3.2.1. Dynamic Mechanical Analysis (DMA)

DMA analyses were carried out on the polymerized samples, with the aim to observe the effect of the functionalization temperature on the mechanical properties of the materials. The tests were performed on the samples Ep, Ep-R-120, and Ep-R-160. DMA investigation was carried out by evaluating the tan δ profile and the storage modulus as a function of the temperature. In Figure 10, the DMA curves of the cured functionalized epoxy matrices Ep-R-120 and Ep-R-160 were compared to the curve of the not functionalized epoxy matrix Ep.

**Figure 10.** DMA curves: (**a**) Tan δ vs. temperature; (**b**) Storage modulus vs. temperature.

All systems present a storage modulus value (see Figure 10b) higher than 1000 MPa at room temperature and in the wide temperature range of 30 ÷ 100 ◦C, thus confirming their reliability if employed, for instance, as structural aeronautical parts generally working in the normal operating temperature range. The trend of the not functionalized epoxy matrix (Ep) shows a progressive decrease of modulus up to 150 ◦C; after that, the principal drop occurs, due to its subsequent attainment of the glass transition temperature (i.e., Tg). The mechanical behavior of this sample, shown in Figure 10a, confirms the glass transition temperature (Tg) value above the 180 ◦C. The highest peak in the mechanical spectrum, related to the glass transition, (i.e., α transition), is centered at 200 ◦C. The introduction of the rubber and the different functionalization temperature value affect both storage modulus and tan δ. The principal drop in the storage modulus is reduced to lower temperature values due to the presence of the functionalized precursor in the resin (see Figure 10b).

One of the big challenges in this work was to improve the resin's dynamic properties and reduce the matrix's rigidity, acting on its phase composition. Therefore, the design of a material containing in a rigid matrix very small domains of rubber phase at higher mobility, finely interpenetrated in the resin, has been considered. A higher mobility is expected of the chains around the elastomeric domains finely distributed in the resin, as highlighted in Section 3.3. Furthermore, DMA analysis evidences a reduction of the rigidity of the matrix (at the macroscopic level). In fact, for the sample which manifests higher healing efficiency (EP-R-160), the peak of Tan δ vs. temperature opens around 60 ◦C. It is worth noting that this peak is shifted to a lower temperature range. It involves a temperature range from 60 ◦C to 180 ◦C (with a max around 132 ◦C), where the sample without the rubber phase shows a transition from 140 ◦C to 250 ◦C. This is clear evidence of a strong toughening effect exerted by the rubber phase finely distributed in the form of small domains in the resin in a range of temperatures closest to ambient temperature.

The more effective the functionalization reaction of the epoxy precursor, the higher the toughening effect of the elastomer on the polymerized resin. This phenomenon results in a reduction of 40 ◦C for the Ep-R-120 system and one of 70 ◦C for the Ep-R-160 system leading the Tg value to 160 ◦C and 132 ◦C, respectively (see Figure 10a). These results agree with those obtained through FT-IR and TGA analyses, confirming the efficacy of the precursor functionalization performed at the temperature of 160 ◦C. To obtain an epoxy resin that shows good auto-repair ability, the composition Ep-R-160 was chosen as the matrix to host self-healing fillers, as the activation of auto-repair mechanism is favored in the presence of higher mobility of the polymeric chains [47,48].

#### 3.2.2. DSC Analyses

DSC investigations were performed to evaluate the influence of the self-healing fillers on the curing process of the functionalized epoxy resin. Figure 11a shows the results of DSC analyses performed on the uncured liquid mixtures.

The presence of the auto-repair agents determines a reduction of the peak temperature (see Table 3) value and a broadening of the peak shape. This phenomenon is more evident in Figure 11b in which the fractional conversion (*α*) as a function of temperature is shown. The fractional conversion (*α*) can be expressed as Equation (4):

$$\alpha(T) = \frac{\Delta H\_T}{\Delta H\_{Tot}}\tag{4}$$

where Δ*HT* is the partial heat of reaction at a certain temperature and Δ*HTot* is the total heat of reaction. The systems filled with the self-healing agents present a fractional conversion curve shifted at lower temperature and a reduced starting curing temperature value, (*Tα*=0.01, temperature value corresponding to *α* = 1%), as shown in Table 3. DSC curves of the samples EP-R-160, EP-R-160-DBA, EP-R-160-2T, EP-R-160-M, before the curing process (after the functionalization reaction) and after the curing process (dashed curves) have been added in Figure S2 of Section S2 of the S.M.

**Figure 11.** (**a**) DSC curves of the analyzed samples before the curing process in the oven; (**b**) Variation of the conversion (α) vs. the temperature.

**Table 3.** DSC results.


Table 3 highlights that the presence of the fillers causes, on the one hand, a reduction of curing temperature, on the other, a decrease of the curing degree value, which remains higher than 90%, allowing to the materials to be suitable for structural applications. Usually a small amount of accelerators, such as tertiary amines or imidazoles, are used in the epoxy/anhydride systems to speed up the curing process [69]. In our case, probably, the fillers act as catalysts in the polymerization mechanism. These results could represent an advantage in reducing processing costs and energy savings.
