing process, the final samples were irradiated for another 30 s. For comparison, specimens were printed without any UV irradiation.

3. Results and Discussion

3.1. Synthesis of Allyl-Functionalized Polysaccharide Carbamates

The first objective of this study was to synthesize allyl-functionalized polysaccharide carbamates (AFCs) with customized properties and solubility. The water solubility of these AFC derivatives was modulated by adjusting the molar ratios of the two key functional groups. Optimizing these characteristics is essential for enhancing printability and expanding the potential uses of AFC-based materials in advanced manufacturing techniques, as will be discussed in the sections below. The preparation of AFCs was carried out through a modular approach in a one-pot reaction, as depicted in Figure 1. This synthetic approach employs polysaccharide phenyl carbonates as activated compounds, which allows for the introduction of multiple amines, such as allylamine (AA) and 3-(dimethylamino)propyl amine (DMAPA). The resulting cellulose- and xylan-based AFCs combine two functionalities: (i) alkene moiety for the thiol–ene functionalization and (ii) a tertiary amine that can induce water solubility and pH responsiveness [41,44].

The mixed AFC derivatives were prepared by one-pot conversion of cellulose phenyl carbonate (CPC; DS of 3.00) or xylan phenyl carbonate (XPC; DS of 2.00) with AA for 1 h, followed by DMAPA overnight without prior isolation of the product. Based on previous knowledge, the total amount of amines was fixed at five equivalents (CPC) or four equivalents (XPC) per phenyl carbonate moiety (

Table 1. Conditions and results of the synthesis of allyl-functionalized polysaccharide carbamates by conversion of polysaccharide carbonates with allyl amine (AA) and 3-(dimethylamino)propyl amine (DMAPA) in N,N-dimethylformamide (DMF) at 60 °C.

Reaction Conditions			Sample ID	Results
Polysaccharide Carbonate (a)	DSPC (b)	Molar Ratio (c)		DSene (b)	DSsol (b)	Solubility (d)
						Buffer pH 5	EtOH/H2O
CPC	3.00	1.00 + 4.00	AFC 01	0.84	1.55	+	+
CPC	3.00	0.80 + 4.20	AFC 02	1.19	0.79	+	+
CPC	3.00	0.55 + 4.45	AFC 03	0.78	1.18	+	+
XPC	2.00	0.55 + 3.45	AFC 04	0.49	1.20	+	+
XPC	2.00	0.30 + 3.70	AFC 05	0.22	1.19	+	+

^(a) CPC: cellulose phenyl carbonate, XPC: xylan phenyl carbonate; ^(b) degree of substitution, PC: phenyl carbonate, ene: N-allyl carbamate, sol: N-[3-(dimethylamino)propyl] carbamate; ^(c) mol AA + DMAPA per mol PC group; ^(d) determined at 10 mg/mL, +: soluble, 0.1 M aq. acetate buffer, ethanol/water (v:v; 1:1).

). Moreover, the relative ratio of AA and DMAPA was adjusted to tune the number of reactive alkene moieties (DS_ene) and solubilizing tertiary amino groups (DS_sol). The synthesized AFC derivatives were characterized in terms of their structure by elemental analysis and NMR spectroscopy. DS values were obtained by using a combination of quantitative ¹³C NMR and elemental analysis [41].

The DS values varied, ranging from 0.22 to 1.19 for the reactive group and from 1.19 to 1.55 for the solubility group (Table 1). These values indicate the versatility of the resulting polymers and their suitability for the preparation of advanced functional materials. Additionally, the presence of reactive double bonds along the polymer backbone makes these derivatives excellent candidates for the synthesis of responsive hydrogels. The AFC derivatives were soluble in organic solvents, such as ethanol and dimethylsulfoxide (DMSO), as well as in aqueous media. This versatility makes them ideal for hydrogel synthesis, where the choice of solvent water, ethanol, or a mixture of both can be tailored to the specific requirements of the intended application. Overall, these properties highlight the potential of AFC derivatives in a wide range of applications, particularly in fields involving stimuli-responsive materials and biomedical hydrogels.

