*2.3. Methods*

The preparation of the samples to carry out FT-IR spectra and the modality of their acquisition are described in Section S1 (S.M.). This last section also reported information on the thermal and dynamic mechanical analysis of the performed elaborations.

Self-healing efficiency (*η*) was evaluated using two different approaches. The first one consists of a fracture test performed with a tapered double cantilever beam (TDCB) geometry sample following a protocol established in literature [12], which allows to calculate the value of the self-healing efficiency by Equation (1),

$$
\eta = \frac{P\_{CH}}{P\_{CV}} \times 100\tag{1}
$$

where *PCH* and *PCV* are the critical fracture load of the healed and virgin sample, respectively. The specimens were tested by INSTRON mod. 5967 Dynamometer, using a load cell of 30 KN and a 250 μm/min displacement rate. The dimensions of the tested samples are reported in Figure S1 of Section S1 of the S.M.

In the second approach, DMA tests were used to evaluate auto-repair ability. The self-healing test was carried out with a continuous dynamic flexural deformation, through which the samples, having dimensions 3 mm × 10 mm × 35 mm and a V-shaped starter notch (1 mm × 2 mm), were analyzed by applying a sinusoidal deformation with a maximum amplitude of 0.1% at a frequency of 1 Hz. An impulsive load of about 25 N induced a pre-crack in the sample. The trend of the mechanical modulus with temperature was considered as representative of the evolution of the healing process in the sample. In this case, *η* was calculated according to Equation (2) [48,63].

$$
\eta = \frac{E\_H}{E\_V} \times 100\tag{2}
$$

where *EH* and *EV* are the storage modulus of the healed and virgin sample, respectively.
