3.1. Optimization of Gel Formulation
CS is an abundant natural source and is already used in health products for many applications as it is safe and classified as G.R.A.S. (generally recognized as safe) by the FDA. Furthermore, it is also cheap, biodegradable, and biocompatible. For these reasons, this raw material was considered interesting as a potential polymer for the development of sustainable patches by 3D printing PAM technique.
In the Farmacopea Ufficiale Italiana F.U. XII Ed., there are monographs of many starch-based formulations for dermatological use. The starting point of this study was the recipe of starch glycerolate consisting of corn starch (CS) 10% w/w, glycerol (Gly) 70% w/w, and bidistilled water (W) 20% w/w.
However, this gel is not suitable for PAM extrusion as it does not allow a regular flow through the syringe nozzle. For this reason, some modifications of the composition were carried out, and six gels were prepared, as reported in
Table 2. CS % was fixed at 10%
w/
w while Gly content was reduced.
Firstly, the extrusion ability through the syringe nozzle was evaluated for the prepared gels (
Table 2). G1 and G2 were very viscous (responsible for nozzle clogging), while G6 was too liquid (leakage from the syringe); for these reasons, these gels were considered not suitable for printing and thus excluded.
The remaining formulations, namely G3, G4, and G5 gels, were more suitable for the purpose and were further characterized. A preliminary printing attitude evaluation showed that these three gels did not allow for the obtaining of a homogeneous object. During the extrusion, in fact, the gel fragmentation occurred with consequent imperfections in the final object. Thus, a further modification of the compositions of gels G3, G4, and G5 was made by introducing a filler from a natural source [
29] (
Table 3). In a previous work [
22], the suitability of a βglucan (βglu) water suspension as filler for polymeric patches was assessed. For this reason, it was decided to use it. Considering that the amount of filler generally useful to improve the mechanical properties of polymeric films is 1%
w/
w [
30], this percentage of βglu was introduced in the formulation composition. The obtained gels (
Table 3) were submitted for a preliminary evaluation of the extrusion ability.
The product printed using both G3-βglu gel and G5-βglu gel did not allow for obtaining uniform objects showing the desired shape (as designed from CAD). This is attributable to the consistency of the two gels. G3-βglu gel is probably too viscous due to the low water content (50%
w/
w), while G5-βglu gel is too fluid due to the high water content (70%
w/
w). In both cases, the consistency was not suitable, and for this reason, G3-βglu and G5-βglu gels were excluded from the study. In the case of G4-βglu gel, it was possible to obtain an object able to maintain the features set by CAD and, for this reason, considered useful for the achievement of fixed objectives. Then, the selected gel was deeply characterized, and in the first instance, the rheological properties were evaluated. The estimation of gel viscosity has a significant impact on the quality of the final printed object [
31]. Generally, for PAM printing, it is preferable to use semisolid formulations having shear-thinning behavior, as this property is responsible for the ability to be pushed through a nozzle [
3,
32].
The G4-βglu gel formation process at 80 °C was evaluated by rheological analysis by SRT, and the obtained rheogram (
Figure 3A) shows that until 420 s (~7 min), the viscosity values change in a wide range; then, at 420 s, the trend line of viscosity as a function of time begins to be constant, indicative of gel formation. The gel cooling process was studied by DTSt, and as shown in
Figure 3B, the viscosity increases as temperature decreases as a consequence of gel consistency improvement during the cooling.
As gel consistency varies as a function of the temperature, it was very important to evaluate its viscosity after preparation and after storage at a fixed time (24 h) and temperature (25 °C). As shown in
Figure 3C, the viscosity measured by RST analysis for the gel after preparation, called G4-βglu fresh gel, is approximately in the range between 10
−2 and 10 Pa·s in the shear rate range 10
1–10
−1 s
−1 and is not linear. This is ascribable to the necessity of a time of stabilization for the newly formed gel. In fact, the viscosity measured for the same gel stored at 25 °C for 24 h is almost linear (
Figure 3C), and the shear rate range 10
1–10
−1 s
−1 resulted in improvements (10
1 and 10
3 Pa·s), demonstrating the stabilization of the polymeric network. The decrease in viscosity as a consequence of shear rate increase is indicative of shear-thinning behavior, in which an increase in shear rates leads to polymeric chain alignment and an ordered structure. Increased shear rate imitates the extrusion of gel via the nozzle, where viscosity reduction results in uniform extrusion [
33].
3.3. Optimization of Post-Printing Process
An important aspect to consider when the PAM technique is used is the selection of the most suitable platform on which the process of extrusion takes place. In fact, this can be constituted by different materials, and, generally, the one that is not capable of establishing interactions with the extruded semisolid must be selected. In this study, three materials were selected: PTFE, PET, and silicone. Moreover, the printed object requires a post-printing drying process in order to remove the solvent (water) and to obtain the final planned object. The drying process is, therefore, crucial to obtain resistant patches that are elastic and intact. For an optimal drying process, both temperature and time must be carefully set; too high temperatures or too long times may lead to excessive solvent evaporation, resulting in patch deformation and shrinking. On the other hand, low temperatures and short heat exposure times lead to inadequate water removal, resulting in sticky patches easily susceptible to microbiological contamination. Using G4-βglu gel, the patches were printed on three different selected supports: PTFE, PET, and silicone. For each support, three samples were printed and submitted to different drying conditions in an oven, as reported in
Table 6. By this study, it was possible to select the most suitable drying conditions as well as the most suitable support for printing. From the observation of the final printed objects (patches), it emerged that the silicone is the most suitable support for performing the drying at 37 °C for 24 h as it allows an easy removal of the patch.
Through these studies, it was possible to select the most suitable gel composition, the support, and the most suitable drying conditions. However, the final object (patch) had some limitations, represented by the limited resistance during the removal from the printing support. Thus, the next studies had the aim of solving this problem.
The first strategy was to print multilayer patches. In particular, bilayer and three-layer patches were prepared using G4-βglu gel that worked using the previously selected parameters: 0.80 mm nozzle diameter, 100% infill, 40% flow, and 15 mm/s printing speed.
Comparing the dried bilayer and three-layer patches with the monolayer, it can be highlighted that in both cases, patch breaking occurred as a consequence of solvent removal by drying (
Table 7).
From these results, it became apparent that this strategy is not suitable, and for this reason, it was decided to perform the extrusion of G4-βglu gel on a polymeric support that, after the post-printing drying process, should represent the backing layer of the final patch.
With this purpose, the polymeric support was prepared by casting method using a gel prepared using a biopolymer sodium alginate (Alg) that had the following composition: Alg 1.5%
w/
w, Gly 10%
w/
w, and water (W) 88.5%
w/
w optimized in a previous work [
27]. Alg is a natural polysaccharide widely used in the health field for its excellent water solubility, biodegradability, biocompatibility, and no toxicity [
36]. Moreover, it was successfully employed for patch preparation by solvent casting method [
27].
The final formulation, Alg_G4-βglu, was a bilayer patch consisting of the Alg backing layer on which a monolayer of G4-βglu gel was printed. After drying, the final object (
Figure 6) is homogeneous, manageable, and resistant.