¹³C NMR spectrum provides clear evidence of the successful incorporation of both the allyl carbamate and the 3-(dimethylamino)propyl carbamate groups into the polymer backbone (Figure 2). This is demonstrated by the appearance of new peaks at approximately 116 ppm and 134 ppm, both corresponding to the double bond of the allyl carbamate. Additionally, a peak around 43 ppm is observed, which is associated with the carbon atom (CII) adjacent to the amide group, while a peak at 155 ppm is related to the carbonyl group. The presence of the 3-(dimethylamino)propyl carbamate group is also confirmed by several peaks in the spectrum at 27 ppm (C9′), 39 ppm (C8′), 45 ppm (C11′), 57 ppm (C10′), and 156 ppm (C7′). Moreover, no residual peaks were detected corresponding to the phenyl carbonate groups, which are typically found in the range from 120 to 130 ppm. This absence supports the proposed structure of the resulting carbamate and confirms that the conversion of the polysaccharide phenyl carbonates into AFCs was quantitative, with no unreacted starting material remaining (see also Figures S7 and S8 in Supporting Information for the NMR spectra of other AFC derivatives). The conversion of carbonates into carbamates was also confirmed by FTIR (see Figures S9 and S10 in Supporting Information). The spectra revealed the characteristic absorption bands around 1540 cm⁻¹ related to N-H deformation vibration and around 1700 cm⁻¹ (C=O), which is characteristic for the carbonyl moiety of carbamate derivative, while a peak around 920 cm⁻¹ related to H-C=C vibration was also observed.

3.2. Preparation of Hydrogels Through Thiol–Ene Click Reaction

In the first set of experiments, hydrogel formation was studied using various polymer concentrations ranging from 0.5 to 7.5 wt% and different crosslinking ratios from 0.1 to 1 mol equivalent. For this purpose, various concentrations of AFCs were dissolved in a buffer solution (pH 5). Afterward, the crosslinker (dithiothreitol/DTT or 2,2′-(ethylenedioxy)diethanethiol/EDT) and photoinitiator (2-hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone/Irgacure 2959) were added and thoroughly mixed until a homogeneous solution was obtained. These solutions were exposed to UV light (365 nm) to initiate the crosslinking process. In previous work, rheological experiments have been employed to gain information on the kinetics of the gelation process by covalent crosslinking [23]. However, this method was not applicable to this work. In a typical rheometer, only a small surface area is potentially accessible for UV irradiation, which is needed to initiate the thiol–ene reaction. Thus, successful hydrogel formation was determined when the solution remained intact and did not flow when inverted (Figure 3).

Using the inverted tube method, the gelation times in aqueous solutions were estimated for different reaction conditions using DTT and EDT (

Table 2. Conditions and results of the formation of hydrogels in aqueous buffer solution (pH value of 5) by crosslinking of allyl-functionalized polysaccharide carbamates with either dithiothreitol (DTT) or 2,2′-(ethylenedioxy)diethanethiol (EDT) and 2-hydroxy-4′-(2-hydroxyethoxy)-2-methylpropiophenone as the photoinitiator with irradiation using UV light (365 nm) at 25 °C.

Polymer		Crosslinking Conditions			Gelt Time
Sample ID	DSene (a)	Polymer Concentration [wt%]	Linker	eq. of Linker [mol] (b)	<10 s	<20 s	<30 s	<40 s
AFC 02	1.19	7.5	DTT	1.00	+
AFC 02	1.19	7.5	DTT	0.50	+
AFC 02	1.19	7.5	DTT	0.25	+
AFC 02	1.19	7.5	DTT	0.10	+
AFC 02	1.19	5	DTT	1.00	+
AFC 02	1.19	5	DTT	0.50	+
AFC 02	1.19	5	DTT	0.25	+
AFC 02	1.19	5	DTT	0.10	+
AFC 02	1.19	3	DTT	1.00	−	+
AFC 02	1.19	3	DTT	0.50	−	+
AFC 02	1.19	3	DTT	0.25	−	+
AFC 02	1.19	3	DTT	0.10	−	−	+
AFC 02	1.19	1.5	DTT	1.00	−	+
AFC 02	1.19	1.5	DTT	0.50	−	−	+
AFC 02	1.19	1.5	DTT	0.25	−	−	+
AFC 02	1.19	1.5	DTT	0.10	−	−	−	+
AFC 02	1.19	1	DTT	1.00	−	+
AFC 02	1.19	1	DTT	0.50	−	−	+
AFC 02	1.19	1	DTT	0.25	−	−	+
AFC 02	1.19	1	DTT	0.10	−	−	−	+
AFC 02	1.19	1	EDT	0.25	−	−	+
AFC 02	1.19	1	EDT	0.10	−	−	−	+
AFC 02	1.19	0.5	DTT	1.00	−	−	−	−
AFC 03	0.78	3	DTT	1.00	−	+
AFC 03	0.78	3	DTT	0.50	−	+
AFC 03	0.78	3	DTT	0.25	−	+
AFC 03	0.78	3	DTT	0.10	−	+
AFC 04	0.49	5	DTT	1.00	+
AFC 04	0.49	3	DTT	1.00	+
AFC 04	0.49	2.5	DTT	1.00	−	−	−	−
AFC 05	0.22	5	DTT	1.00	+

