3.1.2. Fabrication of Electrospun GelMA Matrices

Electrospun GelMA mats were prepared by electrospinning technique. Defect-free nanofibrous matrices were obtained as shown in Figure 1 (NG-0UV). Then, electrospun mats were successfully crosslinked by UV irradiation. As it is well-known, photo-chemical crosslinking takes place only in the presence of a photoinitiator, and thus hydrogel formation is triggered by the external source of UV light. Irgacure 2959 is usually chosen as photoinitiator due to its low toxicity at the used concentration as demonstrated in previous studies [23]. To evaluate the UV irradiation process time, GelMA mats were immersed in ethanol and exposed at different irradiation times. Figure 1 shows SEM micrographs of GelMA mats with 0, 6, 9 and 12 min of UV irradiation. In a first step of UV irradiation, fibers increased their diameter as result of their swelling in polar solvents. After UV irradiation, mats still exhibited their fibrous structures and their diameters were reduced with respect to the diameter of NG-0UV. Probably, this decrease in fiber size is related to the crosslinking of GelMA. The increase of diameter in NG-12UV could be ascribed to the increase in temperature produces by UV device.

The mean fiber diameter is summarized in Table 1.

**Table 1.** Fiber diameter, contact angle, water uptake (WR), and DSC results, melting temperature (Tm), and the onset of glass transition temperature (Tg), of GelMA mats.


**Figure 1.** SEM micrographs of samples at different irradiation times: 0 (NG-OUV), 6 (NG-6UV), 9 (NG-9UV) and 12 min (NG-12UV).

The FTIR spectra of GelMA mats showed typical peaks of gelatin (Figure 2a) as described for GelMA. On the other hand, TGA thermograms in Figure 2b show the thermal degradation behavior of GelMA mats. All samples showed a small loss of mass in the range of 50 ◦C to 120 ◦C, probably caused by the loss of water molecules. At temperatures above 150 ◦C, the degradation of biopolymer started for all samples. In addition, their onset points of degradation increased slightly with the time exposed to UV irradiation. Probably, it is due to an increase in the crosslinking degree.

**Figure 2.** Characterization of samples by (**a**) ATR-FTIR, (**b**) TGA, (**c**) DSC, (**d**) XRD.

Differential scanning calorimetry (DSC) curves of electrospun GelMA mats are shown in Figure 2c. All samples displayed two endothermic peaks at around 50 and 70 ◦C. The first peak could be attributed to glass transition of amino acid blocks in the peptide chain relating to the amorphous regions of gelatin. Uncrosslinked gelatin showed a higher Tg than the observed in crosslinked gelatin due to peptide chains, could interact physically by intermolecular bonds. In addition, Tg of GelMA increased with the crosslinking percentage because of the reduced polymeric chain mobility. As suggested in the literature, the endothermic peak at around 70 ◦C is attributed to denaturation protein [26]. The high Tm of uncrosslinked gelatin could be associated to physical crosslinking of peptide chains. All samples showed endotherm falling at around 97 ◦C; this is attributed to water loss and degradation of protein.

In order to correlate thermal behavior with structural organization, XRD patterns of samples were obtained (Figure 2d). In spite, gelatin has a crystalline structure originated from α-helix and triple helical; the amorphous structures were observed for all the samples. Probably, the electrospinning processing hinders the re-crystallization of gelatin and the XRD pattern showed an amorphous halo. This peak could be associated with the short range order of protein chains. In addition, the intensity of amorphous peak decreased with the increase in crosslinking degree. Probably, the crosslinking hindered the peptide chains re-ordering.

The hydrophilicity of the electrospun meshes was evaluated by measuring their water contact angles (Figure 3a, Table 1). The contact angle of uncrosslinked gelatin mat could not be measured because it was absorbed instantaneously when the PBS solution was in contact. However, crosslinked gelatin nanofibers are stable at these conditions. As expected, the CA values of crosslinked gelatin confirm the hydrophilic character of matrices. UV irradiation time could affect not only crosslinking extension, but could also modify the structure, as can be seen in SEM images. Thus, CA values are determined by both chemical and morphological properties.

**Figure 3.** Results of (**a**) CA, (**b**) WR, (**c**) Tensile properties, (**d**) cell viability of GelMA matrices.

Although water uptake measurements were carried out for all samples, only crosslinked gelatin electrospun mats were stable in the PBS solution (Figure 3b, Table 1). Swelling percentages of the matrices were around 300%, demonstrating the high hydrophilic character and wettability of matrices. There were no significant differences between water uptake values of NG-6UV and NG-9UV samples. However, NG-12UV showed the lowest swelling degree. This fact can be probably ascribed to the major crosslinking extension of this sample, which clearly affects the water uptake capability.

In order to confirm the crosslinking, the mechanical properties of NG-9UV and NG-0UV nanofiber mats were tested (Figure 3c). A uniaxial tensile testing was performed and the Young's modulus (YM) was determined for both matrices. YM value of NG-9UV sample was seven times higher than the measured for NG-0UV (0.717 ± 0.001 MPa and 4.89 ± 0.03 MPa, respectively). Thus, the crosslinking methodology used in this work for GelMA curing was successful. Moreover, a relationship between the morphological/chemical properties and UV irradiation time was observed. Thus, a UV irradiation time of 9 min was chosen for further studies, according to the obtained results.