^(a) degree of substitution of allyl moiety; ^(b) equivalents of crosslinker DTT or EDT per allyl moiety, gelation determined by inverted tube method in seconds, +: gelation occurred, −: gelation did not occur.

). Overall, it can be stated that the gelation occurred rapidly and nearly instantaneously. The process was determined to take less than 40 s for the solution to fully transition into a gel, with no observable flow for any of the polymer concentrations tested. For higher polymer concentrations (5 and 7.5 wt%), the gelation time was even faster, occurring in less than 10 s, regardless of the crosslinking equivalents examined. This is probably because the amount of available crosslinking groups is higher at higher concentrations, i.e., the formation of the first initial crosslinks occurs faster. The number of crosslinkers per reactive allyl group also had a certain effect on the gelation kinetics, especially at lower polymer concentrations where the gelation process proceeded slower. Nevertheless, even at a rather small amount of crosslinker of only 0.1 equivalents, the gelation occurred within less than 40 s, which is still fast when compared to other covalent crosslinking reactions. In addition to DTT, EDT was studied as another multivalent thiol crosslinker. Due to the lower solubility of EDT (1.46 mg/mL), this was only feasible for solutions with a low polymer mass concentration of ≤1 wt%. The same rapid reaction was observed with EDT, suggesting that the gelation kinetics are independent of the crosslinker. However, there was a minimum polymer concentration required for observing a macroscopic gelation. For cellulose-based hydrogels, this threshold was 1 wt%, and 3 wt% was for xylan-based hydrogels, below which gelation could not be achieved, even with the addition of two equivalents of DTT.

The pH value of the polymer solution was found to significantly influence the thiol–ene crosslinking reaction. To investigate this effect, two AFC 02 polymer solutions (3 wt%) were prepared in water without a buffer. After dissolution, the pH value of both solutions was measured and found to have risen to 10. This increase in pH value was attributed to the presence of a tertiary amine group attached to the polymer backbone, which possesses basic properties and influences the pH value of the solution, irrespective of the initial pH value of the solvent. One of the polymer solutions was then neutralized to a pH value of 7 using aqueous HCl, while the other remained unchanged. Hydrogel formation was then tested with the unchanged polymer solution of a pH value of 10, and the gelation was significantly delayed, requiring several hours of UV irradiation to reach completion (see Figure S11 in Supplementary Information). This delay can be attributed to the formation of metastable disulfide radical anion species, which is facilitated by amine deprotonation and leads to thiolate anion generation [45]. The reaction is favored under basic conditions. However, this effect was not observed when the polymer solution was neutralized or acidified to a pH value of 7 or lower using aqueous HCl. Therefore, it was concluded that for proper gelation of AFC derivatives, the polymer solution must be neutralized or acidified.

To summarize, the experiments demonstrated the applicability of thiol–ene reaction for the fast preparation of hydrogels using the novel AFC derivatives. The crosslinking was initiated by UV irradiation. Additionally, it was found that the crosslinking reaction can also be initiated thermally at 70 °C by using potassium persulfate as a thermal initiator for radical formation. This further broadens the applicability, e.g., by enabling the addition of composite materials to the gel-forming solutions that would block the transmission of UV light required for photoinitiated crosslinking.

3.3. Spectroscopic Characterization of Polysaccharide Gels

The gels were characterized by spectroscopic methods to gain further insight into the crosslinking process. For this purpose, hydrogels were prepared using 3 wt% AFC 02 and 1.0 equivalent of DTT. The gels were lyophilized to avoid the influence of water and studied by FTIR spectroscopy. The FTIR spectra revealed several characteristic peaks for the polysaccharide carbamate materials, including a prominent peak near 1700 cm⁻¹, associated with the C=O stretching vibration (see Figure S12 in Supporting Information). When comparing the spectra of the crosslinked gel and the starting material AFC 02, an increase in the intensity of the O–H stretching band around 3320 cm⁻¹ was observed alongside a decrease in the intensity of the H–C=C stretching band near 920 cm⁻¹. These changes in peak intensity indicate the occurrence of a crosslinking reaction between DTT and the AFC derivative. The reduction in the H–C=C peak is specifically attributed to the consumption of double bonds during the thiol–ene reaction, where DTT interacts with the unsaturated bonds in the AFC derivative. HR-MAS ¹H and ¹³C NMR spectroscopy were employed to verify the crosslinking reaction in the native hydrogel and to quantify its crosslinking efficiency (Figure 4).

The HR-MAS ¹³C spectrum of the hydrogel displayed residual peaks corresponding to the C=C double bond, as well as new peaks in the range of 40 to 25 ppm that are attributed to sp³-hybridized carbon atoms introduced by the crosslinking reaction between the allyl group and DTT, confirming the successful crosslinking reaction. A slight shift in the C9′ resonance was observed, which is likely due to the protonation of the tertiary amine during this particular NMR experiment. It should be noted, too, that a broad signal around 110 ppm is present in the spectra, which is caused by the rotor cap (made of Teflon) [46].

The efficiency of the crosslinking reaction was determined by analyzing the HR-MAS ¹H spectra of both the crosslinked hydrogel and the native polymer AFC 02 (Figure 5). The underlying principle is that during the crosslinking process, the allyl groups are consumed, leading to a reduction in the resonance intensity of I_{9 and 10}. This reduction, expressed as the change in resonance intensity ΔI_{9 or 10}, can be obtained by measuring the intensities before and after hydrogel formation, enabling the calculation of the crosslinking efficiency. Since the resonance intensities for the polymer and hydrogel samples are measured separately, it was necessary to normalize them against a reference resonance intensity that remains unaffected by the crosslinking reaction. For this purpose, the resonance intensity I_9′ was chosen. The crosslinking degree was calculated to be 15%, which is consistent with values reported in the literature for similar polymer-based hydrogels prepared by thiol–ene crosslinking [43]. Overall, it can be concluded that the hydrogels still contained allyl moieties, which can be further exploited in terms of surface functionalization for desired applications (e.g., the attachment of dyes or other compounds).

3.4. Swelling Behavior of Thiol–Ene Crosslinked Gels

The swelling behavior of the prepared hydrogels was investigated through a series of experiments. Specifically, the transformation of hydrogels into cryogels via lyophilization was examined to assess any impact on their properties. AFC 02 was used in this context because it showed fast gelation and could be synthesized efficiently in high yields. The hydrogels were synthesized following the same protocol as previously described, with a polymer mass concentration of 3.0 wt%. The concentration of the crosslinker DTT was systematically varied from 0.25 to 1.0 equivalents to study its influence on swelling behavior. For comparison, gels with 1.5 wt% APC 02 and 1.0 equivalent of DTT were included as well. To protect the hydrogels from ice crystal formation during the lyophilization process that involves freezing the samples, a 2 wt% aqueous solution of sucrose was used as a cryoprotectant. This precaution was essential to maintain the intrinsic swelling characteristics of the gels after the drying process [47]. Once the hydrogels were fully immersed in the sucrose solution, they were subjected to lyophilization using a freeze-drying apparatus. The resulting cryogels were rehydrated by immersion in buffer solutions with varying pH values (5, 7, and 8) to evaluate their swelling behavior in different environments. Prior to and after lyophilization, the dimensions and weights of the gel specimens were carefully measured to monitor changes in their physical properties and provide key insights into the structural integrity and swelling capacity of the cryogels under different conditions. All tested samples exhibited swelling and volume expansion upon rehydration while maintaining their structural integrity without any visible deformation or compromise in mechanical strength. The observed volume increase in the dried state can be attributed to the swelling behavior of the native hydrogels when exposed to the sucrose solution. This substantial expansion confirms the inherent swelling capacity of the hydrogels in their hydrated form. Although the shape and size of the hydrogel samples remained largely unchanged by the lyophilization process, there was a significant reduction in weight, which was expected due to the removal of water content during lyophilization. The preservation of shape, despite the significant weight loss, highlights the resilience of the hydrogel structure throughout the drying and rehydration cycles.

The dried cryogels exhibited considerably higher water uptake values of ≥1000% across all tested polymer and crosslinker concentrations (Figure 6). The swelling was most pronounced for cryogels immersed in buffer with a pH value of 5. The volumes and masses were like the original hydrogels before the lyophilization, and the water uptake was 1437% compared to the dried cryogel (3 wt%/0.5 equivalent DTT). At pH values of 7 and 8, the samples still showed considerable water uptake of 991% and 879%, respectively. However, relative to the native hydrogels, a certain shrinkage was observed (see Figure S13 in Supporting Information). This pH responsiveness can likely be attributed to the presence of tertiary amine groups attached to the polymer backbone. When protonated, these amines enhance hydrophilicity and facilitate water permeation, while deprotonation increases hydrophobicity, leading to the observed shrinkage and reduced swelling behavior. This pH-sensitive behavior highlights the potential of these cryogels in applications requiring responsiveness to environmental conditions, such as drug delivery systems or tissue engineering scaffolds, where controlled swelling and deswelling are critical for function [48,49].

In addition to the pH value, the number of crosslinkers showed an effect on the swelling behavior. As a general trend, the lowest relative volume and mass after swelling were observed for the highest crosslinker amount of 1.0 equivalents of DTT. Upon reducing the amount of DTT, the volume increased, as did the amount of water that can be adsorbed. As an example, the water uptake at a pH value of 5 increased from 998% to 1437%, and finally, to 1550% when the amount of DTT decreased from 1.0 to 0.5 and finally to 0.25 equivalents. The inverse relationship between crosslinking concentration and swelling behavior is probably due to the formation of less dense and more flexible polymer networks when the degree of crosslinking decreases, while more rigid structures can restrict the water uptake. These findings suggest that tuning the crosslinking density provides a powerful tool for optimizing the swelling behavior for specific applications. At lower polymer mass concentrations of 1.5 wt%, the absolute mass of water that can be adsorbed was lower compared to 3.0 wt% (Figure 6). This could be a result of the less dense pore structure. However, the relative water uptake was comparable (e.g., 1508% for 1.5 wt% and 1.0 equivalents of DTT). Overall, it can be summarized that the cryogels obtained through thiol–ene crosslinking of AFC derivatives showed strong swelling behavior in aqueous systems. These findings are similar to those that have recently been reported for gels prepared by crosslinking through the Diels–Alder reaction [23]. Moreover, it was possible to tune the swelling behavior by adjusting the crosslinking conditions.

3.5. Morphology and Pore Distribution of Thiol–Ene Crosslinked Gels

To examine the morphology of the hydrogels, scanning electron microscopy (SEM) analysis was conducted. The hydrogels were synthesized by combining AFC 02 at varying polymer mass concentrations with one molar equivalent DTT and Irgacure 2959 as the photoinitiator. The mixture was exposed to UV light at a wavelength of 365 nm. To protect the hydrogel structure during freezing, a 2 wt% aqueous sucrose solution was used as a cryoprotectant before lyophilization. SEM images revealed that the resulting cryogels exhibited a highly porous, interconnected microporous structure (Figure 7). This is attributed to the removal of water molecules during the lyophilization process and the generation of empty vacancies. The pore sizes were influenced by the initial polymer mass concentration. At 1 wt%, pore sizes ranged from 20 to 100 µm. Gels prepared with 3 wt% polymer featured a more compact microstructure, which could be attributed to the overall higher mass content. The pores are partly covered, but these could potentially be artifacts of the drying process, considering the finding that 3 wt% cryogels featured higher swelling. Using a higher mass concentration of 5 wt% yielded slightly larger pore sizes again in the range of 100 to 300 µm. Overall, it can be stated that the cryogels displayed uniformly spherical-shaped porous structures, suggesting a consistent and homogeneous gel microstructure and morphology. The microporous architecture and high porosity of the prepared hydrogels can be beneficial for superabsorbent application due to the facilitation of rapid mass penetration in the prepared gels [50]. Furthermore, the observed morphology characteristics could be potentially exploited for drug delivery applications, as the release of drugs could be facilitated and improved [51].

3.6. Mechanical Properties of Thiol–Ene Crosslinked Gels

The mechanical properties of crosslinked polymer hydrogels play a crucial role in determining their suitability for various application fields, such as pharmaceuticals and biomedicine [52]. To investigate these properties, compression tests were conducted on hydrogels prepared with different polymer mass concentrations (3 wt% and 5 wt%) and amounts of DTT (0.25 to 1.0 equivalents) and reaction media (buffer of a pH value of 5 and a 1:1 water/ethanol mixture). The results indicated that all tested hydrogels exhibited good mechanical stability, demonstrating resilience to high pressures and elastic deformation up to a certain threshold. Beyond this point, the hydrogels experienced irreversible deformation, characterized by the formation of cracks and breaks. The maximum forces that the gels could withstand, as well as the maximum deformation before breaking, were determined (Figure 8).

The mechanical strength of the hydrogels increased with increasing polymer content. As an example, F_max increased from 30.9 kPa (3 wt%) to 57.9 kP (5 wt%) when 1.0 equivalent DTT was employed. Likewise, the deformation at F_max decreased from 38.5% to 31.3% in the same direction, which means that the gels could be compressed to a greater extent before breaking. This finding is reasonable, considering that a higher polymer content results in higher overall density. Decreasing the amount of DTT from 1.0 to 0.25 equivalents resulted in a slight decrease of F_max, which could be the result of a lower crosslinking density. However, the differences were somewhat small when compared to the experimental error of the method. Interestingly, gels prepared using an ethanol–water mixture (v:v; 1:1) exhibited a higher mechanical strength compared to the gels prepared in an aqueous buffer solution. It can be speculated that the differences in the solvent–solute interaction translated into stronger polymer–polymer interactions within the gel materials.

Understanding and controlling the stiffness of gel materials is crucial for various applications. Young’s modulus provides valuable insights into this mechanical property. In this study, Young’s modulus was determined for hydrogels by analyzing stress–strain curves obtained from compression tests. The results demonstrated that all hydrogel samples exhibited moduli within the desirable range of 60 to 198 kPa (Figure 8). This range is, among others, suited for tissue engineering applications [53]. Polymer concentration played a significant role in determining the elastic modulus of the gels. Increasing the mass concentration from 3 wt% to 5 wt% resulted in a three-fold increase in stiffness. For instance, a hydrogel with 3 wt% polymer concentration and 1.0 equivalent of linker displayed a stiffness of 70.9 kPa, while increasing the polymer content to 5 wt% with the same crosslinker mol equivalent resulted in a significant enhancement to 183.8 kPa. Conversely, variations in crosslinking concentration between 1.0 and 0.25 equivalents of linker had a minimal effect, like the results for F_max described above. The only exception was observed when using a small amount of DTT (0.25 equivalents instead of 0.5 or 1.0) for gels with 5 wt% polymer. These gels showed a very strong decrease that could not be fully explained. Moreover, the choice of medium for hydrogel preparation did not noticeably influence the elastic modulus within the experimental error.

The gels prepared in this study by thiol–ene crosslinking of AFCs exhibited improved mechanical strength compared to gels prepared by Diels–Alder crosslinking of polysaccharide carbamates with similar molecular structures that were reported recently [23]. The findings underscore the versatility of the thiol–ene hydrogels in terms of tailoring of mechanical properties. By strategically adjusting the polymer concentration and crosslinking density, the materials can be prepared with desired levels of strength and resilience to compression. This tunability is particularly valuable in applications (e.g., tissue engineering [53] and wound dressing [54]) where specific mechanical characteristics are essential.

3.7. Three-Dimensional Printing and Injectable Hydrogels Based on Polysaccharide Carbamates

Under UV irradiation, the gelation of the AFC derivatives prepared in this study proceeded almost instantaneously, resulting in the formation of gels. In the absence of UV light, the gel-forming solutions remained liquid and could be handled and processed without limitation. Based on these observations, it was postulated that it is possible to employ AFC derivatives for 3D printing of defined gel architectures. AFC 02 was chosen for these experiments because it featured a high DS_ene. The rheological properties of the compound in solution were studied at different concentrations because they are important for the extrusion of the polymer solutions during printing (see Figure S14 in Supplementary Information). Shear-thing behavior was observed for the AFC derivative, which is important for the processing of the liquid during the printing step. The measurements also indicated that polymer mass concentrations ≥ 7.5 wt% were suitable for trial tests. In a preliminary test, a solution of AFC 02 (7.5 wt%) was extruded through a 27-gauge needle in a defined manner using a syringe (see Figure S15 in Supplementary Information). Under UV irradiation, the solution immediately formed a gel, which enables the production of crosslinked gels with unique 3D architecture (see Video S1 in Supplementary Information). In a proof of concept, it was demonstrated that the novel AFC derivatives are well-suited for 3D printing applications in the biomedical field. The compounds are readily soluble in water, which is important for the bioprinting of biological materials such as cells, and rapidly form gels, which is important for the printing process itself.

An approach to further improve printability is to optimize the flow properties and viscosities, e.g., by using additives such as nanofibrillated cellulose (NFC) [55,56,57]. Following the initial proof of concept trials, experiments using a 3D printing device were performed. The viscosity of aqueous solutions of the tested AFC derivatives (around 0.1 to 1 Pa·s; see Figure S14 in Supplementary Information) was too low to facilitate direct printing. Thus, an ink was prepared consisting of the crosslinking components (AFC 02, DTT, and photoinitiator), as well as viscosity modifiers (NFC and alginate). It was shown that the addition of the latter ensures that the prints are stable and do not lose their shape within the time frame of the printing process (several minutes). However, crosslinking of the printed materials is required to maintain the shape for a longer period [57]. The AFC-containing ink was suitable for processing through the printing nozzle, and it was possible to create 3D architectures (Figure 9). After printing each layer, which had a thickness of about 400 µm, as well as at the end of the printing process, the print was exposed to UV irradiation to initiate the thiol–ene crosslinking. This process enabled the creation of mechanically stable AFC-based gels with defined 3D shapes. When stored in aqueous media, the shapes were retained, which is an important aspect for application in biomedical areas (Figure 8b). For comparison, 3D printing was also performed without UV curing. The 3D shapes prepared this way were stable initially but started to disintegrate in aqueous systems due to strong swelling and the lack of stabilizing crosslinks. To summarize, it could be demonstrated that AFC derivatives are well-suited for 3D printing applications. The rapid crosslinking through thiol–ene reaction enables the fabrication of well-defined gel architectures. An interesting approach in this context is to only partially expose the printed shapes to UV curing to fabricate polysaccharide-based gels with isotropic properties.

4. Conclusions

It could be demonstrated that water-soluble allyl-functionalized polysaccharide carbamates (AFCs) can be synthesized by a facile modular synthesis approach. The AFC derivatives formed hydrogels by covalent crosslinking through the thiol–ene reaction within a few seconds after UV irradiation at 365 nm. The gels were characterized comprehensively and demonstrated high mechanical strength and swelling capacity, as well as high porosity and homogeneous pore structure. Due to the fast gelation and the good overall properties, AFCs proved to be well-suited for 3D printing. This will open many potential applications in the biomedical field, in which biobased polymers are highly desired. An interesting and facile approach is to create gel materials with isotropic properties by only partial irradiation with UV light. Currently, comprehensive experiments with 3D printing devices are performed to evaluate process parameters and optimize printability. Moreover, the implementation and release of drug molecules are studied. Another potential field of application is the bioprinting of living cells and tissues. It could also be demonstrated that only about 15% of the reactive allyl moieties are consumed in the crosslinking process. Thus, it is possible to further functionalize the gels, e.g., to include stimuli-responsive properties.